[clang] 9466b49 - [Clang] Convert various tests to opaque pointers (NFC)

Nikita Popov via cfe-commits cfe-commits at lists.llvm.org
Mon Dec 12 08:12:00 PST 2022


Author: Nikita Popov
Date: 2022-12-12T17:11:46+01:00
New Revision: 9466b49171dc4b21f56a48594fc82b1e52f5358a

URL: https://github.com/llvm/llvm-project/commit/9466b49171dc4b21f56a48594fc82b1e52f5358a
DIFF: https://github.com/llvm/llvm-project/commit/9466b49171dc4b21f56a48594fc82b1e52f5358a.diff

LOG: [Clang] Convert various tests to opaque pointers (NFC)

These were all tests where no manual fixup was required.

Added: 
    

Modified: 
    clang/test/CodeGen/X86/avx512f-builtins.c
    clang/test/CodeGen/aarch64-bf16-ldst-intrinsics.c
    clang/test/CodeGen/aarch64-neon-ldst-one.c
    clang/test/CodeGen/arm-vaarg.c
    clang/test/CodeGen/complex-convert.c
    clang/test/CodeGen/instrument-objc-method.m
    clang/test/CodeGen/named_reg_global.c
    clang/test/CodeGen/sanitize-thread-no-checking-at-run-time.m
    clang/test/CodeGen/thinlto-opaque-typed-mix.ll
    clang/test/CodeGen/ubsan-pointer-overflow.m
    clang/test/CodeGenCUDA/Inputs/device-code.ll
    clang/test/CodeGenCUDA/cxx-call-kernel.cpp
    clang/test/CodeGenCXX/amdgcn-automatic-variable.cpp
    clang/test/CodeGenCXX/amdgcn-func-arg.cpp
    clang/test/CodeGenCoroutines/coro-alloc-exp-namespace.cpp
    clang/test/CodeGenCoroutines/coro-alloc.cpp
    clang/test/CodeGenCoroutines/coro-await-exp-namespace.cpp
    clang/test/CodeGenCoroutines/coro-await-resume-eh-exp-namespace.cpp
    clang/test/CodeGenCoroutines/coro-await-resume-eh.cpp
    clang/test/CodeGenCoroutines/coro-await.cpp
    clang/test/CodeGenCoroutines/coro-cleanup-exp-namespace.cpp
    clang/test/CodeGenCoroutines/coro-cleanup.cpp
    clang/test/CodeGenCoroutines/coro-dest-slot-exp-namespace.cpp
    clang/test/CodeGenCoroutines/coro-dest-slot.cpp
    clang/test/CodeGenCoroutines/coro-eh-cleanup-exp-namespace.cpp
    clang/test/CodeGenCoroutines/coro-eh-cleanup.cpp
    clang/test/CodeGenCoroutines/coro-gro-exp-namespace.cpp
    clang/test/CodeGenCoroutines/coro-gro.cpp
    clang/test/CodeGenCoroutines/coro-gro2-exp-namespace.cpp
    clang/test/CodeGenCoroutines/coro-gro2.cpp
    clang/test/CodeGenCoroutines/coro-halo.cpp
    clang/test/CodeGenCoroutines/coro-promise-dtor-exp-namespace.cpp
    clang/test/CodeGenCoroutines/coro-promise-dtor.cpp
    clang/test/CodeGenCoroutines/coro-ret-void-exp-namespace.cpp
    clang/test/CodeGenCoroutines/coro-ret-void.cpp
    clang/test/CodeGenCoroutines/coro-return-exp-namespace.cpp
    clang/test/CodeGenCoroutines/coro-return-voidtype-initlist-exp-namespace.cpp
    clang/test/CodeGenCoroutines/coro-return-voidtype-initlist.cpp
    clang/test/CodeGenCoroutines/coro-return.cpp
    clang/test/CodeGenCoroutines/coro-symmetric-transfer-01-exp-namespace.cpp
    clang/test/CodeGenCoroutines/coro-symmetric-transfer-01.cpp
    clang/test/CodeGenHIP/hipspv-addr-spaces.cpp
    clang/test/CodeGenHIP/hipspv-kernel.cpp
    clang/test/CodeGenHIP/printf.cpp
    clang/test/CodeGenObjC/2010-02-01-utf16-with-null.m
    clang/test/CodeGenObjC/2010-02-09-DbgSelf.m
    clang/test/CodeGenObjC/NSFastEnumeration.m
    clang/test/CodeGenObjC/arc-arm.m
    clang/test/CodeGenObjC/arc-block-copy-escape.m
    clang/test/CodeGenObjC/arc-blocks-exceptions.m
    clang/test/CodeGenObjC/arc-bridged-cast.m
    clang/test/CodeGenObjC/arc-i386.m
    clang/test/CodeGenObjC/arc-linetable.m
    clang/test/CodeGenObjC/arc-no-arc-exceptions.m
    clang/test/CodeGenObjC/arc-rv-attr.m
    clang/test/CodeGenObjC/arc-ternary-op.m
    clang/test/CodeGenObjC/arc-unopt.m
    clang/test/CodeGenObjC/arc-unoptimized-byref-var.m
    clang/test/CodeGenObjC/arc-weak-property.m
    clang/test/CodeGenObjC/arc-weak.m
    clang/test/CodeGenObjC/arc-with-atthrow.m
    clang/test/CodeGenObjC/arc.ll
    clang/test/CodeGenObjC/arm-atomic-scalar-setter-getter.m
    clang/test/CodeGenObjC/arm64-int32-ivar.m
    clang/test/CodeGenObjC/atomic-aggregate-property.m
    clang/test/CodeGenObjC/attr-exception.m
    clang/test/CodeGenObjC/attr-noreturn.m
    clang/test/CodeGenObjC/attr-objc-arc-inert.m
    clang/test/CodeGenObjC/attr-objc-runtime-visible.m
    clang/test/CodeGenObjC/autorelease.m
    clang/test/CodeGenObjC/availability-cf-link-guard.m
    clang/test/CodeGenObjC/bitfield-access.m
    clang/test/CodeGenObjC/block-byref-variable-layout.m
    clang/test/CodeGenObjC/blocks-1.m
    clang/test/CodeGenObjC/builtin-memfns.m
    clang/test/CodeGenObjC/builtins.m
    clang/test/CodeGenObjC/category-super-class-meth.m
    clang/test/CodeGenObjC/compound-literal-property-access.m
    clang/test/CodeGenObjC/constant-non-fragile-ivar-offset.m
    clang/test/CodeGenObjC/constant-strings.m
    clang/test/CodeGenObjC/debug-info-default-synth-ivar.m
    clang/test/CodeGenObjC/debug-info-getter-name.m
    clang/test/CodeGenObjC/debug-info-nested-blocks.m
    clang/test/CodeGenObjC/debug-info-objc-property-dwarf5.m
    clang/test/CodeGenObjC/direct-method-ret-mismatch.m
    clang/test/CodeGenObjC/disable-direct-method.m
    clang/test/CodeGenObjC/encode-test-2.m
    clang/test/CodeGenObjC/encode-test-5.m
    clang/test/CodeGenObjC/encode-test-6.m
    clang/test/CodeGenObjC/encode-test.m
    clang/test/CodeGenObjC/extern-void-class-decl.m
    clang/test/CodeGenObjC/forward-protocol-metadata-symbols.m
    clang/test/CodeGenObjC/gc-weak-attribute.m
    clang/test/CodeGenObjC/gc.m
    clang/test/CodeGenObjC/getter-property-mismatch.m
    clang/test/CodeGenObjC/getter-property-type-mismatch.m
    clang/test/CodeGenObjC/gnu-deterministic-selectors.m
    clang/test/CodeGenObjC/gnu-empty-protocol-v3.m
    clang/test/CodeGenObjC/gnu-exceptions.m
    clang/test/CodeGenObjC/gnu-init.m
    clang/test/CodeGenObjC/gnu-method-only-once.m
    clang/test/CodeGenObjC/gnu-nil-receiver.m
    clang/test/CodeGenObjC/gnustep2-category.m
    clang/test/CodeGenObjC/gnustep2-class.m
    clang/test/CodeGenObjC/gnustep2-ivar-offset.m
    clang/test/CodeGenObjC/gnustep2-proto.m
    clang/test/CodeGenObjC/id-isa-codegen.m
    clang/test/CodeGenObjC/implicit-objc_msgSend.m
    clang/test/CodeGenObjC/instance-method-metadata.m
    clang/test/CodeGenObjC/ivar-base-as-invariant-load.m
    clang/test/CodeGenObjC/ivar-invariant.m
    clang/test/CodeGenObjC/local-static-block.m
    clang/test/CodeGenObjC/matrix-type-builtins.m
    clang/test/CodeGenObjC/matrix-type-operators.m
    clang/test/CodeGenObjC/misc-atomic-property.m
    clang/test/CodeGenObjC/mrc-weak.m
    clang/test/CodeGenObjC/newproperty-nested-synthesis-1.m
    clang/test/CodeGenObjC/next-objc-dispatch.m
    clang/test/CodeGenObjC/non-runtime-protocol.m
    clang/test/CodeGenObjC/nonlazy-msgSend.m
    clang/test/CodeGenObjC/nontrivial-c-struct-within-struct-name.m
    clang/test/CodeGenObjC/nontrivial-struct-param-init.m
    clang/test/CodeGenObjC/null-objc-empty-vtable.m
    clang/test/CodeGenObjC/objc-alloc-init.m
    clang/test/CodeGenObjC/objc-asm-attribute-test.m
    clang/test/CodeGenObjC/objc-dispatch-null-check.m
    clang/test/CodeGenObjC/objc-gc-aggr-assign.m
    clang/test/CodeGenObjC/objc-literal-debugger-test.m
    clang/test/CodeGenObjC/objc-non-trivial-struct-nrvo.m
    clang/test/CodeGenObjC/objc-read-weak-byref.m
    clang/test/CodeGenObjC/objc-runtime-name.m
    clang/test/CodeGenObjC/objc2-protocol-metadata.m
    clang/test/CodeGenObjC/objc2-strong-cast-block-import.m
    clang/test/CodeGenObjC/objc2-weak-block-call.m
    clang/test/CodeGenObjC/objc_copyStruct.m
    clang/test/CodeGenObjC/objfw.m
    clang/test/CodeGenObjC/optimize-ivar-offset-load.m
    clang/test/CodeGenObjC/pass-by-value-noalias.m
    clang/test/CodeGenObjC/personality.m
    clang/test/CodeGenObjC/ppc32-varargs-id.m
    clang/test/CodeGenObjC/property-array-type.m
    clang/test/CodeGenObjC/property-atomic-bool.m
    clang/test/CodeGenObjC/protocol-property-synth.m
    clang/test/CodeGenObjC/related-result-type.m
    clang/test/CodeGenObjC/return-objc-object.mm
    clang/test/CodeGenObjC/runtime-abi-match.m
    clang/test/CodeGenObjC/selector-ref-invariance.m
    clang/test/CodeGenObjC/stret-1.m
    clang/test/CodeGenObjC/super-dotsyntax-struct-property.m
    clang/test/CodeGenObjC/super-message-fragileabi.m
    clang/test/CodeGenObjC/synchronized.m
    clang/test/CodeGenObjC/synthesize_ivar.m
    clang/test/CodeGenObjC/tentative-cfconstantstring.m
    clang/test/CodeGenObjC/terminate.m
    clang/test/CodeGenObjC/ubsan-bool.m
    clang/test/CodeGenObjC/ubsan-nonnull-and-nullability.m
    clang/test/CodeGenObjC/ubsan-nullability-return-unreachable.m
    clang/test/CodeGenObjC/variadic-sends.m
    clang/test/CodeGenObjC/weak-in-c-struct.m
    clang/test/CodeGenObjC/x86_64-struct-return-gc.m
    clang/test/CodeGenObjCXX/arc-cxx11-member-init.mm
    clang/test/CodeGenObjCXX/arc-forwarded-lambda-call.mm
    clang/test/CodeGenObjCXX/arc-indirect.mm
    clang/test/CodeGenObjCXX/arc-mangle.mm
    clang/test/CodeGenObjCXX/arc-marker-funclet.mm
    clang/test/CodeGenObjCXX/arc-pseudo-destructors.mm
    clang/test/CodeGenObjCXX/arc-rv-attr.mm
    clang/test/CodeGenObjCXX/arc-special-member-functions.mm
    clang/test/CodeGenObjCXX/arc-weak.mm
    clang/test/CodeGenObjCXX/auto-release-result-assert.mm
    clang/test/CodeGenObjCXX/block-default-arg.mm
    clang/test/CodeGenObjCXX/boxing.mm
    clang/test/CodeGenObjCXX/catch-id-type.mm
    clang/test/CodeGenObjCXX/copy.mm
    clang/test/CodeGenObjCXX/debug-info-line.mm
    clang/test/CodeGenObjCXX/destroy.mm
    clang/test/CodeGenObjCXX/exception-cxx.mm
    clang/test/CodeGenObjCXX/exceptions.mm
    clang/test/CodeGenObjCXX/implicit-copy-assign-operator.mm
    clang/test/CodeGenObjCXX/implicit-copy-constructor.mm
    clang/test/CodeGenObjCXX/inheriting-constructor-cleanup.mm
    clang/test/CodeGenObjCXX/lambda-expressions.mm
    clang/test/CodeGenObjCXX/mangle-blocks.mm
    clang/test/CodeGenObjCXX/message.mm
    clang/test/CodeGenObjCXX/microsoft-abi-arc-param-order.mm
    clang/test/CodeGenObjCXX/mrc-weak.mm
    clang/test/CodeGenObjCXX/msabi-stret.mm
    clang/test/CodeGenObjCXX/objfw-exceptions.mm
    clang/test/CodeGenObjCXX/personality-abuse.mm
    clang/test/CodeGenObjCXX/personality.mm
    clang/test/CodeGenObjCXX/property-lvalue-capture.mm
    clang/test/CodeGenObjCXX/property-object-reference-2.mm
    clang/test/CodeGenObjCXX/synthesized-property-cleanup.mm
    clang/test/CodeGenObjCXX/ubsan-nullability-return-notypeloc.mm
    clang/test/CodeGenObjCXX/unknown-anytype.mm
    clang/test/CodeGenOpenCL/backend-unsupported-warning.ll
    clang/test/CodeGenOpenCLCXX/address-space-castoperators.cpp
    clang/test/CodeGenSYCL/address-space-conversions.cpp
    clang/test/CodeGenSYCL/address-space-deduction.cpp
    clang/test/CodeGenSYCL/address-space-mangling.cpp
    clang/test/CodeGenSYCL/debug-info-kernel-variables.cpp
    clang/test/CodeGenSYCL/functionptr-addrspace.cpp
    clang/test/Coverage/codegen-gnu.m
    clang/test/Coverage/codegen-next.m
    clang/test/CoverageMapping/ir.c
    clang/test/Frontend/fixed_point_add.c
    clang/test/Frontend/fixed_point_comparisons.c
    clang/test/Frontend/fixed_point_compound.c
    clang/test/Frontend/fixed_point_conversions.c
    clang/test/Frontend/fixed_point_conversions_half.c
    clang/test/Frontend/fixed_point_div.c
    clang/test/Frontend/fixed_point_mul.c
    clang/test/Frontend/fixed_point_same_fbits.c
    clang/test/Frontend/fixed_point_shift.c
    clang/test/Frontend/fixed_point_sub.c
    clang/test/Frontend/fixed_point_to_bool.c
    clang/test/Frontend/fixed_point_unary.c
    clang/test/Headers/stdarg.cpp
    clang/test/Headers/wasm.c
    clang/test/Modules/direct-module-import.m
    clang/test/Modules/templates.mm
    clang/test/PCH/non-trivial-c-compound-literal.m
    clang/test/PCH/objcxx-ivar-class.mm
    clang/test/Sema/ms_class_layout.cpp
    clang/test/SemaCXX/constexpr-vectors.cpp
    clang/test/SemaCXX/linkage.cpp
    clang/test/SemaObjC/debugger-support.m

Removed: 
    


################################################################################
diff  --git a/clang/test/CodeGen/X86/avx512f-builtins.c b/clang/test/CodeGen/X86/avx512f-builtins.c
index 12a406455d5ad..dfafcf812fe24 100644
--- a/clang/test/CodeGen/X86/avx512f-builtins.c
+++ b/clang/test/CodeGen/X86/avx512f-builtins.c
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -flax-vector-conversions=none -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +avx512f -emit-llvm -o - -Wall -Werror -Wsign-conversion | FileCheck %s
-// RUN: %clang_cc1 -no-opaque-pointers -flax-vector-conversions=none -fms-extensions -fms-compatibility -ffreestanding %s -triple=x86_64-windows-msvc -target-feature +avx512f -emit-llvm -o - -Wall -Werror -Wsign-conversion | FileCheck %s
+// RUN: %clang_cc1 -flax-vector-conversions=none -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +avx512f -emit-llvm -o - -Wall -Werror -Wsign-conversion | FileCheck %s
+// RUN: %clang_cc1 -flax-vector-conversions=none -fms-extensions -fms-compatibility -ffreestanding %s -triple=x86_64-windows-msvc -target-feature +avx512f -emit-llvm -o - -Wall -Werror -Wsign-conversion | FileCheck %s
 
 #include <immintrin.h>
 
@@ -176,7 +176,7 @@ __m512d test_mm512_mul_pd(__m512d a, __m512d b)
 void test_mm512_storeu_si512 (void *__P, __m512i __A)
 {
   // CHECK-LABEL: @test_mm512_storeu_si512
-  // CHECK: store <8 x i64> %{{.*}}, <8 x i64>* %{{.*}}, align 1{{$}}
+  // CHECK: store <8 x i64> %{{.*}}, ptr %{{.*}}, align 1{{$}}
   // CHECK-NEXT: ret void
   _mm512_storeu_si512 ( __P,__A);
 }
@@ -184,7 +184,7 @@ void test_mm512_storeu_si512 (void *__P, __m512i __A)
 void test_mm512_storeu_ps(void *p, __m512 a)
 {
   // CHECK-LABEL: @test_mm512_storeu_ps
-  // CHECK: store <16 x float> %{{.*}}, <16 x float>* %{{.*}}, align 1{{$}}
+  // CHECK: store <16 x float> %{{.*}}, ptr %{{.*}}, align 1{{$}}
   // CHECK-NEXT: ret void
   _mm512_storeu_ps(p, a);
 }
@@ -192,7 +192,7 @@ void test_mm512_storeu_ps(void *p, __m512 a)
 void test_mm512_storeu_pd(void *p, __m512d a)
 {
   // CHECK-LABEL: @test_mm512_storeu_pd
-  // CHECK: store <8 x double> %{{.*}}, <8 x double>* %{{.*}}, align 1{{$}}
+  // CHECK: store <8 x double> %{{.*}}, ptr %{{.*}}, align 1{{$}}
   // CHECK-NEXT: ret void
   _mm512_storeu_pd(p, a);
 }
@@ -200,16 +200,15 @@ void test_mm512_storeu_pd(void *p, __m512d a)
 void test_mm512_mask_store_ps(void *p, __m512 a, __mmask16 m)
 {
   // CHECK-LABEL: @test_mm512_mask_store_ps
-  // CHECK: @llvm.masked.store.v16f32.p0v16f32(<16 x float> %{{.*}}, <16 x float>* %{{.*}}, i32 64, <16 x i1> %{{.*}})
+  // CHECK: @llvm.masked.store.v16f32.p0(<16 x float> %{{.*}}, ptr %{{.*}}, i32 64, <16 x i1> %{{.*}})
   _mm512_mask_store_ps(p, m, a);
 }
 
 void test_mm512_store_si512 (void *__P, __m512i __A)
 {
   // CHECK-LABEL: @test_mm512_store_si512 
-  // CHECK: load <8 x i64>, <8 x i64>* %__A.addr.i, align 64{{$}}
-  // CHECK: [[SI512_3:%.+]] = load i8*, i8** %__P.addr.i, align 8{{$}}
-  // CHECK: bitcast i8* [[SI512_3]] to <8 x i64>*
+  // CHECK: load <8 x i64>, ptr %__A.addr.i, align 64{{$}}
+  // CHECK: [[SI512_3:%.+]] = load ptr, ptr %__P.addr.i, align 8{{$}}
   // CHECK: store <8 x i64>  
   _mm512_store_si512 ( __P,__A);
 }
@@ -217,9 +216,8 @@ void test_mm512_store_si512 (void *__P, __m512i __A)
 void test_mm512_store_epi32 (void *__P, __m512i __A)
 {
   // CHECK-LABEL: @test_mm512_store_epi32 
-  // CHECK: load <8 x i64>, <8 x i64>* %__A.addr.i, align 64{{$}}
-  // CHECK: [[Si32_3:%.+]] = load i8*, i8** %__P.addr.i, align 8{{$}}
-  // CHECK: bitcast i8* [[Si32_3]] to <8 x i64>*
+  // CHECK: load <8 x i64>, ptr %__A.addr.i, align 64{{$}}
+  // CHECK: [[Si32_3:%.+]] = load ptr, ptr %__P.addr.i, align 8{{$}}
   // CHECK: store <8 x i64>  
   _mm512_store_epi32 ( __P,__A);
 }
@@ -227,9 +225,8 @@ void test_mm512_store_epi32 (void *__P, __m512i __A)
 void test_mm512_store_epi64 (void *__P, __m512i __A)
 {
   // CHECK-LABEL: @test_mm512_store_epi64 
-  // CHECK: load <8 x i64>, <8 x i64>* %__A.addr.i, align 64{{$}}
-  // CHECK: [[SI64_3:%.+]] = load i8*, i8** %__P.addr.i, align 8{{$}}
-  // CHECK: bitcast i8* [[SI64_3]] to <8 x i64>*
+  // CHECK: load <8 x i64>, ptr %__A.addr.i, align 64{{$}}
+  // CHECK: [[SI64_3:%.+]] = load ptr, ptr %__P.addr.i, align 8{{$}}
   // CHECK: store <8 x i64>  
   _mm512_store_epi64 ( __P,__A);
 }
@@ -251,177 +248,174 @@ void test_mm512_store_pd(void *p, __m512d a)
 void test_mm512_mask_store_pd(void *p, __m512d a, __mmask8 m)
 {
   // CHECK-LABEL: @test_mm512_mask_store_pd
-  // CHECK: @llvm.masked.store.v8f64.p0v8f64(<8 x double> %{{.*}}, <8 x double>* %{{.*}}, i32 64, <8 x i1> %{{.*}})
+  // CHECK: @llvm.masked.store.v8f64.p0(<8 x double> %{{.*}}, ptr %{{.*}}, i32 64, <8 x i1> %{{.*}})
   _mm512_mask_store_pd(p, m, a);
 }
 
 void test_mm512_storeu_epi32(void *__P, __m512i __A) {
   // CHECK-LABEL: @test_mm512_storeu_epi32
-  // CHECK: store <8 x i64> %{{.*}}, <8 x i64>* %{{.*}}, align 1{{$}}
+  // CHECK: store <8 x i64> %{{.*}}, ptr %{{.*}}, align 1{{$}}
   return _mm512_storeu_epi32(__P, __A); 
 }
 
 void test_mm512_mask_storeu_epi32(void *__P, __mmask16 __U, __m512i __A) {
   // CHECK-LABEL: @test_mm512_mask_storeu_epi32
-  // CHECK: @llvm.masked.store.v16i32.p0v16i32(<16 x i32> %{{.*}}, <16 x i32>* %{{.*}}, i32 1, <16 x i1> %{{.*}})
+  // CHECK: @llvm.masked.store.v16i32.p0(<16 x i32> %{{.*}}, ptr %{{.*}}, i32 1, <16 x i1> %{{.*}})
   return _mm512_mask_storeu_epi32(__P, __U, __A); 
 }
 
 void test_mm512_storeu_epi64(void *__P, __m512i __A) {
   // CHECK-LABEL: @test_mm512_storeu_epi64
-  // CHECK: store <8 x i64> %{{.*}}, <8 x i64>* %{{.*}}, align 1{{$}}
+  // CHECK: store <8 x i64> %{{.*}}, ptr %{{.*}}, align 1{{$}}
   return _mm512_storeu_epi64(__P, __A); 
 }
 
 void test_mm512_mask_storeu_epi64(void *__P, __mmask8 __U, __m512i __A) {
   // CHECK-LABEL: @test_mm512_mask_storeu_epi64
-  // CHECK: @llvm.masked.store.v8i64.p0v8i64(<8 x i64> %{{.*}}, <8 x i64>* %{{.*}}, i32 1, <8 x i1> %{{.*}})
+  // CHECK: @llvm.masked.store.v8i64.p0(<8 x i64> %{{.*}}, ptr %{{.*}}, i32 1, <8 x i1> %{{.*}})
   return _mm512_mask_storeu_epi64(__P, __U, __A); 
 }
 
 __m512i test_mm512_loadu_si512 (void *__P)
 {
   // CHECK-LABEL: @test_mm512_loadu_si512 
-  // CHECK: load <8 x i64>, <8 x i64>* %{{.*}}, align 1{{$}}
+  // CHECK: load <8 x i64>, ptr %{{.*}}, align 1{{$}}
   return _mm512_loadu_si512 ( __P);
 }
 
 __m512i test_mm512_loadu_epi32 (void *__P)
 {
   // CHECK-LABEL: @test_mm512_loadu_epi32 
-  // CHECK: load <8 x i64>, <8 x i64>* %{{.*}}, align 1{{$}}
+  // CHECK: load <8 x i64>, ptr %{{.*}}, align 1{{$}}
   return _mm512_loadu_epi32 (__P);
 }
 
 __m512i test_mm512_mask_loadu_epi32 (__m512i __W, __mmask16 __U, void *__P)
 {
   // CHECK-LABEL: @test_mm512_mask_loadu_epi32 
-  // CHECK: @llvm.masked.load.v16i32.p0v16i32(<16 x i32>* %{{.*}}, i32 1, <16 x i1> %{{.*}}, <16 x i32> %{{.*}})
+  // CHECK: @llvm.masked.load.v16i32.p0(ptr %{{.*}}, i32 1, <16 x i1> %{{.*}}, <16 x i32> %{{.*}})
   return _mm512_mask_loadu_epi32 (__W,__U, __P);
 }
 
 __m512i test_mm512_maskz_loadu_epi32 (__mmask16 __U, void *__P)
 {
   // CHECK-LABEL: @test_mm512_maskz_loadu_epi32
-  // CHECK: @llvm.masked.load.v16i32.p0v16i32(<16 x i32>* %{{.*}}, i32 1, <16 x i1> %{{.*}}, <16 x i32> %{{.*}})
+  // CHECK: @llvm.masked.load.v16i32.p0(ptr %{{.*}}, i32 1, <16 x i1> %{{.*}}, <16 x i32> %{{.*}})
   return _mm512_maskz_loadu_epi32 (__U, __P);
 }
 
 __m512i test_mm512_loadu_epi64 (void *__P)
 {
   // CHECK-LABEL: @test_mm512_loadu_epi64 
-  // CHECK: load <8 x i64>, <8 x i64>* %{{.*}}, align 1{{$}}
+  // CHECK: load <8 x i64>, ptr %{{.*}}, align 1{{$}}
   return _mm512_loadu_epi64 (__P);
 }
 
 __m512i test_mm512_mask_loadu_epi64 (__m512i __W, __mmask8 __U, void *__P)
 {
   // CHECK-LABEL: @test_mm512_mask_loadu_epi64 
-  // CHECK: @llvm.masked.load.v8i64.p0v8i64(<8 x i64>* %{{.*}}, i32 1, <8 x i1> %{{.*}}, <8 x i64> %{{.*}})
+  // CHECK: @llvm.masked.load.v8i64.p0(ptr %{{.*}}, i32 1, <8 x i1> %{{.*}}, <8 x i64> %{{.*}})
   return _mm512_mask_loadu_epi64 (__W,__U, __P);
 }
 
 __m512i test_mm512_maskz_loadu_epi64 (__mmask16 __U, void *__P)
 {
   // CHECK-LABEL: @test_mm512_maskz_loadu_epi64
-  // CHECK: @llvm.masked.load.v8i64.p0v8i64(<8 x i64>* %{{.*}}, i32 1, <8 x i1> %{{.*}}, <8 x i64> %{{.*}})
+  // CHECK: @llvm.masked.load.v8i64.p0(ptr %{{.*}}, i32 1, <8 x i1> %{{.*}}, <8 x i64> %{{.*}})
   return _mm512_maskz_loadu_epi64 (__U, __P);
 }
 
 __m512 test_mm512_loadu_ps(void *p)
 {
   // CHECK-LABEL: @test_mm512_loadu_ps
-  // CHECK: load <16 x float>, <16 x float>* {{.*}}, align 1{{$}}
+  // CHECK: load <16 x float>, ptr {{.*}}, align 1{{$}}
   return _mm512_loadu_ps(p);
 }
 
 __m512 test_mm512_mask_loadu_ps (__m512 __W, __mmask16 __U, void *__P)
 {
   // CHECK-LABEL: @test_mm512_mask_loadu_ps 
-  // CHECK: @llvm.masked.load.v16f32.p0v16f32(<16 x float>* %{{.*}}, i32 1, <16 x i1> %{{.*}}, <16 x float> %{{.*}})
+  // CHECK: @llvm.masked.load.v16f32.p0(ptr %{{.*}}, i32 1, <16 x i1> %{{.*}}, <16 x float> %{{.*}})
   return _mm512_mask_loadu_ps (__W,__U, __P);
 }
 
 __m512d test_mm512_loadu_pd(void *p)
 {
   // CHECK-LABEL: @test_mm512_loadu_pd
-  // CHECK: load <8 x double>, <8 x double>* {{.*}}, align 1{{$}}
+  // CHECK: load <8 x double>, ptr {{.*}}, align 1{{$}}
   return _mm512_loadu_pd(p);
 }
 
 __m512d test_mm512_mask_loadu_pd (__m512d __W, __mmask8 __U, void *__P)
 {
   // CHECK-LABEL: @test_mm512_mask_loadu_pd 
-  // CHECK: @llvm.masked.load.v8f64.p0v8f64(<8 x double>* %{{.*}}, i32 1, <8 x i1> %{{.*}}, <8 x double> %{{.*}})
+  // CHECK: @llvm.masked.load.v8f64.p0(ptr %{{.*}}, i32 1, <8 x i1> %{{.*}}, <8 x double> %{{.*}})
   return _mm512_mask_loadu_pd (__W,__U, __P);
 }
 
 __m512i test_mm512_load_si512 (void *__P)
 {
   // CHECK-LABEL: @test_mm512_load_si512 
-  // CHECK: [[LI512_1:%.+]] = load i8*, i8** %__P.addr.i, align 8{{$}}
-  // CHECK: [[LI512_2:%.+]] = bitcast i8* [[LI512_1]] to <8 x i64>*
-  // CHECK: load <8 x i64>, <8 x i64>* [[LI512_2]], align 64{{$}}
+  // CHECK: [[LI512_1:%.+]] = load ptr, ptr %__P.addr.i, align 8{{$}}
+  // CHECK: load <8 x i64>, ptr [[LI512_1]], align 64{{$}}
   return _mm512_load_si512 ( __P);
 }
 
 __m512i test_mm512_load_epi32 (void *__P)
 {
   // CHECK-LABEL: @test_mm512_load_epi32 
-  // CHECK: [[LI32_1:%.+]] = load i8*, i8** %__P.addr.i, align 8{{$}}
-  // CHECK: [[LI32_2:%.+]] = bitcast i8* [[LI32_1]] to <8 x i64>*
-  // CHECK: load <8 x i64>, <8 x i64>* [[LI32_2]], align 64{{$}}
+  // CHECK: [[LI32_1:%.+]] = load ptr, ptr %__P.addr.i, align 8{{$}}
+  // CHECK: load <8 x i64>, ptr [[LI32_1]], align 64{{$}}
   return _mm512_load_epi32 ( __P);
 }
 
 __m512i test_mm512_load_epi64 (void *__P)
 {
   // CHECK-LABEL: @test_mm512_load_epi64 
-  // CHECK: [[LI64_1:%.+]] = load i8*, i8** %__P.addr.i, align 8{{$}}
-  // CHECK: [[LI64_2:%.+]] = bitcast i8* [[LI64_1]] to <8 x i64>*
-  // CHECK: load <8 x i64>, <8 x i64>* [[LI64_2]], align 64{{$}}
+  // CHECK: [[LI64_1:%.+]] = load ptr, ptr %__P.addr.i, align 8{{$}}
+  // CHECK: load <8 x i64>, ptr [[LI64_1]], align 64{{$}}
   return _mm512_load_epi64 ( __P);
 }
 
 __m512 test_mm512_load_ps(void *p)
 {
   // CHECK-LABEL: @test_mm512_load_ps
-  // CHECK: load <16 x float>, <16 x float>* %{{.*}}, align 64{{$}}
+  // CHECK: load <16 x float>, ptr %{{.*}}, align 64{{$}}
   return _mm512_load_ps(p);
 }
 
 __m512 test_mm512_mask_load_ps (__m512 __W, __mmask16 __U, void *__P)
 {
   // CHECK-LABEL: @test_mm512_mask_load_ps 
-  // CHECK: @llvm.masked.load.v16f32.p0v16f32(<16 x float>* %{{.*}}, i32 64, <16 x i1> %{{.*}}, <16 x float> %{{.*}})
+  // CHECK: @llvm.masked.load.v16f32.p0(ptr %{{.*}}, i32 64, <16 x i1> %{{.*}}, <16 x float> %{{.*}})
   return _mm512_mask_load_ps (__W,__U, __P);
 }
 
 __m512 test_mm512_maskz_load_ps(__mmask16 __U, void *__P)
 {
   // CHECK-LABEL: @test_mm512_maskz_load_ps
-  // CHECK: @llvm.masked.load.v16f32.p0v16f32(<16 x float>* %{{.*}}, i32 64, <16 x i1> %{{.*}}, <16 x float> %{{.*}})
+  // CHECK: @llvm.masked.load.v16f32.p0(ptr %{{.*}}, i32 64, <16 x i1> %{{.*}}, <16 x float> %{{.*}})
   return _mm512_maskz_load_ps(__U, __P);
 }
 
 __m512d test_mm512_load_pd(void *p)
 {
   // CHECK-LABEL: @test_mm512_load_pd
-  // CHECK: load <8 x double>, <8 x double>* %{{.*}}, align 64{{$}}
+  // CHECK: load <8 x double>, ptr %{{.*}}, align 64{{$}}
   return _mm512_load_pd(p);
 }
 
 __m512d test_mm512_mask_load_pd (__m512d __W, __mmask8 __U, void *__P)
 {
   // CHECK-LABEL: @test_mm512_mask_load_pd 
-  // CHECK: @llvm.masked.load.v8f64.p0v8f64(<8 x double>* %{{.*}}, i32 64, <8 x i1> %{{.*}}, <8 x double> %{{.*}})
+  // CHECK: @llvm.masked.load.v8f64.p0(ptr %{{.*}}, i32 64, <8 x i1> %{{.*}}, <8 x double> %{{.*}})
   return _mm512_mask_load_pd (__W,__U, __P);
 }
 
 __m512d test_mm512_maskz_load_pd(__mmask8 __U, void *__P)
 {
   // CHECK-LABEL: @test_mm512_maskz_load_pd
-  // CHECK: @llvm.masked.load.v8f64.p0v8f64(<8 x double>* %{{.*}}, i32 64, <8 x i1> %{{.*}}, <8 x double> %{{.*}})
+  // CHECK: @llvm.masked.load.v8f64.p0(ptr %{{.*}}, i32 64, <8 x i1> %{{.*}}, <8 x double> %{{.*}})
   return _mm512_maskz_load_pd(__U, __P);
 }
 
@@ -4326,13 +4320,13 @@ __m512i test_mm512_maskz_srli_epi64_2(__mmask8 __U, __m512i __A, unsigned int __
 
 __m512i test_mm512_mask_load_epi32(__m512i __W, __mmask16 __U, void const *__P) {
   // CHECK-LABEL: @test_mm512_mask_load_epi32
-  // CHECK: @llvm.masked.load.v16i32.p0v16i32(<16 x i32>* %{{.*}}, i32 64, <16 x i1> %{{.*}}, <16 x i32> %{{.*}})
+  // CHECK: @llvm.masked.load.v16i32.p0(ptr %{{.*}}, i32 64, <16 x i1> %{{.*}}, <16 x i32> %{{.*}})
   return _mm512_mask_load_epi32(__W, __U, __P); 
 }
 
 __m512i test_mm512_maskz_load_epi32(__mmask16 __U, void const *__P) {
   // CHECK-LABEL: @test_mm512_maskz_load_epi32
-  // CHECK: @llvm.masked.load.v16i32.p0v16i32(<16 x i32>* %{{.*}}, i32 64, <16 x i1> %{{.*}}, <16 x i32> %{{.*}})
+  // CHECK: @llvm.masked.load.v16i32.p0(ptr %{{.*}}, i32 64, <16 x i1> %{{.*}}, <16 x i32> %{{.*}})
   return _mm512_maskz_load_epi32(__U, __P); 
 }
 
@@ -4362,25 +4356,25 @@ __m512i test_mm512_maskz_mov_epi64(__mmask8 __U, __m512i __A) {
 
 __m512i test_mm512_mask_load_epi64(__m512i __W, __mmask8 __U, void const *__P) {
   // CHECK-LABEL: @test_mm512_mask_load_epi64
-  // CHECK: @llvm.masked.load.v8i64.p0v8i64(<8 x i64>* %{{.*}}, i32 64, <8 x i1> %{{.*}}, <8 x i64> %{{.*}})
+  // CHECK: @llvm.masked.load.v8i64.p0(ptr %{{.*}}, i32 64, <8 x i1> %{{.*}}, <8 x i64> %{{.*}})
   return _mm512_mask_load_epi64(__W, __U, __P); 
 }
 
 __m512i test_mm512_maskz_load_epi64(__mmask8 __U, void const *__P) {
   // CHECK-LABEL: @test_mm512_maskz_load_epi64
-  // CHECK: @llvm.masked.load.v8i64.p0v8i64(<8 x i64>* %{{.*}}, i32 64, <8 x i1> %{{.*}}, <8 x i64> %{{.*}})
+  // CHECK: @llvm.masked.load.v8i64.p0(ptr %{{.*}}, i32 64, <8 x i1> %{{.*}}, <8 x i64> %{{.*}})
   return _mm512_maskz_load_epi64(__U, __P); 
 }
 
 void test_mm512_mask_store_epi32(void *__P, __mmask16 __U, __m512i __A) {
   // CHECK-LABEL: @test_mm512_mask_store_epi32
-  // CHECK: @llvm.masked.store.v16i32.p0v16i32(<16 x i32> %{{.*}}, <16 x i32>* %{{.*}}, i32 64, <16 x i1> %{{.*}})
+  // CHECK: @llvm.masked.store.v16i32.p0(<16 x i32> %{{.*}}, ptr %{{.*}}, i32 64, <16 x i1> %{{.*}})
   return _mm512_mask_store_epi32(__P, __U, __A); 
 }
 
 void test_mm512_mask_store_epi64(void *__P, __mmask8 __U, __m512i __A) {
   // CHECK-LABEL: @test_mm512_mask_store_epi64
-  // CHECK: @llvm.masked.store.v8i64.p0v8i64(<8 x i64> %{{.*}}, <8 x i64>* %{{.*}}, i32 64, <8 x i1> %{{.*}})
+  // CHECK: @llvm.masked.store.v8i64.p0(<8 x i64> %{{.*}}, ptr %{{.*}}, i32 64, <8 x i1> %{{.*}})
   return _mm512_mask_store_epi64(__P, __U, __A); 
 }
 
@@ -4610,7 +4604,7 @@ __m128 test_mm_getmant_ss(__m128 __A, __m128 __B) {
 
 __mmask16 test_mm512_kmov(__mmask16 __A) {
   // CHECK-LABEL: @test_mm512_kmov
-  // CHECK: load i16, i16* %__A.addr.i, align 2{{$}}
+  // CHECK: load i16, ptr %__A.addr.i, align 2{{$}}
   return _mm512_kmov(__A); 
 }
 
@@ -8525,7 +8519,7 @@ __mmask16 test_cvtu32_mask16(__m512i A, __m512i B, unsigned int C) {
 
 __mmask16 test_load_mask16(__mmask16 *A, __m512i B, __m512i C) {
   // CHECK-LABEL: @test_load_mask16
-  // CHECK: [[LOAD:%.*]] = load i16, i16* %{{.*}}{{$}}
+  // CHECK: [[LOAD:%.*]] = load i16, ptr %{{.*}}{{$}}
   // CHECK: bitcast i16 [[LOAD]] to <16 x i1>
   return _mm512_mask_cmpneq_epu32_mask(_load_mask16(A), B, C);
 }
@@ -8533,55 +8527,55 @@ __mmask16 test_load_mask16(__mmask16 *A, __m512i B, __m512i C) {
 void test_store_mask16(__mmask16 *A, __m512i B, __m512i C) {
   // CHECK-LABEL: @test_store_mask16
   // CHECK: bitcast <16 x i1> %{{.*}} to i16
-  // CHECK: store i16 %{{.*}}, i16* %{{.*}}
+  // CHECK: store i16 %{{.*}}, ptr %{{.*}}
   _store_mask16(A, _mm512_cmpneq_epu32_mask(B, C));
 }
 
 void test_mm512_stream_si512(__m512i * __P, __m512i __A) {
   // CHECK-LABEL: @test_mm512_stream_si512
-  // CHECK: store <8 x i64> %{{.*}}, <8 x i64>* %{{.*}}, align 64, !nontemporal [[NONTEMPORAL:![0-9]+]]
+  // CHECK: store <8 x i64> %{{.*}}, ptr %{{.*}}, align 64, !nontemporal [[NONTEMPORAL:![0-9]+]]
   _mm512_stream_si512(__P, __A); 
 }
 
 void test_mm512_stream_si512_2(void * __P, __m512i __A) {
   // CHECK-LABEL: @test_mm512_stream_si512
-  // CHECK: store <8 x i64> %{{.*}}, <8 x i64>* %{{.*}}, align 64, !nontemporal [[NONTEMPORAL]]
+  // CHECK: store <8 x i64> %{{.*}}, ptr %{{.*}}, align 64, !nontemporal [[NONTEMPORAL]]
   _mm512_stream_si512(__P, __A); 
 }
 
 __m512i test_mm512_stream_load_si512(void *__P) {
   // CHECK-LABEL: @test_mm512_stream_load_si512
-  // CHECK: load <8 x i64>, <8 x i64>* %{{.*}}, align 64, !nontemporal [[NONTEMPORAL]]{{$}}
+  // CHECK: load <8 x i64>, ptr %{{.*}}, align 64, !nontemporal [[NONTEMPORAL]]{{$}}
   return _mm512_stream_load_si512(__P); 
 }
 
 __m512i test_mm512_stream_load_si512_const(void const *__P) {
   // CHECK-LABEL: @test_mm512_stream_load_si512_const
-  // CHECK: load <8 x i64>, <8 x i64>* %{{.*}}, align 64, !nontemporal [[NONTEMPORAL]]{{$}}
+  // CHECK: load <8 x i64>, ptr %{{.*}}, align 64, !nontemporal [[NONTEMPORAL]]{{$}}
   return _mm512_stream_load_si512(__P); 
 }
 
 void test_mm512_stream_pd(double *__P, __m512d __A) {
   // CHECK-LABEL: @test_mm512_stream_pd
-  // CHECK: store <8 x double> %{{.*}}, <8 x double>* %{{.*}}, align 64, !nontemporal [[NONTEMPORAL]]
+  // CHECK: store <8 x double> %{{.*}}, ptr %{{.*}}, align 64, !nontemporal [[NONTEMPORAL]]
   return _mm512_stream_pd(__P, __A); 
 }
 
 void test_mm512_stream_pd_2(void *__P, __m512d __A) {
   // CHECK-LABEL: @test_mm512_stream_pd
-  // CHECK: store <8 x double> %{{.*}}, <8 x double>* %{{.*}}, align 64, !nontemporal [[NONTEMPORAL]]
+  // CHECK: store <8 x double> %{{.*}}, ptr %{{.*}}, align 64, !nontemporal [[NONTEMPORAL]]
   return _mm512_stream_pd(__P, __A); 
 }
 
 void test_mm512_stream_ps(float *__P, __m512 __A) {
   // CHECK-LABEL: @test_mm512_stream_ps
-  // CHECK: store <16 x float> %{{.*}}, <16 x float>* %{{.*}}, align 64, !nontemporal [[NONTEMPORAL]]
+  // CHECK: store <16 x float> %{{.*}}, ptr %{{.*}}, align 64, !nontemporal [[NONTEMPORAL]]
   _mm512_stream_ps(__P, __A); 
 }
 
 void test_mm512_stream_ps_2(void *__P, __m512 __A) {
   // CHECK-LABEL: @test_mm512_stream_ps
-  // CHECK: store <16 x float> %{{.*}}, <16 x float>* %{{.*}}, align 64, !nontemporal [[NONTEMPORAL]]
+  // CHECK: store <16 x float> %{{.*}}, ptr %{{.*}}, align 64, !nontemporal [[NONTEMPORAL]]
   _mm512_stream_ps(__P, __A); 
 }
 __m512d test_mm512_mask_compress_pd(__m512d __W, __mmask8 __U, __m512d __A) {
@@ -8765,49 +8759,49 @@ __m512i test_mm512_maskz_expand_epi64(__mmask8 __U, __m512i __A) {
 }
 __m512i test_mm512_mask_expandloadu_epi64(__m512i __W, __mmask8 __U, void const *__P) {
   // CHECK-LABEL: @test_mm512_mask_expandloadu_epi64
-  // CHECK: @llvm.masked.expandload.v8i64(i64* %{{.*}}, <8 x i1> %{{.*}}, <8 x i64> %{{.*}})
+  // CHECK: @llvm.masked.expandload.v8i64(ptr %{{.*}}, <8 x i1> %{{.*}}, <8 x i64> %{{.*}})
   return _mm512_mask_expandloadu_epi64(__W, __U, __P); 
 }
 
 __m512i test_mm512_maskz_expandloadu_epi64(__mmask8 __U, void const *__P) {
   // CHECK-LABEL: @test_mm512_maskz_expandloadu_epi64
-  // CHECK: @llvm.masked.expandload.v8i64(i64* %{{.*}}, <8 x i1> %{{.*}}, <8 x i64> %{{.*}})
+  // CHECK: @llvm.masked.expandload.v8i64(ptr %{{.*}}, <8 x i1> %{{.*}}, <8 x i64> %{{.*}})
   return _mm512_maskz_expandloadu_epi64(__U, __P); 
 }
 
 __m512d test_mm512_mask_expandloadu_pd(__m512d __W, __mmask8 __U, void const *__P) {
   // CHECK-LABEL: @test_mm512_mask_expandloadu_pd
-  // CHECK: @llvm.masked.expandload.v8f64(double* %{{.*}}, <8 x i1> %{{.*}}, <8 x double> %{{.*}})
+  // CHECK: @llvm.masked.expandload.v8f64(ptr %{{.*}}, <8 x i1> %{{.*}}, <8 x double> %{{.*}})
   return _mm512_mask_expandloadu_pd(__W, __U, __P); 
 }
 
 __m512d test_mm512_maskz_expandloadu_pd(__mmask8 __U, void const *__P) {
   // CHECK-LABEL: @test_mm512_maskz_expandloadu_pd
-  // CHECK: @llvm.masked.expandload.v8f64(double* %{{.*}}, <8 x i1> %{{.*}}, <8 x double> %{{.*}})
+  // CHECK: @llvm.masked.expandload.v8f64(ptr %{{.*}}, <8 x i1> %{{.*}}, <8 x double> %{{.*}})
   return _mm512_maskz_expandloadu_pd(__U, __P); 
 }
 
 __m512i test_mm512_mask_expandloadu_epi32(__m512i __W, __mmask16 __U, void const *__P) {
   // CHECK-LABEL: @test_mm512_mask_expandloadu_epi32
-  // CHECK: @llvm.masked.expandload.v16i32(i32* %{{.*}}, <16 x i1> %{{.*}}, <16 x i32> %{{.*}})
+  // CHECK: @llvm.masked.expandload.v16i32(ptr %{{.*}}, <16 x i1> %{{.*}}, <16 x i32> %{{.*}})
   return _mm512_mask_expandloadu_epi32(__W, __U, __P); 
 }
 
 __m512i test_mm512_maskz_expandloadu_epi32(__mmask16 __U, void const *__P) {
   // CHECK-LABEL: @test_mm512_maskz_expandloadu_epi32
-  // CHECK: @llvm.masked.expandload.v16i32(i32* %{{.*}}, <16 x i1> %{{.*}}, <16 x i32> %{{.*}})
+  // CHECK: @llvm.masked.expandload.v16i32(ptr %{{.*}}, <16 x i1> %{{.*}}, <16 x i32> %{{.*}})
   return _mm512_maskz_expandloadu_epi32(__U, __P); 
 }
 
 __m512 test_mm512_mask_expandloadu_ps(__m512 __W, __mmask16 __U, void const *__P) {
   // CHECK-LABEL: @test_mm512_mask_expandloadu_ps
-  // CHECK: @llvm.masked.expandload.v16f32(float* %{{.*}}, <16 x i1> %{{.*}}, <16 x float> %{{.*}})
+  // CHECK: @llvm.masked.expandload.v16f32(ptr %{{.*}}, <16 x i1> %{{.*}}, <16 x float> %{{.*}})
   return _mm512_mask_expandloadu_ps(__W, __U, __P); 
 }
 
 __m512 test_mm512_maskz_expandloadu_ps(__mmask16 __U, void const *__P) {
   // CHECK-LABEL: @test_mm512_maskz_expandloadu_ps
-  // CHECK: @llvm.masked.expandload.v16f32(float* %{{.*}}, <16 x i1> %{{.*}}, <16 x float> %{{.*}})
+  // CHECK: @llvm.masked.expandload.v16f32(ptr %{{.*}}, <16 x i1> %{{.*}}, <16 x float> %{{.*}})
   return _mm512_maskz_expandloadu_ps(__U, __P); 
 }
 
@@ -8912,25 +8906,25 @@ __m512 test_mm512_maskz_mov_ps(__mmask16 __U, __m512 __A) {
 
 void test_mm512_mask_compressstoreu_pd(void *__P, __mmask8 __U, __m512d __A) {
   // CHECK-LABEL: @test_mm512_mask_compressstoreu_pd
-  // CHECK: @llvm.masked.compressstore.v8f64(<8 x double> %{{.*}}, double* %{{.*}}, <8 x i1> %{{.*}})
+  // CHECK: @llvm.masked.compressstore.v8f64(<8 x double> %{{.*}}, ptr %{{.*}}, <8 x i1> %{{.*}})
   return _mm512_mask_compressstoreu_pd(__P, __U, __A); 
 }
 
 void test_mm512_mask_compressstoreu_epi64(void *__P, __mmask8 __U, __m512i __A) {
   // CHECK-LABEL: @test_mm512_mask_compressstoreu_epi64
-  // CHECK: @llvm.masked.compressstore.v8i64(<8 x i64> %{{.*}}, i64* %{{.*}}, <8 x i1> %{{.*}})
+  // CHECK: @llvm.masked.compressstore.v8i64(<8 x i64> %{{.*}}, ptr %{{.*}}, <8 x i1> %{{.*}})
   return _mm512_mask_compressstoreu_epi64(__P, __U, __A); 
 }
 
 void test_mm512_mask_compressstoreu_ps(void *__P, __mmask16 __U, __m512 __A) {
   // CHECK-LABEL: @test_mm512_mask_compressstoreu_ps
-  // CHECK: @llvm.masked.compressstore.v16f32(<16 x float> %{{.*}}, float* %{{.*}}, <16 x i1> %{{.*}})
+  // CHECK: @llvm.masked.compressstore.v16f32(<16 x float> %{{.*}}, ptr %{{.*}}, <16 x i1> %{{.*}})
   return _mm512_mask_compressstoreu_ps(__P, __U, __A); 
 }
 
 void test_mm512_mask_compressstoreu_epi32(void *__P, __mmask16 __U, __m512i __A) {
   // CHECK-LABEL: @test_mm512_mask_compressstoreu_epi32
-  // CHECK: @llvm.masked.compressstore.v16i32(<16 x i32> %{{.*}}, i32* %{{.*}}, <16 x i1> %{{.*}})
+  // CHECK: @llvm.masked.compressstore.v16i32(<16 x i32> %{{.*}}, ptr %{{.*}}, <16 x i1> %{{.*}})
   return _mm512_mask_compressstoreu_epi32(__P, __U, __A); 
 }
 
@@ -10122,70 +10116,70 @@ __m512i test_mm512_set_epi8(char e63, char e62, char e61, char e60, char e59,
     char e1, char e0) {
 
   //CHECK-LABEL: @test_mm512_set_epi8
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
-  //CHECK: load i8, i8* %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
+  //CHECK: load i8, ptr %{{.*}}, align 1
   return _mm512_set_epi8(e63, e62, e61, e60, e59, e58, e57, e56, e55, e54,
       e53, e52, e51, e50, e49, e48,e47, e46, e45, e44, e43, e42, e41, e40,
       e39, e38, e37, e36, e35, e34, e33, e32,e31, e30, e29, e28, e27, e26,
@@ -10711,42 +10705,42 @@ __m128d test_mm_maskz_move_sd (__mmask8 __U, __m128d __A, __m128d __B)
 void test_mm_mask_store_ss(float * __P, __mmask8 __U, __m128 __A)
 {
   // CHECK-LABEL: @test_mm_mask_store_ss
-  // CHECK: call void @llvm.masked.store.v4f32.p0v4f32(<4 x float> %{{.*}}, <4 x float>* %{{.*}}, i32 1, <4 x i1> %{{.*}})
+  // CHECK: call void @llvm.masked.store.v4f32.p0(<4 x float> %{{.*}}, ptr %{{.*}}, i32 1, <4 x i1> %{{.*}})
   _mm_mask_store_ss(__P, __U, __A);
 }
 
 void test_mm_mask_store_sd(double * __P, __mmask8 __U, __m128d __A)
 {
   // CHECK-LABEL: @test_mm_mask_store_sd
-  // CHECK: call void @llvm.masked.store.v2f64.p0v2f64(<2 x double> %{{.*}}, <2 x double>* %{{.*}}, i32 1, <2 x i1> %{{.*}})
+  // CHECK: call void @llvm.masked.store.v2f64.p0(<2 x double> %{{.*}}, ptr %{{.*}}, i32 1, <2 x i1> %{{.*}})
   _mm_mask_store_sd(__P, __U, __A);
 }
 
 __m128 test_mm_mask_load_ss(__m128 __A, __mmask8 __U, const float* __W)
 {
   // CHECK-LABEL: @test_mm_mask_load_ss
-  // CHECK: call <4 x float> @llvm.masked.load.v4f32.p0v4f32(<4 x float>* %{{.*}}, i32 1, <4 x i1> %{{.*}}, <4 x float> %{{.*}})
+  // CHECK: call <4 x float> @llvm.masked.load.v4f32.p0(ptr %{{.*}}, i32 1, <4 x i1> %{{.*}}, <4 x float> %{{.*}})
   return _mm_mask_load_ss(__A, __U, __W);
 }
 
 __m128 test_mm_maskz_load_ss (__mmask8 __U, const float * __W)
 {
   // CHECK-LABEL: @test_mm_maskz_load_ss
-  // CHECK: call <4 x float> @llvm.masked.load.v4f32.p0v4f32(<4 x float>* %{{.*}}, i32 1, <4 x i1> %{{.*}}, <4 x float> %{{.*}})
+  // CHECK: call <4 x float> @llvm.masked.load.v4f32.p0(ptr %{{.*}}, i32 1, <4 x i1> %{{.*}}, <4 x float> %{{.*}})
   return _mm_maskz_load_ss (__U, __W);
 }
 
 __m128d test_mm_mask_load_sd (__m128d __A, __mmask8 __U, const double * __W)
 {
   // CHECK-LABEL: @test_mm_mask_load_sd
-  // CHECK: call <2 x double> @llvm.masked.load.v2f64.p0v2f64(<2 x double>* %{{.*}}, i32 1, <2 x i1> %{{.*}}, <2 x double> %{{.*}})
+  // CHECK: call <2 x double> @llvm.masked.load.v2f64.p0(ptr %{{.*}}, i32 1, <2 x i1> %{{.*}}, <2 x double> %{{.*}})
   return _mm_mask_load_sd (__A, __U, __W);
 }
 
 __m128d test_mm_maskz_load_sd (__mmask8 __U, const double * __W)
 {
   // CHECK-LABEL: @test_mm_maskz_load_sd
-  // CHECK: call <2 x double> @llvm.masked.load.v2f64.p0v2f64(<2 x double>* %{{.*}}, i32 1, <2 x i1> %{{.*}}, <2 x double> %{{.*}})
+  // CHECK: call <2 x double> @llvm.masked.load.v2f64.p0(ptr %{{.*}}, i32 1, <2 x i1> %{{.*}}, <2 x double> %{{.*}})
   return _mm_maskz_load_sd (__U, __W);
 }
 

diff  --git a/clang/test/CodeGen/aarch64-bf16-ldst-intrinsics.c b/clang/test/CodeGen/aarch64-bf16-ldst-intrinsics.c
index 15ccf8d81e580..84d7fd05c276e 100644
--- a/clang/test/CodeGen/aarch64-bf16-ldst-intrinsics.c
+++ b/clang/test/CodeGen/aarch64-bf16-ldst-intrinsics.c
@@ -1,7 +1,7 @@
 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
-// RUN: %clang_cc1 -no-opaque-pointers -triple aarch64-arm-none-eabi -target-feature +neon -target-feature +bf16 \
+// RUN: %clang_cc1 -triple aarch64-arm-none-eabi -target-feature +neon -target-feature +bf16 \
 // RUN:  -O2 -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,CHECK64
-// RUN: %clang_cc1 -no-opaque-pointers -triple armv8.6a-arm-none-eabi -target-feature +neon -target-feature +bf16 -mfloat-abi hard \
+// RUN: %clang_cc1 -triple armv8.6a-arm-none-eabi -target-feature +neon -target-feature +bf16 -mfloat-abi hard \
 // RUN:  -O2 -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,CHECK32
 
 // REQUIRES: arm-registered-target,aarch64-registered-target
@@ -10,8 +10,7 @@
 
 // CHECK-LABEL: @test_vld1_bf16(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to <4 x bfloat>*
-// CHECK-NEXT:    [[TMP1:%.*]] = load <4 x bfloat>, <4 x bfloat>* [[TMP0]], align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load <4 x bfloat>, ptr [[PTR:%.*]], align 2
 // CHECK-NEXT:    ret <4 x bfloat> [[TMP1]]
 //
 bfloat16x4_t test_vld1_bf16(bfloat16_t const *ptr) {
@@ -20,8 +19,7 @@ bfloat16x4_t test_vld1_bf16(bfloat16_t const *ptr) {
 
 // CHECK-LABEL: @test_vld1q_bf16(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to <8 x bfloat>*
-// CHECK-NEXT:    [[TMP1:%.*]] = load <8 x bfloat>, <8 x bfloat>* [[TMP0]], align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load <8 x bfloat>, ptr [[PTR:%.*]], align 2
 // CHECK-NEXT:    ret <8 x bfloat> [[TMP1]]
 //
 bfloat16x8_t test_vld1q_bf16(bfloat16_t const *ptr) {
@@ -30,7 +28,7 @@ bfloat16x8_t test_vld1q_bf16(bfloat16_t const *ptr) {
 
 // CHECK-LABEL: @test_vld1_lane_bf16(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load bfloat, bfloat* [[PTR:%.*]], align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load bfloat, ptr [[PTR:%.*]], align 2
 // CHECK-NEXT:    [[VLD1_LANE:%.*]] = insertelement <4 x bfloat> [[SRC:%.*]], bfloat [[TMP0]], i64 0
 // CHECK-NEXT:    ret <4 x bfloat> [[VLD1_LANE]]
 //
@@ -40,7 +38,7 @@ bfloat16x4_t test_vld1_lane_bf16(bfloat16_t const *ptr, bfloat16x4_t src) {
 
 // CHECK-LABEL: @test_vld1q_lane_bf16(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load bfloat, bfloat* [[PTR:%.*]], align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load bfloat, ptr [[PTR:%.*]], align 2
 // CHECK-NEXT:    [[VLD1_LANE:%.*]] = insertelement <8 x bfloat> [[SRC:%.*]], bfloat [[TMP0]], i64 7
 // CHECK-NEXT:    ret <8 x bfloat> [[VLD1_LANE]]
 //
@@ -50,7 +48,7 @@ bfloat16x8_t test_vld1q_lane_bf16(bfloat16_t const *ptr, bfloat16x8_t src) {
 
 // CHECK-LABEL: @test_vld1_dup_bf16(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load bfloat, bfloat* [[PTR:%.*]], align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load bfloat, ptr [[PTR:%.*]], align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = insertelement <4 x bfloat> poison, bfloat [[TMP0]], i64 0
 // CHECK-NEXT:    [[LANE:%.*]] = shufflevector <4 x bfloat> [[TMP1]], <4 x bfloat> poison, <4 x i32> zeroinitializer
 // CHECK-NEXT:    ret <4 x bfloat> [[LANE]]
@@ -61,7 +59,7 @@ bfloat16x4_t test_vld1_dup_bf16(bfloat16_t const *ptr) {
 
 // CHECK64-LABEL: @test_vld1_bf16_x2(
 // CHECK64-NEXT:  entry:
-// CHECK64-NEXT:    [[VLD1XN:%.*]] = tail call { <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld1x2.v4bf16.p0bf16(bfloat* [[PTR:%.*]])
+// CHECK64-NEXT:    [[VLD1XN:%.*]] = tail call { <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld1x2.v4bf16.p0(ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD1XN_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat> } [[VLD1XN]], 0
 // CHECK64-NEXT:    [[VLD1XN_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat> } [[VLD1XN]], 1
 // CHECK64-NEXT:    [[DOTFCA_0_0_INSERT:%.*]] = insertvalue [[STRUCT_BFLOAT16X4X2_T:%.*]] poison, <4 x bfloat> [[VLD1XN_FCA_0_EXTRACT]], 0, 0
@@ -70,7 +68,7 @@ bfloat16x4_t test_vld1_dup_bf16(bfloat16_t const *ptr) {
 //
 // CHECK32-LABEL: @test_vld1_bf16_x2(
 // CHECK32-NEXT:  entry:
-// CHECK32-NEXT:    [[VLD1XN:%.*]] = tail call { <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld1x2.v4bf16.p0bf16(bfloat* [[PTR:%.*]])
+// CHECK32-NEXT:    [[VLD1XN:%.*]] = tail call { <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld1x2.v4bf16.p0(ptr [[PTR:%.*]])
 // CHECK32-NEXT:    [[VLD1XN_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat> } [[VLD1XN]], 0
 // CHECK32-NEXT:    [[VLD1XN_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat> } [[VLD1XN]], 1
 // CHECK32-NEXT:    [[TMP0:%.*]] = bitcast <4 x bfloat> [[VLD1XN_FCA_0_EXTRACT]] to <2 x i32>
@@ -85,7 +83,7 @@ bfloat16x4x2_t test_vld1_bf16_x2(bfloat16_t const *ptr) {
 
 // CHECK64-LABEL: @test_vld1q_bf16_x2(
 // CHECK64-NEXT:  entry:
-// CHECK64-NEXT:    [[VLD1XN:%.*]] = tail call { <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld1x2.v8bf16.p0bf16(bfloat* [[PTR:%.*]])
+// CHECK64-NEXT:    [[VLD1XN:%.*]] = tail call { <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld1x2.v8bf16.p0(ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD1XN_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat> } [[VLD1XN]], 0
 // CHECK64-NEXT:    [[VLD1XN_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat> } [[VLD1XN]], 1
 // CHECK64-NEXT:    [[DOTFCA_0_0_INSERT:%.*]] = insertvalue [[STRUCT_BFLOAT16X8X2_T:%.*]] poison, <8 x bfloat> [[VLD1XN_FCA_0_EXTRACT]], 0, 0
@@ -94,7 +92,7 @@ bfloat16x4x2_t test_vld1_bf16_x2(bfloat16_t const *ptr) {
 //
 // CHECK32-LABEL: @test_vld1q_bf16_x2(
 // CHECK32-NEXT:  entry:
-// CHECK32-NEXT:    [[VLD1XN:%.*]] = tail call { <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld1x2.v8bf16.p0bf16(bfloat* [[PTR:%.*]])
+// CHECK32-NEXT:    [[VLD1XN:%.*]] = tail call { <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld1x2.v8bf16.p0(ptr [[PTR:%.*]])
 // CHECK32-NEXT:    [[VLD1XN_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat> } [[VLD1XN]], 0
 // CHECK32-NEXT:    [[VLD1XN_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat> } [[VLD1XN]], 1
 // CHECK32-NEXT:    [[TMP0:%.*]] = bitcast <8 x bfloat> [[VLD1XN_FCA_0_EXTRACT]] to <4 x i32>
@@ -109,7 +107,7 @@ bfloat16x8x2_t test_vld1q_bf16_x2(bfloat16_t const *ptr) {
 
 // CHECK64-LABEL: @test_vld1_bf16_x3(
 // CHECK64-NEXT:  entry:
-// CHECK64-NEXT:    [[VLD1XN:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld1x3.v4bf16.p0bf16(bfloat* [[PTR:%.*]])
+// CHECK64-NEXT:    [[VLD1XN:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld1x3.v4bf16.p0(ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD1XN_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD1XN]], 0
 // CHECK64-NEXT:    [[VLD1XN_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD1XN]], 1
 // CHECK64-NEXT:    [[VLD1XN_FCA_2_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD1XN]], 2
@@ -120,7 +118,7 @@ bfloat16x8x2_t test_vld1q_bf16_x2(bfloat16_t const *ptr) {
 //
 // CHECK32-LABEL: @test_vld1_bf16_x3(
 // CHECK32-NEXT:  entry:
-// CHECK32-NEXT:    [[VLD1XN:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld1x3.v4bf16.p0bf16(bfloat* [[PTR:%.*]])
+// CHECK32-NEXT:    [[VLD1XN:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld1x3.v4bf16.p0(ptr [[PTR:%.*]])
 // CHECK32-NEXT:    [[VLD1XN_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD1XN]], 0
 // CHECK32-NEXT:    [[VLD1XN_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD1XN]], 1
 // CHECK32-NEXT:    [[VLD1XN_FCA_2_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD1XN]], 2
@@ -138,7 +136,7 @@ bfloat16x4x3_t test_vld1_bf16_x3(bfloat16_t const *ptr) {
 
 // CHECK64-LABEL: @test_vld1q_bf16_x3(
 // CHECK64-NEXT:  entry:
-// CHECK64-NEXT:    [[VLD1XN:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld1x3.v8bf16.p0bf16(bfloat* [[PTR:%.*]])
+// CHECK64-NEXT:    [[VLD1XN:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld1x3.v8bf16.p0(ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD1XN_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD1XN]], 0
 // CHECK64-NEXT:    [[VLD1XN_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD1XN]], 1
 // CHECK64-NEXT:    [[VLD1XN_FCA_2_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD1XN]], 2
@@ -149,7 +147,7 @@ bfloat16x4x3_t test_vld1_bf16_x3(bfloat16_t const *ptr) {
 //
 // CHECK32-LABEL: @test_vld1q_bf16_x3(
 // CHECK32-NEXT:  entry:
-// CHECK32-NEXT:    [[VLD1XN:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld1x3.v8bf16.p0bf16(bfloat* [[PTR:%.*]])
+// CHECK32-NEXT:    [[VLD1XN:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld1x3.v8bf16.p0(ptr [[PTR:%.*]])
 // CHECK32-NEXT:    [[VLD1XN_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD1XN]], 0
 // CHECK32-NEXT:    [[VLD1XN_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD1XN]], 1
 // CHECK32-NEXT:    [[VLD1XN_FCA_2_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD1XN]], 2
@@ -167,7 +165,7 @@ bfloat16x8x3_t test_vld1q_bf16_x3(bfloat16_t const *ptr) {
 
 // CHECK64-LABEL: @test_vld1_bf16_x4(
 // CHECK64-NEXT:  entry:
-// CHECK64-NEXT:    [[VLD1XN:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld1x4.v4bf16.p0bf16(bfloat* [[PTR:%.*]])
+// CHECK64-NEXT:    [[VLD1XN:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld1x4.v4bf16.p0(ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD1XN_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD1XN]], 0
 // CHECK64-NEXT:    [[VLD1XN_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD1XN]], 1
 // CHECK64-NEXT:    [[VLD1XN_FCA_2_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD1XN]], 2
@@ -180,7 +178,7 @@ bfloat16x8x3_t test_vld1q_bf16_x3(bfloat16_t const *ptr) {
 //
 // CHECK32-LABEL: @test_vld1_bf16_x4(
 // CHECK32-NEXT:  entry:
-// CHECK32-NEXT:    [[VLD1XN:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld1x4.v4bf16.p0bf16(bfloat* [[PTR:%.*]])
+// CHECK32-NEXT:    [[VLD1XN:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld1x4.v4bf16.p0(ptr [[PTR:%.*]])
 // CHECK32-NEXT:    [[VLD1XN_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD1XN]], 0
 // CHECK32-NEXT:    [[VLD1XN_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD1XN]], 1
 // CHECK32-NEXT:    [[VLD1XN_FCA_2_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD1XN]], 2
@@ -201,7 +199,7 @@ bfloat16x4x4_t test_vld1_bf16_x4(bfloat16_t const *ptr) {
 
 // CHECK64-LABEL: @test_vld1q_bf16_x4(
 // CHECK64-NEXT:  entry:
-// CHECK64-NEXT:    [[VLD1XN:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld1x4.v8bf16.p0bf16(bfloat* [[PTR:%.*]])
+// CHECK64-NEXT:    [[VLD1XN:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld1x4.v8bf16.p0(ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD1XN_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD1XN]], 0
 // CHECK64-NEXT:    [[VLD1XN_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD1XN]], 1
 // CHECK64-NEXT:    [[VLD1XN_FCA_2_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD1XN]], 2
@@ -214,7 +212,7 @@ bfloat16x4x4_t test_vld1_bf16_x4(bfloat16_t const *ptr) {
 //
 // CHECK32-LABEL: @test_vld1q_bf16_x4(
 // CHECK32-NEXT:  entry:
-// CHECK32-NEXT:    [[VLD1XN:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld1x4.v8bf16.p0bf16(bfloat* [[PTR:%.*]])
+// CHECK32-NEXT:    [[VLD1XN:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld1x4.v8bf16.p0(ptr [[PTR:%.*]])
 // CHECK32-NEXT:    [[VLD1XN_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD1XN]], 0
 // CHECK32-NEXT:    [[VLD1XN_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD1XN]], 1
 // CHECK32-NEXT:    [[VLD1XN_FCA_2_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD1XN]], 2
@@ -235,7 +233,7 @@ bfloat16x8x4_t test_vld1q_bf16_x4(bfloat16_t const *ptr) {
 
 // CHECK-LABEL: @test_vld1q_dup_bf16(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load bfloat, bfloat* [[PTR:%.*]], align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load bfloat, ptr [[PTR:%.*]], align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = insertelement <8 x bfloat> poison, bfloat [[TMP0]], i64 0
 // CHECK-NEXT:    [[LANE:%.*]] = shufflevector <8 x bfloat> [[TMP1]], <8 x bfloat> poison, <8 x i32> zeroinitializer
 // CHECK-NEXT:    ret <8 x bfloat> [[LANE]]
@@ -246,8 +244,7 @@ bfloat16x8_t test_vld1q_dup_bf16(bfloat16_t const *ptr) {
 
 // CHECK64-LABEL: @test_vld2_bf16(
 // CHECK64-NEXT:  entry:
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to <4 x bfloat>*
-// CHECK64-NEXT:    [[VLD2:%.*]] = tail call { <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld2.v4bf16.p0v4bf16(<4 x bfloat>* [[TMP0]])
+// CHECK64-NEXT:    [[VLD2:%.*]] = tail call { <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld2.v4bf16.p0(ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD2_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat> } [[VLD2]], 0
 // CHECK64-NEXT:    [[VLD2_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat> } [[VLD2]], 1
 // CHECK64-NEXT:    [[DOTFCA_0_0_INSERT:%.*]] = insertvalue [[STRUCT_BFLOAT16X4X2_T:%.*]] poison, <4 x bfloat> [[VLD2_FCA_0_EXTRACT]], 0, 0
@@ -256,8 +253,7 @@ bfloat16x8_t test_vld1q_dup_bf16(bfloat16_t const *ptr) {
 //
 // CHECK32-LABEL: @test_vld2_bf16(
 // CHECK32-NEXT:  entry:
-// CHECK32-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    [[VLD2_V:%.*]] = tail call { <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld2.v4bf16.p0i8(i8* [[TMP0]], i32 2)
+// CHECK32-NEXT:    [[VLD2_V:%.*]] = tail call { <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld2.v4bf16.p0(ptr [[PTR:%.*]], i32 2)
 // CHECK32-NEXT:    [[VLD2_V_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat> } [[VLD2_V]], 0
 // CHECK32-NEXT:    [[VLD2_V_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat> } [[VLD2_V]], 1
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <4 x bfloat> [[VLD2_V_FCA_0_EXTRACT]] to <2 x i32>
@@ -272,8 +268,7 @@ bfloat16x4x2_t test_vld2_bf16(bfloat16_t const *ptr) {
 
 // CHECK64-LABEL: @test_vld2q_bf16(
 // CHECK64-NEXT:  entry:
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to <8 x bfloat>*
-// CHECK64-NEXT:    [[VLD2:%.*]] = tail call { <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld2.v8bf16.p0v8bf16(<8 x bfloat>* [[TMP0]])
+// CHECK64-NEXT:    [[VLD2:%.*]] = tail call { <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld2.v8bf16.p0(ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD2_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat> } [[VLD2]], 0
 // CHECK64-NEXT:    [[VLD2_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat> } [[VLD2]], 1
 // CHECK64-NEXT:    [[DOTFCA_0_0_INSERT:%.*]] = insertvalue [[STRUCT_BFLOAT16X8X2_T:%.*]] poison, <8 x bfloat> [[VLD2_FCA_0_EXTRACT]], 0, 0
@@ -282,8 +277,7 @@ bfloat16x4x2_t test_vld2_bf16(bfloat16_t const *ptr) {
 //
 // CHECK32-LABEL: @test_vld2q_bf16(
 // CHECK32-NEXT:  entry:
-// CHECK32-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    [[VLD2Q_V:%.*]] = tail call { <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld2.v8bf16.p0i8(i8* [[TMP0]], i32 2)
+// CHECK32-NEXT:    [[VLD2Q_V:%.*]] = tail call { <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld2.v8bf16.p0(ptr [[PTR:%.*]], i32 2)
 // CHECK32-NEXT:    [[VLD2Q_V_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat> } [[VLD2Q_V]], 0
 // CHECK32-NEXT:    [[VLD2Q_V_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat> } [[VLD2Q_V]], 1
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <8 x bfloat> [[VLD2Q_V_FCA_0_EXTRACT]] to <4 x i32>
@@ -300,8 +294,7 @@ bfloat16x8x2_t test_vld2q_bf16(bfloat16_t const *ptr) {
 // CHECK64-NEXT:  entry:
 // CHECK64-NEXT:    [[SRC_COERCE_FCA_0_EXTRACT:%.*]] = extractvalue [2 x <4 x bfloat>] [[SRC_COERCE:%.*]], 0
 // CHECK64-NEXT:    [[SRC_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [2 x <4 x bfloat>] [[SRC_COERCE]], 1
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK64-NEXT:    [[VLD2_LANE:%.*]] = tail call { <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld2lane.v4bf16.p0i8(<4 x bfloat> [[SRC_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[SRC_COERCE_FCA_1_EXTRACT]], i64 1, i8* [[TMP0]])
+// CHECK64-NEXT:    [[VLD2_LANE:%.*]] = tail call { <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld2lane.v4bf16.p0(<4 x bfloat> [[SRC_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[SRC_COERCE_FCA_1_EXTRACT]], i64 1, ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD2_LANE_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat> } [[VLD2_LANE]], 0
 // CHECK64-NEXT:    [[VLD2_LANE_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat> } [[VLD2_LANE]], 1
 // CHECK64-NEXT:    [[DOTFCA_0_0_INSERT:%.*]] = insertvalue [[STRUCT_BFLOAT16X4X2_T:%.*]] poison, <4 x bfloat> [[VLD2_LANE_FCA_0_EXTRACT]], 0, 0
@@ -314,8 +307,7 @@ bfloat16x8x2_t test_vld2q_bf16(bfloat16_t const *ptr) {
 // CHECK32-NEXT:    [[SRC_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [2 x <2 x i32>] [[SRC_COERCE]], 1
 // CHECK32-NEXT:    [[TMP0:%.*]] = bitcast <2 x i32> [[SRC_COERCE_FCA_0_EXTRACT]] to <4 x bfloat>
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <2 x i32> [[SRC_COERCE_FCA_1_EXTRACT]] to <4 x bfloat>
-// CHECK32-NEXT:    [[TMP2:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    [[VLD2_LANE_V:%.*]] = tail call { <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld2lane.v4bf16.p0i8(i8* [[TMP2]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], i32 1, i32 2)
+// CHECK32-NEXT:    [[VLD2_LANE_V:%.*]] = tail call { <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld2lane.v4bf16.p0(ptr [[PTR:%.*]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], i32 1, i32 2)
 // CHECK32-NEXT:    [[VLD2_LANE_V_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat> } [[VLD2_LANE_V]], 0
 // CHECK32-NEXT:    [[VLD2_LANE_V_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat> } [[VLD2_LANE_V]], 1
 // CHECK32-NEXT:    [[TMP3:%.*]] = bitcast <4 x bfloat> [[VLD2_LANE_V_FCA_0_EXTRACT]] to <2 x i32>
@@ -332,8 +324,7 @@ bfloat16x4x2_t test_vld2_lane_bf16(bfloat16_t const *ptr, bfloat16x4x2_t src) {
 // CHECK64-NEXT:  entry:
 // CHECK64-NEXT:    [[SRC_COERCE_FCA_0_EXTRACT:%.*]] = extractvalue [2 x <8 x bfloat>] [[SRC_COERCE:%.*]], 0
 // CHECK64-NEXT:    [[SRC_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [2 x <8 x bfloat>] [[SRC_COERCE]], 1
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK64-NEXT:    [[VLD2_LANE:%.*]] = tail call { <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld2lane.v8bf16.p0i8(<8 x bfloat> [[SRC_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[SRC_COERCE_FCA_1_EXTRACT]], i64 7, i8* [[TMP0]])
+// CHECK64-NEXT:    [[VLD2_LANE:%.*]] = tail call { <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld2lane.v8bf16.p0(<8 x bfloat> [[SRC_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[SRC_COERCE_FCA_1_EXTRACT]], i64 7, ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD2_LANE_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat> } [[VLD2_LANE]], 0
 // CHECK64-NEXT:    [[VLD2_LANE_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat> } [[VLD2_LANE]], 1
 // CHECK64-NEXT:    [[DOTFCA_0_0_INSERT:%.*]] = insertvalue [[STRUCT_BFLOAT16X8X2_T:%.*]] poison, <8 x bfloat> [[VLD2_LANE_FCA_0_EXTRACT]], 0, 0
@@ -346,8 +337,7 @@ bfloat16x4x2_t test_vld2_lane_bf16(bfloat16_t const *ptr, bfloat16x4x2_t src) {
 // CHECK32-NEXT:    [[SRC_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [2 x <4 x i32>] [[SRC_COERCE]], 1
 // CHECK32-NEXT:    [[TMP0:%.*]] = bitcast <4 x i32> [[SRC_COERCE_FCA_0_EXTRACT]] to <8 x bfloat>
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <4 x i32> [[SRC_COERCE_FCA_1_EXTRACT]] to <8 x bfloat>
-// CHECK32-NEXT:    [[TMP2:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    [[VLD2Q_LANE_V:%.*]] = tail call { <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld2lane.v8bf16.p0i8(i8* [[TMP2]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], i32 7, i32 2)
+// CHECK32-NEXT:    [[VLD2Q_LANE_V:%.*]] = tail call { <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld2lane.v8bf16.p0(ptr [[PTR:%.*]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], i32 7, i32 2)
 // CHECK32-NEXT:    [[VLD2Q_LANE_V_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat> } [[VLD2Q_LANE_V]], 0
 // CHECK32-NEXT:    [[VLD2Q_LANE_V_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat> } [[VLD2Q_LANE_V]], 1
 // CHECK32-NEXT:    [[TMP3:%.*]] = bitcast <8 x bfloat> [[VLD2Q_LANE_V_FCA_0_EXTRACT]] to <4 x i32>
@@ -362,8 +352,7 @@ bfloat16x8x2_t test_vld2q_lane_bf16(bfloat16_t const *ptr, bfloat16x8x2_t src) {
 
 // CHECK64-LABEL: @test_vld3_bf16(
 // CHECK64-NEXT:  entry:
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to <4 x bfloat>*
-// CHECK64-NEXT:    [[VLD3:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld3.v4bf16.p0v4bf16(<4 x bfloat>* [[TMP0]])
+// CHECK64-NEXT:    [[VLD3:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld3.v4bf16.p0(ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD3_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD3]], 0
 // CHECK64-NEXT:    [[VLD3_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD3]], 1
 // CHECK64-NEXT:    [[VLD3_FCA_2_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD3]], 2
@@ -374,8 +363,7 @@ bfloat16x8x2_t test_vld2q_lane_bf16(bfloat16_t const *ptr, bfloat16x8x2_t src) {
 //
 // CHECK32-LABEL: @test_vld3_bf16(
 // CHECK32-NEXT:  entry:
-// CHECK32-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    [[VLD3_V:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld3.v4bf16.p0i8(i8* [[TMP0]], i32 2)
+// CHECK32-NEXT:    [[VLD3_V:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld3.v4bf16.p0(ptr [[PTR:%.*]], i32 2)
 // CHECK32-NEXT:    [[VLD3_V_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD3_V]], 0
 // CHECK32-NEXT:    [[VLD3_V_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD3_V]], 1
 // CHECK32-NEXT:    [[VLD3_V_FCA_2_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD3_V]], 2
@@ -393,8 +381,7 @@ bfloat16x4x3_t test_vld3_bf16(bfloat16_t const *ptr) {
 
 // CHECK64-LABEL: @test_vld3q_bf16(
 // CHECK64-NEXT:  entry:
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to <8 x bfloat>*
-// CHECK64-NEXT:    [[VLD3:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld3.v8bf16.p0v8bf16(<8 x bfloat>* [[TMP0]])
+// CHECK64-NEXT:    [[VLD3:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld3.v8bf16.p0(ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD3_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD3]], 0
 // CHECK64-NEXT:    [[VLD3_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD3]], 1
 // CHECK64-NEXT:    [[VLD3_FCA_2_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD3]], 2
@@ -405,8 +392,7 @@ bfloat16x4x3_t test_vld3_bf16(bfloat16_t const *ptr) {
 //
 // CHECK32-LABEL: @test_vld3q_bf16(
 // CHECK32-NEXT:  entry:
-// CHECK32-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    [[VLD3Q_V:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld3.v8bf16.p0i8(i8* [[TMP0]], i32 2)
+// CHECK32-NEXT:    [[VLD3Q_V:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld3.v8bf16.p0(ptr [[PTR:%.*]], i32 2)
 // CHECK32-NEXT:    [[VLD3Q_V_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD3Q_V]], 0
 // CHECK32-NEXT:    [[VLD3Q_V_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD3Q_V]], 1
 // CHECK32-NEXT:    [[VLD3Q_V_FCA_2_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD3Q_V]], 2
@@ -427,8 +413,7 @@ bfloat16x8x3_t test_vld3q_bf16(bfloat16_t const *ptr) {
 // CHECK64-NEXT:    [[SRC_COERCE_FCA_0_EXTRACT:%.*]] = extractvalue [3 x <4 x bfloat>] [[SRC_COERCE:%.*]], 0
 // CHECK64-NEXT:    [[SRC_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [3 x <4 x bfloat>] [[SRC_COERCE]], 1
 // CHECK64-NEXT:    [[SRC_COERCE_FCA_2_EXTRACT:%.*]] = extractvalue [3 x <4 x bfloat>] [[SRC_COERCE]], 2
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK64-NEXT:    [[VLD3_LANE:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld3lane.v4bf16.p0i8(<4 x bfloat> [[SRC_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[SRC_COERCE_FCA_1_EXTRACT]], <4 x bfloat> [[SRC_COERCE_FCA_2_EXTRACT]], i64 1, i8* [[TMP0]])
+// CHECK64-NEXT:    [[VLD3_LANE:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld3lane.v4bf16.p0(<4 x bfloat> [[SRC_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[SRC_COERCE_FCA_1_EXTRACT]], <4 x bfloat> [[SRC_COERCE_FCA_2_EXTRACT]], i64 1, ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD3_LANE_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD3_LANE]], 0
 // CHECK64-NEXT:    [[VLD3_LANE_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD3_LANE]], 1
 // CHECK64-NEXT:    [[VLD3_LANE_FCA_2_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD3_LANE]], 2
@@ -445,8 +430,7 @@ bfloat16x8x3_t test_vld3q_bf16(bfloat16_t const *ptr) {
 // CHECK32-NEXT:    [[TMP0:%.*]] = bitcast <2 x i32> [[SRC_COERCE_FCA_0_EXTRACT]] to <4 x bfloat>
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <2 x i32> [[SRC_COERCE_FCA_1_EXTRACT]] to <4 x bfloat>
 // CHECK32-NEXT:    [[TMP2:%.*]] = bitcast <2 x i32> [[SRC_COERCE_FCA_2_EXTRACT]] to <4 x bfloat>
-// CHECK32-NEXT:    [[TMP3:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    [[VLD3_LANE_V:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld3lane.v4bf16.p0i8(i8* [[TMP3]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], <4 x bfloat> [[TMP2]], i32 1, i32 2)
+// CHECK32-NEXT:    [[VLD3_LANE_V:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld3lane.v4bf16.p0(ptr [[PTR:%.*]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], <4 x bfloat> [[TMP2]], i32 1, i32 2)
 // CHECK32-NEXT:    [[VLD3_LANE_V_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD3_LANE_V]], 0
 // CHECK32-NEXT:    [[VLD3_LANE_V_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD3_LANE_V]], 1
 // CHECK32-NEXT:    [[VLD3_LANE_V_FCA_2_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD3_LANE_V]], 2
@@ -467,8 +451,7 @@ bfloat16x4x3_t test_vld3_lane_bf16(bfloat16_t const *ptr, bfloat16x4x3_t src) {
 // CHECK64-NEXT:    [[SRC_COERCE_FCA_0_EXTRACT:%.*]] = extractvalue [3 x <8 x bfloat>] [[SRC_COERCE:%.*]], 0
 // CHECK64-NEXT:    [[SRC_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [3 x <8 x bfloat>] [[SRC_COERCE]], 1
 // CHECK64-NEXT:    [[SRC_COERCE_FCA_2_EXTRACT:%.*]] = extractvalue [3 x <8 x bfloat>] [[SRC_COERCE]], 2
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK64-NEXT:    [[VLD3_LANE:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld3lane.v8bf16.p0i8(<8 x bfloat> [[SRC_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[SRC_COERCE_FCA_1_EXTRACT]], <8 x bfloat> [[SRC_COERCE_FCA_2_EXTRACT]], i64 7, i8* [[TMP0]])
+// CHECK64-NEXT:    [[VLD3_LANE:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld3lane.v8bf16.p0(<8 x bfloat> [[SRC_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[SRC_COERCE_FCA_1_EXTRACT]], <8 x bfloat> [[SRC_COERCE_FCA_2_EXTRACT]], i64 7, ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD3_LANE_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD3_LANE]], 0
 // CHECK64-NEXT:    [[VLD3_LANE_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD3_LANE]], 1
 // CHECK64-NEXT:    [[VLD3_LANE_FCA_2_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD3_LANE]], 2
@@ -485,8 +468,7 @@ bfloat16x4x3_t test_vld3_lane_bf16(bfloat16_t const *ptr, bfloat16x4x3_t src) {
 // CHECK32-NEXT:    [[TMP0:%.*]] = bitcast <4 x i32> [[SRC_COERCE_FCA_0_EXTRACT]] to <8 x bfloat>
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <4 x i32> [[SRC_COERCE_FCA_1_EXTRACT]] to <8 x bfloat>
 // CHECK32-NEXT:    [[TMP2:%.*]] = bitcast <4 x i32> [[SRC_COERCE_FCA_2_EXTRACT]] to <8 x bfloat>
-// CHECK32-NEXT:    [[TMP3:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    [[VLD3Q_LANE_V:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld3lane.v8bf16.p0i8(i8* [[TMP3]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], <8 x bfloat> [[TMP2]], i32 7, i32 2)
+// CHECK32-NEXT:    [[VLD3Q_LANE_V:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld3lane.v8bf16.p0(ptr [[PTR:%.*]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], <8 x bfloat> [[TMP2]], i32 7, i32 2)
 // CHECK32-NEXT:    [[VLD3Q_LANE_V_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD3Q_LANE_V]], 0
 // CHECK32-NEXT:    [[VLD3Q_LANE_V_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD3Q_LANE_V]], 1
 // CHECK32-NEXT:    [[VLD3Q_LANE_V_FCA_2_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD3Q_LANE_V]], 2
@@ -505,8 +487,7 @@ bfloat16x8x3_t test_vld3q_lane_bf16(bfloat16_t const *ptr, bfloat16x8x3_t src) {
 
 // CHECK64-LABEL: @test_vld4_bf16(
 // CHECK64-NEXT:  entry:
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to <4 x bfloat>*
-// CHECK64-NEXT:    [[VLD4:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld4.v4bf16.p0v4bf16(<4 x bfloat>* [[TMP0]])
+// CHECK64-NEXT:    [[VLD4:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld4.v4bf16.p0(ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD4_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD4]], 0
 // CHECK64-NEXT:    [[VLD4_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD4]], 1
 // CHECK64-NEXT:    [[VLD4_FCA_2_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD4]], 2
@@ -519,8 +500,7 @@ bfloat16x8x3_t test_vld3q_lane_bf16(bfloat16_t const *ptr, bfloat16x8x3_t src) {
 //
 // CHECK32-LABEL: @test_vld4_bf16(
 // CHECK32-NEXT:  entry:
-// CHECK32-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    [[VLD4_V:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld4.v4bf16.p0i8(i8* [[TMP0]], i32 2)
+// CHECK32-NEXT:    [[VLD4_V:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld4.v4bf16.p0(ptr [[PTR:%.*]], i32 2)
 // CHECK32-NEXT:    [[VLD4_V_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD4_V]], 0
 // CHECK32-NEXT:    [[VLD4_V_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD4_V]], 1
 // CHECK32-NEXT:    [[VLD4_V_FCA_2_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD4_V]], 2
@@ -541,8 +521,7 @@ bfloat16x4x4_t test_vld4_bf16(bfloat16_t const *ptr) {
 
 // CHECK64-LABEL: @test_vld4q_bf16(
 // CHECK64-NEXT:  entry:
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to <8 x bfloat>*
-// CHECK64-NEXT:    [[VLD4:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld4.v8bf16.p0v8bf16(<8 x bfloat>* [[TMP0]])
+// CHECK64-NEXT:    [[VLD4:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld4.v8bf16.p0(ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD4_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD4]], 0
 // CHECK64-NEXT:    [[VLD4_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD4]], 1
 // CHECK64-NEXT:    [[VLD4_FCA_2_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD4]], 2
@@ -555,8 +534,7 @@ bfloat16x4x4_t test_vld4_bf16(bfloat16_t const *ptr) {
 //
 // CHECK32-LABEL: @test_vld4q_bf16(
 // CHECK32-NEXT:  entry:
-// CHECK32-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    [[VLD4Q_V:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld4.v8bf16.p0i8(i8* [[TMP0]], i32 2)
+// CHECK32-NEXT:    [[VLD4Q_V:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld4.v8bf16.p0(ptr [[PTR:%.*]], i32 2)
 // CHECK32-NEXT:    [[VLD4Q_V_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD4Q_V]], 0
 // CHECK32-NEXT:    [[VLD4Q_V_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD4Q_V]], 1
 // CHECK32-NEXT:    [[VLD4Q_V_FCA_2_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD4Q_V]], 2
@@ -581,8 +559,7 @@ bfloat16x8x4_t test_vld4q_bf16(bfloat16_t const *ptr) {
 // CHECK64-NEXT:    [[SRC_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [4 x <4 x bfloat>] [[SRC_COERCE]], 1
 // CHECK64-NEXT:    [[SRC_COERCE_FCA_2_EXTRACT:%.*]] = extractvalue [4 x <4 x bfloat>] [[SRC_COERCE]], 2
 // CHECK64-NEXT:    [[SRC_COERCE_FCA_3_EXTRACT:%.*]] = extractvalue [4 x <4 x bfloat>] [[SRC_COERCE]], 3
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK64-NEXT:    [[VLD4_LANE:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld4lane.v4bf16.p0i8(<4 x bfloat> [[SRC_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[SRC_COERCE_FCA_1_EXTRACT]], <4 x bfloat> [[SRC_COERCE_FCA_2_EXTRACT]], <4 x bfloat> [[SRC_COERCE_FCA_3_EXTRACT]], i64 1, i8* [[TMP0]])
+// CHECK64-NEXT:    [[VLD4_LANE:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld4lane.v4bf16.p0(<4 x bfloat> [[SRC_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[SRC_COERCE_FCA_1_EXTRACT]], <4 x bfloat> [[SRC_COERCE_FCA_2_EXTRACT]], <4 x bfloat> [[SRC_COERCE_FCA_3_EXTRACT]], i64 1, ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD4_LANE_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD4_LANE]], 0
 // CHECK64-NEXT:    [[VLD4_LANE_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD4_LANE]], 1
 // CHECK64-NEXT:    [[VLD4_LANE_FCA_2_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD4_LANE]], 2
@@ -603,8 +580,7 @@ bfloat16x8x4_t test_vld4q_bf16(bfloat16_t const *ptr) {
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <2 x i32> [[SRC_COERCE_FCA_1_EXTRACT]] to <4 x bfloat>
 // CHECK32-NEXT:    [[TMP2:%.*]] = bitcast <2 x i32> [[SRC_COERCE_FCA_2_EXTRACT]] to <4 x bfloat>
 // CHECK32-NEXT:    [[TMP3:%.*]] = bitcast <2 x i32> [[SRC_COERCE_FCA_3_EXTRACT]] to <4 x bfloat>
-// CHECK32-NEXT:    [[TMP4:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    [[VLD4_LANE_V:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld4lane.v4bf16.p0i8(i8* [[TMP4]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], <4 x bfloat> [[TMP2]], <4 x bfloat> [[TMP3]], i32 1, i32 2)
+// CHECK32-NEXT:    [[VLD4_LANE_V:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld4lane.v4bf16.p0(ptr [[PTR:%.*]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], <4 x bfloat> [[TMP2]], <4 x bfloat> [[TMP3]], i32 1, i32 2)
 // CHECK32-NEXT:    [[VLD4_LANE_V_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD4_LANE_V]], 0
 // CHECK32-NEXT:    [[VLD4_LANE_V_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD4_LANE_V]], 1
 // CHECK32-NEXT:    [[VLD4_LANE_V_FCA_2_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD4_LANE_V]], 2
@@ -629,8 +605,7 @@ bfloat16x4x4_t test_vld4_lane_bf16(bfloat16_t const *ptr, bfloat16x4x4_t src) {
 // CHECK64-NEXT:    [[SRC_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [4 x <8 x bfloat>] [[SRC_COERCE]], 1
 // CHECK64-NEXT:    [[SRC_COERCE_FCA_2_EXTRACT:%.*]] = extractvalue [4 x <8 x bfloat>] [[SRC_COERCE]], 2
 // CHECK64-NEXT:    [[SRC_COERCE_FCA_3_EXTRACT:%.*]] = extractvalue [4 x <8 x bfloat>] [[SRC_COERCE]], 3
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK64-NEXT:    [[VLD4_LANE:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld4lane.v8bf16.p0i8(<8 x bfloat> [[SRC_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[SRC_COERCE_FCA_1_EXTRACT]], <8 x bfloat> [[SRC_COERCE_FCA_2_EXTRACT]], <8 x bfloat> [[SRC_COERCE_FCA_3_EXTRACT]], i64 7, i8* [[TMP0]])
+// CHECK64-NEXT:    [[VLD4_LANE:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld4lane.v8bf16.p0(<8 x bfloat> [[SRC_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[SRC_COERCE_FCA_1_EXTRACT]], <8 x bfloat> [[SRC_COERCE_FCA_2_EXTRACT]], <8 x bfloat> [[SRC_COERCE_FCA_3_EXTRACT]], i64 7, ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD4_LANE_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD4_LANE]], 0
 // CHECK64-NEXT:    [[VLD4_LANE_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD4_LANE]], 1
 // CHECK64-NEXT:    [[VLD4_LANE_FCA_2_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD4_LANE]], 2
@@ -651,8 +626,7 @@ bfloat16x4x4_t test_vld4_lane_bf16(bfloat16_t const *ptr, bfloat16x4x4_t src) {
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <4 x i32> [[SRC_COERCE_FCA_1_EXTRACT]] to <8 x bfloat>
 // CHECK32-NEXT:    [[TMP2:%.*]] = bitcast <4 x i32> [[SRC_COERCE_FCA_2_EXTRACT]] to <8 x bfloat>
 // CHECK32-NEXT:    [[TMP3:%.*]] = bitcast <4 x i32> [[SRC_COERCE_FCA_3_EXTRACT]] to <8 x bfloat>
-// CHECK32-NEXT:    [[TMP4:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    [[VLD4Q_LANE_V:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld4lane.v8bf16.p0i8(i8* [[TMP4]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], <8 x bfloat> [[TMP2]], <8 x bfloat> [[TMP3]], i32 7, i32 2)
+// CHECK32-NEXT:    [[VLD4Q_LANE_V:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld4lane.v8bf16.p0(ptr [[PTR:%.*]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], <8 x bfloat> [[TMP2]], <8 x bfloat> [[TMP3]], i32 7, i32 2)
 // CHECK32-NEXT:    [[VLD4Q_LANE_V_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD4Q_LANE_V]], 0
 // CHECK32-NEXT:    [[VLD4Q_LANE_V_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD4Q_LANE_V]], 1
 // CHECK32-NEXT:    [[VLD4Q_LANE_V_FCA_2_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD4Q_LANE_V]], 2
@@ -673,7 +647,7 @@ bfloat16x8x4_t test_vld4q_lane_bf16(bfloat16_t const *ptr, bfloat16x8x4_t src) {
 
 // CHECK64-LABEL: @test_vld2_dup_bf16(
 // CHECK64-NEXT:  entry:
-// CHECK64-NEXT:    [[VLD2:%.*]] = tail call { <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld2r.v4bf16.p0bf16(bfloat* [[PTR:%.*]])
+// CHECK64-NEXT:    [[VLD2:%.*]] = tail call { <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld2r.v4bf16.p0(ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD2_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat> } [[VLD2]], 0
 // CHECK64-NEXT:    [[VLD2_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat> } [[VLD2]], 1
 // CHECK64-NEXT:    [[DOTFCA_0_0_INSERT:%.*]] = insertvalue [[STRUCT_BFLOAT16X4X2_T:%.*]] poison, <4 x bfloat> [[VLD2_FCA_0_EXTRACT]], 0, 0
@@ -682,8 +656,7 @@ bfloat16x8x4_t test_vld4q_lane_bf16(bfloat16_t const *ptr, bfloat16x8x4_t src) {
 //
 // CHECK32-LABEL: @test_vld2_dup_bf16(
 // CHECK32-NEXT:  entry:
-// CHECK32-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    [[VLD2_DUP_V:%.*]] = tail call { <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld2dup.v4bf16.p0i8(i8* [[TMP0]], i32 2)
+// CHECK32-NEXT:    [[VLD2_DUP_V:%.*]] = tail call { <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld2dup.v4bf16.p0(ptr [[PTR:%.*]], i32 2)
 // CHECK32-NEXT:    [[VLD2_DUP_V_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat> } [[VLD2_DUP_V]], 0
 // CHECK32-NEXT:    [[VLD2_DUP_V_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat> } [[VLD2_DUP_V]], 1
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <4 x bfloat> [[VLD2_DUP_V_FCA_0_EXTRACT]] to <2 x i32>
@@ -698,7 +671,7 @@ bfloat16x4x2_t test_vld2_dup_bf16(bfloat16_t const *ptr) {
 
 // CHECK64-LABEL: @test_vld2q_dup_bf16(
 // CHECK64-NEXT:  entry:
-// CHECK64-NEXT:    [[VLD2:%.*]] = tail call { <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld2r.v8bf16.p0bf16(bfloat* [[PTR:%.*]])
+// CHECK64-NEXT:    [[VLD2:%.*]] = tail call { <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld2r.v8bf16.p0(ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD2_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat> } [[VLD2]], 0
 // CHECK64-NEXT:    [[VLD2_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat> } [[VLD2]], 1
 // CHECK64-NEXT:    [[DOTFCA_0_0_INSERT:%.*]] = insertvalue [[STRUCT_BFLOAT16X8X2_T:%.*]] poison, <8 x bfloat> [[VLD2_FCA_0_EXTRACT]], 0, 0
@@ -707,8 +680,7 @@ bfloat16x4x2_t test_vld2_dup_bf16(bfloat16_t const *ptr) {
 //
 // CHECK32-LABEL: @test_vld2q_dup_bf16(
 // CHECK32-NEXT:  entry:
-// CHECK32-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    [[VLD2Q_DUP_V:%.*]] = tail call { <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld2dup.v8bf16.p0i8(i8* [[TMP0]], i32 2)
+// CHECK32-NEXT:    [[VLD2Q_DUP_V:%.*]] = tail call { <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld2dup.v8bf16.p0(ptr [[PTR:%.*]], i32 2)
 // CHECK32-NEXT:    [[VLD2Q_DUP_V_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat> } [[VLD2Q_DUP_V]], 0
 // CHECK32-NEXT:    [[VLD2Q_DUP_V_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat> } [[VLD2Q_DUP_V]], 1
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <8 x bfloat> [[VLD2Q_DUP_V_FCA_0_EXTRACT]] to <4 x i32>
@@ -723,7 +695,7 @@ bfloat16x8x2_t test_vld2q_dup_bf16(bfloat16_t const *ptr) {
 
 // CHECK64-LABEL: @test_vld3_dup_bf16(
 // CHECK64-NEXT:  entry:
-// CHECK64-NEXT:    [[VLD3:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld3r.v4bf16.p0bf16(bfloat* [[PTR:%.*]])
+// CHECK64-NEXT:    [[VLD3:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld3r.v4bf16.p0(ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD3_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD3]], 0
 // CHECK64-NEXT:    [[VLD3_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD3]], 1
 // CHECK64-NEXT:    [[VLD3_FCA_2_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD3]], 2
@@ -734,8 +706,7 @@ bfloat16x8x2_t test_vld2q_dup_bf16(bfloat16_t const *ptr) {
 //
 // CHECK32-LABEL: @test_vld3_dup_bf16(
 // CHECK32-NEXT:  entry:
-// CHECK32-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    [[VLD3_DUP_V:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld3dup.v4bf16.p0i8(i8* [[TMP0]], i32 2)
+// CHECK32-NEXT:    [[VLD3_DUP_V:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld3dup.v4bf16.p0(ptr [[PTR:%.*]], i32 2)
 // CHECK32-NEXT:    [[VLD3_DUP_V_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD3_DUP_V]], 0
 // CHECK32-NEXT:    [[VLD3_DUP_V_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD3_DUP_V]], 1
 // CHECK32-NEXT:    [[VLD3_DUP_V_FCA_2_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD3_DUP_V]], 2
@@ -753,7 +724,7 @@ bfloat16x4x3_t test_vld3_dup_bf16(bfloat16_t const *ptr) {
 
 // CHECK64-LABEL: @test_vld3q_dup_bf16(
 // CHECK64-NEXT:  entry:
-// CHECK64-NEXT:    [[VLD3:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld3r.v8bf16.p0bf16(bfloat* [[PTR:%.*]])
+// CHECK64-NEXT:    [[VLD3:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld3r.v8bf16.p0(ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD3_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD3]], 0
 // CHECK64-NEXT:    [[VLD3_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD3]], 1
 // CHECK64-NEXT:    [[VLD3_FCA_2_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD3]], 2
@@ -764,8 +735,7 @@ bfloat16x4x3_t test_vld3_dup_bf16(bfloat16_t const *ptr) {
 //
 // CHECK32-LABEL: @test_vld3q_dup_bf16(
 // CHECK32-NEXT:  entry:
-// CHECK32-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    [[VLD3Q_DUP_V:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld3dup.v8bf16.p0i8(i8* [[TMP0]], i32 2)
+// CHECK32-NEXT:    [[VLD3Q_DUP_V:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld3dup.v8bf16.p0(ptr [[PTR:%.*]], i32 2)
 // CHECK32-NEXT:    [[VLD3Q_DUP_V_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD3Q_DUP_V]], 0
 // CHECK32-NEXT:    [[VLD3Q_DUP_V_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD3Q_DUP_V]], 1
 // CHECK32-NEXT:    [[VLD3Q_DUP_V_FCA_2_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD3Q_DUP_V]], 2
@@ -783,7 +753,7 @@ bfloat16x8x3_t test_vld3q_dup_bf16(bfloat16_t const *ptr) {
 
 // CHECK64-LABEL: @test_vld4_dup_bf16(
 // CHECK64-NEXT:  entry:
-// CHECK64-NEXT:    [[VLD4:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld4r.v4bf16.p0bf16(bfloat* [[PTR:%.*]])
+// CHECK64-NEXT:    [[VLD4:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.aarch64.neon.ld4r.v4bf16.p0(ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD4_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD4]], 0
 // CHECK64-NEXT:    [[VLD4_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD4]], 1
 // CHECK64-NEXT:    [[VLD4_FCA_2_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD4]], 2
@@ -796,8 +766,7 @@ bfloat16x8x3_t test_vld3q_dup_bf16(bfloat16_t const *ptr) {
 //
 // CHECK32-LABEL: @test_vld4_dup_bf16(
 // CHECK32-NEXT:  entry:
-// CHECK32-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    [[VLD4_DUP_V:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld4dup.v4bf16.p0i8(i8* [[TMP0]], i32 2)
+// CHECK32-NEXT:    [[VLD4_DUP_V:%.*]] = tail call { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } @llvm.arm.neon.vld4dup.v4bf16.p0(ptr [[PTR:%.*]], i32 2)
 // CHECK32-NEXT:    [[VLD4_DUP_V_FCA_0_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD4_DUP_V]], 0
 // CHECK32-NEXT:    [[VLD4_DUP_V_FCA_1_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD4_DUP_V]], 1
 // CHECK32-NEXT:    [[VLD4_DUP_V_FCA_2_EXTRACT:%.*]] = extractvalue { <4 x bfloat>, <4 x bfloat>, <4 x bfloat>, <4 x bfloat> } [[VLD4_DUP_V]], 2
@@ -818,7 +787,7 @@ bfloat16x4x4_t test_vld4_dup_bf16(bfloat16_t const *ptr) {
 
 // CHECK64-LABEL: @test_vld4q_dup_bf16(
 // CHECK64-NEXT:  entry:
-// CHECK64-NEXT:    [[VLD4:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld4r.v8bf16.p0bf16(bfloat* [[PTR:%.*]])
+// CHECK64-NEXT:    [[VLD4:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.aarch64.neon.ld4r.v8bf16.p0(ptr [[PTR:%.*]])
 // CHECK64-NEXT:    [[VLD4_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD4]], 0
 // CHECK64-NEXT:    [[VLD4_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD4]], 1
 // CHECK64-NEXT:    [[VLD4_FCA_2_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD4]], 2
@@ -831,8 +800,7 @@ bfloat16x4x4_t test_vld4_dup_bf16(bfloat16_t const *ptr) {
 //
 // CHECK32-LABEL: @test_vld4q_dup_bf16(
 // CHECK32-NEXT:  entry:
-// CHECK32-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    [[VLD4Q_DUP_V:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld4dup.v8bf16.p0i8(i8* [[TMP0]], i32 2)
+// CHECK32-NEXT:    [[VLD4Q_DUP_V:%.*]] = tail call { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } @llvm.arm.neon.vld4dup.v8bf16.p0(ptr [[PTR:%.*]], i32 2)
 // CHECK32-NEXT:    [[VLD4Q_DUP_V_FCA_0_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD4Q_DUP_V]], 0
 // CHECK32-NEXT:    [[VLD4Q_DUP_V_FCA_1_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD4Q_DUP_V]], 1
 // CHECK32-NEXT:    [[VLD4Q_DUP_V_FCA_2_EXTRACT:%.*]] = extractvalue { <8 x bfloat>, <8 x bfloat>, <8 x bfloat>, <8 x bfloat> } [[VLD4Q_DUP_V]], 2
@@ -853,14 +821,12 @@ bfloat16x8x4_t test_vld4q_dup_bf16(bfloat16_t const *ptr) {
 
 // CHECK64-LABEL: @test_vst1_bf16(
 // CHECK64-NEXT:  entry:
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to <4 x bfloat>*
-// CHECK64-NEXT:    store <4 x bfloat> [[VAL:%.*]], <4 x bfloat>* [[TMP0]], align 2
+// CHECK64-NEXT:    store <4 x bfloat> [[VAL:%.*]], ptr [[PTR:%.*]], align 2
 // CHECK64-NEXT:    ret void
 //
 // CHECK32-LABEL: @test_vst1_bf16(
 // CHECK32-NEXT:  entry:
-// CHECK32-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst1.p0i8.v4bf16(i8* [[TMP0]], <4 x bfloat> [[VAL:%.*]], i32 2)
+// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst1.p0.v4bf16(ptr [[PTR:%.*]], <4 x bfloat> [[VAL:%.*]], i32 2)
 // CHECK32-NEXT:    ret void
 //
 void test_vst1_bf16(bfloat16_t *ptr, bfloat16x4_t val) {
@@ -869,14 +835,12 @@ void test_vst1_bf16(bfloat16_t *ptr, bfloat16x4_t val) {
 
 // CHECK64-LABEL: @test_vst1q_bf16(
 // CHECK64-NEXT:  entry:
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to <8 x bfloat>*
-// CHECK64-NEXT:    store <8 x bfloat> [[VAL:%.*]], <8 x bfloat>* [[TMP0]], align 2
+// CHECK64-NEXT:    store <8 x bfloat> [[VAL:%.*]], ptr [[PTR:%.*]], align 2
 // CHECK64-NEXT:    ret void
 //
 // CHECK32-LABEL: @test_vst1q_bf16(
 // CHECK32-NEXT:  entry:
-// CHECK32-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst1.p0i8.v8bf16(i8* [[TMP0]], <8 x bfloat> [[VAL:%.*]], i32 2)
+// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst1.p0.v8bf16(ptr [[PTR:%.*]], <8 x bfloat> [[VAL:%.*]], i32 2)
 // CHECK32-NEXT:    ret void
 //
 void test_vst1q_bf16(bfloat16_t *ptr, bfloat16x8_t val) {
@@ -886,7 +850,7 @@ void test_vst1q_bf16(bfloat16_t *ptr, bfloat16x8_t val) {
 // CHECK-LABEL: @test_vst1_lane_bf16(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = extractelement <4 x bfloat> [[VAL:%.*]], i64 1
-// CHECK-NEXT:    store bfloat [[TMP0]], bfloat* [[PTR:%.*]], align 2
+// CHECK-NEXT:    store bfloat [[TMP0]], ptr [[PTR:%.*]], align 2
 // CHECK-NEXT:    ret void
 //
 void test_vst1_lane_bf16(bfloat16_t *ptr, bfloat16x4_t val) {
@@ -896,7 +860,7 @@ void test_vst1_lane_bf16(bfloat16_t *ptr, bfloat16x4_t val) {
 // CHECK-LABEL: @test_vst1q_lane_bf16(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = extractelement <8 x bfloat> [[VAL:%.*]], i64 7
-// CHECK-NEXT:    store bfloat [[TMP0]], bfloat* [[PTR:%.*]], align 2
+// CHECK-NEXT:    store bfloat [[TMP0]], ptr [[PTR:%.*]], align 2
 // CHECK-NEXT:    ret void
 //
 void test_vst1q_lane_bf16(bfloat16_t *ptr, bfloat16x8_t val) {
@@ -907,7 +871,7 @@ void test_vst1q_lane_bf16(bfloat16_t *ptr, bfloat16x8_t val) {
 // CHECK64-NEXT:  entry:
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_0_EXTRACT:%.*]] = extractvalue [2 x <4 x bfloat>] [[VAL_COERCE:%.*]], 0
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [2 x <4 x bfloat>] [[VAL_COERCE]], 1
-// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st1x2.v4bf16.p0bf16(<4 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], bfloat* [[PTR:%.*]])
+// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st1x2.v4bf16.p0(<4 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], ptr [[PTR:%.*]])
 // CHECK64-NEXT:    ret void
 //
 // CHECK32-LABEL: @test_vst1_bf16_x2(
@@ -916,7 +880,7 @@ void test_vst1q_lane_bf16(bfloat16_t *ptr, bfloat16x8_t val) {
 // CHECK32-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [2 x <2 x i32>] [[VAL_COERCE]], 1
 // CHECK32-NEXT:    [[TMP0:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_0_EXTRACT]] to <4 x bfloat>
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_1_EXTRACT]] to <4 x bfloat>
-// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst1x2.p0bf16.v4bf16(bfloat* [[PTR:%.*]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]])
+// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst1x2.p0.v4bf16(ptr [[PTR:%.*]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]])
 // CHECK32-NEXT:    ret void
 //
 void test_vst1_bf16_x2(bfloat16_t *ptr, bfloat16x4x2_t val) {
@@ -927,7 +891,7 @@ void test_vst1_bf16_x2(bfloat16_t *ptr, bfloat16x4x2_t val) {
 // CHECK64-NEXT:  entry:
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_0_EXTRACT:%.*]] = extractvalue [2 x <8 x bfloat>] [[VAL_COERCE:%.*]], 0
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [2 x <8 x bfloat>] [[VAL_COERCE]], 1
-// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st1x2.v8bf16.p0bf16(<8 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], bfloat* [[PTR:%.*]])
+// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st1x2.v8bf16.p0(<8 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], ptr [[PTR:%.*]])
 // CHECK64-NEXT:    ret void
 //
 // CHECK32-LABEL: @test_vst1q_bf16_x2(
@@ -936,7 +900,7 @@ void test_vst1_bf16_x2(bfloat16_t *ptr, bfloat16x4x2_t val) {
 // CHECK32-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [2 x <4 x i32>] [[VAL_COERCE]], 1
 // CHECK32-NEXT:    [[TMP0:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_0_EXTRACT]] to <8 x bfloat>
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_1_EXTRACT]] to <8 x bfloat>
-// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst1x2.p0bf16.v8bf16(bfloat* [[PTR:%.*]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]])
+// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst1x2.p0.v8bf16(ptr [[PTR:%.*]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]])
 // CHECK32-NEXT:    ret void
 //
 void test_vst1q_bf16_x2(bfloat16_t *ptr, bfloat16x8x2_t val) {
@@ -948,7 +912,7 @@ void test_vst1q_bf16_x2(bfloat16_t *ptr, bfloat16x8x2_t val) {
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_0_EXTRACT:%.*]] = extractvalue [3 x <4 x bfloat>] [[VAL_COERCE:%.*]], 0
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [3 x <4 x bfloat>] [[VAL_COERCE]], 1
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_2_EXTRACT:%.*]] = extractvalue [3 x <4 x bfloat>] [[VAL_COERCE]], 2
-// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st1x3.v4bf16.p0bf16(<4 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], bfloat* [[PTR:%.*]])
+// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st1x3.v4bf16.p0(<4 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], ptr [[PTR:%.*]])
 // CHECK64-NEXT:    ret void
 //
 // CHECK32-LABEL: @test_vst1_bf16_x3(
@@ -959,7 +923,7 @@ void test_vst1q_bf16_x2(bfloat16_t *ptr, bfloat16x8x2_t val) {
 // CHECK32-NEXT:    [[TMP0:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_0_EXTRACT]] to <4 x bfloat>
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_1_EXTRACT]] to <4 x bfloat>
 // CHECK32-NEXT:    [[TMP2:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_2_EXTRACT]] to <4 x bfloat>
-// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst1x3.p0bf16.v4bf16(bfloat* [[PTR:%.*]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], <4 x bfloat> [[TMP2]])
+// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst1x3.p0.v4bf16(ptr [[PTR:%.*]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], <4 x bfloat> [[TMP2]])
 // CHECK32-NEXT:    ret void
 //
 void test_vst1_bf16_x3(bfloat16_t *ptr, bfloat16x4x3_t val) {
@@ -971,7 +935,7 @@ void test_vst1_bf16_x3(bfloat16_t *ptr, bfloat16x4x3_t val) {
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_0_EXTRACT:%.*]] = extractvalue [3 x <8 x bfloat>] [[VAL_COERCE:%.*]], 0
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [3 x <8 x bfloat>] [[VAL_COERCE]], 1
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_2_EXTRACT:%.*]] = extractvalue [3 x <8 x bfloat>] [[VAL_COERCE]], 2
-// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st1x3.v8bf16.p0bf16(<8 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], bfloat* [[PTR:%.*]])
+// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st1x3.v8bf16.p0(<8 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], ptr [[PTR:%.*]])
 // CHECK64-NEXT:    ret void
 //
 // CHECK32-LABEL: @test_vst1q_bf16_x3(
@@ -982,7 +946,7 @@ void test_vst1_bf16_x3(bfloat16_t *ptr, bfloat16x4x3_t val) {
 // CHECK32-NEXT:    [[TMP0:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_0_EXTRACT]] to <8 x bfloat>
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_1_EXTRACT]] to <8 x bfloat>
 // CHECK32-NEXT:    [[TMP2:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_2_EXTRACT]] to <8 x bfloat>
-// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst1x3.p0bf16.v8bf16(bfloat* [[PTR:%.*]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], <8 x bfloat> [[TMP2]])
+// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst1x3.p0.v8bf16(ptr [[PTR:%.*]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], <8 x bfloat> [[TMP2]])
 // CHECK32-NEXT:    ret void
 //
 void test_vst1q_bf16_x3(bfloat16_t *ptr, bfloat16x8x3_t val) {
@@ -995,7 +959,7 @@ void test_vst1q_bf16_x3(bfloat16_t *ptr, bfloat16x8x3_t val) {
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [4 x <4 x bfloat>] [[VAL_COERCE]], 1
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_2_EXTRACT:%.*]] = extractvalue [4 x <4 x bfloat>] [[VAL_COERCE]], 2
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_3_EXTRACT:%.*]] = extractvalue [4 x <4 x bfloat>] [[VAL_COERCE]], 3
-// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st1x4.v4bf16.p0bf16(<4 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_3_EXTRACT]], bfloat* [[PTR:%.*]])
+// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st1x4.v4bf16.p0(<4 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_3_EXTRACT]], ptr [[PTR:%.*]])
 // CHECK64-NEXT:    ret void
 //
 // CHECK32-LABEL: @test_vst1_bf16_x4(
@@ -1008,7 +972,7 @@ void test_vst1q_bf16_x3(bfloat16_t *ptr, bfloat16x8x3_t val) {
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_1_EXTRACT]] to <4 x bfloat>
 // CHECK32-NEXT:    [[TMP2:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_2_EXTRACT]] to <4 x bfloat>
 // CHECK32-NEXT:    [[TMP3:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_3_EXTRACT]] to <4 x bfloat>
-// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst1x4.p0bf16.v4bf16(bfloat* [[PTR:%.*]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], <4 x bfloat> [[TMP2]], <4 x bfloat> [[TMP3]])
+// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst1x4.p0.v4bf16(ptr [[PTR:%.*]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], <4 x bfloat> [[TMP2]], <4 x bfloat> [[TMP3]])
 // CHECK32-NEXT:    ret void
 //
 void test_vst1_bf16_x4(bfloat16_t *ptr, bfloat16x4x4_t val) {
@@ -1021,7 +985,7 @@ void test_vst1_bf16_x4(bfloat16_t *ptr, bfloat16x4x4_t val) {
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [4 x <8 x bfloat>] [[VAL_COERCE]], 1
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_2_EXTRACT:%.*]] = extractvalue [4 x <8 x bfloat>] [[VAL_COERCE]], 2
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_3_EXTRACT:%.*]] = extractvalue [4 x <8 x bfloat>] [[VAL_COERCE]], 3
-// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st1x4.v8bf16.p0bf16(<8 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_3_EXTRACT]], bfloat* [[PTR:%.*]])
+// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st1x4.v8bf16.p0(<8 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_3_EXTRACT]], ptr [[PTR:%.*]])
 // CHECK64-NEXT:    ret void
 //
 // CHECK32-LABEL: @test_vst1q_bf16_x4(
@@ -1034,7 +998,7 @@ void test_vst1_bf16_x4(bfloat16_t *ptr, bfloat16x4x4_t val) {
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_1_EXTRACT]] to <8 x bfloat>
 // CHECK32-NEXT:    [[TMP2:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_2_EXTRACT]] to <8 x bfloat>
 // CHECK32-NEXT:    [[TMP3:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_3_EXTRACT]] to <8 x bfloat>
-// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst1x4.p0bf16.v8bf16(bfloat* [[PTR:%.*]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], <8 x bfloat> [[TMP2]], <8 x bfloat> [[TMP3]])
+// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst1x4.p0.v8bf16(ptr [[PTR:%.*]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], <8 x bfloat> [[TMP2]], <8 x bfloat> [[TMP3]])
 // CHECK32-NEXT:    ret void
 //
 void test_vst1q_bf16_x4(bfloat16_t *ptr, bfloat16x8x4_t val) {
@@ -1045,8 +1009,7 @@ void test_vst1q_bf16_x4(bfloat16_t *ptr, bfloat16x8x4_t val) {
 // CHECK64-NEXT:  entry:
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_0_EXTRACT:%.*]] = extractvalue [2 x <4 x bfloat>] [[VAL_COERCE:%.*]], 0
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [2 x <4 x bfloat>] [[VAL_COERCE]], 1
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st2.v4bf16.p0i8(<4 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], i8* [[TMP0]])
+// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st2.v4bf16.p0(<4 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], ptr [[PTR:%.*]])
 // CHECK64-NEXT:    ret void
 //
 // CHECK32-LABEL: @test_vst2_bf16(
@@ -1055,8 +1018,7 @@ void test_vst1q_bf16_x4(bfloat16_t *ptr, bfloat16x8x4_t val) {
 // CHECK32-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [2 x <2 x i32>] [[VAL_COERCE]], 1
 // CHECK32-NEXT:    [[TMP0:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_0_EXTRACT]] to <4 x bfloat>
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_1_EXTRACT]] to <4 x bfloat>
-// CHECK32-NEXT:    [[TMP2:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst2.p0i8.v4bf16(i8* [[TMP2]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], i32 2)
+// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst2.p0.v4bf16(ptr [[PTR:%.*]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], i32 2)
 // CHECK32-NEXT:    ret void
 //
 void test_vst2_bf16(bfloat16_t *ptr, bfloat16x4x2_t val) {
@@ -1067,8 +1029,7 @@ void test_vst2_bf16(bfloat16_t *ptr, bfloat16x4x2_t val) {
 // CHECK64-NEXT:  entry:
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_0_EXTRACT:%.*]] = extractvalue [2 x <8 x bfloat>] [[VAL_COERCE:%.*]], 0
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [2 x <8 x bfloat>] [[VAL_COERCE]], 1
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st2.v8bf16.p0i8(<8 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], i8* [[TMP0]])
+// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st2.v8bf16.p0(<8 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], ptr [[PTR:%.*]])
 // CHECK64-NEXT:    ret void
 //
 // CHECK32-LABEL: @test_vst2q_bf16(
@@ -1077,8 +1038,7 @@ void test_vst2_bf16(bfloat16_t *ptr, bfloat16x4x2_t val) {
 // CHECK32-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [2 x <4 x i32>] [[VAL_COERCE]], 1
 // CHECK32-NEXT:    [[TMP0:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_0_EXTRACT]] to <8 x bfloat>
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_1_EXTRACT]] to <8 x bfloat>
-// CHECK32-NEXT:    [[TMP2:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst2.p0i8.v8bf16(i8* [[TMP2]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], i32 2)
+// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst2.p0.v8bf16(ptr [[PTR:%.*]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], i32 2)
 // CHECK32-NEXT:    ret void
 //
 void test_vst2q_bf16(bfloat16_t *ptr, bfloat16x8x2_t val) {
@@ -1089,8 +1049,7 @@ void test_vst2q_bf16(bfloat16_t *ptr, bfloat16x8x2_t val) {
 // CHECK64-NEXT:  entry:
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_0_EXTRACT:%.*]] = extractvalue [2 x <4 x bfloat>] [[VAL_COERCE:%.*]], 0
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [2 x <4 x bfloat>] [[VAL_COERCE]], 1
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st2lane.v4bf16.p0i8(<4 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], i64 1, i8* [[TMP0]])
+// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st2lane.v4bf16.p0(<4 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], i64 1, ptr [[PTR:%.*]])
 // CHECK64-NEXT:    ret void
 //
 // CHECK32-LABEL: @test_vst2_lane_bf16(
@@ -1099,8 +1058,7 @@ void test_vst2q_bf16(bfloat16_t *ptr, bfloat16x8x2_t val) {
 // CHECK32-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [2 x <2 x i32>] [[VAL_COERCE]], 1
 // CHECK32-NEXT:    [[TMP0:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_0_EXTRACT]] to <4 x bfloat>
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_1_EXTRACT]] to <4 x bfloat>
-// CHECK32-NEXT:    [[TMP2:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst2lane.p0i8.v4bf16(i8* [[TMP2]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], i32 1, i32 2)
+// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst2lane.p0.v4bf16(ptr [[PTR:%.*]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], i32 1, i32 2)
 // CHECK32-NEXT:    ret void
 //
 void test_vst2_lane_bf16(bfloat16_t *ptr, bfloat16x4x2_t val) {
@@ -1111,8 +1069,7 @@ void test_vst2_lane_bf16(bfloat16_t *ptr, bfloat16x4x2_t val) {
 // CHECK64-NEXT:  entry:
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_0_EXTRACT:%.*]] = extractvalue [2 x <8 x bfloat>] [[VAL_COERCE:%.*]], 0
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [2 x <8 x bfloat>] [[VAL_COERCE]], 1
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st2lane.v8bf16.p0i8(<8 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], i64 7, i8* [[TMP0]])
+// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st2lane.v8bf16.p0(<8 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], i64 7, ptr [[PTR:%.*]])
 // CHECK64-NEXT:    ret void
 //
 // CHECK32-LABEL: @test_vst2q_lane_bf16(
@@ -1121,8 +1078,7 @@ void test_vst2_lane_bf16(bfloat16_t *ptr, bfloat16x4x2_t val) {
 // CHECK32-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [2 x <4 x i32>] [[VAL_COERCE]], 1
 // CHECK32-NEXT:    [[TMP0:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_0_EXTRACT]] to <8 x bfloat>
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_1_EXTRACT]] to <8 x bfloat>
-// CHECK32-NEXT:    [[TMP2:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst2lane.p0i8.v8bf16(i8* [[TMP2]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], i32 7, i32 2)
+// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst2lane.p0.v8bf16(ptr [[PTR:%.*]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], i32 7, i32 2)
 // CHECK32-NEXT:    ret void
 //
 void test_vst2q_lane_bf16(bfloat16_t *ptr, bfloat16x8x2_t val) {
@@ -1134,8 +1090,7 @@ void test_vst2q_lane_bf16(bfloat16_t *ptr, bfloat16x8x2_t val) {
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_0_EXTRACT:%.*]] = extractvalue [3 x <4 x bfloat>] [[VAL_COERCE:%.*]], 0
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [3 x <4 x bfloat>] [[VAL_COERCE]], 1
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_2_EXTRACT:%.*]] = extractvalue [3 x <4 x bfloat>] [[VAL_COERCE]], 2
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st3.v4bf16.p0i8(<4 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], i8* [[TMP0]])
+// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st3.v4bf16.p0(<4 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], ptr [[PTR:%.*]])
 // CHECK64-NEXT:    ret void
 //
 // CHECK32-LABEL: @test_vst3_bf16(
@@ -1146,8 +1101,7 @@ void test_vst2q_lane_bf16(bfloat16_t *ptr, bfloat16x8x2_t val) {
 // CHECK32-NEXT:    [[TMP0:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_0_EXTRACT]] to <4 x bfloat>
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_1_EXTRACT]] to <4 x bfloat>
 // CHECK32-NEXT:    [[TMP2:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_2_EXTRACT]] to <4 x bfloat>
-// CHECK32-NEXT:    [[TMP3:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst3.p0i8.v4bf16(i8* [[TMP3]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], <4 x bfloat> [[TMP2]], i32 2)
+// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst3.p0.v4bf16(ptr [[PTR:%.*]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], <4 x bfloat> [[TMP2]], i32 2)
 // CHECK32-NEXT:    ret void
 //
 void test_vst3_bf16(bfloat16_t *ptr, bfloat16x4x3_t val) {
@@ -1159,8 +1113,7 @@ void test_vst3_bf16(bfloat16_t *ptr, bfloat16x4x3_t val) {
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_0_EXTRACT:%.*]] = extractvalue [3 x <8 x bfloat>] [[VAL_COERCE:%.*]], 0
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [3 x <8 x bfloat>] [[VAL_COERCE]], 1
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_2_EXTRACT:%.*]] = extractvalue [3 x <8 x bfloat>] [[VAL_COERCE]], 2
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st3.v8bf16.p0i8(<8 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], i8* [[TMP0]])
+// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st3.v8bf16.p0(<8 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], ptr [[PTR:%.*]])
 // CHECK64-NEXT:    ret void
 //
 // CHECK32-LABEL: @test_vst3q_bf16(
@@ -1171,8 +1124,7 @@ void test_vst3_bf16(bfloat16_t *ptr, bfloat16x4x3_t val) {
 // CHECK32-NEXT:    [[TMP0:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_0_EXTRACT]] to <8 x bfloat>
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_1_EXTRACT]] to <8 x bfloat>
 // CHECK32-NEXT:    [[TMP2:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_2_EXTRACT]] to <8 x bfloat>
-// CHECK32-NEXT:    [[TMP3:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst3.p0i8.v8bf16(i8* [[TMP3]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], <8 x bfloat> [[TMP2]], i32 2)
+// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst3.p0.v8bf16(ptr [[PTR:%.*]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], <8 x bfloat> [[TMP2]], i32 2)
 // CHECK32-NEXT:    ret void
 //
 void test_vst3q_bf16(bfloat16_t *ptr, bfloat16x8x3_t val) {
@@ -1184,8 +1136,7 @@ void test_vst3q_bf16(bfloat16_t *ptr, bfloat16x8x3_t val) {
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_0_EXTRACT:%.*]] = extractvalue [3 x <4 x bfloat>] [[VAL_COERCE:%.*]], 0
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [3 x <4 x bfloat>] [[VAL_COERCE]], 1
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_2_EXTRACT:%.*]] = extractvalue [3 x <4 x bfloat>] [[VAL_COERCE]], 2
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st3lane.v4bf16.p0i8(<4 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], i64 1, i8* [[TMP0]])
+// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st3lane.v4bf16.p0(<4 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], i64 1, ptr [[PTR:%.*]])
 // CHECK64-NEXT:    ret void
 //
 // CHECK32-LABEL: @test_vst3_lane_bf16(
@@ -1196,8 +1147,7 @@ void test_vst3q_bf16(bfloat16_t *ptr, bfloat16x8x3_t val) {
 // CHECK32-NEXT:    [[TMP0:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_0_EXTRACT]] to <4 x bfloat>
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_1_EXTRACT]] to <4 x bfloat>
 // CHECK32-NEXT:    [[TMP2:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_2_EXTRACT]] to <4 x bfloat>
-// CHECK32-NEXT:    [[TMP3:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst3lane.p0i8.v4bf16(i8* [[TMP3]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], <4 x bfloat> [[TMP2]], i32 1, i32 2)
+// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst3lane.p0.v4bf16(ptr [[PTR:%.*]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], <4 x bfloat> [[TMP2]], i32 1, i32 2)
 // CHECK32-NEXT:    ret void
 //
 void test_vst3_lane_bf16(bfloat16_t *ptr, bfloat16x4x3_t val) {
@@ -1209,8 +1159,7 @@ void test_vst3_lane_bf16(bfloat16_t *ptr, bfloat16x4x3_t val) {
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_0_EXTRACT:%.*]] = extractvalue [3 x <8 x bfloat>] [[VAL_COERCE:%.*]], 0
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [3 x <8 x bfloat>] [[VAL_COERCE]], 1
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_2_EXTRACT:%.*]] = extractvalue [3 x <8 x bfloat>] [[VAL_COERCE]], 2
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st3lane.v8bf16.p0i8(<8 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], i64 7, i8* [[TMP0]])
+// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st3lane.v8bf16.p0(<8 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], i64 7, ptr [[PTR:%.*]])
 // CHECK64-NEXT:    ret void
 //
 // CHECK32-LABEL: @test_vst3q_lane_bf16(
@@ -1221,8 +1170,7 @@ void test_vst3_lane_bf16(bfloat16_t *ptr, bfloat16x4x3_t val) {
 // CHECK32-NEXT:    [[TMP0:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_0_EXTRACT]] to <8 x bfloat>
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_1_EXTRACT]] to <8 x bfloat>
 // CHECK32-NEXT:    [[TMP2:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_2_EXTRACT]] to <8 x bfloat>
-// CHECK32-NEXT:    [[TMP3:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst3lane.p0i8.v8bf16(i8* [[TMP3]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], <8 x bfloat> [[TMP2]], i32 7, i32 2)
+// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst3lane.p0.v8bf16(ptr [[PTR:%.*]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], <8 x bfloat> [[TMP2]], i32 7, i32 2)
 // CHECK32-NEXT:    ret void
 //
 void test_vst3q_lane_bf16(bfloat16_t *ptr, bfloat16x8x3_t val) {
@@ -1235,8 +1183,7 @@ void test_vst3q_lane_bf16(bfloat16_t *ptr, bfloat16x8x3_t val) {
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [4 x <4 x bfloat>] [[VAL_COERCE]], 1
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_2_EXTRACT:%.*]] = extractvalue [4 x <4 x bfloat>] [[VAL_COERCE]], 2
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_3_EXTRACT:%.*]] = extractvalue [4 x <4 x bfloat>] [[VAL_COERCE]], 3
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st4.v4bf16.p0i8(<4 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_3_EXTRACT]], i8* [[TMP0]])
+// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st4.v4bf16.p0(<4 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_3_EXTRACT]], ptr [[PTR:%.*]])
 // CHECK64-NEXT:    ret void
 //
 // CHECK32-LABEL: @test_vst4_bf16(
@@ -1249,8 +1196,7 @@ void test_vst3q_lane_bf16(bfloat16_t *ptr, bfloat16x8x3_t val) {
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_1_EXTRACT]] to <4 x bfloat>
 // CHECK32-NEXT:    [[TMP2:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_2_EXTRACT]] to <4 x bfloat>
 // CHECK32-NEXT:    [[TMP3:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_3_EXTRACT]] to <4 x bfloat>
-// CHECK32-NEXT:    [[TMP4:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst4.p0i8.v4bf16(i8* [[TMP4]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], <4 x bfloat> [[TMP2]], <4 x bfloat> [[TMP3]], i32 2)
+// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst4.p0.v4bf16(ptr [[PTR:%.*]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], <4 x bfloat> [[TMP2]], <4 x bfloat> [[TMP3]], i32 2)
 // CHECK32-NEXT:    ret void
 //
 void test_vst4_bf16(bfloat16_t *ptr, bfloat16x4x4_t val) {
@@ -1263,8 +1209,7 @@ void test_vst4_bf16(bfloat16_t *ptr, bfloat16x4x4_t val) {
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [4 x <8 x bfloat>] [[VAL_COERCE]], 1
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_2_EXTRACT:%.*]] = extractvalue [4 x <8 x bfloat>] [[VAL_COERCE]], 2
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_3_EXTRACT:%.*]] = extractvalue [4 x <8 x bfloat>] [[VAL_COERCE]], 3
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st4.v8bf16.p0i8(<8 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_3_EXTRACT]], i8* [[TMP0]])
+// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st4.v8bf16.p0(<8 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_3_EXTRACT]], ptr [[PTR:%.*]])
 // CHECK64-NEXT:    ret void
 //
 // CHECK32-LABEL: @test_vst4q_bf16(
@@ -1277,8 +1222,7 @@ void test_vst4_bf16(bfloat16_t *ptr, bfloat16x4x4_t val) {
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_1_EXTRACT]] to <8 x bfloat>
 // CHECK32-NEXT:    [[TMP2:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_2_EXTRACT]] to <8 x bfloat>
 // CHECK32-NEXT:    [[TMP3:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_3_EXTRACT]] to <8 x bfloat>
-// CHECK32-NEXT:    [[TMP4:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst4.p0i8.v8bf16(i8* [[TMP4]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], <8 x bfloat> [[TMP2]], <8 x bfloat> [[TMP3]], i32 2)
+// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst4.p0.v8bf16(ptr [[PTR:%.*]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], <8 x bfloat> [[TMP2]], <8 x bfloat> [[TMP3]], i32 2)
 // CHECK32-NEXT:    ret void
 //
 void test_vst4q_bf16(bfloat16_t *ptr, bfloat16x8x4_t val) {
@@ -1291,8 +1235,7 @@ void test_vst4q_bf16(bfloat16_t *ptr, bfloat16x8x4_t val) {
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [4 x <4 x bfloat>] [[VAL_COERCE]], 1
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_2_EXTRACT:%.*]] = extractvalue [4 x <4 x bfloat>] [[VAL_COERCE]], 2
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_3_EXTRACT:%.*]] = extractvalue [4 x <4 x bfloat>] [[VAL_COERCE]], 3
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st4lane.v4bf16.p0i8(<4 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_3_EXTRACT]], i64 1, i8* [[TMP0]])
+// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st4lane.v4bf16.p0(<4 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], <4 x bfloat> [[VAL_COERCE_FCA_3_EXTRACT]], i64 1, ptr [[PTR:%.*]])
 // CHECK64-NEXT:    ret void
 //
 // CHECK32-LABEL: @test_vst4_lane_bf16(
@@ -1305,8 +1248,7 @@ void test_vst4q_bf16(bfloat16_t *ptr, bfloat16x8x4_t val) {
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_1_EXTRACT]] to <4 x bfloat>
 // CHECK32-NEXT:    [[TMP2:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_2_EXTRACT]] to <4 x bfloat>
 // CHECK32-NEXT:    [[TMP3:%.*]] = bitcast <2 x i32> [[VAL_COERCE_FCA_3_EXTRACT]] to <4 x bfloat>
-// CHECK32-NEXT:    [[TMP4:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst4lane.p0i8.v4bf16(i8* [[TMP4]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], <4 x bfloat> [[TMP2]], <4 x bfloat> [[TMP3]], i32 1, i32 2)
+// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst4lane.p0.v4bf16(ptr [[PTR:%.*]], <4 x bfloat> [[TMP0]], <4 x bfloat> [[TMP1]], <4 x bfloat> [[TMP2]], <4 x bfloat> [[TMP3]], i32 1, i32 2)
 // CHECK32-NEXT:    ret void
 //
 void test_vst4_lane_bf16(bfloat16_t *ptr, bfloat16x4x4_t val) {
@@ -1319,8 +1261,7 @@ void test_vst4_lane_bf16(bfloat16_t *ptr, bfloat16x4x4_t val) {
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_1_EXTRACT:%.*]] = extractvalue [4 x <8 x bfloat>] [[VAL_COERCE]], 1
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_2_EXTRACT:%.*]] = extractvalue [4 x <8 x bfloat>] [[VAL_COERCE]], 2
 // CHECK64-NEXT:    [[VAL_COERCE_FCA_3_EXTRACT:%.*]] = extractvalue [4 x <8 x bfloat>] [[VAL_COERCE]], 3
-// CHECK64-NEXT:    [[TMP0:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st4lane.v8bf16.p0i8(<8 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_3_EXTRACT]], i64 7, i8* [[TMP0]])
+// CHECK64-NEXT:    tail call void @llvm.aarch64.neon.st4lane.v8bf16.p0(<8 x bfloat> [[VAL_COERCE_FCA_0_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_1_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_2_EXTRACT]], <8 x bfloat> [[VAL_COERCE_FCA_3_EXTRACT]], i64 7, ptr [[PTR:%.*]])
 // CHECK64-NEXT:    ret void
 //
 // CHECK32-LABEL: @test_vst4q_lane_bf16(
@@ -1333,8 +1274,7 @@ void test_vst4_lane_bf16(bfloat16_t *ptr, bfloat16x4x4_t val) {
 // CHECK32-NEXT:    [[TMP1:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_1_EXTRACT]] to <8 x bfloat>
 // CHECK32-NEXT:    [[TMP2:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_2_EXTRACT]] to <8 x bfloat>
 // CHECK32-NEXT:    [[TMP3:%.*]] = bitcast <4 x i32> [[VAL_COERCE_FCA_3_EXTRACT]] to <8 x bfloat>
-// CHECK32-NEXT:    [[TMP4:%.*]] = bitcast bfloat* [[PTR:%.*]] to i8*
-// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst4lane.p0i8.v8bf16(i8* [[TMP4]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], <8 x bfloat> [[TMP2]], <8 x bfloat> [[TMP3]], i32 7, i32 2)
+// CHECK32-NEXT:    tail call void @llvm.arm.neon.vst4lane.p0.v8bf16(ptr [[PTR:%.*]], <8 x bfloat> [[TMP0]], <8 x bfloat> [[TMP1]], <8 x bfloat> [[TMP2]], <8 x bfloat> [[TMP3]], i32 7, i32 2)
 // CHECK32-NEXT:    ret void
 //
 void test_vst4q_lane_bf16(bfloat16_t *ptr, bfloat16x8x4_t val) {

diff  --git a/clang/test/CodeGen/aarch64-neon-ldst-one.c b/clang/test/CodeGen/aarch64-neon-ldst-one.c
index 328a7204855a6..9dfb3c4a9d3d0 100644
--- a/clang/test/CodeGen/aarch64-neon-ldst-one.c
+++ b/clang/test/CodeGen/aarch64-neon-ldst-one.c
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple arm64-none-linux-gnu -target-feature +neon \
+// RUN: %clang_cc1 -triple arm64-none-linux-gnu -target-feature +neon \
 // RUN:  -disable-O0-optnone -emit-llvm -o - %s \
 // RUN: | opt -S -passes=mem2reg | FileCheck %s
 
@@ -6,8 +6,8 @@
 
 #include <arm_neon.h>
 
-// CHECK-LABEL: define{{.*}} <16 x i8> @test_vld1q_dup_u8(i8* noundef %a) #0 {
-// CHECK:   [[TMP0:%.*]] = load i8, i8* %a
+// CHECK-LABEL: define{{.*}} <16 x i8> @test_vld1q_dup_u8(ptr noundef %a) #0 {
+// CHECK:   [[TMP0:%.*]] = load i8, ptr %a
 // CHECK:   [[TMP1:%.*]] = insertelement <16 x i8> poison, i8 [[TMP0]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <16 x i8> [[TMP1]], <16 x i8> [[TMP1]], <16 x i32> zeroinitializer
 // CHECK:   ret <16 x i8> [[LANE]]
@@ -15,10 +15,8 @@ uint8x16_t test_vld1q_dup_u8(uint8_t  *a) {
   return vld1q_dup_u8(a);
 }
 
-// CHECK-LABEL: define{{.*}} <8 x i16> @test_vld1q_dup_u16(i16* noundef %a) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to i16*
-// CHECK:   [[TMP2:%.*]] = load i16, i16* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <8 x i16> @test_vld1q_dup_u16(ptr noundef %a) #0 {
+// CHECK:   [[TMP2:%.*]] = load i16, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <8 x i16> poison, i16 [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <8 x i16> [[TMP3]], <8 x i16> [[TMP3]], <8 x i32> zeroinitializer
 // CHECK:   ret <8 x i16> [[LANE]]
@@ -26,10 +24,8 @@ uint16x8_t test_vld1q_dup_u16(uint16_t  *a) {
   return vld1q_dup_u16(a);
 }
 
-// CHECK-LABEL: define{{.*}} <4 x i32> @test_vld1q_dup_u32(i32* noundef %a) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to i32*
-// CHECK:   [[TMP2:%.*]] = load i32, i32* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <4 x i32> @test_vld1q_dup_u32(ptr noundef %a) #0 {
+// CHECK:   [[TMP2:%.*]] = load i32, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <4 x i32> poison, i32 [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <4 x i32> [[TMP3]], <4 x i32> [[TMP3]], <4 x i32> zeroinitializer
 // CHECK:   ret <4 x i32> [[LANE]]
@@ -37,10 +33,8 @@ uint32x4_t test_vld1q_dup_u32(uint32_t  *a) {
   return vld1q_dup_u32(a);
 }
 
-// CHECK-LABEL: define{{.*}} <2 x i64> @test_vld1q_dup_u64(i64* noundef %a) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to i64*
-// CHECK:   [[TMP2:%.*]] = load i64, i64* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <2 x i64> @test_vld1q_dup_u64(ptr noundef %a) #0 {
+// CHECK:   [[TMP2:%.*]] = load i64, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <2 x i64> poison, i64 [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <2 x i64> [[TMP3]], <2 x i64> [[TMP3]], <2 x i32> zeroinitializer
 // CHECK:   ret <2 x i64> [[LANE]]
@@ -48,8 +42,8 @@ uint64x2_t test_vld1q_dup_u64(uint64_t  *a) {
   return vld1q_dup_u64(a);
 }
 
-// CHECK-LABEL: define{{.*}} <16 x i8> @test_vld1q_dup_s8(i8* noundef %a) #0 {
-// CHECK:   [[TMP0:%.*]] = load i8, i8* %a
+// CHECK-LABEL: define{{.*}} <16 x i8> @test_vld1q_dup_s8(ptr noundef %a) #0 {
+// CHECK:   [[TMP0:%.*]] = load i8, ptr %a
 // CHECK:   [[TMP1:%.*]] = insertelement <16 x i8> poison, i8 [[TMP0]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <16 x i8> [[TMP1]], <16 x i8> [[TMP1]], <16 x i32> zeroinitializer
 // CHECK:   ret <16 x i8> [[LANE]]
@@ -57,10 +51,8 @@ int8x16_t test_vld1q_dup_s8(int8_t  *a) {
   return vld1q_dup_s8(a);
 }
 
-// CHECK-LABEL: define{{.*}} <8 x i16> @test_vld1q_dup_s16(i16* noundef %a) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to i16*
-// CHECK:   [[TMP2:%.*]] = load i16, i16* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <8 x i16> @test_vld1q_dup_s16(ptr noundef %a) #0 {
+// CHECK:   [[TMP2:%.*]] = load i16, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <8 x i16> poison, i16 [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <8 x i16> [[TMP3]], <8 x i16> [[TMP3]], <8 x i32> zeroinitializer
 // CHECK:   ret <8 x i16> [[LANE]]
@@ -68,10 +60,8 @@ int16x8_t test_vld1q_dup_s16(int16_t  *a) {
   return vld1q_dup_s16(a);
 }
 
-// CHECK-LABEL: define{{.*}} <4 x i32> @test_vld1q_dup_s32(i32* noundef %a) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to i32*
-// CHECK:   [[TMP2:%.*]] = load i32, i32* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <4 x i32> @test_vld1q_dup_s32(ptr noundef %a) #0 {
+// CHECK:   [[TMP2:%.*]] = load i32, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <4 x i32> poison, i32 [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <4 x i32> [[TMP3]], <4 x i32> [[TMP3]], <4 x i32> zeroinitializer
 // CHECK:   ret <4 x i32> [[LANE]]
@@ -79,10 +69,8 @@ int32x4_t test_vld1q_dup_s32(int32_t  *a) {
   return vld1q_dup_s32(a);
 }
 
-// CHECK-LABEL: define{{.*}} <2 x i64> @test_vld1q_dup_s64(i64* noundef %a) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to i64*
-// CHECK:   [[TMP2:%.*]] = load i64, i64* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <2 x i64> @test_vld1q_dup_s64(ptr noundef %a) #0 {
+// CHECK:   [[TMP2:%.*]] = load i64, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <2 x i64> poison, i64 [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <2 x i64> [[TMP3]], <2 x i64> [[TMP3]], <2 x i32> zeroinitializer
 // CHECK:   ret <2 x i64> [[LANE]]
@@ -90,10 +78,8 @@ int64x2_t test_vld1q_dup_s64(int64_t  *a) {
   return vld1q_dup_s64(a);
 }
 
-// CHECK-LABEL: define{{.*}} <8 x half> @test_vld1q_dup_f16(half* noundef %a) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast half* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to half*
-// CHECK:   [[TMP2:%.*]] = load half, half* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <8 x half> @test_vld1q_dup_f16(ptr noundef %a) #0 {
+// CHECK:   [[TMP2:%.*]] = load half, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <8 x half> poison, half [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <8 x half> [[TMP3]], <8 x half> [[TMP3]], <8 x i32> zeroinitializer
 // CHECK:   ret <8 x half> [[LANE]]
@@ -101,10 +87,8 @@ float16x8_t test_vld1q_dup_f16(float16_t  *a) {
   return vld1q_dup_f16(a);
 }
 
-// CHECK-LABEL: define{{.*}} <4 x float> @test_vld1q_dup_f32(float* noundef %a) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast float* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to float*
-// CHECK:   [[TMP2:%.*]] = load float, float* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <4 x float> @test_vld1q_dup_f32(ptr noundef %a) #0 {
+// CHECK:   [[TMP2:%.*]] = load float, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <4 x float> poison, float [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <4 x float> [[TMP3]], <4 x float> [[TMP3]], <4 x i32> zeroinitializer
 // CHECK:   ret <4 x float> [[LANE]]
@@ -112,10 +96,8 @@ float32x4_t test_vld1q_dup_f32(float32_t  *a) {
   return vld1q_dup_f32(a);
 }
 
-// CHECK-LABEL: define{{.*}} <2 x double> @test_vld1q_dup_f64(double* noundef %a) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast double* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to double*
-// CHECK:   [[TMP2:%.*]] = load double, double* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <2 x double> @test_vld1q_dup_f64(ptr noundef %a) #0 {
+// CHECK:   [[TMP2:%.*]] = load double, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <2 x double> poison, double [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <2 x double> [[TMP3]], <2 x double> [[TMP3]], <2 x i32> zeroinitializer
 // CHECK:   ret <2 x double> [[LANE]]
@@ -123,8 +105,8 @@ float64x2_t test_vld1q_dup_f64(float64_t  *a) {
   return vld1q_dup_f64(a);
 }
 
-// CHECK-LABEL: define{{.*}} <16 x i8> @test_vld1q_dup_p8(i8* noundef %a) #0 {
-// CHECK:   [[TMP0:%.*]] = load i8, i8* %a
+// CHECK-LABEL: define{{.*}} <16 x i8> @test_vld1q_dup_p8(ptr noundef %a) #0 {
+// CHECK:   [[TMP0:%.*]] = load i8, ptr %a
 // CHECK:   [[TMP1:%.*]] = insertelement <16 x i8> poison, i8 [[TMP0]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <16 x i8> [[TMP1]], <16 x i8> [[TMP1]], <16 x i32> zeroinitializer
 // CHECK:   ret <16 x i8> [[LANE]]
@@ -132,10 +114,8 @@ poly8x16_t test_vld1q_dup_p8(poly8_t  *a) {
   return vld1q_dup_p8(a);
 }
 
-// CHECK-LABEL: define{{.*}} <8 x i16> @test_vld1q_dup_p16(i16* noundef %a) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to i16*
-// CHECK:   [[TMP2:%.*]] = load i16, i16* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <8 x i16> @test_vld1q_dup_p16(ptr noundef %a) #0 {
+// CHECK:   [[TMP2:%.*]] = load i16, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <8 x i16> poison, i16 [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <8 x i16> [[TMP3]], <8 x i16> [[TMP3]], <8 x i32> zeroinitializer
 // CHECK:   ret <8 x i16> [[LANE]]
@@ -143,10 +123,8 @@ poly16x8_t test_vld1q_dup_p16(poly16_t  *a) {
   return vld1q_dup_p16(a);
 }
 
-// CHECK-LABEL: define{{.*}} <2 x i64> @test_vld1q_dup_p64(i64* noundef %a) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to i64*
-// CHECK:   [[TMP2:%.*]] = load i64, i64* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <2 x i64> @test_vld1q_dup_p64(ptr noundef %a) #0 {
+// CHECK:   [[TMP2:%.*]] = load i64, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <2 x i64> poison, i64 [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <2 x i64> [[TMP3]], <2 x i64> [[TMP3]], <2 x i32> zeroinitializer
 // CHECK:   ret <2 x i64> [[LANE]]
@@ -154,8 +132,8 @@ poly64x2_t test_vld1q_dup_p64(poly64_t  *a) {
   return vld1q_dup_p64(a);
 }
 
-// CHECK-LABEL: define{{.*}} <8 x i8> @test_vld1_dup_u8(i8* noundef %a) #1 {
-// CHECK:   [[TMP0:%.*]] = load i8, i8* %a
+// CHECK-LABEL: define{{.*}} <8 x i8> @test_vld1_dup_u8(ptr noundef %a) #1 {
+// CHECK:   [[TMP0:%.*]] = load i8, ptr %a
 // CHECK:   [[TMP1:%.*]] = insertelement <8 x i8> poison, i8 [[TMP0]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <8 x i8> [[TMP1]], <8 x i8> [[TMP1]], <8 x i32> zeroinitializer
 // CHECK:   ret <8 x i8> [[LANE]]
@@ -163,10 +141,8 @@ uint8x8_t test_vld1_dup_u8(uint8_t  *a) {
   return vld1_dup_u8(a);
 }
 
-// CHECK-LABEL: define{{.*}} <4 x i16> @test_vld1_dup_u16(i16* noundef %a) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to i16*
-// CHECK:   [[TMP2:%.*]] = load i16, i16* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <4 x i16> @test_vld1_dup_u16(ptr noundef %a) #1 {
+// CHECK:   [[TMP2:%.*]] = load i16, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <4 x i16> poison, i16 [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <4 x i16> [[TMP3]], <4 x i16> [[TMP3]], <4 x i32> zeroinitializer
 // CHECK:   ret <4 x i16> [[LANE]]
@@ -174,10 +150,8 @@ uint16x4_t test_vld1_dup_u16(uint16_t  *a) {
   return vld1_dup_u16(a);
 }
 
-// CHECK-LABEL: define{{.*}} <2 x i32> @test_vld1_dup_u32(i32* noundef %a) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to i32*
-// CHECK:   [[TMP2:%.*]] = load i32, i32* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <2 x i32> @test_vld1_dup_u32(ptr noundef %a) #1 {
+// CHECK:   [[TMP2:%.*]] = load i32, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <2 x i32> poison, i32 [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <2 x i32> [[TMP3]], <2 x i32> [[TMP3]], <2 x i32> zeroinitializer
 // CHECK:   ret <2 x i32> [[LANE]]
@@ -185,10 +159,8 @@ uint32x2_t test_vld1_dup_u32(uint32_t  *a) {
   return vld1_dup_u32(a);
 }
 
-// CHECK-LABEL: define{{.*}} <1 x i64> @test_vld1_dup_u64(i64* noundef %a) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to i64*
-// CHECK:   [[TMP2:%.*]] = load i64, i64* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <1 x i64> @test_vld1_dup_u64(ptr noundef %a) #1 {
+// CHECK:   [[TMP2:%.*]] = load i64, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <1 x i64> poison, i64 [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <1 x i64> [[TMP3]], <1 x i64> [[TMP3]], <1 x i32> zeroinitializer
 // CHECK:   ret <1 x i64> [[LANE]]
@@ -196,8 +168,8 @@ uint64x1_t test_vld1_dup_u64(uint64_t  *a) {
   return vld1_dup_u64(a);
 }
 
-// CHECK-LABEL: define{{.*}} <8 x i8> @test_vld1_dup_s8(i8* noundef %a) #1 {
-// CHECK:   [[TMP0:%.*]] = load i8, i8* %a
+// CHECK-LABEL: define{{.*}} <8 x i8> @test_vld1_dup_s8(ptr noundef %a) #1 {
+// CHECK:   [[TMP0:%.*]] = load i8, ptr %a
 // CHECK:   [[TMP1:%.*]] = insertelement <8 x i8> poison, i8 [[TMP0]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <8 x i8> [[TMP1]], <8 x i8> [[TMP1]], <8 x i32> zeroinitializer
 // CHECK:   ret <8 x i8> [[LANE]]
@@ -205,10 +177,8 @@ int8x8_t test_vld1_dup_s8(int8_t  *a) {
   return vld1_dup_s8(a);
 }
 
-// CHECK-LABEL: define{{.*}} <4 x i16> @test_vld1_dup_s16(i16* noundef %a) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to i16*
-// CHECK:   [[TMP2:%.*]] = load i16, i16* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <4 x i16> @test_vld1_dup_s16(ptr noundef %a) #1 {
+// CHECK:   [[TMP2:%.*]] = load i16, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <4 x i16> poison, i16 [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <4 x i16> [[TMP3]], <4 x i16> [[TMP3]], <4 x i32> zeroinitializer
 // CHECK:   ret <4 x i16> [[LANE]]
@@ -216,10 +186,8 @@ int16x4_t test_vld1_dup_s16(int16_t  *a) {
   return vld1_dup_s16(a);
 }
 
-// CHECK-LABEL: define{{.*}} <2 x i32> @test_vld1_dup_s32(i32* noundef %a) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to i32*
-// CHECK:   [[TMP2:%.*]] = load i32, i32* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <2 x i32> @test_vld1_dup_s32(ptr noundef %a) #1 {
+// CHECK:   [[TMP2:%.*]] = load i32, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <2 x i32> poison, i32 [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <2 x i32> [[TMP3]], <2 x i32> [[TMP3]], <2 x i32> zeroinitializer
 // CHECK:   ret <2 x i32> [[LANE]]
@@ -227,10 +195,8 @@ int32x2_t test_vld1_dup_s32(int32_t  *a) {
   return vld1_dup_s32(a);
 }
 
-// CHECK-LABEL: define{{.*}} <1 x i64> @test_vld1_dup_s64(i64* noundef %a) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to i64*
-// CHECK:   [[TMP2:%.*]] = load i64, i64* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <1 x i64> @test_vld1_dup_s64(ptr noundef %a) #1 {
+// CHECK:   [[TMP2:%.*]] = load i64, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <1 x i64> poison, i64 [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <1 x i64> [[TMP3]], <1 x i64> [[TMP3]], <1 x i32> zeroinitializer
 // CHECK:   ret <1 x i64> [[LANE]]
@@ -238,10 +204,8 @@ int64x1_t test_vld1_dup_s64(int64_t  *a) {
   return vld1_dup_s64(a);
 }
 
-// CHECK-LABEL: define{{.*}} <4 x half> @test_vld1_dup_f16(half* noundef %a) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast half* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to half*
-// CHECK:   [[TMP2:%.*]] = load half, half* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <4 x half> @test_vld1_dup_f16(ptr noundef %a) #1 {
+// CHECK:   [[TMP2:%.*]] = load half, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <4 x half> poison, half [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <4 x half> [[TMP3]], <4 x half> [[TMP3]], <4 x i32> zeroinitializer
 // CHECK:   ret <4 x half> [[LANE]]
@@ -249,10 +213,8 @@ float16x4_t test_vld1_dup_f16(float16_t  *a) {
   return vld1_dup_f16(a);
 }
 
-// CHECK-LABEL: define{{.*}} <2 x float> @test_vld1_dup_f32(float* noundef %a) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast float* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to float*
-// CHECK:   [[TMP2:%.*]] = load float, float* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <2 x float> @test_vld1_dup_f32(ptr noundef %a) #1 {
+// CHECK:   [[TMP2:%.*]] = load float, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <2 x float> poison, float [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <2 x float> [[TMP3]], <2 x float> [[TMP3]], <2 x i32> zeroinitializer
 // CHECK:   ret <2 x float> [[LANE]]
@@ -260,10 +222,8 @@ float32x2_t test_vld1_dup_f32(float32_t  *a) {
   return vld1_dup_f32(a);
 }
 
-// CHECK-LABEL: define{{.*}} <1 x double> @test_vld1_dup_f64(double* noundef %a) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast double* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to double*
-// CHECK:   [[TMP2:%.*]] = load double, double* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <1 x double> @test_vld1_dup_f64(ptr noundef %a) #1 {
+// CHECK:   [[TMP2:%.*]] = load double, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <1 x double> poison, double [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <1 x double> [[TMP3]], <1 x double> [[TMP3]], <1 x i32> zeroinitializer
 // CHECK:   ret <1 x double> [[LANE]]
@@ -271,8 +231,8 @@ float64x1_t test_vld1_dup_f64(float64_t  *a) {
   return vld1_dup_f64(a);
 }
 
-// CHECK-LABEL: define{{.*}} <8 x i8> @test_vld1_dup_p8(i8* noundef %a) #1 {
-// CHECK:   [[TMP0:%.*]] = load i8, i8* %a
+// CHECK-LABEL: define{{.*}} <8 x i8> @test_vld1_dup_p8(ptr noundef %a) #1 {
+// CHECK:   [[TMP0:%.*]] = load i8, ptr %a
 // CHECK:   [[TMP1:%.*]] = insertelement <8 x i8> poison, i8 [[TMP0]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <8 x i8> [[TMP1]], <8 x i8> [[TMP1]], <8 x i32> zeroinitializer
 // CHECK:   ret <8 x i8> [[LANE]]
@@ -280,10 +240,8 @@ poly8x8_t test_vld1_dup_p8(poly8_t  *a) {
   return vld1_dup_p8(a);
 }
 
-// CHECK-LABEL: define{{.*}} <4 x i16> @test_vld1_dup_p16(i16* noundef %a) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to i16*
-// CHECK:   [[TMP2:%.*]] = load i16, i16* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <4 x i16> @test_vld1_dup_p16(ptr noundef %a) #1 {
+// CHECK:   [[TMP2:%.*]] = load i16, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <4 x i16> poison, i16 [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <4 x i16> [[TMP3]], <4 x i16> [[TMP3]], <4 x i32> zeroinitializer
 // CHECK:   ret <4 x i16> [[LANE]]
@@ -291,10 +249,8 @@ poly16x4_t test_vld1_dup_p16(poly16_t  *a) {
   return vld1_dup_p16(a);
 }
 
-// CHECK-LABEL: define{{.*}} <1 x i64> @test_vld1_dup_p64(i64* noundef %a) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i8* [[TMP0]] to i64*
-// CHECK:   [[TMP2:%.*]] = load i64, i64* [[TMP1]]
+// CHECK-LABEL: define{{.*}} <1 x i64> @test_vld1_dup_p64(ptr noundef %a) #1 {
+// CHECK:   [[TMP2:%.*]] = load i64, ptr %a
 // CHECK:   [[TMP3:%.*]] = insertelement <1 x i64> poison, i64 [[TMP2]], i32 0
 // CHECK:   [[LANE:%.*]] = shufflevector <1 x i64> [[TMP3]], <1 x i64> [[TMP3]], <1 x i32> zeroinitializer
 // CHECK:   ret <1 x i64> [[LANE]]
@@ -302,6499 +258,5499 @@ poly64x1_t test_vld1_dup_p64(poly64_t  *a) {
   return vld1_dup_p64(a);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint64x2x2_t @test_vld2q_dup_u64(i64* noundef %a) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint64x2x2_t @test_vld2q_dup_u64(ptr noundef %a) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint64x2x2_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint64x2x2_t, align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint64x2x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[TMP2:%.*]] = bitcast i8* [[TMP1]] to i64*
-// CHECK:   [[VLD2:%.*]] = call { <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld2r.v2i64.p0i64(i64* [[TMP2]])
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to { <2 x i64>, <2 x i64> }*
-// CHECK:   store { <2 x i64>, <2 x i64> } [[VLD2]], { <2 x i64>, <2 x i64> }* [[TMP3]]
-// CHECK:   [[TMP4:%.*]] = bitcast %struct.uint64x2x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP5:%.*]] = bitcast %struct.uint64x2x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP4]], i8* align 16 [[TMP5]], i64 32, i1 false)
-// CHECK:   [[TMP6:%.*]] = load %struct.uint64x2x2_t, %struct.uint64x2x2_t* [[RETVAL]], align 16
+// CHECK:   [[VLD2:%.*]] = call { <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld2r.v2i64.p0(ptr %a)
+// CHECK:   store { <2 x i64>, <2 x i64> } [[VLD2]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP6:%.*]] = load %struct.uint64x2x2_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.uint64x2x2_t [[TMP6]]
 uint64x2x2_t test_vld2q_dup_u64(uint64_t  *a) {
   return vld2q_dup_u64(a);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int64x2x2_t @test_vld2q_dup_s64(i64* noundef %a) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int64x2x2_t @test_vld2q_dup_s64(ptr noundef %a) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int64x2x2_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.int64x2x2_t, align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int64x2x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[TMP2:%.*]] = bitcast i8* [[TMP1]] to i64*
-// CHECK:   [[VLD2:%.*]] = call { <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld2r.v2i64.p0i64(i64* [[TMP2]])
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to { <2 x i64>, <2 x i64> }*
-// CHECK:   store { <2 x i64>, <2 x i64> } [[VLD2]], { <2 x i64>, <2 x i64> }* [[TMP3]]
-// CHECK:   [[TMP4:%.*]] = bitcast %struct.int64x2x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP5:%.*]] = bitcast %struct.int64x2x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP4]], i8* align 16 [[TMP5]], i64 32, i1 false)
-// CHECK:   [[TMP6:%.*]] = load %struct.int64x2x2_t, %struct.int64x2x2_t* [[RETVAL]], align 16
+// CHECK:   [[VLD2:%.*]] = call { <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld2r.v2i64.p0(ptr %a)
+// CHECK:   store { <2 x i64>, <2 x i64> } [[VLD2]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP6:%.*]] = load %struct.int64x2x2_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.int64x2x2_t [[TMP6]]
 int64x2x2_t test_vld2q_dup_s64(int64_t  *a) {
   return vld2q_dup_s64(a);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float64x2x2_t @test_vld2q_dup_f64(double* noundef %a) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float64x2x2_t @test_vld2q_dup_f64(ptr noundef %a) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float64x2x2_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.float64x2x2_t, align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float64x2x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast double* %a to i8*
-// CHECK:   [[TMP2:%.*]] = bitcast i8* [[TMP1]] to double*
-// CHECK:   [[VLD2:%.*]] = call { <2 x double>, <2 x double> } @llvm.aarch64.neon.ld2r.v2f64.p0f64(double* [[TMP2]])
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to { <2 x double>, <2 x double> }*
-// CHECK:   store { <2 x double>, <2 x double> } [[VLD2]], { <2 x double>, <2 x double> }* [[TMP3]]
-// CHECK:   [[TMP4:%.*]] = bitcast %struct.float64x2x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP5:%.*]] = bitcast %struct.float64x2x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP4]], i8* align 16 [[TMP5]], i64 32, i1 false)
-// CHECK:   [[TMP6:%.*]] = load %struct.float64x2x2_t, %struct.float64x2x2_t* [[RETVAL]], align 16
+// CHECK:   [[VLD2:%.*]] = call { <2 x double>, <2 x double> } @llvm.aarch64.neon.ld2r.v2f64.p0(ptr %a)
+// CHECK:   store { <2 x double>, <2 x double> } [[VLD2]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP6:%.*]] = load %struct.float64x2x2_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.float64x2x2_t [[TMP6]]
 float64x2x2_t test_vld2q_dup_f64(float64_t  *a) {
   return vld2q_dup_f64(a);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly64x2x2_t @test_vld2q_dup_p64(i64* noundef %a) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly64x2x2_t @test_vld2q_dup_p64(ptr noundef %a) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly64x2x2_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly64x2x2_t, align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly64x2x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[TMP2:%.*]] = bitcast i8* [[TMP1]] to i64*
-// CHECK:   [[VLD2:%.*]] = call { <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld2r.v2i64.p0i64(i64* [[TMP2]])
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to { <2 x i64>, <2 x i64> }*
-// CHECK:   store { <2 x i64>, <2 x i64> } [[VLD2]], { <2 x i64>, <2 x i64> }* [[TMP3]]
-// CHECK:   [[TMP4:%.*]] = bitcast %struct.poly64x2x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP5:%.*]] = bitcast %struct.poly64x2x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP4]], i8* align 16 [[TMP5]], i64 32, i1 false)
-// CHECK:   [[TMP6:%.*]] = load %struct.poly64x2x2_t, %struct.poly64x2x2_t* [[RETVAL]], align 16
+// CHECK:   [[VLD2:%.*]] = call { <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld2r.v2i64.p0(ptr %a)
+// CHECK:   store { <2 x i64>, <2 x i64> } [[VLD2]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP6:%.*]] = load %struct.poly64x2x2_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.poly64x2x2_t [[TMP6]]
 poly64x2x2_t test_vld2q_dup_p64(poly64_t  *a) {
   return vld2q_dup_p64(a);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float64x1x2_t @test_vld2_dup_f64(double* noundef %a) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float64x1x2_t @test_vld2_dup_f64(ptr noundef %a) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float64x1x2_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.float64x1x2_t, align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float64x1x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast double* %a to i8*
-// CHECK:   [[TMP2:%.*]] = bitcast i8* [[TMP1]] to double*
-// CHECK:   [[VLD2:%.*]] = call { <1 x double>, <1 x double> } @llvm.aarch64.neon.ld2r.v1f64.p0f64(double* [[TMP2]])
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to { <1 x double>, <1 x double> }*
-// CHECK:   store { <1 x double>, <1 x double> } [[VLD2]], { <1 x double>, <1 x double> }* [[TMP3]]
-// CHECK:   [[TMP4:%.*]] = bitcast %struct.float64x1x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP5:%.*]] = bitcast %struct.float64x1x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP4]], i8* align 8 [[TMP5]], i64 16, i1 false)
-// CHECK:   [[TMP6:%.*]] = load %struct.float64x1x2_t, %struct.float64x1x2_t* [[RETVAL]], align 8
+// CHECK:   [[VLD2:%.*]] = call { <1 x double>, <1 x double> } @llvm.aarch64.neon.ld2r.v1f64.p0(ptr %a)
+// CHECK:   store { <1 x double>, <1 x double> } [[VLD2]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 16, i1 false)
+// CHECK:   [[TMP6:%.*]] = load %struct.float64x1x2_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.float64x1x2_t [[TMP6]]
 float64x1x2_t test_vld2_dup_f64(float64_t  *a) {
   return vld2_dup_f64(a);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly64x1x2_t @test_vld2_dup_p64(i64* noundef %a) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly64x1x2_t @test_vld2_dup_p64(ptr noundef %a) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly64x1x2_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly64x1x2_t, align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly64x1x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[TMP2:%.*]] = bitcast i8* [[TMP1]] to i64*
-// CHECK:   [[VLD2:%.*]] = call { <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld2r.v1i64.p0i64(i64* [[TMP2]])
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to { <1 x i64>, <1 x i64> }*
-// CHECK:   store { <1 x i64>, <1 x i64> } [[VLD2]], { <1 x i64>, <1 x i64> }* [[TMP3]]
-// CHECK:   [[TMP4:%.*]] = bitcast %struct.poly64x1x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP5:%.*]] = bitcast %struct.poly64x1x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP4]], i8* align 8 [[TMP5]], i64 16, i1 false)
-// CHECK:   [[TMP6:%.*]] = load %struct.poly64x1x2_t, %struct.poly64x1x2_t* [[RETVAL]], align 8
+// CHECK:   [[VLD2:%.*]] = call { <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld2r.v1i64.p0(ptr %a)
+// CHECK:   store { <1 x i64>, <1 x i64> } [[VLD2]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 16, i1 false)
+// CHECK:   [[TMP6:%.*]] = load %struct.poly64x1x2_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.poly64x1x2_t [[TMP6]]
 poly64x1x2_t test_vld2_dup_p64(poly64_t  *a) {
   return vld2_dup_p64(a);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint64x2x3_t @test_vld3q_dup_u64(i64* noundef %a) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint64x2x3_t @test_vld3q_dup_u64(ptr noundef %a) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint64x2x3_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint64x2x3_t, align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint64x2x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[TMP2:%.*]] = bitcast i8* [[TMP1]] to i64*
-// CHECK:   [[VLD3:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld3r.v2i64.p0i64(i64* [[TMP2]])
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to { <2 x i64>, <2 x i64>, <2 x i64> }*
-// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64> } [[VLD3]], { <2 x i64>, <2 x i64>, <2 x i64> }* [[TMP3]]
-// CHECK:   [[TMP4:%.*]] = bitcast %struct.uint64x2x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP5:%.*]] = bitcast %struct.uint64x2x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP4]], i8* align 16 [[TMP5]], i64 48, i1 false)
-// CHECK:   [[TMP6:%.*]] = load %struct.uint64x2x3_t, %struct.uint64x2x3_t* [[RETVAL]], align 16
+// CHECK:   [[VLD3:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld3r.v2i64.p0(ptr %a)
+// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64> } [[VLD3]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 48, i1 false)
+// CHECK:   [[TMP6:%.*]] = load %struct.uint64x2x3_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.uint64x2x3_t [[TMP6]]
 uint64x2x3_t test_vld3q_dup_u64(uint64_t  *a) {
   return vld3q_dup_u64(a);
   // [{{x[0-9]+|sp}}]
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int64x2x3_t @test_vld3q_dup_s64(i64* noundef %a) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int64x2x3_t @test_vld3q_dup_s64(ptr noundef %a) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int64x2x3_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.int64x2x3_t, align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int64x2x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[TMP2:%.*]] = bitcast i8* [[TMP1]] to i64*
-// CHECK:   [[VLD3:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld3r.v2i64.p0i64(i64* [[TMP2]])
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to { <2 x i64>, <2 x i64>, <2 x i64> }*
-// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64> } [[VLD3]], { <2 x i64>, <2 x i64>, <2 x i64> }* [[TMP3]]
-// CHECK:   [[TMP4:%.*]] = bitcast %struct.int64x2x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP5:%.*]] = bitcast %struct.int64x2x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP4]], i8* align 16 [[TMP5]], i64 48, i1 false)
-// CHECK:   [[TMP6:%.*]] = load %struct.int64x2x3_t, %struct.int64x2x3_t* [[RETVAL]], align 16
+// CHECK:   [[VLD3:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld3r.v2i64.p0(ptr %a)
+// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64> } [[VLD3]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 48, i1 false)
+// CHECK:   [[TMP6:%.*]] = load %struct.int64x2x3_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.int64x2x3_t [[TMP6]]
 int64x2x3_t test_vld3q_dup_s64(int64_t  *a) {
   return vld3q_dup_s64(a);
   // [{{x[0-9]+|sp}}]
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float64x2x3_t @test_vld3q_dup_f64(double* noundef %a) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float64x2x3_t @test_vld3q_dup_f64(ptr noundef %a) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float64x2x3_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.float64x2x3_t, align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float64x2x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast double* %a to i8*
-// CHECK:   [[TMP2:%.*]] = bitcast i8* [[TMP1]] to double*
-// CHECK:   [[VLD3:%.*]] = call { <2 x double>, <2 x double>, <2 x double> } @llvm.aarch64.neon.ld3r.v2f64.p0f64(double* [[TMP2]])
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to { <2 x double>, <2 x double>, <2 x double> }*
-// CHECK:   store { <2 x double>, <2 x double>, <2 x double> } [[VLD3]], { <2 x double>, <2 x double>, <2 x double> }* [[TMP3]]
-// CHECK:   [[TMP4:%.*]] = bitcast %struct.float64x2x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP5:%.*]] = bitcast %struct.float64x2x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP4]], i8* align 16 [[TMP5]], i64 48, i1 false)
-// CHECK:   [[TMP6:%.*]] = load %struct.float64x2x3_t, %struct.float64x2x3_t* [[RETVAL]], align 16
+// CHECK:   [[VLD3:%.*]] = call { <2 x double>, <2 x double>, <2 x double> } @llvm.aarch64.neon.ld3r.v2f64.p0(ptr %a)
+// CHECK:   store { <2 x double>, <2 x double>, <2 x double> } [[VLD3]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 48, i1 false)
+// CHECK:   [[TMP6:%.*]] = load %struct.float64x2x3_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.float64x2x3_t [[TMP6]]
 float64x2x3_t test_vld3q_dup_f64(float64_t  *a) {
   return vld3q_dup_f64(a);
   // [{{x[0-9]+|sp}}]
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly64x2x3_t @test_vld3q_dup_p64(i64* noundef %a) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly64x2x3_t @test_vld3q_dup_p64(ptr noundef %a) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly64x2x3_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly64x2x3_t, align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly64x2x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[TMP2:%.*]] = bitcast i8* [[TMP1]] to i64*
-// CHECK:   [[VLD3:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld3r.v2i64.p0i64(i64* [[TMP2]])
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to { <2 x i64>, <2 x i64>, <2 x i64> }*
-// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64> } [[VLD3]], { <2 x i64>, <2 x i64>, <2 x i64> }* [[TMP3]]
-// CHECK:   [[TMP4:%.*]] = bitcast %struct.poly64x2x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP5:%.*]] = bitcast %struct.poly64x2x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP4]], i8* align 16 [[TMP5]], i64 48, i1 false)
-// CHECK:   [[TMP6:%.*]] = load %struct.poly64x2x3_t, %struct.poly64x2x3_t* [[RETVAL]], align 16
+// CHECK:   [[VLD3:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld3r.v2i64.p0(ptr %a)
+// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64> } [[VLD3]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 48, i1 false)
+// CHECK:   [[TMP6:%.*]] = load %struct.poly64x2x3_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.poly64x2x3_t [[TMP6]]
 poly64x2x3_t test_vld3q_dup_p64(poly64_t  *a) {
   return vld3q_dup_p64(a);
   // [{{x[0-9]+|sp}}]
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float64x1x3_t @test_vld3_dup_f64(double* noundef %a) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float64x1x3_t @test_vld3_dup_f64(ptr noundef %a) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float64x1x3_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.float64x1x3_t, align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float64x1x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast double* %a to i8*
-// CHECK:   [[TMP2:%.*]] = bitcast i8* [[TMP1]] to double*
-// CHECK:   [[VLD3:%.*]] = call { <1 x double>, <1 x double>, <1 x double> } @llvm.aarch64.neon.ld3r.v1f64.p0f64(double* [[TMP2]])
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to { <1 x double>, <1 x double>, <1 x double> }*
-// CHECK:   store { <1 x double>, <1 x double>, <1 x double> } [[VLD3]], { <1 x double>, <1 x double>, <1 x double> }* [[TMP3]]
-// CHECK:   [[TMP4:%.*]] = bitcast %struct.float64x1x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP5:%.*]] = bitcast %struct.float64x1x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP4]], i8* align 8 [[TMP5]], i64 24, i1 false)
-// CHECK:   [[TMP6:%.*]] = load %struct.float64x1x3_t, %struct.float64x1x3_t* [[RETVAL]], align 8
+// CHECK:   [[VLD3:%.*]] = call { <1 x double>, <1 x double>, <1 x double> } @llvm.aarch64.neon.ld3r.v1f64.p0(ptr %a)
+// CHECK:   store { <1 x double>, <1 x double>, <1 x double> } [[VLD3]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 24, i1 false)
+// CHECK:   [[TMP6:%.*]] = load %struct.float64x1x3_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.float64x1x3_t [[TMP6]]
 float64x1x3_t test_vld3_dup_f64(float64_t  *a) {
   return vld3_dup_f64(a);
   // [{{x[0-9]+|sp}}]
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly64x1x3_t @test_vld3_dup_p64(i64* noundef %a) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly64x1x3_t @test_vld3_dup_p64(ptr noundef %a) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly64x1x3_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly64x1x3_t, align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly64x1x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[TMP2:%.*]] = bitcast i8* [[TMP1]] to i64*
-// CHECK:   [[VLD3:%.*]] = call { <1 x i64>, <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld3r.v1i64.p0i64(i64* [[TMP2]])
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to { <1 x i64>, <1 x i64>, <1 x i64> }*
-// CHECK:   store { <1 x i64>, <1 x i64>, <1 x i64> } [[VLD3]], { <1 x i64>, <1 x i64>, <1 x i64> }* [[TMP3]]
-// CHECK:   [[TMP4:%.*]] = bitcast %struct.poly64x1x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP5:%.*]] = bitcast %struct.poly64x1x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP4]], i8* align 8 [[TMP5]], i64 24, i1 false)
-// CHECK:   [[TMP6:%.*]] = load %struct.poly64x1x3_t, %struct.poly64x1x3_t* [[RETVAL]], align 8
+// CHECK:   [[VLD3:%.*]] = call { <1 x i64>, <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld3r.v1i64.p0(ptr %a)
+// CHECK:   store { <1 x i64>, <1 x i64>, <1 x i64> } [[VLD3]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 24, i1 false)
+// CHECK:   [[TMP6:%.*]] = load %struct.poly64x1x3_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.poly64x1x3_t [[TMP6]]
 poly64x1x3_t test_vld3_dup_p64(poly64_t  *a) {
   return vld3_dup_p64(a);
   // [{{x[0-9]+|sp}}]
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint64x2x4_t @test_vld4q_dup_u64(i64* noundef %a) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint64x2x4_t @test_vld4q_dup_u64(ptr noundef %a) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint64x2x4_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint64x2x4_t, align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint64x2x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[TMP2:%.*]] = bitcast i8* [[TMP1]] to i64*
-// CHECK:   [[VLD4:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld4r.v2i64.p0i64(i64* [[TMP2]])
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> }*
-// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } [[VLD4]], { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> }* [[TMP3]]
-// CHECK:   [[TMP4:%.*]] = bitcast %struct.uint64x2x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP5:%.*]] = bitcast %struct.uint64x2x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP4]], i8* align 16 [[TMP5]], i64 64, i1 false)
-// CHECK:   [[TMP6:%.*]] = load %struct.uint64x2x4_t, %struct.uint64x2x4_t* [[RETVAL]], align 16
+// CHECK:   [[VLD4:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld4r.v2i64.p0(ptr %a)
+// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } [[VLD4]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 64, i1 false)
+// CHECK:   [[TMP6:%.*]] = load %struct.uint64x2x4_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.uint64x2x4_t [[TMP6]]
 uint64x2x4_t test_vld4q_dup_u64(uint64_t  *a) {
   return vld4q_dup_u64(a);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int64x2x4_t @test_vld4q_dup_s64(i64* noundef %a) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int64x2x4_t @test_vld4q_dup_s64(ptr noundef %a) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int64x2x4_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.int64x2x4_t, align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int64x2x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[TMP2:%.*]] = bitcast i8* [[TMP1]] to i64*
-// CHECK:   [[VLD4:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld4r.v2i64.p0i64(i64* [[TMP2]])
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> }*
-// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } [[VLD4]], { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> }* [[TMP3]]
-// CHECK:   [[TMP4:%.*]] = bitcast %struct.int64x2x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP5:%.*]] = bitcast %struct.int64x2x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP4]], i8* align 16 [[TMP5]], i64 64, i1 false)
-// CHECK:   [[TMP6:%.*]] = load %struct.int64x2x4_t, %struct.int64x2x4_t* [[RETVAL]], align 16
+// CHECK:   [[VLD4:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld4r.v2i64.p0(ptr %a)
+// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } [[VLD4]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 64, i1 false)
+// CHECK:   [[TMP6:%.*]] = load %struct.int64x2x4_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.int64x2x4_t [[TMP6]]
 int64x2x4_t test_vld4q_dup_s64(int64_t  *a) {
   return vld4q_dup_s64(a);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float64x2x4_t @test_vld4q_dup_f64(double* noundef %a) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float64x2x4_t @test_vld4q_dup_f64(ptr noundef %a) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float64x2x4_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.float64x2x4_t, align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float64x2x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast double* %a to i8*
-// CHECK:   [[TMP2:%.*]] = bitcast i8* [[TMP1]] to double*
-// CHECK:   [[VLD4:%.*]] = call { <2 x double>, <2 x double>, <2 x double>, <2 x double> } @llvm.aarch64.neon.ld4r.v2f64.p0f64(double* [[TMP2]])
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to { <2 x double>, <2 x double>, <2 x double>, <2 x double> }*
-// CHECK:   store { <2 x double>, <2 x double>, <2 x double>, <2 x double> } [[VLD4]], { <2 x double>, <2 x double>, <2 x double>, <2 x double> }* [[TMP3]]
-// CHECK:   [[TMP4:%.*]] = bitcast %struct.float64x2x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP5:%.*]] = bitcast %struct.float64x2x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP4]], i8* align 16 [[TMP5]], i64 64, i1 false)
-// CHECK:   [[TMP6:%.*]] = load %struct.float64x2x4_t, %struct.float64x2x4_t* [[RETVAL]], align 16
+// CHECK:   [[VLD4:%.*]] = call { <2 x double>, <2 x double>, <2 x double>, <2 x double> } @llvm.aarch64.neon.ld4r.v2f64.p0(ptr %a)
+// CHECK:   store { <2 x double>, <2 x double>, <2 x double>, <2 x double> } [[VLD4]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 64, i1 false)
+// CHECK:   [[TMP6:%.*]] = load %struct.float64x2x4_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.float64x2x4_t [[TMP6]]
 float64x2x4_t test_vld4q_dup_f64(float64_t  *a) {
   return vld4q_dup_f64(a);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly64x2x4_t @test_vld4q_dup_p64(i64* noundef %a) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly64x2x4_t @test_vld4q_dup_p64(ptr noundef %a) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly64x2x4_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly64x2x4_t, align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly64x2x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[TMP2:%.*]] = bitcast i8* [[TMP1]] to i64*
-// CHECK:   [[VLD4:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld4r.v2i64.p0i64(i64* [[TMP2]])
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> }*
-// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } [[VLD4]], { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> }* [[TMP3]]
-// CHECK:   [[TMP4:%.*]] = bitcast %struct.poly64x2x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP5:%.*]] = bitcast %struct.poly64x2x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP4]], i8* align 16 [[TMP5]], i64 64, i1 false)
-// CHECK:   [[TMP6:%.*]] = load %struct.poly64x2x4_t, %struct.poly64x2x4_t* [[RETVAL]], align 16
+// CHECK:   [[VLD4:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld4r.v2i64.p0(ptr %a)
+// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } [[VLD4]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 64, i1 false)
+// CHECK:   [[TMP6:%.*]] = load %struct.poly64x2x4_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.poly64x2x4_t [[TMP6]]
 poly64x2x4_t test_vld4q_dup_p64(poly64_t  *a) {
   return vld4q_dup_p64(a);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float64x1x4_t @test_vld4_dup_f64(double* noundef %a) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float64x1x4_t @test_vld4_dup_f64(ptr noundef %a) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float64x1x4_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.float64x1x4_t, align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float64x1x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast double* %a to i8*
-// CHECK:   [[TMP2:%.*]] = bitcast i8* [[TMP1]] to double*
-// CHECK:   [[VLD4:%.*]] = call { <1 x double>, <1 x double>, <1 x double>, <1 x double> } @llvm.aarch64.neon.ld4r.v1f64.p0f64(double* [[TMP2]])
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to { <1 x double>, <1 x double>, <1 x double>, <1 x double> }*
-// CHECK:   store { <1 x double>, <1 x double>, <1 x double>, <1 x double> } [[VLD4]], { <1 x double>, <1 x double>, <1 x double>, <1 x double> }* [[TMP3]]
-// CHECK:   [[TMP4:%.*]] = bitcast %struct.float64x1x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP5:%.*]] = bitcast %struct.float64x1x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP4]], i8* align 8 [[TMP5]], i64 32, i1 false)
-// CHECK:   [[TMP6:%.*]] = load %struct.float64x1x4_t, %struct.float64x1x4_t* [[RETVAL]], align 8
+// CHECK:   [[VLD4:%.*]] = call { <1 x double>, <1 x double>, <1 x double>, <1 x double> } @llvm.aarch64.neon.ld4r.v1f64.p0(ptr %a)
+// CHECK:   store { <1 x double>, <1 x double>, <1 x double>, <1 x double> } [[VLD4]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP6:%.*]] = load %struct.float64x1x4_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.float64x1x4_t [[TMP6]]
 float64x1x4_t test_vld4_dup_f64(float64_t  *a) {
   return vld4_dup_f64(a);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly64x1x4_t @test_vld4_dup_p64(i64* noundef %a) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly64x1x4_t @test_vld4_dup_p64(ptr noundef %a) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly64x1x4_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly64x1x4_t, align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly64x1x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[TMP2:%.*]] = bitcast i8* [[TMP1]] to i64*
-// CHECK:   [[VLD4:%.*]] = call { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld4r.v1i64.p0i64(i64* [[TMP2]])
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> }*
-// CHECK:   store { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> } [[VLD4]], { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> }* [[TMP3]]
-// CHECK:   [[TMP4:%.*]] = bitcast %struct.poly64x1x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP5:%.*]] = bitcast %struct.poly64x1x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP4]], i8* align 8 [[TMP5]], i64 32, i1 false)
-// CHECK:   [[TMP6:%.*]] = load %struct.poly64x1x4_t, %struct.poly64x1x4_t* [[RETVAL]], align 8
+// CHECK:   [[VLD4:%.*]] = call { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld4r.v1i64.p0(ptr %a)
+// CHECK:   store { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> } [[VLD4]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP6:%.*]] = load %struct.poly64x1x4_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.poly64x1x4_t [[TMP6]]
 poly64x1x4_t test_vld4_dup_p64(poly64_t  *a) {
   return vld4_dup_p64(a);
 }
 
-// CHECK-LABEL: define{{.*}} <16 x i8> @test_vld1q_lane_u8(i8* noundef %a, <16 x i8> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = load i8, i8* %a
+// CHECK-LABEL: define{{.*}} <16 x i8> @test_vld1q_lane_u8(ptr noundef %a, <16 x i8> noundef %b) #0 {
+// CHECK:   [[TMP0:%.*]] = load i8, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <16 x i8> %b, i8 [[TMP0]], i32 15
 // CHECK:   ret <16 x i8> [[VLD1_LANE]]
 uint8x16_t test_vld1q_lane_u8(uint8_t  *a, uint8x16_t b) {
   return vld1q_lane_u8(a, b, 15);
 }
 
-// CHECK-LABEL: define{{.*}} <8 x i16> @test_vld1q_lane_u16(i16* noundef %a, <8 x i16> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i16* %a to i8*
+// CHECK-LABEL: define{{.*}} <8 x i16> @test_vld1q_lane_u16(ptr noundef %a, <8 x i16> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <8 x i16> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <8 x i16>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to i16*
-// CHECK:   [[TMP4:%.*]] = load i16, i16* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load i16, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <8 x i16> [[TMP2]], i16 [[TMP4]], i32 7
 // CHECK:   ret <8 x i16> [[VLD1_LANE]]
 uint16x8_t test_vld1q_lane_u16(uint16_t  *a, uint16x8_t b) {
   return vld1q_lane_u16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} <4 x i32> @test_vld1q_lane_u32(i32* noundef %a, <4 x i32> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i32* %a to i8*
+// CHECK-LABEL: define{{.*}} <4 x i32> @test_vld1q_lane_u32(ptr noundef %a, <4 x i32> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i32> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <4 x i32>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to i32*
-// CHECK:   [[TMP4:%.*]] = load i32, i32* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load i32, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <4 x i32> [[TMP2]], i32 [[TMP4]], i32 3
 // CHECK:   ret <4 x i32> [[VLD1_LANE]]
 uint32x4_t test_vld1q_lane_u32(uint32_t  *a, uint32x4_t b) {
   return vld1q_lane_u32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} <2 x i64> @test_vld1q_lane_u64(i64* noundef %a, <2 x i64> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i64* %a to i8*
+// CHECK-LABEL: define{{.*}} <2 x i64> @test_vld1q_lane_u64(ptr noundef %a, <2 x i64> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x i64> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <2 x i64>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to i64*
-// CHECK:   [[TMP4:%.*]] = load i64, i64* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load i64, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <2 x i64> [[TMP2]], i64 [[TMP4]], i32 1
 // CHECK:   ret <2 x i64> [[VLD1_LANE]]
 uint64x2_t test_vld1q_lane_u64(uint64_t  *a, uint64x2_t b) {
   return vld1q_lane_u64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} <16 x i8> @test_vld1q_lane_s8(i8* noundef %a, <16 x i8> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = load i8, i8* %a
+// CHECK-LABEL: define{{.*}} <16 x i8> @test_vld1q_lane_s8(ptr noundef %a, <16 x i8> noundef %b) #0 {
+// CHECK:   [[TMP0:%.*]] = load i8, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <16 x i8> %b, i8 [[TMP0]], i32 15
 // CHECK:   ret <16 x i8> [[VLD1_LANE]]
 int8x16_t test_vld1q_lane_s8(int8_t  *a, int8x16_t b) {
   return vld1q_lane_s8(a, b, 15);
 }
 
-// CHECK-LABEL: define{{.*}} <8 x i16> @test_vld1q_lane_s16(i16* noundef %a, <8 x i16> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i16* %a to i8*
+// CHECK-LABEL: define{{.*}} <8 x i16> @test_vld1q_lane_s16(ptr noundef %a, <8 x i16> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <8 x i16> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <8 x i16>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to i16*
-// CHECK:   [[TMP4:%.*]] = load i16, i16* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load i16, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <8 x i16> [[TMP2]], i16 [[TMP4]], i32 7
 // CHECK:   ret <8 x i16> [[VLD1_LANE]]
 int16x8_t test_vld1q_lane_s16(int16_t  *a, int16x8_t b) {
   return vld1q_lane_s16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} <4 x i32> @test_vld1q_lane_s32(i32* noundef %a, <4 x i32> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i32* %a to i8*
+// CHECK-LABEL: define{{.*}} <4 x i32> @test_vld1q_lane_s32(ptr noundef %a, <4 x i32> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i32> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <4 x i32>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to i32*
-// CHECK:   [[TMP4:%.*]] = load i32, i32* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load i32, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <4 x i32> [[TMP2]], i32 [[TMP4]], i32 3
 // CHECK:   ret <4 x i32> [[VLD1_LANE]]
 int32x4_t test_vld1q_lane_s32(int32_t  *a, int32x4_t b) {
   return vld1q_lane_s32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} <2 x i64> @test_vld1q_lane_s64(i64* noundef %a, <2 x i64> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i64* %a to i8*
+// CHECK-LABEL: define{{.*}} <2 x i64> @test_vld1q_lane_s64(ptr noundef %a, <2 x i64> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x i64> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <2 x i64>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to i64*
-// CHECK:   [[TMP4:%.*]] = load i64, i64* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load i64, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <2 x i64> [[TMP2]], i64 [[TMP4]], i32 1
 // CHECK:   ret <2 x i64> [[VLD1_LANE]]
 int64x2_t test_vld1q_lane_s64(int64_t  *a, int64x2_t b) {
   return vld1q_lane_s64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} <8 x half> @test_vld1q_lane_f16(half* noundef %a, <8 x half> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast half* %a to i8*
+// CHECK-LABEL: define{{.*}} <8 x half> @test_vld1q_lane_f16(ptr noundef %a, <8 x half> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <8 x half> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <8 x half>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to half*
-// CHECK:   [[TMP4:%.*]] = load half, half* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load half, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <8 x half> [[TMP2]], half [[TMP4]], i32 7
 // CHECK:   ret <8 x half> [[VLD1_LANE]]
 float16x8_t test_vld1q_lane_f16(float16_t  *a, float16x8_t b) {
   return vld1q_lane_f16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} <4 x float> @test_vld1q_lane_f32(float* noundef %a, <4 x float> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast float* %a to i8*
+// CHECK-LABEL: define{{.*}} <4 x float> @test_vld1q_lane_f32(ptr noundef %a, <4 x float> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x float> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <4 x float>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to float*
-// CHECK:   [[TMP4:%.*]] = load float, float* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load float, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <4 x float> [[TMP2]], float [[TMP4]], i32 3
 // CHECK:   ret <4 x float> [[VLD1_LANE]]
 float32x4_t test_vld1q_lane_f32(float32_t  *a, float32x4_t b) {
   return vld1q_lane_f32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} <2 x double> @test_vld1q_lane_f64(double* noundef %a, <2 x double> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast double* %a to i8*
+// CHECK-LABEL: define{{.*}} <2 x double> @test_vld1q_lane_f64(ptr noundef %a, <2 x double> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x double> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <2 x double>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to double*
-// CHECK:   [[TMP4:%.*]] = load double, double* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load double, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <2 x double> [[TMP2]], double [[TMP4]], i32 1
 // CHECK:   ret <2 x double> [[VLD1_LANE]]
 float64x2_t test_vld1q_lane_f64(float64_t  *a, float64x2_t b) {
   return vld1q_lane_f64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} <16 x i8> @test_vld1q_lane_p8(i8* noundef %a, <16 x i8> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = load i8, i8* %a
+// CHECK-LABEL: define{{.*}} <16 x i8> @test_vld1q_lane_p8(ptr noundef %a, <16 x i8> noundef %b) #0 {
+// CHECK:   [[TMP0:%.*]] = load i8, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <16 x i8> %b, i8 [[TMP0]], i32 15
 // CHECK:   ret <16 x i8> [[VLD1_LANE]]
 poly8x16_t test_vld1q_lane_p8(poly8_t  *a, poly8x16_t b) {
   return vld1q_lane_p8(a, b, 15);
 }
 
-// CHECK-LABEL: define{{.*}} <8 x i16> @test_vld1q_lane_p16(i16* noundef %a, <8 x i16> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i16* %a to i8*
+// CHECK-LABEL: define{{.*}} <8 x i16> @test_vld1q_lane_p16(ptr noundef %a, <8 x i16> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <8 x i16> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <8 x i16>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to i16*
-// CHECK:   [[TMP4:%.*]] = load i16, i16* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load i16, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <8 x i16> [[TMP2]], i16 [[TMP4]], i32 7
 // CHECK:   ret <8 x i16> [[VLD1_LANE]]
 poly16x8_t test_vld1q_lane_p16(poly16_t  *a, poly16x8_t b) {
   return vld1q_lane_p16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} <2 x i64> @test_vld1q_lane_p64(i64* noundef %a, <2 x i64> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i64* %a to i8*
+// CHECK-LABEL: define{{.*}} <2 x i64> @test_vld1q_lane_p64(ptr noundef %a, <2 x i64> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x i64> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <2 x i64>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to i64*
-// CHECK:   [[TMP4:%.*]] = load i64, i64* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load i64, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <2 x i64> [[TMP2]], i64 [[TMP4]], i32 1
 // CHECK:   ret <2 x i64> [[VLD1_LANE]]
 poly64x2_t test_vld1q_lane_p64(poly64_t  *a, poly64x2_t b) {
   return vld1q_lane_p64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} <8 x i8> @test_vld1_lane_u8(i8* noundef %a, <8 x i8> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = load i8, i8* %a
+// CHECK-LABEL: define{{.*}} <8 x i8> @test_vld1_lane_u8(ptr noundef %a, <8 x i8> noundef %b) #1 {
+// CHECK:   [[TMP0:%.*]] = load i8, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <8 x i8> %b, i8 [[TMP0]], i32 7
 // CHECK:   ret <8 x i8> [[VLD1_LANE]]
 uint8x8_t test_vld1_lane_u8(uint8_t  *a, uint8x8_t b) {
   return vld1_lane_u8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} <4 x i16> @test_vld1_lane_u16(i16* noundef %a, <4 x i16> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i16* %a to i8*
+// CHECK-LABEL: define{{.*}} <4 x i16> @test_vld1_lane_u16(ptr noundef %a, <4 x i16> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i16> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <4 x i16>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to i16*
-// CHECK:   [[TMP4:%.*]] = load i16, i16* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load i16, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <4 x i16> [[TMP2]], i16 [[TMP4]], i32 3
 // CHECK:   ret <4 x i16> [[VLD1_LANE]]
 uint16x4_t test_vld1_lane_u16(uint16_t  *a, uint16x4_t b) {
   return vld1_lane_u16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} <2 x i32> @test_vld1_lane_u32(i32* noundef %a, <2 x i32> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i32* %a to i8*
+// CHECK-LABEL: define{{.*}} <2 x i32> @test_vld1_lane_u32(ptr noundef %a, <2 x i32> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x i32> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <2 x i32>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to i32*
-// CHECK:   [[TMP4:%.*]] = load i32, i32* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load i32, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <2 x i32> [[TMP2]], i32 [[TMP4]], i32 1
 // CHECK:   ret <2 x i32> [[VLD1_LANE]]
 uint32x2_t test_vld1_lane_u32(uint32_t  *a, uint32x2_t b) {
   return vld1_lane_u32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} <1 x i64> @test_vld1_lane_u64(i64* noundef %a, <1 x i64> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i64* %a to i8*
+// CHECK-LABEL: define{{.*}} <1 x i64> @test_vld1_lane_u64(ptr noundef %a, <1 x i64> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <1 x i64> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <1 x i64>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to i64*
-// CHECK:   [[TMP4:%.*]] = load i64, i64* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load i64, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <1 x i64> [[TMP2]], i64 [[TMP4]], i32 0
 // CHECK:   ret <1 x i64> [[VLD1_LANE]]
 uint64x1_t test_vld1_lane_u64(uint64_t  *a, uint64x1_t b) {
   return vld1_lane_u64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} <8 x i8> @test_vld1_lane_s8(i8* noundef %a, <8 x i8> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = load i8, i8* %a
+// CHECK-LABEL: define{{.*}} <8 x i8> @test_vld1_lane_s8(ptr noundef %a, <8 x i8> noundef %b) #1 {
+// CHECK:   [[TMP0:%.*]] = load i8, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <8 x i8> %b, i8 [[TMP0]], i32 7
 // CHECK:   ret <8 x i8> [[VLD1_LANE]]
 int8x8_t test_vld1_lane_s8(int8_t  *a, int8x8_t b) {
   return vld1_lane_s8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} <4 x i16> @test_vld1_lane_s16(i16* noundef %a, <4 x i16> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i16* %a to i8*
+// CHECK-LABEL: define{{.*}} <4 x i16> @test_vld1_lane_s16(ptr noundef %a, <4 x i16> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i16> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <4 x i16>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to i16*
-// CHECK:   [[TMP4:%.*]] = load i16, i16* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load i16, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <4 x i16> [[TMP2]], i16 [[TMP4]], i32 3
 // CHECK:   ret <4 x i16> [[VLD1_LANE]]
 int16x4_t test_vld1_lane_s16(int16_t  *a, int16x4_t b) {
   return vld1_lane_s16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} <2 x i32> @test_vld1_lane_s32(i32* noundef %a, <2 x i32> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i32* %a to i8*
+// CHECK-LABEL: define{{.*}} <2 x i32> @test_vld1_lane_s32(ptr noundef %a, <2 x i32> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x i32> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <2 x i32>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to i32*
-// CHECK:   [[TMP4:%.*]] = load i32, i32* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load i32, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <2 x i32> [[TMP2]], i32 [[TMP4]], i32 1
 // CHECK:   ret <2 x i32> [[VLD1_LANE]]
 int32x2_t test_vld1_lane_s32(int32_t  *a, int32x2_t b) {
   return vld1_lane_s32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} <1 x i64> @test_vld1_lane_s64(i64* noundef %a, <1 x i64> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i64* %a to i8*
+// CHECK-LABEL: define{{.*}} <1 x i64> @test_vld1_lane_s64(ptr noundef %a, <1 x i64> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <1 x i64> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <1 x i64>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to i64*
-// CHECK:   [[TMP4:%.*]] = load i64, i64* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load i64, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <1 x i64> [[TMP2]], i64 [[TMP4]], i32 0
 // CHECK:   ret <1 x i64> [[VLD1_LANE]]
 int64x1_t test_vld1_lane_s64(int64_t  *a, int64x1_t b) {
   return vld1_lane_s64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} <4 x half> @test_vld1_lane_f16(half* noundef %a, <4 x half> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast half* %a to i8*
+// CHECK-LABEL: define{{.*}} <4 x half> @test_vld1_lane_f16(ptr noundef %a, <4 x half> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x half> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <4 x half>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to half*
-// CHECK:   [[TMP4:%.*]] = load half, half* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load half, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <4 x half> [[TMP2]], half [[TMP4]], i32 3
 // CHECK:   ret <4 x half> [[VLD1_LANE]]
 float16x4_t test_vld1_lane_f16(float16_t  *a, float16x4_t b) {
   return vld1_lane_f16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} <2 x float> @test_vld1_lane_f32(float* noundef %a, <2 x float> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast float* %a to i8*
+// CHECK-LABEL: define{{.*}} <2 x float> @test_vld1_lane_f32(ptr noundef %a, <2 x float> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x float> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <2 x float>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to float*
-// CHECK:   [[TMP4:%.*]] = load float, float* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load float, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <2 x float> [[TMP2]], float [[TMP4]], i32 1
 // CHECK:   ret <2 x float> [[VLD1_LANE]]
 float32x2_t test_vld1_lane_f32(float32_t  *a, float32x2_t b) {
   return vld1_lane_f32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} <1 x double> @test_vld1_lane_f64(double* noundef %a, <1 x double> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast double* %a to i8*
+// CHECK-LABEL: define{{.*}} <1 x double> @test_vld1_lane_f64(ptr noundef %a, <1 x double> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <1 x double> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <1 x double>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to double*
-// CHECK:   [[TMP4:%.*]] = load double, double* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load double, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <1 x double> [[TMP2]], double [[TMP4]], i32 0
 // CHECK:   ret <1 x double> [[VLD1_LANE]]
 float64x1_t test_vld1_lane_f64(float64_t  *a, float64x1_t b) {
   return vld1_lane_f64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} <8 x i8> @test_vld1_lane_p8(i8* noundef %a, <8 x i8> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = load i8, i8* %a
+// CHECK-LABEL: define{{.*}} <8 x i8> @test_vld1_lane_p8(ptr noundef %a, <8 x i8> noundef %b) #1 {
+// CHECK:   [[TMP0:%.*]] = load i8, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <8 x i8> %b, i8 [[TMP0]], i32 7
 // CHECK:   ret <8 x i8> [[VLD1_LANE]]
 poly8x8_t test_vld1_lane_p8(poly8_t  *a, poly8x8_t b) {
   return vld1_lane_p8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} <4 x i16> @test_vld1_lane_p16(i16* noundef %a, <4 x i16> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i16* %a to i8*
+// CHECK-LABEL: define{{.*}} <4 x i16> @test_vld1_lane_p16(ptr noundef %a, <4 x i16> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i16> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <4 x i16>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to i16*
-// CHECK:   [[TMP4:%.*]] = load i16, i16* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load i16, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <4 x i16> [[TMP2]], i16 [[TMP4]], i32 3
 // CHECK:   ret <4 x i16> [[VLD1_LANE]]
 poly16x4_t test_vld1_lane_p16(poly16_t  *a, poly16x4_t b) {
   return vld1_lane_p16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} <1 x i64> @test_vld1_lane_p64(i64* noundef %a, <1 x i64> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i64* %a to i8*
+// CHECK-LABEL: define{{.*}} <1 x i64> @test_vld1_lane_p64(ptr noundef %a, <1 x i64> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <1 x i64> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <1 x i64>
-// CHECK:   [[TMP3:%.*]] = bitcast i8* [[TMP0]] to i64*
-// CHECK:   [[TMP4:%.*]] = load i64, i64* [[TMP3]]
+// CHECK:   [[TMP4:%.*]] = load i64, ptr %a
 // CHECK:   [[VLD1_LANE:%.*]] = insertelement <1 x i64> [[TMP2]], i64 [[TMP4]], i32 0
 // CHECK:   ret <1 x i64> [[VLD1_LANE]]
 poly64x1_t test_vld1_lane_p64(poly64_t  *a, poly64x1_t b) {
   return vld1_lane_p64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int8x16x2_t @test_vld2q_lane_s8(i8* noundef %ptr, [2 x <16 x i8>] %src.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int8x16x2_t @test_vld2q_lane_s8(ptr noundef %ptr, [2 x <16 x i8>] %src.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int8x16x2_t, align 16
 // CHECK:   [[SRC:%.*]] = alloca %struct.int8x16x2_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.int8x16x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int8x16x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x16x2_t, %struct.int8x16x2_t* [[SRC]], i32 0, i32 0
-// CHECK:   store [2 x <16 x i8>] [[SRC]].coerce, [2 x <16 x i8>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int8x16x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int8x16x2_t* [[SRC]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int8x16x2_t* [[__RET]] to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x16x2_t, %struct.int8x16x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <16 x i8>], [2 x <16 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x16x2_t, %struct.int8x16x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <16 x i8>], [2 x <16 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16
-// CHECK:   [[VLD2_LANE:%.*]] = call { <16 x i8>, <16 x i8> } @llvm.aarch64.neon.ld2lane.v16i8.p0i8(<16 x i8> [[TMP3]], <16 x i8> [[TMP4]], i64 15, i8* %ptr)
-// CHECK:   [[TMP5:%.*]] = bitcast i8* [[TMP2]] to { <16 x i8>, <16 x i8> }*
-// CHECK:   store { <16 x i8>, <16 x i8> } [[VLD2_LANE]], { <16 x i8>, <16 x i8> }* [[TMP5]]
-// CHECK:   [[TMP6:%.*]] = bitcast %struct.int8x16x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP7:%.*]] = bitcast %struct.int8x16x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP6]], i8* align 16 [[TMP7]], i64 32, i1 false)
-// CHECK:   [[TMP8:%.*]] = load %struct.int8x16x2_t, %struct.int8x16x2_t* [[RETVAL]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x16x2_t, ptr [[SRC]], i32 0, i32 0
+// CHECK:   store [2 x <16 x i8>] [[SRC]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[SRC]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x16x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <16 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <16 x i8>, ptr [[ARRAYIDX]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x16x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <16 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP4:%.*]] = load <16 x i8>, ptr [[ARRAYIDX2]], align 16
+// CHECK:   [[VLD2_LANE:%.*]] = call { <16 x i8>, <16 x i8> } @llvm.aarch64.neon.ld2lane.v16i8.p0(<16 x i8> [[TMP3]], <16 x i8> [[TMP4]], i64 15, ptr %ptr)
+// CHECK:   store { <16 x i8>, <16 x i8> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP8:%.*]] = load %struct.int8x16x2_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.int8x16x2_t [[TMP8]]
 int8x16x2_t test_vld2q_lane_s8(int8_t const * ptr, int8x16x2_t src) {
   return vld2q_lane_s8(ptr, src, 15);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint8x16x2_t @test_vld2q_lane_u8(i8* noundef %ptr, [2 x <16 x i8>] %src.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint8x16x2_t @test_vld2q_lane_u8(ptr noundef %ptr, [2 x <16 x i8>] %src.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint8x16x2_t, align 16
 // CHECK:   [[SRC:%.*]] = alloca %struct.uint8x16x2_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint8x16x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint8x16x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x16x2_t, %struct.uint8x16x2_t* [[SRC]], i32 0, i32 0
-// CHECK:   store [2 x <16 x i8>] [[SRC]].coerce, [2 x <16 x i8>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint8x16x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint8x16x2_t* [[SRC]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint8x16x2_t* [[__RET]] to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x16x2_t, %struct.uint8x16x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <16 x i8>], [2 x <16 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x16x2_t, %struct.uint8x16x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <16 x i8>], [2 x <16 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16
-// CHECK:   [[VLD2_LANE:%.*]] = call { <16 x i8>, <16 x i8> } @llvm.aarch64.neon.ld2lane.v16i8.p0i8(<16 x i8> [[TMP3]], <16 x i8> [[TMP4]], i64 15, i8* %ptr)
-// CHECK:   [[TMP5:%.*]] = bitcast i8* [[TMP2]] to { <16 x i8>, <16 x i8> }*
-// CHECK:   store { <16 x i8>, <16 x i8> } [[VLD2_LANE]], { <16 x i8>, <16 x i8> }* [[TMP5]]
-// CHECK:   [[TMP6:%.*]] = bitcast %struct.uint8x16x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP7:%.*]] = bitcast %struct.uint8x16x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP6]], i8* align 16 [[TMP7]], i64 32, i1 false)
-// CHECK:   [[TMP8:%.*]] = load %struct.uint8x16x2_t, %struct.uint8x16x2_t* [[RETVAL]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x16x2_t, ptr [[SRC]], i32 0, i32 0
+// CHECK:   store [2 x <16 x i8>] [[SRC]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[SRC]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x16x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <16 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <16 x i8>, ptr [[ARRAYIDX]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x16x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <16 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP4:%.*]] = load <16 x i8>, ptr [[ARRAYIDX2]], align 16
+// CHECK:   [[VLD2_LANE:%.*]] = call { <16 x i8>, <16 x i8> } @llvm.aarch64.neon.ld2lane.v16i8.p0(<16 x i8> [[TMP3]], <16 x i8> [[TMP4]], i64 15, ptr %ptr)
+// CHECK:   store { <16 x i8>, <16 x i8> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP8:%.*]] = load %struct.uint8x16x2_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.uint8x16x2_t [[TMP8]]
 uint8x16x2_t test_vld2q_lane_u8(uint8_t const * ptr, uint8x16x2_t src) {
   return vld2q_lane_u8(ptr, src, 15);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly8x16x2_t @test_vld2q_lane_p8(i8* noundef %ptr, [2 x <16 x i8>] %src.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly8x16x2_t @test_vld2q_lane_p8(ptr noundef %ptr, [2 x <16 x i8>] %src.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly8x16x2_t, align 16
 // CHECK:   [[SRC:%.*]] = alloca %struct.poly8x16x2_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly8x16x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly8x16x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x16x2_t, %struct.poly8x16x2_t* [[SRC]], i32 0, i32 0
-// CHECK:   store [2 x <16 x i8>] [[SRC]].coerce, [2 x <16 x i8>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly8x16x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly8x16x2_t* [[SRC]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.poly8x16x2_t* [[__RET]] to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x16x2_t, %struct.poly8x16x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <16 x i8>], [2 x <16 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x16x2_t, %struct.poly8x16x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <16 x i8>], [2 x <16 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16
-// CHECK:   [[VLD2_LANE:%.*]] = call { <16 x i8>, <16 x i8> } @llvm.aarch64.neon.ld2lane.v16i8.p0i8(<16 x i8> [[TMP3]], <16 x i8> [[TMP4]], i64 15, i8* %ptr)
-// CHECK:   [[TMP5:%.*]] = bitcast i8* [[TMP2]] to { <16 x i8>, <16 x i8> }*
-// CHECK:   store { <16 x i8>, <16 x i8> } [[VLD2_LANE]], { <16 x i8>, <16 x i8> }* [[TMP5]]
-// CHECK:   [[TMP6:%.*]] = bitcast %struct.poly8x16x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP7:%.*]] = bitcast %struct.poly8x16x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP6]], i8* align 16 [[TMP7]], i64 32, i1 false)
-// CHECK:   [[TMP8:%.*]] = load %struct.poly8x16x2_t, %struct.poly8x16x2_t* [[RETVAL]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x16x2_t, ptr [[SRC]], i32 0, i32 0
+// CHECK:   store [2 x <16 x i8>] [[SRC]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[SRC]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x16x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <16 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <16 x i8>, ptr [[ARRAYIDX]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x16x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <16 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP4:%.*]] = load <16 x i8>, ptr [[ARRAYIDX2]], align 16
+// CHECK:   [[VLD2_LANE:%.*]] = call { <16 x i8>, <16 x i8> } @llvm.aarch64.neon.ld2lane.v16i8.p0(<16 x i8> [[TMP3]], <16 x i8> [[TMP4]], i64 15, ptr %ptr)
+// CHECK:   store { <16 x i8>, <16 x i8> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP8:%.*]] = load %struct.poly8x16x2_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.poly8x16x2_t [[TMP8]]
 poly8x16x2_t test_vld2q_lane_p8(poly8_t const * ptr, poly8x16x2_t src) {
   return vld2q_lane_p8(ptr, src, 15);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int8x16x3_t @test_vld3q_lane_s8(i8* noundef %ptr, [3 x <16 x i8>] %src.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int8x16x3_t @test_vld3q_lane_s8(ptr noundef %ptr, [3 x <16 x i8>] %src.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int8x16x3_t, align 16
 // CHECK:   [[SRC:%.*]] = alloca %struct.int8x16x3_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.int8x16x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int8x16x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x16x3_t, %struct.int8x16x3_t* [[SRC]], i32 0, i32 0
-// CHECK:   store [3 x <16 x i8>] [[SRC]].coerce, [3 x <16 x i8>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int8x16x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int8x16x3_t* [[SRC]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int8x16x3_t* [[__RET]] to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x16x3_t, %struct.int8x16x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x16x3_t, %struct.int8x16x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int8x16x3_t, %struct.int8x16x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 16
-// CHECK:   [[VLD3_LANE:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8> } @llvm.aarch64.neon.ld3lane.v16i8.p0i8(<16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], i64 15, i8* %ptr)
-// CHECK:   [[TMP6:%.*]] = bitcast i8* [[TMP2]] to { <16 x i8>, <16 x i8>, <16 x i8> }*
-// CHECK:   store { <16 x i8>, <16 x i8>, <16 x i8> } [[VLD3_LANE]], { <16 x i8>, <16 x i8>, <16 x i8> }* [[TMP6]]
-// CHECK:   [[TMP7:%.*]] = bitcast %struct.int8x16x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP8:%.*]] = bitcast %struct.int8x16x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP7]], i8* align 16 [[TMP8]], i64 48, i1 false)
-// CHECK:   [[TMP9:%.*]] = load %struct.int8x16x3_t, %struct.int8x16x3_t* [[RETVAL]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x16x3_t, ptr [[SRC]], i32 0, i32 0
+// CHECK:   store [3 x <16 x i8>] [[SRC]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[SRC]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x16x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <16 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <16 x i8>, ptr [[ARRAYIDX]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x16x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <16 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP4:%.*]] = load <16 x i8>, ptr [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int8x16x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <16 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP5:%.*]] = load <16 x i8>, ptr [[ARRAYIDX4]], align 16
+// CHECK:   [[VLD3_LANE:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8> } @llvm.aarch64.neon.ld3lane.v16i8.p0(<16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], i64 15, ptr %ptr)
+// CHECK:   store { <16 x i8>, <16 x i8>, <16 x i8> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 48, i1 false)
+// CHECK:   [[TMP9:%.*]] = load %struct.int8x16x3_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.int8x16x3_t [[TMP9]]
 int8x16x3_t test_vld3q_lane_s8(int8_t const * ptr, int8x16x3_t src) {
   return vld3q_lane_s8(ptr, src, 15);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint8x16x3_t @test_vld3q_lane_u8(i8* noundef %ptr, [3 x <16 x i8>] %src.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint8x16x3_t @test_vld3q_lane_u8(ptr noundef %ptr, [3 x <16 x i8>] %src.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint8x16x3_t, align 16
 // CHECK:   [[SRC:%.*]] = alloca %struct.uint8x16x3_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint8x16x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint8x16x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x16x3_t, %struct.uint8x16x3_t* [[SRC]], i32 0, i32 0
-// CHECK:   store [3 x <16 x i8>] [[SRC]].coerce, [3 x <16 x i8>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint8x16x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint8x16x3_t* [[SRC]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint8x16x3_t* [[__RET]] to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x16x3_t, %struct.uint8x16x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x16x3_t, %struct.uint8x16x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint8x16x3_t, %struct.uint8x16x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 16
-// CHECK:   [[VLD3_LANE:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8> } @llvm.aarch64.neon.ld3lane.v16i8.p0i8(<16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], i64 15, i8* %ptr)
-// CHECK:   [[TMP6:%.*]] = bitcast i8* [[TMP2]] to { <16 x i8>, <16 x i8>, <16 x i8> }*
-// CHECK:   store { <16 x i8>, <16 x i8>, <16 x i8> } [[VLD3_LANE]], { <16 x i8>, <16 x i8>, <16 x i8> }* [[TMP6]]
-// CHECK:   [[TMP7:%.*]] = bitcast %struct.uint8x16x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP8:%.*]] = bitcast %struct.uint8x16x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP7]], i8* align 16 [[TMP8]], i64 48, i1 false)
-// CHECK:   [[TMP9:%.*]] = load %struct.uint8x16x3_t, %struct.uint8x16x3_t* [[RETVAL]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x16x3_t, ptr [[SRC]], i32 0, i32 0
+// CHECK:   store [3 x <16 x i8>] [[SRC]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[SRC]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x16x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <16 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <16 x i8>, ptr [[ARRAYIDX]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x16x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <16 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP4:%.*]] = load <16 x i8>, ptr [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint8x16x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <16 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP5:%.*]] = load <16 x i8>, ptr [[ARRAYIDX4]], align 16
+// CHECK:   [[VLD3_LANE:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8> } @llvm.aarch64.neon.ld3lane.v16i8.p0(<16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], i64 15, ptr %ptr)
+// CHECK:   store { <16 x i8>, <16 x i8>, <16 x i8> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 48, i1 false)
+// CHECK:   [[TMP9:%.*]] = load %struct.uint8x16x3_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.uint8x16x3_t [[TMP9]]
 uint8x16x3_t test_vld3q_lane_u8(uint8_t const * ptr, uint8x16x3_t src) {
   return vld3q_lane_u8(ptr, src, 15);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint16x8x2_t @test_vld2q_lane_u16(i16* noundef %a, [2 x <8 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint16x8x2_t @test_vld2q_lane_u16(ptr noundef %a, [2 x <8 x i16>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint16x8x2_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.uint16x8x2_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint16x8x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint16x8x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, %struct.uint16x8x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <8 x i16>] [[B]].coerce, [2 x <8 x i16>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint16x8x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint16x8x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint16x8x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, %struct.uint16x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <8 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <8 x i16>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, %struct.uint16x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <8 x i16>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <8 x i16>, <8 x i16> } @llvm.aarch64.neon.ld2lane.v8i16.p0i8(<8 x i16> [[TMP8]], <8 x i16> [[TMP9]], i64 7, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <8 x i16>, <8 x i16> }*
-// CHECK:   store { <8 x i16>, <8 x i16> } [[VLD2_LANE]], { <8 x i16>, <8 x i16> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.uint16x8x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.uint16x8x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP11]], i8* align 16 [[TMP12]], i64 32, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.uint16x8x2_t, %struct.uint16x8x2_t* [[RETVAL]], align 16
+// CHECK:   [[VLD2_LANE:%.*]] = call { <8 x i16>, <8 x i16> } @llvm.aarch64.neon.ld2lane.v8i16.p0(<8 x i16> [[TMP8]], <8 x i16> [[TMP9]], i64 7, ptr %a)
+// CHECK:   store { <8 x i16>, <8 x i16> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.uint16x8x2_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.uint16x8x2_t [[TMP13]]
 uint16x8x2_t test_vld2q_lane_u16(uint16_t  *a, uint16x8x2_t b) {
   return vld2q_lane_u16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint32x4x2_t @test_vld2q_lane_u32(i32* noundef %a, [2 x <4 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint32x4x2_t @test_vld2q_lane_u32(ptr noundef %a, [2 x <4 x i32>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint32x4x2_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.uint32x4x2_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint32x4x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint32x4x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, %struct.uint32x4x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <4 x i32>] [[B]].coerce, [2 x <4 x i32>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint32x4x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint32x4x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint32x4x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, %struct.uint32x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <4 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x i32>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x i32> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, %struct.uint32x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x i32>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x i32> [[TMP6]] to <16 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x i32>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x i32>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2lane.v4i32.p0i8(<4 x i32> [[TMP8]], <4 x i32> [[TMP9]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <4 x i32>, <4 x i32> }*
-// CHECK:   store { <4 x i32>, <4 x i32> } [[VLD2_LANE]], { <4 x i32>, <4 x i32> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.uint32x4x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.uint32x4x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP11]], i8* align 16 [[TMP12]], i64 32, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.uint32x4x2_t, %struct.uint32x4x2_t* [[RETVAL]], align 16
+// CHECK:   [[VLD2_LANE:%.*]] = call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2lane.v4i32.p0(<4 x i32> [[TMP8]], <4 x i32> [[TMP9]], i64 3, ptr %a)
+// CHECK:   store { <4 x i32>, <4 x i32> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.uint32x4x2_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.uint32x4x2_t [[TMP13]]
 uint32x4x2_t test_vld2q_lane_u32(uint32_t  *a, uint32x4x2_t b) {
   return vld2q_lane_u32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint64x2x2_t @test_vld2q_lane_u64(i64* noundef %a, [2 x <2 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint64x2x2_t @test_vld2q_lane_u64(ptr noundef %a, [2 x <2 x i64>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint64x2x2_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.uint64x2x2_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint64x2x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint64x2x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x2x2_t, %struct.uint64x2x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <2 x i64>] [[B]].coerce, [2 x <2 x i64>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint64x2x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint64x2x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint64x2x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x2x2_t, %struct.uint64x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x i64>], [2 x <2 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x2x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <2 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x i64>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x i64> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x2x2_t, %struct.uint64x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x i64>], [2 x <2 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x i64>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x i64> [[TMP6]] to <16 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <2 x i64>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <2 x i64>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld2lane.v2i64.p0i8(<2 x i64> [[TMP8]], <2 x i64> [[TMP9]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <2 x i64>, <2 x i64> }*
-// CHECK:   store { <2 x i64>, <2 x i64> } [[VLD2_LANE]], { <2 x i64>, <2 x i64> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.uint64x2x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.uint64x2x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP11]], i8* align 16 [[TMP12]], i64 32, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.uint64x2x2_t, %struct.uint64x2x2_t* [[RETVAL]], align 16
+// CHECK:   [[VLD2_LANE:%.*]] = call { <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld2lane.v2i64.p0(<2 x i64> [[TMP8]], <2 x i64> [[TMP9]], i64 1, ptr %a)
+// CHECK:   store { <2 x i64>, <2 x i64> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.uint64x2x2_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.uint64x2x2_t [[TMP13]]
 uint64x2x2_t test_vld2q_lane_u64(uint64_t  *a, uint64x2x2_t b) {
   return vld2q_lane_u64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int16x8x2_t @test_vld2q_lane_s16(i16* noundef %a, [2 x <8 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int16x8x2_t @test_vld2q_lane_s16(ptr noundef %a, [2 x <8 x i16>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int16x8x2_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.int16x8x2_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.int16x8x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int16x8x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x2_t, %struct.int16x8x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <8 x i16>] [[B]].coerce, [2 x <8 x i16>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int16x8x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int16x8x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int16x8x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x2_t, %struct.int16x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <8 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <8 x i16>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x2_t, %struct.int16x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <8 x i16>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <8 x i16>, <8 x i16> } @llvm.aarch64.neon.ld2lane.v8i16.p0i8(<8 x i16> [[TMP8]], <8 x i16> [[TMP9]], i64 7, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <8 x i16>, <8 x i16> }*
-// CHECK:   store { <8 x i16>, <8 x i16> } [[VLD2_LANE]], { <8 x i16>, <8 x i16> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.int16x8x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.int16x8x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP11]], i8* align 16 [[TMP12]], i64 32, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.int16x8x2_t, %struct.int16x8x2_t* [[RETVAL]], align 16
+// CHECK:   [[VLD2_LANE:%.*]] = call { <8 x i16>, <8 x i16> } @llvm.aarch64.neon.ld2lane.v8i16.p0(<8 x i16> [[TMP8]], <8 x i16> [[TMP9]], i64 7, ptr %a)
+// CHECK:   store { <8 x i16>, <8 x i16> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.int16x8x2_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.int16x8x2_t [[TMP13]]
 int16x8x2_t test_vld2q_lane_s16(int16_t  *a, int16x8x2_t b) {
   return vld2q_lane_s16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int32x4x2_t @test_vld2q_lane_s32(i32* noundef %a, [2 x <4 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int32x4x2_t @test_vld2q_lane_s32(ptr noundef %a, [2 x <4 x i32>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int32x4x2_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.int32x4x2_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.int32x4x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int32x4x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x2_t, %struct.int32x4x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <4 x i32>] [[B]].coerce, [2 x <4 x i32>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int32x4x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int32x4x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int32x4x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x2_t, %struct.int32x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <4 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x i32>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x i32> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x2_t, %struct.int32x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x i32>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x i32> [[TMP6]] to <16 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x i32>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x i32>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2lane.v4i32.p0i8(<4 x i32> [[TMP8]], <4 x i32> [[TMP9]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <4 x i32>, <4 x i32> }*
-// CHECK:   store { <4 x i32>, <4 x i32> } [[VLD2_LANE]], { <4 x i32>, <4 x i32> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.int32x4x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.int32x4x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP11]], i8* align 16 [[TMP12]], i64 32, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.int32x4x2_t, %struct.int32x4x2_t* [[RETVAL]], align 16
+// CHECK:   [[VLD2_LANE:%.*]] = call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2lane.v4i32.p0(<4 x i32> [[TMP8]], <4 x i32> [[TMP9]], i64 3, ptr %a)
+// CHECK:   store { <4 x i32>, <4 x i32> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.int32x4x2_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.int32x4x2_t [[TMP13]]
 int32x4x2_t test_vld2q_lane_s32(int32_t  *a, int32x4x2_t b) {
   return vld2q_lane_s32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int64x2x2_t @test_vld2q_lane_s64(i64* noundef %a, [2 x <2 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int64x2x2_t @test_vld2q_lane_s64(ptr noundef %a, [2 x <2 x i64>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int64x2x2_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.int64x2x2_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.int64x2x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int64x2x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x2x2_t, %struct.int64x2x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <2 x i64>] [[B]].coerce, [2 x <2 x i64>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int64x2x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int64x2x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int64x2x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x2x2_t, %struct.int64x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x i64>], [2 x <2 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x2x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <2 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x i64>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x i64> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x2x2_t, %struct.int64x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x i64>], [2 x <2 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x i64>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x i64> [[TMP6]] to <16 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <2 x i64>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <2 x i64>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld2lane.v2i64.p0i8(<2 x i64> [[TMP8]], <2 x i64> [[TMP9]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <2 x i64>, <2 x i64> }*
-// CHECK:   store { <2 x i64>, <2 x i64> } [[VLD2_LANE]], { <2 x i64>, <2 x i64> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.int64x2x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.int64x2x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP11]], i8* align 16 [[TMP12]], i64 32, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.int64x2x2_t, %struct.int64x2x2_t* [[RETVAL]], align 16
+// CHECK:   [[VLD2_LANE:%.*]] = call { <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld2lane.v2i64.p0(<2 x i64> [[TMP8]], <2 x i64> [[TMP9]], i64 1, ptr %a)
+// CHECK:   store { <2 x i64>, <2 x i64> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.int64x2x2_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.int64x2x2_t [[TMP13]]
 int64x2x2_t test_vld2q_lane_s64(int64_t  *a, int64x2x2_t b) {
   return vld2q_lane_s64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float16x8x2_t @test_vld2q_lane_f16(half* noundef %a, [2 x <8 x half>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float16x8x2_t @test_vld2q_lane_f16(ptr noundef %a, [2 x <8 x half>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float16x8x2_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.float16x8x2_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.float16x8x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.float16x8x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x2_t, %struct.float16x8x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <8 x half>] [[B]].coerce, [2 x <8 x half>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float16x8x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float16x8x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.float16x8x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast half* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x2_t, %struct.float16x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x half>], [2 x <8 x half>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <8 x half>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x half>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <8 x half>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <8 x half> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x2_t, %struct.float16x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x half>], [2 x <8 x half>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x half>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <8 x half>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x half> [[TMP6]] to <16 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x half>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x half>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <8 x half>, <8 x half> } @llvm.aarch64.neon.ld2lane.v8f16.p0i8(<8 x half> [[TMP8]], <8 x half> [[TMP9]], i64 7, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <8 x half>, <8 x half> }*
-// CHECK:   store { <8 x half>, <8 x half> } [[VLD2_LANE]], { <8 x half>, <8 x half> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.float16x8x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.float16x8x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP11]], i8* align 16 [[TMP12]], i64 32, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.float16x8x2_t, %struct.float16x8x2_t* [[RETVAL]], align 16
+// CHECK:   [[VLD2_LANE:%.*]] = call { <8 x half>, <8 x half> } @llvm.aarch64.neon.ld2lane.v8f16.p0(<8 x half> [[TMP8]], <8 x half> [[TMP9]], i64 7, ptr %a)
+// CHECK:   store { <8 x half>, <8 x half> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.float16x8x2_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.float16x8x2_t [[TMP13]]
 float16x8x2_t test_vld2q_lane_f16(float16_t  *a, float16x8x2_t b) {
   return vld2q_lane_f16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float32x4x2_t @test_vld2q_lane_f32(float* noundef %a, [2 x <4 x float>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float32x4x2_t @test_vld2q_lane_f32(ptr noundef %a, [2 x <4 x float>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float32x4x2_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.float32x4x2_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.float32x4x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.float32x4x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x2_t, %struct.float32x4x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <4 x float>] [[B]].coerce, [2 x <4 x float>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float32x4x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float32x4x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.float32x4x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast float* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x2_t, %struct.float32x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x float>], [2 x <4 x float>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <4 x float>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x float>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x float>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x float> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x2_t, %struct.float32x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x float>], [2 x <4 x float>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x float>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x float>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x float> [[TMP6]] to <16 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x float>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x float>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <4 x float>, <4 x float> } @llvm.aarch64.neon.ld2lane.v4f32.p0i8(<4 x float> [[TMP8]], <4 x float> [[TMP9]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <4 x float>, <4 x float> }*
-// CHECK:   store { <4 x float>, <4 x float> } [[VLD2_LANE]], { <4 x float>, <4 x float> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.float32x4x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.float32x4x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP11]], i8* align 16 [[TMP12]], i64 32, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.float32x4x2_t, %struct.float32x4x2_t* [[RETVAL]], align 16
+// CHECK:   [[VLD2_LANE:%.*]] = call { <4 x float>, <4 x float> } @llvm.aarch64.neon.ld2lane.v4f32.p0(<4 x float> [[TMP8]], <4 x float> [[TMP9]], i64 3, ptr %a)
+// CHECK:   store { <4 x float>, <4 x float> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.float32x4x2_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.float32x4x2_t [[TMP13]]
 float32x4x2_t test_vld2q_lane_f32(float32_t  *a, float32x4x2_t b) {
   return vld2q_lane_f32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float64x2x2_t @test_vld2q_lane_f64(double* noundef %a, [2 x <2 x double>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float64x2x2_t @test_vld2q_lane_f64(ptr noundef %a, [2 x <2 x double>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float64x2x2_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.float64x2x2_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.float64x2x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.float64x2x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x2x2_t, %struct.float64x2x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <2 x double>] [[B]].coerce, [2 x <2 x double>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float64x2x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float64x2x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.float64x2x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast double* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x2x2_t, %struct.float64x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x double>], [2 x <2 x double>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x double>, <2 x double>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x2x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <2 x double>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x double>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x double>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x double> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x2x2_t, %struct.float64x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x double>], [2 x <2 x double>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x double>, <2 x double>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x double>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x double>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x double> [[TMP6]] to <16 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <2 x double>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <2 x double>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <2 x double>, <2 x double> } @llvm.aarch64.neon.ld2lane.v2f64.p0i8(<2 x double> [[TMP8]], <2 x double> [[TMP9]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <2 x double>, <2 x double> }*
-// CHECK:   store { <2 x double>, <2 x double> } [[VLD2_LANE]], { <2 x double>, <2 x double> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.float64x2x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.float64x2x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP11]], i8* align 16 [[TMP12]], i64 32, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.float64x2x2_t, %struct.float64x2x2_t* [[RETVAL]], align 16
+// CHECK:   [[VLD2_LANE:%.*]] = call { <2 x double>, <2 x double> } @llvm.aarch64.neon.ld2lane.v2f64.p0(<2 x double> [[TMP8]], <2 x double> [[TMP9]], i64 1, ptr %a)
+// CHECK:   store { <2 x double>, <2 x double> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.float64x2x2_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.float64x2x2_t [[TMP13]]
 float64x2x2_t test_vld2q_lane_f64(float64_t  *a, float64x2x2_t b) {
   return vld2q_lane_f64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly16x8x2_t @test_vld2q_lane_p16(i16* noundef %a, [2 x <8 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly16x8x2_t @test_vld2q_lane_p16(ptr noundef %a, [2 x <8 x i16>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly16x8x2_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.poly16x8x2_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly16x8x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly16x8x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, %struct.poly16x8x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <8 x i16>] [[B]].coerce, [2 x <8 x i16>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly16x8x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly16x8x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.poly16x8x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, %struct.poly16x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <8 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <8 x i16>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, %struct.poly16x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <8 x i16>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <8 x i16>, <8 x i16> } @llvm.aarch64.neon.ld2lane.v8i16.p0i8(<8 x i16> [[TMP8]], <8 x i16> [[TMP9]], i64 7, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <8 x i16>, <8 x i16> }*
-// CHECK:   store { <8 x i16>, <8 x i16> } [[VLD2_LANE]], { <8 x i16>, <8 x i16> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.poly16x8x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.poly16x8x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP11]], i8* align 16 [[TMP12]], i64 32, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.poly16x8x2_t, %struct.poly16x8x2_t* [[RETVAL]], align 16
+// CHECK:   [[VLD2_LANE:%.*]] = call { <8 x i16>, <8 x i16> } @llvm.aarch64.neon.ld2lane.v8i16.p0(<8 x i16> [[TMP8]], <8 x i16> [[TMP9]], i64 7, ptr %a)
+// CHECK:   store { <8 x i16>, <8 x i16> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.poly16x8x2_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.poly16x8x2_t [[TMP13]]
 poly16x8x2_t test_vld2q_lane_p16(poly16_t  *a, poly16x8x2_t b) {
   return vld2q_lane_p16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly64x2x2_t @test_vld2q_lane_p64(i64* noundef %a, [2 x <2 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly64x2x2_t @test_vld2q_lane_p64(ptr noundef %a, [2 x <2 x i64>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly64x2x2_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.poly64x2x2_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly64x2x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly64x2x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x2x2_t, %struct.poly64x2x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <2 x i64>] [[B]].coerce, [2 x <2 x i64>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly64x2x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly64x2x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.poly64x2x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x2x2_t, %struct.poly64x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x i64>], [2 x <2 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x2x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <2 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x i64>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x i64> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x2x2_t, %struct.poly64x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x i64>], [2 x <2 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x i64>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x i64> [[TMP6]] to <16 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP5]] to <2 x i64>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP7]] to <2 x i64>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld2lane.v2i64.p0i8(<2 x i64> [[TMP8]], <2 x i64> [[TMP9]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <2 x i64>, <2 x i64> }*
-// CHECK:   store { <2 x i64>, <2 x i64> } [[VLD2_LANE]], { <2 x i64>, <2 x i64> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.poly64x2x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.poly64x2x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP11]], i8* align 16 [[TMP12]], i64 32, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.poly64x2x2_t, %struct.poly64x2x2_t* [[RETVAL]], align 16
+// CHECK:   [[VLD2_LANE:%.*]] = call { <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld2lane.v2i64.p0(<2 x i64> [[TMP8]], <2 x i64> [[TMP9]], i64 1, ptr %a)
+// CHECK:   store { <2 x i64>, <2 x i64> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.poly64x2x2_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.poly64x2x2_t [[TMP13]]
 poly64x2x2_t test_vld2q_lane_p64(poly64_t  *a, poly64x2x2_t b) {
   return vld2q_lane_p64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint8x8x2_t @test_vld2_lane_u8(i8* noundef %a, [2 x <8 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint8x8x2_t @test_vld2_lane_u8(ptr noundef %a, [2 x <8 x i8>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint8x8x2_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.uint8x8x2_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint8x8x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint8x8x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x8x2_t, %struct.uint8x8x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <8 x i8>] [[B]].coerce, [2 x <8 x i8>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint8x8x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint8x8x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint8x8x2_t* [[__RET]] to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x8x2_t, %struct.uint8x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i8>], [2 x <8 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX]], align 8
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x8x2_t, %struct.uint8x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i8>], [2 x <8 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP4:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX2]], align 8
-// CHECK:   [[VLD2_LANE:%.*]] = call { <8 x i8>, <8 x i8> } @llvm.aarch64.neon.ld2lane.v8i8.p0i8(<8 x i8> [[TMP3]], <8 x i8> [[TMP4]], i64 7, i8* %a)
-// CHECK:   [[TMP5:%.*]] = bitcast i8* [[TMP2]] to { <8 x i8>, <8 x i8> }*
-// CHECK:   store { <8 x i8>, <8 x i8> } [[VLD2_LANE]], { <8 x i8>, <8 x i8> }* [[TMP5]]
-// CHECK:   [[TMP6:%.*]] = bitcast %struct.uint8x8x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP7:%.*]] = bitcast %struct.uint8x8x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP6]], i8* align 8 [[TMP7]], i64 16, i1 false)
-// CHECK:   [[TMP8:%.*]] = load %struct.uint8x8x2_t, %struct.uint8x8x2_t* [[RETVAL]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x8x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <8 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x i8>, ptr [[ARRAYIDX]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP4:%.*]] = load <8 x i8>, ptr [[ARRAYIDX2]], align 8
+// CHECK:   [[VLD2_LANE:%.*]] = call { <8 x i8>, <8 x i8> } @llvm.aarch64.neon.ld2lane.v8i8.p0(<8 x i8> [[TMP3]], <8 x i8> [[TMP4]], i64 7, ptr %a)
+// CHECK:   store { <8 x i8>, <8 x i8> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 16, i1 false)
+// CHECK:   [[TMP8:%.*]] = load %struct.uint8x8x2_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.uint8x8x2_t [[TMP8]]
 uint8x8x2_t test_vld2_lane_u8(uint8_t  *a, uint8x8x2_t b) {
   return vld2_lane_u8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint16x4x2_t @test_vld2_lane_u16(i16* noundef %a, [2 x <4 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint16x4x2_t @test_vld2_lane_u16(ptr noundef %a, [2 x <4 x i16>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint16x4x2_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.uint16x4x2_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint16x4x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint16x4x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x4x2_t, %struct.uint16x4x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <4 x i16>] [[B]].coerce, [2 x <4 x i16>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint16x4x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint16x4x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint16x4x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x4x2_t, %struct.uint16x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i16>], [2 x <4 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x4x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <4 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x i16>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x i16> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x4x2_t, %struct.uint16x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i16>], [2 x <4 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x i16>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x i16> [[TMP6]] to <8 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP5]] to <4 x i16>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP7]] to <4 x i16>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <4 x i16>, <4 x i16> } @llvm.aarch64.neon.ld2lane.v4i16.p0i8(<4 x i16> [[TMP8]], <4 x i16> [[TMP9]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <4 x i16>, <4 x i16> }*
-// CHECK:   store { <4 x i16>, <4 x i16> } [[VLD2_LANE]], { <4 x i16>, <4 x i16> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.uint16x4x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.uint16x4x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP11]], i8* align 8 [[TMP12]], i64 16, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.uint16x4x2_t, %struct.uint16x4x2_t* [[RETVAL]], align 8
+// CHECK:   [[VLD2_LANE:%.*]] = call { <4 x i16>, <4 x i16> } @llvm.aarch64.neon.ld2lane.v4i16.p0(<4 x i16> [[TMP8]], <4 x i16> [[TMP9]], i64 3, ptr %a)
+// CHECK:   store { <4 x i16>, <4 x i16> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 16, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.uint16x4x2_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.uint16x4x2_t [[TMP13]]
 uint16x4x2_t test_vld2_lane_u16(uint16_t  *a, uint16x4x2_t b) {
   return vld2_lane_u16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint32x2x2_t @test_vld2_lane_u32(i32* noundef %a, [2 x <2 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint32x2x2_t @test_vld2_lane_u32(ptr noundef %a, [2 x <2 x i32>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint32x2x2_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.uint32x2x2_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint32x2x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint32x2x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x2x2_t, %struct.uint32x2x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <2 x i32>] [[B]].coerce, [2 x <2 x i32>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint32x2x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint32x2x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint32x2x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x2x2_t, %struct.uint32x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x i32>], [2 x <2 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x2x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <2 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x i32>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x i32> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x2x2_t, %struct.uint32x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x i32>], [2 x <2 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x i32>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x i32> [[TMP6]] to <8 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP5]] to <2 x i32>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP7]] to <2 x i32>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <2 x i32>, <2 x i32> } @llvm.aarch64.neon.ld2lane.v2i32.p0i8(<2 x i32> [[TMP8]], <2 x i32> [[TMP9]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <2 x i32>, <2 x i32> }*
-// CHECK:   store { <2 x i32>, <2 x i32> } [[VLD2_LANE]], { <2 x i32>, <2 x i32> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.uint32x2x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.uint32x2x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP11]], i8* align 8 [[TMP12]], i64 16, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.uint32x2x2_t, %struct.uint32x2x2_t* [[RETVAL]], align 8
+// CHECK:   [[VLD2_LANE:%.*]] = call { <2 x i32>, <2 x i32> } @llvm.aarch64.neon.ld2lane.v2i32.p0(<2 x i32> [[TMP8]], <2 x i32> [[TMP9]], i64 1, ptr %a)
+// CHECK:   store { <2 x i32>, <2 x i32> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 16, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.uint32x2x2_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.uint32x2x2_t [[TMP13]]
 uint32x2x2_t test_vld2_lane_u32(uint32_t  *a, uint32x2x2_t b) {
   return vld2_lane_u32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint64x1x2_t @test_vld2_lane_u64(i64* noundef %a, [2 x <1 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint64x1x2_t @test_vld2_lane_u64(ptr noundef %a, [2 x <1 x i64>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint64x1x2_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.uint64x1x2_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint64x1x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint64x1x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x1x2_t, %struct.uint64x1x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <1 x i64>] [[B]].coerce, [2 x <1 x i64>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint64x1x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint64x1x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint64x1x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x1x2_t, %struct.uint64x1x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <1 x i64>], [2 x <1 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x1x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <1 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x1x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <1 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <1 x i64>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <1 x i64> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x1x2_t, %struct.uint64x1x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <1 x i64>], [2 x <1 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x1x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <1 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <1 x i64>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <1 x i64> [[TMP6]] to <8 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP5]] to <1 x i64>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP7]] to <1 x i64>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld2lane.v1i64.p0i8(<1 x i64> [[TMP8]], <1 x i64> [[TMP9]], i64 0, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <1 x i64>, <1 x i64> }*
-// CHECK:   store { <1 x i64>, <1 x i64> } [[VLD2_LANE]], { <1 x i64>, <1 x i64> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.uint64x1x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.uint64x1x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP11]], i8* align 8 [[TMP12]], i64 16, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.uint64x1x2_t, %struct.uint64x1x2_t* [[RETVAL]], align 8
+// CHECK:   [[VLD2_LANE:%.*]] = call { <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld2lane.v1i64.p0(<1 x i64> [[TMP8]], <1 x i64> [[TMP9]], i64 0, ptr %a)
+// CHECK:   store { <1 x i64>, <1 x i64> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 16, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.uint64x1x2_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.uint64x1x2_t [[TMP13]]
 uint64x1x2_t test_vld2_lane_u64(uint64_t  *a, uint64x1x2_t b) {
   return vld2_lane_u64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int8x8x2_t @test_vld2_lane_s8(i8* noundef %a, [2 x <8 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int8x8x2_t @test_vld2_lane_s8(ptr noundef %a, [2 x <8 x i8>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int8x8x2_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.int8x8x2_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.int8x8x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int8x8x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x8x2_t, %struct.int8x8x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <8 x i8>] [[B]].coerce, [2 x <8 x i8>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int8x8x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int8x8x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int8x8x2_t* [[__RET]] to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x8x2_t, %struct.int8x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i8>], [2 x <8 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX]], align 8
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x8x2_t, %struct.int8x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i8>], [2 x <8 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP4:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX2]], align 8
-// CHECK:   [[VLD2_LANE:%.*]] = call { <8 x i8>, <8 x i8> } @llvm.aarch64.neon.ld2lane.v8i8.p0i8(<8 x i8> [[TMP3]], <8 x i8> [[TMP4]], i64 7, i8* %a)
-// CHECK:   [[TMP5:%.*]] = bitcast i8* [[TMP2]] to { <8 x i8>, <8 x i8> }*
-// CHECK:   store { <8 x i8>, <8 x i8> } [[VLD2_LANE]], { <8 x i8>, <8 x i8> }* [[TMP5]]
-// CHECK:   [[TMP6:%.*]] = bitcast %struct.int8x8x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP7:%.*]] = bitcast %struct.int8x8x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP6]], i8* align 8 [[TMP7]], i64 16, i1 false)
-// CHECK:   [[TMP8:%.*]] = load %struct.int8x8x2_t, %struct.int8x8x2_t* [[RETVAL]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x8x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <8 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x i8>, ptr [[ARRAYIDX]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP4:%.*]] = load <8 x i8>, ptr [[ARRAYIDX2]], align 8
+// CHECK:   [[VLD2_LANE:%.*]] = call { <8 x i8>, <8 x i8> } @llvm.aarch64.neon.ld2lane.v8i8.p0(<8 x i8> [[TMP3]], <8 x i8> [[TMP4]], i64 7, ptr %a)
+// CHECK:   store { <8 x i8>, <8 x i8> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 16, i1 false)
+// CHECK:   [[TMP8:%.*]] = load %struct.int8x8x2_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.int8x8x2_t [[TMP8]]
 int8x8x2_t test_vld2_lane_s8(int8_t  *a, int8x8x2_t b) {
   return vld2_lane_s8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int16x4x2_t @test_vld2_lane_s16(i16* noundef %a, [2 x <4 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int16x4x2_t @test_vld2_lane_s16(ptr noundef %a, [2 x <4 x i16>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int16x4x2_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.int16x4x2_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.int16x4x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int16x4x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x4x2_t, %struct.int16x4x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <4 x i16>] [[B]].coerce, [2 x <4 x i16>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int16x4x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int16x4x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int16x4x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x4x2_t, %struct.int16x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i16>], [2 x <4 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x4x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <4 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x i16>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x i16> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x4x2_t, %struct.int16x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i16>], [2 x <4 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x i16>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x i16> [[TMP6]] to <8 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP5]] to <4 x i16>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP7]] to <4 x i16>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <4 x i16>, <4 x i16> } @llvm.aarch64.neon.ld2lane.v4i16.p0i8(<4 x i16> [[TMP8]], <4 x i16> [[TMP9]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <4 x i16>, <4 x i16> }*
-// CHECK:   store { <4 x i16>, <4 x i16> } [[VLD2_LANE]], { <4 x i16>, <4 x i16> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.int16x4x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.int16x4x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP11]], i8* align 8 [[TMP12]], i64 16, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.int16x4x2_t, %struct.int16x4x2_t* [[RETVAL]], align 8
+// CHECK:   [[VLD2_LANE:%.*]] = call { <4 x i16>, <4 x i16> } @llvm.aarch64.neon.ld2lane.v4i16.p0(<4 x i16> [[TMP8]], <4 x i16> [[TMP9]], i64 3, ptr %a)
+// CHECK:   store { <4 x i16>, <4 x i16> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 16, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.int16x4x2_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.int16x4x2_t [[TMP13]]
 int16x4x2_t test_vld2_lane_s16(int16_t  *a, int16x4x2_t b) {
   return vld2_lane_s16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int32x2x2_t @test_vld2_lane_s32(i32* noundef %a, [2 x <2 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int32x2x2_t @test_vld2_lane_s32(ptr noundef %a, [2 x <2 x i32>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int32x2x2_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.int32x2x2_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.int32x2x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int32x2x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x2x2_t, %struct.int32x2x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <2 x i32>] [[B]].coerce, [2 x <2 x i32>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int32x2x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int32x2x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int32x2x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x2x2_t, %struct.int32x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x i32>], [2 x <2 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x2x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <2 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x i32>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x i32> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x2x2_t, %struct.int32x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x i32>], [2 x <2 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x i32>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x i32> [[TMP6]] to <8 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP5]] to <2 x i32>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP7]] to <2 x i32>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <2 x i32>, <2 x i32> } @llvm.aarch64.neon.ld2lane.v2i32.p0i8(<2 x i32> [[TMP8]], <2 x i32> [[TMP9]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <2 x i32>, <2 x i32> }*
-// CHECK:   store { <2 x i32>, <2 x i32> } [[VLD2_LANE]], { <2 x i32>, <2 x i32> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.int32x2x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.int32x2x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP11]], i8* align 8 [[TMP12]], i64 16, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.int32x2x2_t, %struct.int32x2x2_t* [[RETVAL]], align 8
+// CHECK:   [[VLD2_LANE:%.*]] = call { <2 x i32>, <2 x i32> } @llvm.aarch64.neon.ld2lane.v2i32.p0(<2 x i32> [[TMP8]], <2 x i32> [[TMP9]], i64 1, ptr %a)
+// CHECK:   store { <2 x i32>, <2 x i32> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 16, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.int32x2x2_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.int32x2x2_t [[TMP13]]
 int32x2x2_t test_vld2_lane_s32(int32_t  *a, int32x2x2_t b) {
   return vld2_lane_s32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int64x1x2_t @test_vld2_lane_s64(i64* noundef %a, [2 x <1 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int64x1x2_t @test_vld2_lane_s64(ptr noundef %a, [2 x <1 x i64>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int64x1x2_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.int64x1x2_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.int64x1x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int64x1x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x1x2_t, %struct.int64x1x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <1 x i64>] [[B]].coerce, [2 x <1 x i64>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int64x1x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int64x1x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int64x1x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x1x2_t, %struct.int64x1x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <1 x i64>], [2 x <1 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x1x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <1 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x1x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <1 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <1 x i64>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <1 x i64> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x1x2_t, %struct.int64x1x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <1 x i64>], [2 x <1 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x1x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <1 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <1 x i64>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <1 x i64> [[TMP6]] to <8 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP5]] to <1 x i64>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP7]] to <1 x i64>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld2lane.v1i64.p0i8(<1 x i64> [[TMP8]], <1 x i64> [[TMP9]], i64 0, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <1 x i64>, <1 x i64> }*
-// CHECK:   store { <1 x i64>, <1 x i64> } [[VLD2_LANE]], { <1 x i64>, <1 x i64> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.int64x1x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.int64x1x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP11]], i8* align 8 [[TMP12]], i64 16, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.int64x1x2_t, %struct.int64x1x2_t* [[RETVAL]], align 8
+// CHECK:   [[VLD2_LANE:%.*]] = call { <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld2lane.v1i64.p0(<1 x i64> [[TMP8]], <1 x i64> [[TMP9]], i64 0, ptr %a)
+// CHECK:   store { <1 x i64>, <1 x i64> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 16, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.int64x1x2_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.int64x1x2_t [[TMP13]]
 int64x1x2_t test_vld2_lane_s64(int64_t  *a, int64x1x2_t b) {
   return vld2_lane_s64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float16x4x2_t @test_vld2_lane_f16(half* noundef %a, [2 x <4 x half>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float16x4x2_t @test_vld2_lane_f16(ptr noundef %a, [2 x <4 x half>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float16x4x2_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.float16x4x2_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.float16x4x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.float16x4x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x4x2_t, %struct.float16x4x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <4 x half>] [[B]].coerce, [2 x <4 x half>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float16x4x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float16x4x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.float16x4x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast half* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x4x2_t, %struct.float16x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x half>], [2 x <4 x half>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x half>, <4 x half>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x4x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <4 x half>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x half>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x half>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x half> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x4x2_t, %struct.float16x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x half>], [2 x <4 x half>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x half>, <4 x half>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x half>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x half>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x half> [[TMP6]] to <8 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP5]] to <4 x half>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP7]] to <4 x half>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <4 x half>, <4 x half> } @llvm.aarch64.neon.ld2lane.v4f16.p0i8(<4 x half> [[TMP8]], <4 x half> [[TMP9]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <4 x half>, <4 x half> }*
-// CHECK:   store { <4 x half>, <4 x half> } [[VLD2_LANE]], { <4 x half>, <4 x half> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.float16x4x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.float16x4x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP11]], i8* align 8 [[TMP12]], i64 16, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.float16x4x2_t, %struct.float16x4x2_t* [[RETVAL]], align 8
+// CHECK:   [[VLD2_LANE:%.*]] = call { <4 x half>, <4 x half> } @llvm.aarch64.neon.ld2lane.v4f16.p0(<4 x half> [[TMP8]], <4 x half> [[TMP9]], i64 3, ptr %a)
+// CHECK:   store { <4 x half>, <4 x half> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 16, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.float16x4x2_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.float16x4x2_t [[TMP13]]
 float16x4x2_t test_vld2_lane_f16(float16_t  *a, float16x4x2_t b) {
   return vld2_lane_f16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float32x2x2_t @test_vld2_lane_f32(float* noundef %a, [2 x <2 x float>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float32x2x2_t @test_vld2_lane_f32(ptr noundef %a, [2 x <2 x float>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float32x2x2_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.float32x2x2_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.float32x2x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.float32x2x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x2x2_t, %struct.float32x2x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <2 x float>] [[B]].coerce, [2 x <2 x float>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float32x2x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float32x2x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.float32x2x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast float* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x2x2_t, %struct.float32x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x float>], [2 x <2 x float>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x float>, <2 x float>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x2x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <2 x float>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x float>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x float>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x float> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x2x2_t, %struct.float32x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x float>], [2 x <2 x float>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x float>, <2 x float>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x float>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x float>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x float> [[TMP6]] to <8 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP5]] to <2 x float>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP7]] to <2 x float>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <2 x float>, <2 x float> } @llvm.aarch64.neon.ld2lane.v2f32.p0i8(<2 x float> [[TMP8]], <2 x float> [[TMP9]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <2 x float>, <2 x float> }*
-// CHECK:   store { <2 x float>, <2 x float> } [[VLD2_LANE]], { <2 x float>, <2 x float> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.float32x2x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.float32x2x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP11]], i8* align 8 [[TMP12]], i64 16, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.float32x2x2_t, %struct.float32x2x2_t* [[RETVAL]], align 8
+// CHECK:   [[VLD2_LANE:%.*]] = call { <2 x float>, <2 x float> } @llvm.aarch64.neon.ld2lane.v2f32.p0(<2 x float> [[TMP8]], <2 x float> [[TMP9]], i64 1, ptr %a)
+// CHECK:   store { <2 x float>, <2 x float> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 16, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.float32x2x2_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.float32x2x2_t [[TMP13]]
 float32x2x2_t test_vld2_lane_f32(float32_t  *a, float32x2x2_t b) {
   return vld2_lane_f32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float64x1x2_t @test_vld2_lane_f64(double* noundef %a, [2 x <1 x double>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float64x1x2_t @test_vld2_lane_f64(ptr noundef %a, [2 x <1 x double>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float64x1x2_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.float64x1x2_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.float64x1x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.float64x1x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x1x2_t, %struct.float64x1x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <1 x double>] [[B]].coerce, [2 x <1 x double>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float64x1x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float64x1x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.float64x1x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast double* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x1x2_t, %struct.float64x1x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <1 x double>], [2 x <1 x double>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <1 x double>, <1 x double>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x1x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <1 x double>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x1x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <1 x double>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <1 x double>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <1 x double> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x1x2_t, %struct.float64x1x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <1 x double>], [2 x <1 x double>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <1 x double>, <1 x double>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x1x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <1 x double>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <1 x double>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <1 x double> [[TMP6]] to <8 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP5]] to <1 x double>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP7]] to <1 x double>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <1 x double>, <1 x double> } @llvm.aarch64.neon.ld2lane.v1f64.p0i8(<1 x double> [[TMP8]], <1 x double> [[TMP9]], i64 0, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <1 x double>, <1 x double> }*
-// CHECK:   store { <1 x double>, <1 x double> } [[VLD2_LANE]], { <1 x double>, <1 x double> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.float64x1x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.float64x1x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP11]], i8* align 8 [[TMP12]], i64 16, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.float64x1x2_t, %struct.float64x1x2_t* [[RETVAL]], align 8
+// CHECK:   [[VLD2_LANE:%.*]] = call { <1 x double>, <1 x double> } @llvm.aarch64.neon.ld2lane.v1f64.p0(<1 x double> [[TMP8]], <1 x double> [[TMP9]], i64 0, ptr %a)
+// CHECK:   store { <1 x double>, <1 x double> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 16, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.float64x1x2_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.float64x1x2_t [[TMP13]]
 float64x1x2_t test_vld2_lane_f64(float64_t  *a, float64x1x2_t b) {
   return vld2_lane_f64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly8x8x2_t @test_vld2_lane_p8(i8* noundef %a, [2 x <8 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly8x8x2_t @test_vld2_lane_p8(ptr noundef %a, [2 x <8 x i8>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly8x8x2_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.poly8x8x2_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly8x8x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly8x8x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x8x2_t, %struct.poly8x8x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <8 x i8>] [[B]].coerce, [2 x <8 x i8>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly8x8x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly8x8x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.poly8x8x2_t* [[__RET]] to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x8x2_t, %struct.poly8x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i8>], [2 x <8 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX]], align 8
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x8x2_t, %struct.poly8x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i8>], [2 x <8 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP4:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX2]], align 8
-// CHECK:   [[VLD2_LANE:%.*]] = call { <8 x i8>, <8 x i8> } @llvm.aarch64.neon.ld2lane.v8i8.p0i8(<8 x i8> [[TMP3]], <8 x i8> [[TMP4]], i64 7, i8* %a)
-// CHECK:   [[TMP5:%.*]] = bitcast i8* [[TMP2]] to { <8 x i8>, <8 x i8> }*
-// CHECK:   store { <8 x i8>, <8 x i8> } [[VLD2_LANE]], { <8 x i8>, <8 x i8> }* [[TMP5]]
-// CHECK:   [[TMP6:%.*]] = bitcast %struct.poly8x8x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP7:%.*]] = bitcast %struct.poly8x8x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP6]], i8* align 8 [[TMP7]], i64 16, i1 false)
-// CHECK:   [[TMP8:%.*]] = load %struct.poly8x8x2_t, %struct.poly8x8x2_t* [[RETVAL]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x8x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <8 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x i8>, ptr [[ARRAYIDX]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP4:%.*]] = load <8 x i8>, ptr [[ARRAYIDX2]], align 8
+// CHECK:   [[VLD2_LANE:%.*]] = call { <8 x i8>, <8 x i8> } @llvm.aarch64.neon.ld2lane.v8i8.p0(<8 x i8> [[TMP3]], <8 x i8> [[TMP4]], i64 7, ptr %a)
+// CHECK:   store { <8 x i8>, <8 x i8> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 16, i1 false)
+// CHECK:   [[TMP8:%.*]] = load %struct.poly8x8x2_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.poly8x8x2_t [[TMP8]]
 poly8x8x2_t test_vld2_lane_p8(poly8_t  *a, poly8x8x2_t b) {
   return vld2_lane_p8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly16x4x2_t @test_vld2_lane_p16(i16* noundef %a, [2 x <4 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly16x4x2_t @test_vld2_lane_p16(ptr noundef %a, [2 x <4 x i16>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly16x4x2_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.poly16x4x2_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly16x4x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly16x4x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x4x2_t, %struct.poly16x4x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <4 x i16>] [[B]].coerce, [2 x <4 x i16>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly16x4x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly16x4x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.poly16x4x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x4x2_t, %struct.poly16x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i16>], [2 x <4 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x4x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <4 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x i16>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x i16> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x4x2_t, %struct.poly16x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i16>], [2 x <4 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x i16>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x i16> [[TMP6]] to <8 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP5]] to <4 x i16>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP7]] to <4 x i16>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <4 x i16>, <4 x i16> } @llvm.aarch64.neon.ld2lane.v4i16.p0i8(<4 x i16> [[TMP8]], <4 x i16> [[TMP9]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <4 x i16>, <4 x i16> }*
-// CHECK:   store { <4 x i16>, <4 x i16> } [[VLD2_LANE]], { <4 x i16>, <4 x i16> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.poly16x4x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.poly16x4x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP11]], i8* align 8 [[TMP12]], i64 16, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.poly16x4x2_t, %struct.poly16x4x2_t* [[RETVAL]], align 8
+// CHECK:   [[VLD2_LANE:%.*]] = call { <4 x i16>, <4 x i16> } @llvm.aarch64.neon.ld2lane.v4i16.p0(<4 x i16> [[TMP8]], <4 x i16> [[TMP9]], i64 3, ptr %a)
+// CHECK:   store { <4 x i16>, <4 x i16> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 16, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.poly16x4x2_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.poly16x4x2_t [[TMP13]]
 poly16x4x2_t test_vld2_lane_p16(poly16_t  *a, poly16x4x2_t b) {
   return vld2_lane_p16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly64x1x2_t @test_vld2_lane_p64(i64* noundef %a, [2 x <1 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly64x1x2_t @test_vld2_lane_p64(ptr noundef %a, [2 x <1 x i64>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly64x1x2_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.poly64x1x2_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly64x1x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly64x1x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x1x2_t, %struct.poly64x1x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <1 x i64>] [[B]].coerce, [2 x <1 x i64>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly64x1x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly64x1x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.poly64x1x2_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x1x2_t, %struct.poly64x1x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <1 x i64>], [2 x <1 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x1x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <1 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x1x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <1 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <1 x i64>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <1 x i64> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x1x2_t, %struct.poly64x1x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <1 x i64>], [2 x <1 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x1x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <1 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <1 x i64>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <1 x i64> [[TMP6]] to <8 x i8>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP5]] to <1 x i64>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP7]] to <1 x i64>
-// CHECK:   [[VLD2_LANE:%.*]] = call { <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld2lane.v1i64.p0i8(<1 x i64> [[TMP8]], <1 x i64> [[TMP9]], i64 0, i8* [[TMP3]])
-// CHECK:   [[TMP10:%.*]] = bitcast i8* [[TMP2]] to { <1 x i64>, <1 x i64> }*
-// CHECK:   store { <1 x i64>, <1 x i64> } [[VLD2_LANE]], { <1 x i64>, <1 x i64> }* [[TMP10]]
-// CHECK:   [[TMP11:%.*]] = bitcast %struct.poly64x1x2_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP12:%.*]] = bitcast %struct.poly64x1x2_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP11]], i8* align 8 [[TMP12]], i64 16, i1 false)
-// CHECK:   [[TMP13:%.*]] = load %struct.poly64x1x2_t, %struct.poly64x1x2_t* [[RETVAL]], align 8
+// CHECK:   [[VLD2_LANE:%.*]] = call { <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld2lane.v1i64.p0(<1 x i64> [[TMP8]], <1 x i64> [[TMP9]], i64 0, ptr %a)
+// CHECK:   store { <1 x i64>, <1 x i64> } [[VLD2_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 16, i1 false)
+// CHECK:   [[TMP13:%.*]] = load %struct.poly64x1x2_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.poly64x1x2_t [[TMP13]]
 poly64x1x2_t test_vld2_lane_p64(poly64_t  *a, poly64x1x2_t b) {
   return vld2_lane_p64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint16x8x3_t @test_vld3q_lane_u16(i16* noundef %a, [3 x <8 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint16x8x3_t @test_vld3q_lane_u16(ptr noundef %a, [3 x <8 x i16>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint16x8x3_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.uint16x8x3_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint16x8x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint16x8x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <8 x i16>] [[B]].coerce, [3 x <8 x i16>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint16x8x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint16x8x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint16x8x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <8 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <8 x i16>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <8 x i16>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <8 x i16>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i16> [[TMP8]] to <16 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP9]] to <8 x i16>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <8 x i16>, <8 x i16>, <8 x i16> } @llvm.aarch64.neon.ld3lane.v8i16.p0i8(<8 x i16> [[TMP10]], <8 x i16> [[TMP11]], <8 x i16> [[TMP12]], i64 7, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <8 x i16>, <8 x i16>, <8 x i16> }*
-// CHECK:   store { <8 x i16>, <8 x i16>, <8 x i16> } [[VLD3_LANE]], { <8 x i16>, <8 x i16>, <8 x i16> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.uint16x8x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.uint16x8x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP14]], i8* align 16 [[TMP15]], i64 48, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[RETVAL]], align 16
+// CHECK:   [[VLD3_LANE:%.*]] = call { <8 x i16>, <8 x i16>, <8 x i16> } @llvm.aarch64.neon.ld3lane.v8i16.p0(<8 x i16> [[TMP10]], <8 x i16> [[TMP11]], <8 x i16> [[TMP12]], i64 7, ptr %a)
+// CHECK:   store { <8 x i16>, <8 x i16>, <8 x i16> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 48, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.uint16x8x3_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.uint16x8x3_t [[TMP16]]
 uint16x8x3_t test_vld3q_lane_u16(uint16_t  *a, uint16x8x3_t b) {
   return vld3q_lane_u16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint32x4x3_t @test_vld3q_lane_u32(i32* noundef %a, [3 x <4 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint32x4x3_t @test_vld3q_lane_u32(ptr noundef %a, [3 x <4 x i32>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint32x4x3_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.uint32x4x3_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint32x4x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint32x4x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <4 x i32>] [[B]].coerce, [3 x <4 x i32>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint32x4x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint32x4x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint32x4x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <4 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x i32>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x i32> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x i32>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x i32> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i32>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <4 x i32>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <4 x i32> [[TMP8]] to <16 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x i32>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x i32>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP9]] to <4 x i32>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld3lane.v4i32.p0i8(<4 x i32> [[TMP10]], <4 x i32> [[TMP11]], <4 x i32> [[TMP12]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <4 x i32>, <4 x i32>, <4 x i32> }*
-// CHECK:   store { <4 x i32>, <4 x i32>, <4 x i32> } [[VLD3_LANE]], { <4 x i32>, <4 x i32>, <4 x i32> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.uint32x4x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.uint32x4x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP14]], i8* align 16 [[TMP15]], i64 48, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[RETVAL]], align 16
+// CHECK:   [[VLD3_LANE:%.*]] = call { <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld3lane.v4i32.p0(<4 x i32> [[TMP10]], <4 x i32> [[TMP11]], <4 x i32> [[TMP12]], i64 3, ptr %a)
+// CHECK:   store { <4 x i32>, <4 x i32>, <4 x i32> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 48, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.uint32x4x3_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.uint32x4x3_t [[TMP16]]
 uint32x4x3_t test_vld3q_lane_u32(uint32_t  *a, uint32x4x3_t b) {
   return vld3q_lane_u32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint64x2x3_t @test_vld3q_lane_u64(i64* noundef %a, [3 x <2 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint64x2x3_t @test_vld3q_lane_u64(ptr noundef %a, [3 x <2 x i64>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint64x2x3_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.uint64x2x3_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint64x2x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint64x2x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x2x3_t, %struct.uint64x2x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <2 x i64>] [[B]].coerce, [3 x <2 x i64>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint64x2x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint64x2x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint64x2x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x2x3_t, %struct.uint64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x i64>], [3 x <2 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x2x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <2 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x i64>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x i64> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x2x3_t, %struct.uint64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x i64>], [3 x <2 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x i64>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x i64> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint64x2x3_t, %struct.uint64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x i64>], [3 x <2 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <2 x i64>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <2 x i64> [[TMP8]] to <16 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <2 x i64>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <2 x i64>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP9]] to <2 x i64>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld3lane.v2i64.p0i8(<2 x i64> [[TMP10]], <2 x i64> [[TMP11]], <2 x i64> [[TMP12]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <2 x i64>, <2 x i64>, <2 x i64> }*
-// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64> } [[VLD3_LANE]], { <2 x i64>, <2 x i64>, <2 x i64> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.uint64x2x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.uint64x2x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP14]], i8* align 16 [[TMP15]], i64 48, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.uint64x2x3_t, %struct.uint64x2x3_t* [[RETVAL]], align 16
+// CHECK:   [[VLD3_LANE:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld3lane.v2i64.p0(<2 x i64> [[TMP10]], <2 x i64> [[TMP11]], <2 x i64> [[TMP12]], i64 1, ptr %a)
+// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 48, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.uint64x2x3_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.uint64x2x3_t [[TMP16]]
 uint64x2x3_t test_vld3q_lane_u64(uint64_t  *a, uint64x2x3_t b) {
   return vld3q_lane_u64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int16x8x3_t @test_vld3q_lane_s16(i16* noundef %a, [3 x <8 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int16x8x3_t @test_vld3q_lane_s16(ptr noundef %a, [3 x <8 x i16>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int16x8x3_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.int16x8x3_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.int16x8x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int16x8x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x3_t, %struct.int16x8x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <8 x i16>] [[B]].coerce, [3 x <8 x i16>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int16x8x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int16x8x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int16x8x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x3_t, %struct.int16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <8 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <8 x i16>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x3_t, %struct.int16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <8 x i16>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int16x8x3_t, %struct.int16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <8 x i16>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i16> [[TMP8]] to <16 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP9]] to <8 x i16>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <8 x i16>, <8 x i16>, <8 x i16> } @llvm.aarch64.neon.ld3lane.v8i16.p0i8(<8 x i16> [[TMP10]], <8 x i16> [[TMP11]], <8 x i16> [[TMP12]], i64 7, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <8 x i16>, <8 x i16>, <8 x i16> }*
-// CHECK:   store { <8 x i16>, <8 x i16>, <8 x i16> } [[VLD3_LANE]], { <8 x i16>, <8 x i16>, <8 x i16> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.int16x8x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.int16x8x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP14]], i8* align 16 [[TMP15]], i64 48, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.int16x8x3_t, %struct.int16x8x3_t* [[RETVAL]], align 16
+// CHECK:   [[VLD3_LANE:%.*]] = call { <8 x i16>, <8 x i16>, <8 x i16> } @llvm.aarch64.neon.ld3lane.v8i16.p0(<8 x i16> [[TMP10]], <8 x i16> [[TMP11]], <8 x i16> [[TMP12]], i64 7, ptr %a)
+// CHECK:   store { <8 x i16>, <8 x i16>, <8 x i16> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 48, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.int16x8x3_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.int16x8x3_t [[TMP16]]
 int16x8x3_t test_vld3q_lane_s16(int16_t  *a, int16x8x3_t b) {
   return vld3q_lane_s16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int32x4x3_t @test_vld3q_lane_s32(i32* noundef %a, [3 x <4 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int32x4x3_t @test_vld3q_lane_s32(ptr noundef %a, [3 x <4 x i32>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int32x4x3_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.int32x4x3_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.int32x4x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int32x4x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x3_t, %struct.int32x4x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <4 x i32>] [[B]].coerce, [3 x <4 x i32>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int32x4x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int32x4x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int32x4x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x3_t, %struct.int32x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <4 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x i32>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x i32> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x3_t, %struct.int32x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x i32>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x i32> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int32x4x3_t, %struct.int32x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int32x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i32>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <4 x i32>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <4 x i32> [[TMP8]] to <16 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x i32>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x i32>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP9]] to <4 x i32>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld3lane.v4i32.p0i8(<4 x i32> [[TMP10]], <4 x i32> [[TMP11]], <4 x i32> [[TMP12]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <4 x i32>, <4 x i32>, <4 x i32> }*
-// CHECK:   store { <4 x i32>, <4 x i32>, <4 x i32> } [[VLD3_LANE]], { <4 x i32>, <4 x i32>, <4 x i32> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.int32x4x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.int32x4x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP14]], i8* align 16 [[TMP15]], i64 48, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.int32x4x3_t, %struct.int32x4x3_t* [[RETVAL]], align 16
+// CHECK:   [[VLD3_LANE:%.*]] = call { <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld3lane.v4i32.p0(<4 x i32> [[TMP10]], <4 x i32> [[TMP11]], <4 x i32> [[TMP12]], i64 3, ptr %a)
+// CHECK:   store { <4 x i32>, <4 x i32>, <4 x i32> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 48, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.int32x4x3_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.int32x4x3_t [[TMP16]]
 int32x4x3_t test_vld3q_lane_s32(int32_t  *a, int32x4x3_t b) {
   return vld3q_lane_s32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int64x2x3_t @test_vld3q_lane_s64(i64* noundef %a, [3 x <2 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int64x2x3_t @test_vld3q_lane_s64(ptr noundef %a, [3 x <2 x i64>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int64x2x3_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.int64x2x3_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.int64x2x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int64x2x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x2x3_t, %struct.int64x2x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <2 x i64>] [[B]].coerce, [3 x <2 x i64>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int64x2x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int64x2x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int64x2x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x2x3_t, %struct.int64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x i64>], [3 x <2 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x2x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <2 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x i64>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x i64> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x2x3_t, %struct.int64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x i64>], [3 x <2 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x i64>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x i64> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int64x2x3_t, %struct.int64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x i64>], [3 x <2 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <2 x i64>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <2 x i64> [[TMP8]] to <16 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <2 x i64>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <2 x i64>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP9]] to <2 x i64>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld3lane.v2i64.p0i8(<2 x i64> [[TMP10]], <2 x i64> [[TMP11]], <2 x i64> [[TMP12]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <2 x i64>, <2 x i64>, <2 x i64> }*
-// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64> } [[VLD3_LANE]], { <2 x i64>, <2 x i64>, <2 x i64> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.int64x2x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.int64x2x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP14]], i8* align 16 [[TMP15]], i64 48, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.int64x2x3_t, %struct.int64x2x3_t* [[RETVAL]], align 16
+// CHECK:   [[VLD3_LANE:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld3lane.v2i64.p0(<2 x i64> [[TMP10]], <2 x i64> [[TMP11]], <2 x i64> [[TMP12]], i64 1, ptr %a)
+// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 48, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.int64x2x3_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.int64x2x3_t [[TMP16]]
 int64x2x3_t test_vld3q_lane_s64(int64_t  *a, int64x2x3_t b) {
   return vld3q_lane_s64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float16x8x3_t @test_vld3q_lane_f16(half* noundef %a, [3 x <8 x half>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float16x8x3_t @test_vld3q_lane_f16(ptr noundef %a, [3 x <8 x half>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float16x8x3_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.float16x8x3_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.float16x8x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.float16x8x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x3_t, %struct.float16x8x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <8 x half>] [[B]].coerce, [3 x <8 x half>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float16x8x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float16x8x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.float16x8x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast half* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x3_t, %struct.float16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x half>], [3 x <8 x half>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <8 x half>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x half>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <8 x half>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <8 x half> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x3_t, %struct.float16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x half>], [3 x <8 x half>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x half>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <8 x half>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x half> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float16x8x3_t, %struct.float16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x half>], [3 x <8 x half>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x half>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <8 x half>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x half> [[TMP8]] to <16 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x half>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x half>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP9]] to <8 x half>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <8 x half>, <8 x half>, <8 x half> } @llvm.aarch64.neon.ld3lane.v8f16.p0i8(<8 x half> [[TMP10]], <8 x half> [[TMP11]], <8 x half> [[TMP12]], i64 7, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <8 x half>, <8 x half>, <8 x half> }*
-// CHECK:   store { <8 x half>, <8 x half>, <8 x half> } [[VLD3_LANE]], { <8 x half>, <8 x half>, <8 x half> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.float16x8x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.float16x8x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP14]], i8* align 16 [[TMP15]], i64 48, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.float16x8x3_t, %struct.float16x8x3_t* [[RETVAL]], align 16
+// CHECK:   [[VLD3_LANE:%.*]] = call { <8 x half>, <8 x half>, <8 x half> } @llvm.aarch64.neon.ld3lane.v8f16.p0(<8 x half> [[TMP10]], <8 x half> [[TMP11]], <8 x half> [[TMP12]], i64 7, ptr %a)
+// CHECK:   store { <8 x half>, <8 x half>, <8 x half> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 48, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.float16x8x3_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.float16x8x3_t [[TMP16]]
 float16x8x3_t test_vld3q_lane_f16(float16_t  *a, float16x8x3_t b) {
   return vld3q_lane_f16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float32x4x3_t @test_vld3q_lane_f32(float* noundef %a, [3 x <4 x float>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float32x4x3_t @test_vld3q_lane_f32(ptr noundef %a, [3 x <4 x float>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float32x4x3_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.float32x4x3_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.float32x4x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.float32x4x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x3_t, %struct.float32x4x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <4 x float>] [[B]].coerce, [3 x <4 x float>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float32x4x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float32x4x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.float32x4x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast float* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x3_t, %struct.float32x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x float>], [3 x <4 x float>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <4 x float>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x float>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x float>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x float> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x3_t, %struct.float32x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x float>], [3 x <4 x float>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x float>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x float>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x float> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float32x4x3_t, %struct.float32x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x float>], [3 x <4 x float>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float32x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x float>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <4 x float>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <4 x float> [[TMP8]] to <16 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x float>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x float>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP9]] to <4 x float>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <4 x float>, <4 x float>, <4 x float> } @llvm.aarch64.neon.ld3lane.v4f32.p0i8(<4 x float> [[TMP10]], <4 x float> [[TMP11]], <4 x float> [[TMP12]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <4 x float>, <4 x float>, <4 x float> }*
-// CHECK:   store { <4 x float>, <4 x float>, <4 x float> } [[VLD3_LANE]], { <4 x float>, <4 x float>, <4 x float> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.float32x4x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.float32x4x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP14]], i8* align 16 [[TMP15]], i64 48, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.float32x4x3_t, %struct.float32x4x3_t* [[RETVAL]], align 16
+// CHECK:   [[VLD3_LANE:%.*]] = call { <4 x float>, <4 x float>, <4 x float> } @llvm.aarch64.neon.ld3lane.v4f32.p0(<4 x float> [[TMP10]], <4 x float> [[TMP11]], <4 x float> [[TMP12]], i64 3, ptr %a)
+// CHECK:   store { <4 x float>, <4 x float>, <4 x float> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 48, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.float32x4x3_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.float32x4x3_t [[TMP16]]
 float32x4x3_t test_vld3q_lane_f32(float32_t  *a, float32x4x3_t b) {
   return vld3q_lane_f32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float64x2x3_t @test_vld3q_lane_f64(double* noundef %a, [3 x <2 x double>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float64x2x3_t @test_vld3q_lane_f64(ptr noundef %a, [3 x <2 x double>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float64x2x3_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.float64x2x3_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.float64x2x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.float64x2x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x2x3_t, %struct.float64x2x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <2 x double>] [[B]].coerce, [3 x <2 x double>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float64x2x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float64x2x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.float64x2x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast double* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x2x3_t, %struct.float64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x double>], [3 x <2 x double>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x double>, <2 x double>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x2x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <2 x double>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x double>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x double>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x double> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x2x3_t, %struct.float64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x double>], [3 x <2 x double>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x double>, <2 x double>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x double>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x double>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x double> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float64x2x3_t, %struct.float64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x double>], [3 x <2 x double>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <2 x double>, <2 x double>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x double>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <2 x double>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <2 x double> [[TMP8]] to <16 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <2 x double>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <2 x double>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP9]] to <2 x double>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <2 x double>, <2 x double>, <2 x double> } @llvm.aarch64.neon.ld3lane.v2f64.p0i8(<2 x double> [[TMP10]], <2 x double> [[TMP11]], <2 x double> [[TMP12]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <2 x double>, <2 x double>, <2 x double> }*
-// CHECK:   store { <2 x double>, <2 x double>, <2 x double> } [[VLD3_LANE]], { <2 x double>, <2 x double>, <2 x double> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.float64x2x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.float64x2x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP14]], i8* align 16 [[TMP15]], i64 48, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.float64x2x3_t, %struct.float64x2x3_t* [[RETVAL]], align 16
+// CHECK:   [[VLD3_LANE:%.*]] = call { <2 x double>, <2 x double>, <2 x double> } @llvm.aarch64.neon.ld3lane.v2f64.p0(<2 x double> [[TMP10]], <2 x double> [[TMP11]], <2 x double> [[TMP12]], i64 1, ptr %a)
+// CHECK:   store { <2 x double>, <2 x double>, <2 x double> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 48, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.float64x2x3_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.float64x2x3_t [[TMP16]]
 float64x2x3_t test_vld3q_lane_f64(float64_t  *a, float64x2x3_t b) {
   return vld3q_lane_f64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly8x16x3_t @test_vld3q_lane_p8(i8* noundef %a, [3 x <16 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly8x16x3_t @test_vld3q_lane_p8(ptr noundef %a, [3 x <16 x i8>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly8x16x3_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.poly8x16x3_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly8x16x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly8x16x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x16x3_t, %struct.poly8x16x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <16 x i8>] [[B]].coerce, [3 x <16 x i8>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly8x16x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly8x16x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.poly8x16x3_t* [[__RET]] to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x16x3_t, %struct.poly8x16x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x16x3_t, %struct.poly8x16x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly8x16x3_t, %struct.poly8x16x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 16
-// CHECK:   [[VLD3_LANE:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8> } @llvm.aarch64.neon.ld3lane.v16i8.p0i8(<16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], i64 15, i8* %a)
-// CHECK:   [[TMP6:%.*]] = bitcast i8* [[TMP2]] to { <16 x i8>, <16 x i8>, <16 x i8> }*
-// CHECK:   store { <16 x i8>, <16 x i8>, <16 x i8> } [[VLD3_LANE]], { <16 x i8>, <16 x i8>, <16 x i8> }* [[TMP6]]
-// CHECK:   [[TMP7:%.*]] = bitcast %struct.poly8x16x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP8:%.*]] = bitcast %struct.poly8x16x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP7]], i8* align 16 [[TMP8]], i64 48, i1 false)
-// CHECK:   [[TMP9:%.*]] = load %struct.poly8x16x3_t, %struct.poly8x16x3_t* [[RETVAL]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x16x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <16 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x16x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <16 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <16 x i8>, ptr [[ARRAYIDX]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x16x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <16 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP4:%.*]] = load <16 x i8>, ptr [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly8x16x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <16 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP5:%.*]] = load <16 x i8>, ptr [[ARRAYIDX4]], align 16
+// CHECK:   [[VLD3_LANE:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8> } @llvm.aarch64.neon.ld3lane.v16i8.p0(<16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], i64 15, ptr %a)
+// CHECK:   store { <16 x i8>, <16 x i8>, <16 x i8> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 48, i1 false)
+// CHECK:   [[TMP9:%.*]] = load %struct.poly8x16x3_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.poly8x16x3_t [[TMP9]]
 poly8x16x3_t test_vld3q_lane_p8(poly8_t  *a, poly8x16x3_t b) {
   return vld3q_lane_p8(a, b, 15);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly16x8x3_t @test_vld3q_lane_p16(i16* noundef %a, [3 x <8 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly16x8x3_t @test_vld3q_lane_p16(ptr noundef %a, [3 x <8 x i16>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly16x8x3_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.poly16x8x3_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly16x8x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly16x8x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <8 x i16>] [[B]].coerce, [3 x <8 x i16>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly16x8x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly16x8x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.poly16x8x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <8 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <8 x i16>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <8 x i16>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <8 x i16>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i16> [[TMP8]] to <16 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP9]] to <8 x i16>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <8 x i16>, <8 x i16>, <8 x i16> } @llvm.aarch64.neon.ld3lane.v8i16.p0i8(<8 x i16> [[TMP10]], <8 x i16> [[TMP11]], <8 x i16> [[TMP12]], i64 7, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <8 x i16>, <8 x i16>, <8 x i16> }*
-// CHECK:   store { <8 x i16>, <8 x i16>, <8 x i16> } [[VLD3_LANE]], { <8 x i16>, <8 x i16>, <8 x i16> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.poly16x8x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.poly16x8x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP14]], i8* align 16 [[TMP15]], i64 48, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[RETVAL]], align 16
+// CHECK:   [[VLD3_LANE:%.*]] = call { <8 x i16>, <8 x i16>, <8 x i16> } @llvm.aarch64.neon.ld3lane.v8i16.p0(<8 x i16> [[TMP10]], <8 x i16> [[TMP11]], <8 x i16> [[TMP12]], i64 7, ptr %a)
+// CHECK:   store { <8 x i16>, <8 x i16>, <8 x i16> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 48, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.poly16x8x3_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.poly16x8x3_t [[TMP16]]
 poly16x8x3_t test_vld3q_lane_p16(poly16_t  *a, poly16x8x3_t b) {
   return vld3q_lane_p16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly64x2x3_t @test_vld3q_lane_p64(i64* noundef %a, [3 x <2 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly64x2x3_t @test_vld3q_lane_p64(ptr noundef %a, [3 x <2 x i64>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly64x2x3_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.poly64x2x3_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly64x2x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly64x2x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x2x3_t, %struct.poly64x2x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <2 x i64>] [[B]].coerce, [3 x <2 x i64>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly64x2x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly64x2x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.poly64x2x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x2x3_t, %struct.poly64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x i64>], [3 x <2 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x2x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <2 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x i64>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x i64> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x2x3_t, %struct.poly64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x i64>], [3 x <2 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x i64>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x i64> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly64x2x3_t, %struct.poly64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x i64>], [3 x <2 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <2 x i64>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <2 x i64> [[TMP8]] to <16 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP5]] to <2 x i64>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP7]] to <2 x i64>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP9]] to <2 x i64>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld3lane.v2i64.p0i8(<2 x i64> [[TMP10]], <2 x i64> [[TMP11]], <2 x i64> [[TMP12]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <2 x i64>, <2 x i64>, <2 x i64> }*
-// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64> } [[VLD3_LANE]], { <2 x i64>, <2 x i64>, <2 x i64> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.poly64x2x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.poly64x2x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP14]], i8* align 16 [[TMP15]], i64 48, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.poly64x2x3_t, %struct.poly64x2x3_t* [[RETVAL]], align 16
+// CHECK:   [[VLD3_LANE:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld3lane.v2i64.p0(<2 x i64> [[TMP10]], <2 x i64> [[TMP11]], <2 x i64> [[TMP12]], i64 1, ptr %a)
+// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 48, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.poly64x2x3_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.poly64x2x3_t [[TMP16]]
 poly64x2x3_t test_vld3q_lane_p64(poly64_t  *a, poly64x2x3_t b) {
   return vld3q_lane_p64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint8x8x3_t @test_vld3_lane_u8(i8* noundef %a, [3 x <8 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint8x8x3_t @test_vld3_lane_u8(ptr noundef %a, [3 x <8 x i8>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint8x8x3_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.uint8x8x3_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint8x8x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint8x8x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x8x3_t, %struct.uint8x8x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <8 x i8>] [[B]].coerce, [3 x <8 x i8>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint8x8x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint8x8x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint8x8x3_t* [[__RET]] to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x8x3_t, %struct.uint8x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i8>], [3 x <8 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX]], align 8
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x8x3_t, %struct.uint8x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i8>], [3 x <8 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP4:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX2]], align 8
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint8x8x3_t, %struct.uint8x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i8>], [3 x <8 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP5:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX4]], align 8
-// CHECK:   [[VLD3_LANE:%.*]] = call { <8 x i8>, <8 x i8>, <8 x i8> } @llvm.aarch64.neon.ld3lane.v8i8.p0i8(<8 x i8> [[TMP3]], <8 x i8> [[TMP4]], <8 x i8> [[TMP5]], i64 7, i8* %a)
-// CHECK:   [[TMP6:%.*]] = bitcast i8* [[TMP2]] to { <8 x i8>, <8 x i8>, <8 x i8> }*
-// CHECK:   store { <8 x i8>, <8 x i8>, <8 x i8> } [[VLD3_LANE]], { <8 x i8>, <8 x i8>, <8 x i8> }* [[TMP6]]
-// CHECK:   [[TMP7:%.*]] = bitcast %struct.uint8x8x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP8:%.*]] = bitcast %struct.uint8x8x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP7]], i8* align 8 [[TMP8]], i64 24, i1 false)
-// CHECK:   [[TMP9:%.*]] = load %struct.uint8x8x3_t, %struct.uint8x8x3_t* [[RETVAL]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x8x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <8 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x i8>, ptr [[ARRAYIDX]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP4:%.*]] = load <8 x i8>, ptr [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint8x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP5:%.*]] = load <8 x i8>, ptr [[ARRAYIDX4]], align 8
+// CHECK:   [[VLD3_LANE:%.*]] = call { <8 x i8>, <8 x i8>, <8 x i8> } @llvm.aarch64.neon.ld3lane.v8i8.p0(<8 x i8> [[TMP3]], <8 x i8> [[TMP4]], <8 x i8> [[TMP5]], i64 7, ptr %a)
+// CHECK:   store { <8 x i8>, <8 x i8>, <8 x i8> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 24, i1 false)
+// CHECK:   [[TMP9:%.*]] = load %struct.uint8x8x3_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.uint8x8x3_t [[TMP9]]
 uint8x8x3_t test_vld3_lane_u8(uint8_t  *a, uint8x8x3_t b) {
   return vld3_lane_u8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint16x4x3_t @test_vld3_lane_u16(i16* noundef %a, [3 x <4 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint16x4x3_t @test_vld3_lane_u16(ptr noundef %a, [3 x <4 x i16>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint16x4x3_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.uint16x4x3_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint16x4x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint16x4x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x4x3_t, %struct.uint16x4x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <4 x i16>] [[B]].coerce, [3 x <4 x i16>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint16x4x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint16x4x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint16x4x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x4x3_t, %struct.uint16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i16>], [3 x <4 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x4x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <4 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x i16>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x i16> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x4x3_t, %struct.uint16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i16>], [3 x <4 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x i16>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x i16> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x4x3_t, %struct.uint16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i16>], [3 x <4 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <4 x i16>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <4 x i16> [[TMP8]] to <8 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP5]] to <4 x i16>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP7]] to <4 x i16>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP9]] to <4 x i16>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <4 x i16>, <4 x i16>, <4 x i16> } @llvm.aarch64.neon.ld3lane.v4i16.p0i8(<4 x i16> [[TMP10]], <4 x i16> [[TMP11]], <4 x i16> [[TMP12]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <4 x i16>, <4 x i16>, <4 x i16> }*
-// CHECK:   store { <4 x i16>, <4 x i16>, <4 x i16> } [[VLD3_LANE]], { <4 x i16>, <4 x i16>, <4 x i16> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.uint16x4x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.uint16x4x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP14]], i8* align 8 [[TMP15]], i64 24, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.uint16x4x3_t, %struct.uint16x4x3_t* [[RETVAL]], align 8
+// CHECK:   [[VLD3_LANE:%.*]] = call { <4 x i16>, <4 x i16>, <4 x i16> } @llvm.aarch64.neon.ld3lane.v4i16.p0(<4 x i16> [[TMP10]], <4 x i16> [[TMP11]], <4 x i16> [[TMP12]], i64 3, ptr %a)
+// CHECK:   store { <4 x i16>, <4 x i16>, <4 x i16> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 24, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.uint16x4x3_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.uint16x4x3_t [[TMP16]]
 uint16x4x3_t test_vld3_lane_u16(uint16_t  *a, uint16x4x3_t b) {
   return vld3_lane_u16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint32x2x3_t @test_vld3_lane_u32(i32* noundef %a, [3 x <2 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint32x2x3_t @test_vld3_lane_u32(ptr noundef %a, [3 x <2 x i32>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint32x2x3_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.uint32x2x3_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint32x2x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint32x2x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x2x3_t, %struct.uint32x2x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <2 x i32>] [[B]].coerce, [3 x <2 x i32>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint32x2x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint32x2x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint32x2x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x2x3_t, %struct.uint32x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x i32>], [3 x <2 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x2x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <2 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x i32>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x i32> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x2x3_t, %struct.uint32x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x i32>], [3 x <2 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x i32>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x i32> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x2x3_t, %struct.uint32x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x i32>], [3 x <2 x i32>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x i32>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <2 x i32>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <2 x i32> [[TMP8]] to <8 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP5]] to <2 x i32>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP7]] to <2 x i32>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP9]] to <2 x i32>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <2 x i32>, <2 x i32>, <2 x i32> } @llvm.aarch64.neon.ld3lane.v2i32.p0i8(<2 x i32> [[TMP10]], <2 x i32> [[TMP11]], <2 x i32> [[TMP12]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <2 x i32>, <2 x i32>, <2 x i32> }*
-// CHECK:   store { <2 x i32>, <2 x i32>, <2 x i32> } [[VLD3_LANE]], { <2 x i32>, <2 x i32>, <2 x i32> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.uint32x2x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.uint32x2x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP14]], i8* align 8 [[TMP15]], i64 24, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.uint32x2x3_t, %struct.uint32x2x3_t* [[RETVAL]], align 8
+// CHECK:   [[VLD3_LANE:%.*]] = call { <2 x i32>, <2 x i32>, <2 x i32> } @llvm.aarch64.neon.ld3lane.v2i32.p0(<2 x i32> [[TMP10]], <2 x i32> [[TMP11]], <2 x i32> [[TMP12]], i64 1, ptr %a)
+// CHECK:   store { <2 x i32>, <2 x i32>, <2 x i32> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 24, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.uint32x2x3_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.uint32x2x3_t [[TMP16]]
 uint32x2x3_t test_vld3_lane_u32(uint32_t  *a, uint32x2x3_t b) {
   return vld3_lane_u32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint64x1x3_t @test_vld3_lane_u64(i64* noundef %a, [3 x <1 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint64x1x3_t @test_vld3_lane_u64(ptr noundef %a, [3 x <1 x i64>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint64x1x3_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.uint64x1x3_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint64x1x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint64x1x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x1x3_t, %struct.uint64x1x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <1 x i64>] [[B]].coerce, [3 x <1 x i64>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint64x1x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint64x1x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint64x1x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x1x3_t, %struct.uint64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <1 x i64>], [3 x <1 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x1x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <1 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <1 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <1 x i64>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <1 x i64> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x1x3_t, %struct.uint64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <1 x i64>], [3 x <1 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <1 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <1 x i64>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <1 x i64> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint64x1x3_t, %struct.uint64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <1 x i64>], [3 x <1 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <1 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <1 x i64>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <1 x i64> [[TMP8]] to <8 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP5]] to <1 x i64>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP7]] to <1 x i64>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x i64>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <1 x i64>, <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld3lane.v1i64.p0i8(<1 x i64> [[TMP10]], <1 x i64> [[TMP11]], <1 x i64> [[TMP12]], i64 0, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <1 x i64>, <1 x i64>, <1 x i64> }*
-// CHECK:   store { <1 x i64>, <1 x i64>, <1 x i64> } [[VLD3_LANE]], { <1 x i64>, <1 x i64>, <1 x i64> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.uint64x1x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.uint64x1x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP14]], i8* align 8 [[TMP15]], i64 24, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.uint64x1x3_t, %struct.uint64x1x3_t* [[RETVAL]], align 8
+// CHECK:   [[VLD3_LANE:%.*]] = call { <1 x i64>, <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld3lane.v1i64.p0(<1 x i64> [[TMP10]], <1 x i64> [[TMP11]], <1 x i64> [[TMP12]], i64 0, ptr %a)
+// CHECK:   store { <1 x i64>, <1 x i64>, <1 x i64> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 24, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.uint64x1x3_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.uint64x1x3_t [[TMP16]]
 uint64x1x3_t test_vld3_lane_u64(uint64_t  *a, uint64x1x3_t b) {
   return vld3_lane_u64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int8x8x3_t @test_vld3_lane_s8(i8* noundef %a, [3 x <8 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int8x8x3_t @test_vld3_lane_s8(ptr noundef %a, [3 x <8 x i8>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int8x8x3_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.int8x8x3_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.int8x8x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int8x8x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x8x3_t, %struct.int8x8x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <8 x i8>] [[B]].coerce, [3 x <8 x i8>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int8x8x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int8x8x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int8x8x3_t* [[__RET]] to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x8x3_t, %struct.int8x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i8>], [3 x <8 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX]], align 8
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x8x3_t, %struct.int8x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i8>], [3 x <8 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP4:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX2]], align 8
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int8x8x3_t, %struct.int8x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i8>], [3 x <8 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP5:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX4]], align 8
-// CHECK:   [[VLD3_LANE:%.*]] = call { <8 x i8>, <8 x i8>, <8 x i8> } @llvm.aarch64.neon.ld3lane.v8i8.p0i8(<8 x i8> [[TMP3]], <8 x i8> [[TMP4]], <8 x i8> [[TMP5]], i64 7, i8* %a)
-// CHECK:   [[TMP6:%.*]] = bitcast i8* [[TMP2]] to { <8 x i8>, <8 x i8>, <8 x i8> }*
-// CHECK:   store { <8 x i8>, <8 x i8>, <8 x i8> } [[VLD3_LANE]], { <8 x i8>, <8 x i8>, <8 x i8> }* [[TMP6]]
-// CHECK:   [[TMP7:%.*]] = bitcast %struct.int8x8x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP8:%.*]] = bitcast %struct.int8x8x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP7]], i8* align 8 [[TMP8]], i64 24, i1 false)
-// CHECK:   [[TMP9:%.*]] = load %struct.int8x8x3_t, %struct.int8x8x3_t* [[RETVAL]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x8x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <8 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x i8>, ptr [[ARRAYIDX]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP4:%.*]] = load <8 x i8>, ptr [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int8x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP5:%.*]] = load <8 x i8>, ptr [[ARRAYIDX4]], align 8
+// CHECK:   [[VLD3_LANE:%.*]] = call { <8 x i8>, <8 x i8>, <8 x i8> } @llvm.aarch64.neon.ld3lane.v8i8.p0(<8 x i8> [[TMP3]], <8 x i8> [[TMP4]], <8 x i8> [[TMP5]], i64 7, ptr %a)
+// CHECK:   store { <8 x i8>, <8 x i8>, <8 x i8> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 24, i1 false)
+// CHECK:   [[TMP9:%.*]] = load %struct.int8x8x3_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.int8x8x3_t [[TMP9]]
 int8x8x3_t test_vld3_lane_s8(int8_t  *a, int8x8x3_t b) {
   return vld3_lane_s8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int16x4x3_t @test_vld3_lane_s16(i16* noundef %a, [3 x <4 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int16x4x3_t @test_vld3_lane_s16(ptr noundef %a, [3 x <4 x i16>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int16x4x3_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.int16x4x3_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.int16x4x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int16x4x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x4x3_t, %struct.int16x4x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <4 x i16>] [[B]].coerce, [3 x <4 x i16>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int16x4x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int16x4x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int16x4x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x4x3_t, %struct.int16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i16>], [3 x <4 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x4x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <4 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x i16>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x i16> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x4x3_t, %struct.int16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i16>], [3 x <4 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x i16>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x i16> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int16x4x3_t, %struct.int16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i16>], [3 x <4 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <4 x i16>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <4 x i16> [[TMP8]] to <8 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP5]] to <4 x i16>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP7]] to <4 x i16>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP9]] to <4 x i16>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <4 x i16>, <4 x i16>, <4 x i16> } @llvm.aarch64.neon.ld3lane.v4i16.p0i8(<4 x i16> [[TMP10]], <4 x i16> [[TMP11]], <4 x i16> [[TMP12]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <4 x i16>, <4 x i16>, <4 x i16> }*
-// CHECK:   store { <4 x i16>, <4 x i16>, <4 x i16> } [[VLD3_LANE]], { <4 x i16>, <4 x i16>, <4 x i16> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.int16x4x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.int16x4x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP14]], i8* align 8 [[TMP15]], i64 24, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.int16x4x3_t, %struct.int16x4x3_t* [[RETVAL]], align 8
+// CHECK:   [[VLD3_LANE:%.*]] = call { <4 x i16>, <4 x i16>, <4 x i16> } @llvm.aarch64.neon.ld3lane.v4i16.p0(<4 x i16> [[TMP10]], <4 x i16> [[TMP11]], <4 x i16> [[TMP12]], i64 3, ptr %a)
+// CHECK:   store { <4 x i16>, <4 x i16>, <4 x i16> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 24, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.int16x4x3_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.int16x4x3_t [[TMP16]]
 int16x4x3_t test_vld3_lane_s16(int16_t  *a, int16x4x3_t b) {
   return vld3_lane_s16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int32x2x3_t @test_vld3_lane_s32(i32* noundef %a, [3 x <2 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int32x2x3_t @test_vld3_lane_s32(ptr noundef %a, [3 x <2 x i32>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int32x2x3_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.int32x2x3_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.int32x2x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int32x2x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x2x3_t, %struct.int32x2x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <2 x i32>] [[B]].coerce, [3 x <2 x i32>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int32x2x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int32x2x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int32x2x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x2x3_t, %struct.int32x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x i32>], [3 x <2 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x2x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <2 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x i32>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x i32> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x2x3_t, %struct.int32x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x i32>], [3 x <2 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x i32>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x i32> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int32x2x3_t, %struct.int32x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x i32>], [3 x <2 x i32>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int32x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x i32>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <2 x i32>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <2 x i32> [[TMP8]] to <8 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP5]] to <2 x i32>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP7]] to <2 x i32>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP9]] to <2 x i32>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <2 x i32>, <2 x i32>, <2 x i32> } @llvm.aarch64.neon.ld3lane.v2i32.p0i8(<2 x i32> [[TMP10]], <2 x i32> [[TMP11]], <2 x i32> [[TMP12]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <2 x i32>, <2 x i32>, <2 x i32> }*
-// CHECK:   store { <2 x i32>, <2 x i32>, <2 x i32> } [[VLD3_LANE]], { <2 x i32>, <2 x i32>, <2 x i32> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.int32x2x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.int32x2x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP14]], i8* align 8 [[TMP15]], i64 24, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.int32x2x3_t, %struct.int32x2x3_t* [[RETVAL]], align 8
+// CHECK:   [[VLD3_LANE:%.*]] = call { <2 x i32>, <2 x i32>, <2 x i32> } @llvm.aarch64.neon.ld3lane.v2i32.p0(<2 x i32> [[TMP10]], <2 x i32> [[TMP11]], <2 x i32> [[TMP12]], i64 1, ptr %a)
+// CHECK:   store { <2 x i32>, <2 x i32>, <2 x i32> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 24, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.int32x2x3_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.int32x2x3_t [[TMP16]]
 int32x2x3_t test_vld3_lane_s32(int32_t  *a, int32x2x3_t b) {
   return vld3_lane_s32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int64x1x3_t @test_vld3_lane_s64(i64* noundef %a, [3 x <1 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int64x1x3_t @test_vld3_lane_s64(ptr noundef %a, [3 x <1 x i64>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int64x1x3_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.int64x1x3_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.int64x1x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int64x1x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x1x3_t, %struct.int64x1x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <1 x i64>] [[B]].coerce, [3 x <1 x i64>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int64x1x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int64x1x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int64x1x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x1x3_t, %struct.int64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <1 x i64>], [3 x <1 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x1x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <1 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <1 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <1 x i64>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <1 x i64> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x1x3_t, %struct.int64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <1 x i64>], [3 x <1 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <1 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <1 x i64>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <1 x i64> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int64x1x3_t, %struct.int64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <1 x i64>], [3 x <1 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <1 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <1 x i64>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <1 x i64> [[TMP8]] to <8 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP5]] to <1 x i64>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP7]] to <1 x i64>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x i64>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <1 x i64>, <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld3lane.v1i64.p0i8(<1 x i64> [[TMP10]], <1 x i64> [[TMP11]], <1 x i64> [[TMP12]], i64 0, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <1 x i64>, <1 x i64>, <1 x i64> }*
-// CHECK:   store { <1 x i64>, <1 x i64>, <1 x i64> } [[VLD3_LANE]], { <1 x i64>, <1 x i64>, <1 x i64> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.int64x1x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.int64x1x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP14]], i8* align 8 [[TMP15]], i64 24, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.int64x1x3_t, %struct.int64x1x3_t* [[RETVAL]], align 8
+// CHECK:   [[VLD3_LANE:%.*]] = call { <1 x i64>, <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld3lane.v1i64.p0(<1 x i64> [[TMP10]], <1 x i64> [[TMP11]], <1 x i64> [[TMP12]], i64 0, ptr %a)
+// CHECK:   store { <1 x i64>, <1 x i64>, <1 x i64> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 24, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.int64x1x3_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.int64x1x3_t [[TMP16]]
 int64x1x3_t test_vld3_lane_s64(int64_t  *a, int64x1x3_t b) {
   return vld3_lane_s64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float16x4x3_t @test_vld3_lane_f16(half* noundef %a, [3 x <4 x half>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float16x4x3_t @test_vld3_lane_f16(ptr noundef %a, [3 x <4 x half>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float16x4x3_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.float16x4x3_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.float16x4x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.float16x4x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x4x3_t, %struct.float16x4x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <4 x half>] [[B]].coerce, [3 x <4 x half>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float16x4x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float16x4x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.float16x4x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast half* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x4x3_t, %struct.float16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x half>], [3 x <4 x half>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x half>, <4 x half>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x4x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <4 x half>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x half>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x half>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x half> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x4x3_t, %struct.float16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x half>], [3 x <4 x half>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x half>, <4 x half>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x half>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x half>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x half> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float16x4x3_t, %struct.float16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x half>], [3 x <4 x half>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <4 x half>, <4 x half>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x half>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <4 x half>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <4 x half> [[TMP8]] to <8 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP5]] to <4 x half>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP7]] to <4 x half>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP9]] to <4 x half>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <4 x half>, <4 x half>, <4 x half> } @llvm.aarch64.neon.ld3lane.v4f16.p0i8(<4 x half> [[TMP10]], <4 x half> [[TMP11]], <4 x half> [[TMP12]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <4 x half>, <4 x half>, <4 x half> }*
-// CHECK:   store { <4 x half>, <4 x half>, <4 x half> } [[VLD3_LANE]], { <4 x half>, <4 x half>, <4 x half> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.float16x4x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.float16x4x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP14]], i8* align 8 [[TMP15]], i64 24, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.float16x4x3_t, %struct.float16x4x3_t* [[RETVAL]], align 8
+// CHECK:   [[VLD3_LANE:%.*]] = call { <4 x half>, <4 x half>, <4 x half> } @llvm.aarch64.neon.ld3lane.v4f16.p0(<4 x half> [[TMP10]], <4 x half> [[TMP11]], <4 x half> [[TMP12]], i64 3, ptr %a)
+// CHECK:   store { <4 x half>, <4 x half>, <4 x half> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 24, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.float16x4x3_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.float16x4x3_t [[TMP16]]
 float16x4x3_t test_vld3_lane_f16(float16_t  *a, float16x4x3_t b) {
   return vld3_lane_f16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float32x2x3_t @test_vld3_lane_f32(float* noundef %a, [3 x <2 x float>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float32x2x3_t @test_vld3_lane_f32(ptr noundef %a, [3 x <2 x float>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float32x2x3_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.float32x2x3_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.float32x2x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.float32x2x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x2x3_t, %struct.float32x2x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <2 x float>] [[B]].coerce, [3 x <2 x float>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float32x2x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float32x2x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.float32x2x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast float* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x2x3_t, %struct.float32x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x float>], [3 x <2 x float>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x float>, <2 x float>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x2x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <2 x float>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x float>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x float>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x float> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x2x3_t, %struct.float32x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x float>], [3 x <2 x float>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x float>, <2 x float>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x float>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x float>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x float> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float32x2x3_t, %struct.float32x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x float>], [3 x <2 x float>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <2 x float>, <2 x float>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float32x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x float>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <2 x float>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <2 x float> [[TMP8]] to <8 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP5]] to <2 x float>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP7]] to <2 x float>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP9]] to <2 x float>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <2 x float>, <2 x float>, <2 x float> } @llvm.aarch64.neon.ld3lane.v2f32.p0i8(<2 x float> [[TMP10]], <2 x float> [[TMP11]], <2 x float> [[TMP12]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <2 x float>, <2 x float>, <2 x float> }*
-// CHECK:   store { <2 x float>, <2 x float>, <2 x float> } [[VLD3_LANE]], { <2 x float>, <2 x float>, <2 x float> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.float32x2x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.float32x2x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP14]], i8* align 8 [[TMP15]], i64 24, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.float32x2x3_t, %struct.float32x2x3_t* [[RETVAL]], align 8
+// CHECK:   [[VLD3_LANE:%.*]] = call { <2 x float>, <2 x float>, <2 x float> } @llvm.aarch64.neon.ld3lane.v2f32.p0(<2 x float> [[TMP10]], <2 x float> [[TMP11]], <2 x float> [[TMP12]], i64 1, ptr %a)
+// CHECK:   store { <2 x float>, <2 x float>, <2 x float> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 24, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.float32x2x3_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.float32x2x3_t [[TMP16]]
 float32x2x3_t test_vld3_lane_f32(float32_t  *a, float32x2x3_t b) {
   return vld3_lane_f32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float64x1x3_t @test_vld3_lane_f64(double* noundef %a, [3 x <1 x double>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float64x1x3_t @test_vld3_lane_f64(ptr noundef %a, [3 x <1 x double>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float64x1x3_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.float64x1x3_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.float64x1x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.float64x1x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x1x3_t, %struct.float64x1x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <1 x double>] [[B]].coerce, [3 x <1 x double>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float64x1x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float64x1x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.float64x1x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast double* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x1x3_t, %struct.float64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <1 x double>], [3 x <1 x double>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <1 x double>, <1 x double>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x1x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <1 x double>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <1 x double>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <1 x double>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <1 x double> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x1x3_t, %struct.float64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <1 x double>], [3 x <1 x double>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <1 x double>, <1 x double>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <1 x double>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <1 x double>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <1 x double> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float64x1x3_t, %struct.float64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <1 x double>], [3 x <1 x double>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <1 x double>, <1 x double>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <1 x double>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <1 x double>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <1 x double> [[TMP8]] to <8 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP5]] to <1 x double>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP7]] to <1 x double>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x double>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <1 x double>, <1 x double>, <1 x double> } @llvm.aarch64.neon.ld3lane.v1f64.p0i8(<1 x double> [[TMP10]], <1 x double> [[TMP11]], <1 x double> [[TMP12]], i64 0, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <1 x double>, <1 x double>, <1 x double> }*
-// CHECK:   store { <1 x double>, <1 x double>, <1 x double> } [[VLD3_LANE]], { <1 x double>, <1 x double>, <1 x double> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.float64x1x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.float64x1x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP14]], i8* align 8 [[TMP15]], i64 24, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.float64x1x3_t, %struct.float64x1x3_t* [[RETVAL]], align 8
+// CHECK:   [[VLD3_LANE:%.*]] = call { <1 x double>, <1 x double>, <1 x double> } @llvm.aarch64.neon.ld3lane.v1f64.p0(<1 x double> [[TMP10]], <1 x double> [[TMP11]], <1 x double> [[TMP12]], i64 0, ptr %a)
+// CHECK:   store { <1 x double>, <1 x double>, <1 x double> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 24, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.float64x1x3_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.float64x1x3_t [[TMP16]]
 float64x1x3_t test_vld3_lane_f64(float64_t  *a, float64x1x3_t b) {
   return vld3_lane_f64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly8x8x3_t @test_vld3_lane_p8(i8* noundef %a, [3 x <8 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly8x8x3_t @test_vld3_lane_p8(ptr noundef %a, [3 x <8 x i8>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly8x8x3_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.poly8x8x3_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly8x8x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly8x8x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x8x3_t, %struct.poly8x8x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <8 x i8>] [[B]].coerce, [3 x <8 x i8>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly8x8x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly8x8x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.poly8x8x3_t* [[__RET]] to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x8x3_t, %struct.poly8x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i8>], [3 x <8 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX]], align 8
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x8x3_t, %struct.poly8x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i8>], [3 x <8 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP4:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX2]], align 8
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly8x8x3_t, %struct.poly8x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i8>], [3 x <8 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP5:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX4]], align 8
-// CHECK:   [[VLD3_LANE:%.*]] = call { <8 x i8>, <8 x i8>, <8 x i8> } @llvm.aarch64.neon.ld3lane.v8i8.p0i8(<8 x i8> [[TMP3]], <8 x i8> [[TMP4]], <8 x i8> [[TMP5]], i64 7, i8* %a)
-// CHECK:   [[TMP6:%.*]] = bitcast i8* [[TMP2]] to { <8 x i8>, <8 x i8>, <8 x i8> }*
-// CHECK:   store { <8 x i8>, <8 x i8>, <8 x i8> } [[VLD3_LANE]], { <8 x i8>, <8 x i8>, <8 x i8> }* [[TMP6]]
-// CHECK:   [[TMP7:%.*]] = bitcast %struct.poly8x8x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP8:%.*]] = bitcast %struct.poly8x8x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP7]], i8* align 8 [[TMP8]], i64 24, i1 false)
-// CHECK:   [[TMP9:%.*]] = load %struct.poly8x8x3_t, %struct.poly8x8x3_t* [[RETVAL]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x8x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <8 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x i8>, ptr [[ARRAYIDX]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP4:%.*]] = load <8 x i8>, ptr [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly8x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP5:%.*]] = load <8 x i8>, ptr [[ARRAYIDX4]], align 8
+// CHECK:   [[VLD3_LANE:%.*]] = call { <8 x i8>, <8 x i8>, <8 x i8> } @llvm.aarch64.neon.ld3lane.v8i8.p0(<8 x i8> [[TMP3]], <8 x i8> [[TMP4]], <8 x i8> [[TMP5]], i64 7, ptr %a)
+// CHECK:   store { <8 x i8>, <8 x i8>, <8 x i8> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 24, i1 false)
+// CHECK:   [[TMP9:%.*]] = load %struct.poly8x8x3_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.poly8x8x3_t [[TMP9]]
 poly8x8x3_t test_vld3_lane_p8(poly8_t  *a, poly8x8x3_t b) {
   return vld3_lane_p8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly16x4x3_t @test_vld3_lane_p16(i16* noundef %a, [3 x <4 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly16x4x3_t @test_vld3_lane_p16(ptr noundef %a, [3 x <4 x i16>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly16x4x3_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.poly16x4x3_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly16x4x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly16x4x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x4x3_t, %struct.poly16x4x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <4 x i16>] [[B]].coerce, [3 x <4 x i16>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly16x4x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly16x4x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.poly16x4x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x4x3_t, %struct.poly16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i16>], [3 x <4 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x4x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <4 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x i16>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x i16> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x4x3_t, %struct.poly16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i16>], [3 x <4 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x i16>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x i16> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x4x3_t, %struct.poly16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i16>], [3 x <4 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <4 x i16>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <4 x i16> [[TMP8]] to <8 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP5]] to <4 x i16>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP7]] to <4 x i16>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP9]] to <4 x i16>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <4 x i16>, <4 x i16>, <4 x i16> } @llvm.aarch64.neon.ld3lane.v4i16.p0i8(<4 x i16> [[TMP10]], <4 x i16> [[TMP11]], <4 x i16> [[TMP12]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <4 x i16>, <4 x i16>, <4 x i16> }*
-// CHECK:   store { <4 x i16>, <4 x i16>, <4 x i16> } [[VLD3_LANE]], { <4 x i16>, <4 x i16>, <4 x i16> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.poly16x4x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.poly16x4x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP14]], i8* align 8 [[TMP15]], i64 24, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.poly16x4x3_t, %struct.poly16x4x3_t* [[RETVAL]], align 8
+// CHECK:   [[VLD3_LANE:%.*]] = call { <4 x i16>, <4 x i16>, <4 x i16> } @llvm.aarch64.neon.ld3lane.v4i16.p0(<4 x i16> [[TMP10]], <4 x i16> [[TMP11]], <4 x i16> [[TMP12]], i64 3, ptr %a)
+// CHECK:   store { <4 x i16>, <4 x i16>, <4 x i16> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 24, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.poly16x4x3_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.poly16x4x3_t [[TMP16]]
 poly16x4x3_t test_vld3_lane_p16(poly16_t  *a, poly16x4x3_t b) {
   return vld3_lane_p16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly64x1x3_t @test_vld3_lane_p64(i64* noundef %a, [3 x <1 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly64x1x3_t @test_vld3_lane_p64(ptr noundef %a, [3 x <1 x i64>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly64x1x3_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.poly64x1x3_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly64x1x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly64x1x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x1x3_t, %struct.poly64x1x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <1 x i64>] [[B]].coerce, [3 x <1 x i64>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly64x1x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly64x1x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.poly64x1x3_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x1x3_t, %struct.poly64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <1 x i64>], [3 x <1 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x1x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <1 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <1 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <1 x i64>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <1 x i64> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x1x3_t, %struct.poly64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <1 x i64>], [3 x <1 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <1 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <1 x i64>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <1 x i64> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly64x1x3_t, %struct.poly64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <1 x i64>], [3 x <1 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <1 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <1 x i64>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <1 x i64> [[TMP8]] to <8 x i8>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP5]] to <1 x i64>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP7]] to <1 x i64>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x i64>
-// CHECK:   [[VLD3_LANE:%.*]] = call { <1 x i64>, <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld3lane.v1i64.p0i8(<1 x i64> [[TMP10]], <1 x i64> [[TMP11]], <1 x i64> [[TMP12]], i64 0, i8* [[TMP3]])
-// CHECK:   [[TMP13:%.*]] = bitcast i8* [[TMP2]] to { <1 x i64>, <1 x i64>, <1 x i64> }*
-// CHECK:   store { <1 x i64>, <1 x i64>, <1 x i64> } [[VLD3_LANE]], { <1 x i64>, <1 x i64>, <1 x i64> }* [[TMP13]]
-// CHECK:   [[TMP14:%.*]] = bitcast %struct.poly64x1x3_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP15:%.*]] = bitcast %struct.poly64x1x3_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP14]], i8* align 8 [[TMP15]], i64 24, i1 false)
-// CHECK:   [[TMP16:%.*]] = load %struct.poly64x1x3_t, %struct.poly64x1x3_t* [[RETVAL]], align 8
+// CHECK:   [[VLD3_LANE:%.*]] = call { <1 x i64>, <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld3lane.v1i64.p0(<1 x i64> [[TMP10]], <1 x i64> [[TMP11]], <1 x i64> [[TMP12]], i64 0, ptr %a)
+// CHECK:   store { <1 x i64>, <1 x i64>, <1 x i64> } [[VLD3_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 24, i1 false)
+// CHECK:   [[TMP16:%.*]] = load %struct.poly64x1x3_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.poly64x1x3_t [[TMP16]]
 poly64x1x3_t test_vld3_lane_p64(poly64_t  *a, poly64x1x3_t b) {
   return vld3_lane_p64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint8x16x4_t @test_vld4q_lane_u8(i8* noundef %a, [4 x <16 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint8x16x4_t @test_vld4q_lane_u8(ptr noundef %a, [4 x <16 x i8>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint8x16x4_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.uint8x16x4_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint8x16x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint8x16x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, %struct.uint8x16x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <16 x i8>] [[B]].coerce, [4 x <16 x i8>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint8x16x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint8x16x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint8x16x4_t* [[__RET]] to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, %struct.uint8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, %struct.uint8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, %struct.uint8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 16
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, %struct.uint8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP6:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX6]], align 16
-// CHECK:   [[VLD4_LANE:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> } @llvm.aarch64.neon.ld4lane.v16i8.p0i8(<16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], <16 x i8> [[TMP6]], i64 15, i8* %a)
-// CHECK:   [[TMP7:%.*]] = bitcast i8* [[TMP2]] to { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> }*
-// CHECK:   store { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> } [[VLD4_LANE]], { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> }* [[TMP7]]
-// CHECK:   [[TMP8:%.*]] = bitcast %struct.uint8x16x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP9:%.*]] = bitcast %struct.uint8x16x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP8]], i8* align 16 [[TMP9]], i64 64, i1 false)
-// CHECK:   [[TMP10:%.*]] = load %struct.uint8x16x4_t, %struct.uint8x16x4_t* [[RETVAL]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <16 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <16 x i8>, ptr [[ARRAYIDX]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP4:%.*]] = load <16 x i8>, ptr [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP5:%.*]] = load <16 x i8>, ptr [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP6:%.*]] = load <16 x i8>, ptr [[ARRAYIDX6]], align 16
+// CHECK:   [[VLD4_LANE:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> } @llvm.aarch64.neon.ld4lane.v16i8.p0(<16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], <16 x i8> [[TMP6]], i64 15, ptr %a)
+// CHECK:   store { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 64, i1 false)
+// CHECK:   [[TMP10:%.*]] = load %struct.uint8x16x4_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.uint8x16x4_t [[TMP10]]
 uint8x16x4_t test_vld4q_lane_u8(uint8_t  *a, uint8x16x4_t b) {
   return vld4q_lane_u8(a, b, 15);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint16x8x4_t @test_vld4q_lane_u16(i16* noundef %a, [4 x <8 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint16x8x4_t @test_vld4q_lane_u16(ptr noundef %a, [4 x <8 x i16>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint16x8x4_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.uint16x8x4_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint16x8x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint16x8x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <8 x i16>] [[B]].coerce, [4 x <8 x i16>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint16x8x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint16x8x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint16x8x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <8 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <8 x i16>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <8 x i16>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <8 x i16>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i16> [[TMP8]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <8 x i16>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i16> [[TMP10]] to <16 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP9]] to <8 x i16>
 // CHECK:   [[TMP15:%.*]] = bitcast <16 x i8> [[TMP11]] to <8 x i16>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> } @llvm.aarch64.neon.ld4lane.v8i16.p0i8(<8 x i16> [[TMP12]], <8 x i16> [[TMP13]], <8 x i16> [[TMP14]], <8 x i16> [[TMP15]], i64 7, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> }*
-// CHECK:   store { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> } [[VLD4_LANE]], { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.uint16x8x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.uint16x8x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP17]], i8* align 16 [[TMP18]], i64 64, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[RETVAL]], align 16
+// CHECK:   [[VLD4_LANE:%.*]] = call { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> } @llvm.aarch64.neon.ld4lane.v8i16.p0(<8 x i16> [[TMP12]], <8 x i16> [[TMP13]], <8 x i16> [[TMP14]], <8 x i16> [[TMP15]], i64 7, ptr %a)
+// CHECK:   store { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 64, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.uint16x8x4_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.uint16x8x4_t [[TMP19]]
 uint16x8x4_t test_vld4q_lane_u16(uint16_t  *a, uint16x8x4_t b) {
   return vld4q_lane_u16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint32x4x4_t @test_vld4q_lane_u32(i32* noundef %a, [4 x <4 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint32x4x4_t @test_vld4q_lane_u32(ptr noundef %a, [4 x <4 x i32>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint32x4x4_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.uint32x4x4_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint32x4x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint32x4x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <4 x i32>] [[B]].coerce, [4 x <4 x i32>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint32x4x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint32x4x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint32x4x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <4 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x i32>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x i32> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x i32>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x i32> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i32>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <4 x i32>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <4 x i32> [[TMP8]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i32>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <4 x i32>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP11:%.*]] = bitcast <4 x i32> [[TMP10]] to <16 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x i32>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x i32>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP9]] to <4 x i32>
 // CHECK:   [[TMP15:%.*]] = bitcast <16 x i8> [[TMP11]] to <4 x i32>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld4lane.v4i32.p0i8(<4 x i32> [[TMP12]], <4 x i32> [[TMP13]], <4 x i32> [[TMP14]], <4 x i32> [[TMP15]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> }*
-// CHECK:   store { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } [[VLD4_LANE]], { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.uint32x4x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.uint32x4x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP17]], i8* align 16 [[TMP18]], i64 64, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[RETVAL]], align 16
+// CHECK:   [[VLD4_LANE:%.*]] = call { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld4lane.v4i32.p0(<4 x i32> [[TMP12]], <4 x i32> [[TMP13]], <4 x i32> [[TMP14]], <4 x i32> [[TMP15]], i64 3, ptr %a)
+// CHECK:   store { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 64, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.uint32x4x4_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.uint32x4x4_t [[TMP19]]
 uint32x4x4_t test_vld4q_lane_u32(uint32_t  *a, uint32x4x4_t b) {
   return vld4q_lane_u32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint64x2x4_t @test_vld4q_lane_u64(i64* noundef %a, [4 x <2 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint64x2x4_t @test_vld4q_lane_u64(ptr noundef %a, [4 x <2 x i64>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint64x2x4_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.uint64x2x4_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint64x2x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint64x2x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x2x4_t, %struct.uint64x2x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <2 x i64>] [[B]].coerce, [4 x <2 x i64>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint64x2x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint64x2x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint64x2x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x2x4_t, %struct.uint64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x2x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <2 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x i64>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x i64> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x2x4_t, %struct.uint64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x i64>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x i64> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint64x2x4_t, %struct.uint64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <2 x i64>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <2 x i64> [[TMP8]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint64x2x4_t, %struct.uint64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <2 x i64>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP11:%.*]] = bitcast <2 x i64> [[TMP10]] to <16 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <2 x i64>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <2 x i64>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP9]] to <2 x i64>
 // CHECK:   [[TMP15:%.*]] = bitcast <16 x i8> [[TMP11]] to <2 x i64>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld4lane.v2i64.p0i8(<2 x i64> [[TMP12]], <2 x i64> [[TMP13]], <2 x i64> [[TMP14]], <2 x i64> [[TMP15]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> }*
-// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } [[VLD4_LANE]], { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.uint64x2x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.uint64x2x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP17]], i8* align 16 [[TMP18]], i64 64, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.uint64x2x4_t, %struct.uint64x2x4_t* [[RETVAL]], align 16
+// CHECK:   [[VLD4_LANE:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld4lane.v2i64.p0(<2 x i64> [[TMP12]], <2 x i64> [[TMP13]], <2 x i64> [[TMP14]], <2 x i64> [[TMP15]], i64 1, ptr %a)
+// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 64, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.uint64x2x4_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.uint64x2x4_t [[TMP19]]
 uint64x2x4_t test_vld4q_lane_u64(uint64_t  *a, uint64x2x4_t b) {
   return vld4q_lane_u64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int8x16x4_t @test_vld4q_lane_s8(i8* noundef %a, [4 x <16 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int8x16x4_t @test_vld4q_lane_s8(ptr noundef %a, [4 x <16 x i8>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int8x16x4_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.int8x16x4_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.int8x16x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int8x16x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x16x4_t, %struct.int8x16x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <16 x i8>] [[B]].coerce, [4 x <16 x i8>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int8x16x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int8x16x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int8x16x4_t* [[__RET]] to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x16x4_t, %struct.int8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x16x4_t, %struct.int8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int8x16x4_t, %struct.int8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 16
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int8x16x4_t, %struct.int8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP6:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX6]], align 16
-// CHECK:   [[VLD4_LANE:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> } @llvm.aarch64.neon.ld4lane.v16i8.p0i8(<16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], <16 x i8> [[TMP6]], i64 15, i8* %a)
-// CHECK:   [[TMP7:%.*]] = bitcast i8* [[TMP2]] to { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> }*
-// CHECK:   store { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> } [[VLD4_LANE]], { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> }* [[TMP7]]
-// CHECK:   [[TMP8:%.*]] = bitcast %struct.int8x16x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP9:%.*]] = bitcast %struct.int8x16x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP8]], i8* align 16 [[TMP9]], i64 64, i1 false)
-// CHECK:   [[TMP10:%.*]] = load %struct.int8x16x4_t, %struct.int8x16x4_t* [[RETVAL]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x16x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <16 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <16 x i8>, ptr [[ARRAYIDX]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP4:%.*]] = load <16 x i8>, ptr [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP5:%.*]] = load <16 x i8>, ptr [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP6:%.*]] = load <16 x i8>, ptr [[ARRAYIDX6]], align 16
+// CHECK:   [[VLD4_LANE:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> } @llvm.aarch64.neon.ld4lane.v16i8.p0(<16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], <16 x i8> [[TMP6]], i64 15, ptr %a)
+// CHECK:   store { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 64, i1 false)
+// CHECK:   [[TMP10:%.*]] = load %struct.int8x16x4_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.int8x16x4_t [[TMP10]]
 int8x16x4_t test_vld4q_lane_s8(int8_t  *a, int8x16x4_t b) {
   return vld4q_lane_s8(a, b, 15);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int16x8x4_t @test_vld4q_lane_s16(i16* noundef %a, [4 x <8 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int16x8x4_t @test_vld4q_lane_s16(ptr noundef %a, [4 x <8 x i16>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int16x8x4_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.int16x8x4_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.int16x8x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int16x8x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <8 x i16>] [[B]].coerce, [4 x <8 x i16>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int16x8x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int16x8x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int16x8x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <8 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <8 x i16>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <8 x i16>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <8 x i16>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i16> [[TMP8]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <8 x i16>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i16> [[TMP10]] to <16 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP9]] to <8 x i16>
 // CHECK:   [[TMP15:%.*]] = bitcast <16 x i8> [[TMP11]] to <8 x i16>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> } @llvm.aarch64.neon.ld4lane.v8i16.p0i8(<8 x i16> [[TMP12]], <8 x i16> [[TMP13]], <8 x i16> [[TMP14]], <8 x i16> [[TMP15]], i64 7, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> }*
-// CHECK:   store { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> } [[VLD4_LANE]], { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.int16x8x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.int16x8x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP17]], i8* align 16 [[TMP18]], i64 64, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.int16x8x4_t, %struct.int16x8x4_t* [[RETVAL]], align 16
+// CHECK:   [[VLD4_LANE:%.*]] = call { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> } @llvm.aarch64.neon.ld4lane.v8i16.p0(<8 x i16> [[TMP12]], <8 x i16> [[TMP13]], <8 x i16> [[TMP14]], <8 x i16> [[TMP15]], i64 7, ptr %a)
+// CHECK:   store { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 64, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.int16x8x4_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.int16x8x4_t [[TMP19]]
 int16x8x4_t test_vld4q_lane_s16(int16_t  *a, int16x8x4_t b) {
   return vld4q_lane_s16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int32x4x4_t @test_vld4q_lane_s32(i32* noundef %a, [4 x <4 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int32x4x4_t @test_vld4q_lane_s32(ptr noundef %a, [4 x <4 x i32>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int32x4x4_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.int32x4x4_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.int32x4x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int32x4x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <4 x i32>] [[B]].coerce, [4 x <4 x i32>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int32x4x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int32x4x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int32x4x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <4 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x i32>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x i32> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x i32>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x i32> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i32>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <4 x i32>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <4 x i32> [[TMP8]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i32>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <4 x i32>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP11:%.*]] = bitcast <4 x i32> [[TMP10]] to <16 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x i32>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x i32>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP9]] to <4 x i32>
 // CHECK:   [[TMP15:%.*]] = bitcast <16 x i8> [[TMP11]] to <4 x i32>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld4lane.v4i32.p0i8(<4 x i32> [[TMP12]], <4 x i32> [[TMP13]], <4 x i32> [[TMP14]], <4 x i32> [[TMP15]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> }*
-// CHECK:   store { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } [[VLD4_LANE]], { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.int32x4x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.int32x4x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP17]], i8* align 16 [[TMP18]], i64 64, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.int32x4x4_t, %struct.int32x4x4_t* [[RETVAL]], align 16
+// CHECK:   [[VLD4_LANE:%.*]] = call { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld4lane.v4i32.p0(<4 x i32> [[TMP12]], <4 x i32> [[TMP13]], <4 x i32> [[TMP14]], <4 x i32> [[TMP15]], i64 3, ptr %a)
+// CHECK:   store { <4 x i32>, <4 x i32>, <4 x i32>, <4 x i32> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 64, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.int32x4x4_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.int32x4x4_t [[TMP19]]
 int32x4x4_t test_vld4q_lane_s32(int32_t  *a, int32x4x4_t b) {
   return vld4q_lane_s32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int64x2x4_t @test_vld4q_lane_s64(i64* noundef %a, [4 x <2 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int64x2x4_t @test_vld4q_lane_s64(ptr noundef %a, [4 x <2 x i64>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int64x2x4_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.int64x2x4_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.int64x2x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int64x2x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x2x4_t, %struct.int64x2x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <2 x i64>] [[B]].coerce, [4 x <2 x i64>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int64x2x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int64x2x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int64x2x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x2x4_t, %struct.int64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x2x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <2 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x i64>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x i64> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x2x4_t, %struct.int64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x i64>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x i64> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int64x2x4_t, %struct.int64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <2 x i64>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <2 x i64> [[TMP8]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int64x2x4_t, %struct.int64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <2 x i64>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP11:%.*]] = bitcast <2 x i64> [[TMP10]] to <16 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <2 x i64>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <2 x i64>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP9]] to <2 x i64>
 // CHECK:   [[TMP15:%.*]] = bitcast <16 x i8> [[TMP11]] to <2 x i64>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld4lane.v2i64.p0i8(<2 x i64> [[TMP12]], <2 x i64> [[TMP13]], <2 x i64> [[TMP14]], <2 x i64> [[TMP15]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> }*
-// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } [[VLD4_LANE]], { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.int64x2x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.int64x2x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP17]], i8* align 16 [[TMP18]], i64 64, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.int64x2x4_t, %struct.int64x2x4_t* [[RETVAL]], align 16
+// CHECK:   [[VLD4_LANE:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld4lane.v2i64.p0(<2 x i64> [[TMP12]], <2 x i64> [[TMP13]], <2 x i64> [[TMP14]], <2 x i64> [[TMP15]], i64 1, ptr %a)
+// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 64, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.int64x2x4_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.int64x2x4_t [[TMP19]]
 int64x2x4_t test_vld4q_lane_s64(int64_t  *a, int64x2x4_t b) {
   return vld4q_lane_s64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float16x8x4_t @test_vld4q_lane_f16(half* noundef %a, [4 x <8 x half>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float16x8x4_t @test_vld4q_lane_f16(ptr noundef %a, [4 x <8 x half>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float16x8x4_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.float16x8x4_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.float16x8x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.float16x8x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <8 x half>] [[B]].coerce, [4 x <8 x half>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float16x8x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float16x8x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.float16x8x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast half* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x half>], [4 x <8 x half>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <8 x half>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x half>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <8 x half>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <8 x half> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x half>], [4 x <8 x half>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x half>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <8 x half>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x half> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x half>], [4 x <8 x half>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x half>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <8 x half>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x half> [[TMP8]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x half>], [4 x <8 x half>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x half>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <8 x half>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x half> [[TMP10]] to <16 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x half>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x half>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP9]] to <8 x half>
 // CHECK:   [[TMP15:%.*]] = bitcast <16 x i8> [[TMP11]] to <8 x half>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <8 x half>, <8 x half>, <8 x half>, <8 x half> } @llvm.aarch64.neon.ld4lane.v8f16.p0i8(<8 x half> [[TMP12]], <8 x half> [[TMP13]], <8 x half> [[TMP14]], <8 x half> [[TMP15]], i64 7, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <8 x half>, <8 x half>, <8 x half>, <8 x half> }*
-// CHECK:   store { <8 x half>, <8 x half>, <8 x half>, <8 x half> } [[VLD4_LANE]], { <8 x half>, <8 x half>, <8 x half>, <8 x half> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.float16x8x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.float16x8x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP17]], i8* align 16 [[TMP18]], i64 64, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.float16x8x4_t, %struct.float16x8x4_t* [[RETVAL]], align 16
+// CHECK:   [[VLD4_LANE:%.*]] = call { <8 x half>, <8 x half>, <8 x half>, <8 x half> } @llvm.aarch64.neon.ld4lane.v8f16.p0(<8 x half> [[TMP12]], <8 x half> [[TMP13]], <8 x half> [[TMP14]], <8 x half> [[TMP15]], i64 7, ptr %a)
+// CHECK:   store { <8 x half>, <8 x half>, <8 x half>, <8 x half> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 64, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.float16x8x4_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.float16x8x4_t [[TMP19]]
 float16x8x4_t test_vld4q_lane_f16(float16_t  *a, float16x8x4_t b) {
   return vld4q_lane_f16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float32x4x4_t @test_vld4q_lane_f32(float* noundef %a, [4 x <4 x float>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float32x4x4_t @test_vld4q_lane_f32(ptr noundef %a, [4 x <4 x float>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float32x4x4_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.float32x4x4_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.float32x4x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.float32x4x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <4 x float>] [[B]].coerce, [4 x <4 x float>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float32x4x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float32x4x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.float32x4x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast float* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x float>], [4 x <4 x float>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <4 x float>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x float>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x float>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x float> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x float>], [4 x <4 x float>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x float>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x float>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x float> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x float>], [4 x <4 x float>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x float>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <4 x float>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <4 x float> [[TMP8]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x float>], [4 x <4 x float>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x float>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <4 x float>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP11:%.*]] = bitcast <4 x float> [[TMP10]] to <16 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <4 x float>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <4 x float>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP9]] to <4 x float>
 // CHECK:   [[TMP15:%.*]] = bitcast <16 x i8> [[TMP11]] to <4 x float>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <4 x float>, <4 x float>, <4 x float>, <4 x float> } @llvm.aarch64.neon.ld4lane.v4f32.p0i8(<4 x float> [[TMP12]], <4 x float> [[TMP13]], <4 x float> [[TMP14]], <4 x float> [[TMP15]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <4 x float>, <4 x float>, <4 x float>, <4 x float> }*
-// CHECK:   store { <4 x float>, <4 x float>, <4 x float>, <4 x float> } [[VLD4_LANE]], { <4 x float>, <4 x float>, <4 x float>, <4 x float> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.float32x4x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.float32x4x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP17]], i8* align 16 [[TMP18]], i64 64, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.float32x4x4_t, %struct.float32x4x4_t* [[RETVAL]], align 16
+// CHECK:   [[VLD4_LANE:%.*]] = call { <4 x float>, <4 x float>, <4 x float>, <4 x float> } @llvm.aarch64.neon.ld4lane.v4f32.p0(<4 x float> [[TMP12]], <4 x float> [[TMP13]], <4 x float> [[TMP14]], <4 x float> [[TMP15]], i64 3, ptr %a)
+// CHECK:   store { <4 x float>, <4 x float>, <4 x float>, <4 x float> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 64, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.float32x4x4_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.float32x4x4_t [[TMP19]]
 float32x4x4_t test_vld4q_lane_f32(float32_t  *a, float32x4x4_t b) {
   return vld4q_lane_f32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float64x2x4_t @test_vld4q_lane_f64(double* noundef %a, [4 x <2 x double>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float64x2x4_t @test_vld4q_lane_f64(ptr noundef %a, [4 x <2 x double>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float64x2x4_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.float64x2x4_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.float64x2x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.float64x2x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x2x4_t, %struct.float64x2x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <2 x double>] [[B]].coerce, [4 x <2 x double>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float64x2x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float64x2x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.float64x2x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast double* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x2x4_t, %struct.float64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x double>], [4 x <2 x double>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x double>, <2 x double>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x2x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <2 x double>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x double>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x double>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x double> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x2x4_t, %struct.float64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x double>], [4 x <2 x double>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x double>, <2 x double>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x double>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x double>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x double> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float64x2x4_t, %struct.float64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x double>], [4 x <2 x double>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <2 x double>, <2 x double>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x double>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <2 x double>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <2 x double> [[TMP8]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float64x2x4_t, %struct.float64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x double>], [4 x <2 x double>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <2 x double>, <2 x double>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x double>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <2 x double>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP11:%.*]] = bitcast <2 x double> [[TMP10]] to <16 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <2 x double>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <2 x double>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP9]] to <2 x double>
 // CHECK:   [[TMP15:%.*]] = bitcast <16 x i8> [[TMP11]] to <2 x double>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <2 x double>, <2 x double>, <2 x double>, <2 x double> } @llvm.aarch64.neon.ld4lane.v2f64.p0i8(<2 x double> [[TMP12]], <2 x double> [[TMP13]], <2 x double> [[TMP14]], <2 x double> [[TMP15]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <2 x double>, <2 x double>, <2 x double>, <2 x double> }*
-// CHECK:   store { <2 x double>, <2 x double>, <2 x double>, <2 x double> } [[VLD4_LANE]], { <2 x double>, <2 x double>, <2 x double>, <2 x double> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.float64x2x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.float64x2x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP17]], i8* align 16 [[TMP18]], i64 64, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.float64x2x4_t, %struct.float64x2x4_t* [[RETVAL]], align 16
+// CHECK:   [[VLD4_LANE:%.*]] = call { <2 x double>, <2 x double>, <2 x double>, <2 x double> } @llvm.aarch64.neon.ld4lane.v2f64.p0(<2 x double> [[TMP12]], <2 x double> [[TMP13]], <2 x double> [[TMP14]], <2 x double> [[TMP15]], i64 1, ptr %a)
+// CHECK:   store { <2 x double>, <2 x double>, <2 x double>, <2 x double> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 64, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.float64x2x4_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.float64x2x4_t [[TMP19]]
 float64x2x4_t test_vld4q_lane_f64(float64_t  *a, float64x2x4_t b) {
   return vld4q_lane_f64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly8x16x4_t @test_vld4q_lane_p8(i8* noundef %a, [4 x <16 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly8x16x4_t @test_vld4q_lane_p8(ptr noundef %a, [4 x <16 x i8>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly8x16x4_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.poly8x16x4_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly8x16x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly8x16x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, %struct.poly8x16x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <16 x i8>] [[B]].coerce, [4 x <16 x i8>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly8x16x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly8x16x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.poly8x16x4_t* [[__RET]] to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, %struct.poly8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, %struct.poly8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, %struct.poly8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 16
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, %struct.poly8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP6:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX6]], align 16
-// CHECK:   [[VLD4_LANE:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> } @llvm.aarch64.neon.ld4lane.v16i8.p0i8(<16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], <16 x i8> [[TMP6]], i64 15, i8* %a)
-// CHECK:   [[TMP7:%.*]] = bitcast i8* [[TMP2]] to { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> }*
-// CHECK:   store { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> } [[VLD4_LANE]], { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> }* [[TMP7]]
-// CHECK:   [[TMP8:%.*]] = bitcast %struct.poly8x16x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP9:%.*]] = bitcast %struct.poly8x16x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP8]], i8* align 16 [[TMP9]], i64 64, i1 false)
-// CHECK:   [[TMP10:%.*]] = load %struct.poly8x16x4_t, %struct.poly8x16x4_t* [[RETVAL]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <16 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <16 x i8>, ptr [[ARRAYIDX]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP4:%.*]] = load <16 x i8>, ptr [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP5:%.*]] = load <16 x i8>, ptr [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP6:%.*]] = load <16 x i8>, ptr [[ARRAYIDX6]], align 16
+// CHECK:   [[VLD4_LANE:%.*]] = call { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> } @llvm.aarch64.neon.ld4lane.v16i8.p0(<16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], <16 x i8> [[TMP6]], i64 15, ptr %a)
+// CHECK:   store { <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 64, i1 false)
+// CHECK:   [[TMP10:%.*]] = load %struct.poly8x16x4_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.poly8x16x4_t [[TMP10]]
 poly8x16x4_t test_vld4q_lane_p8(poly8_t  *a, poly8x16x4_t b) {
   return vld4q_lane_p8(a, b, 15);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly16x8x4_t @test_vld4q_lane_p16(i16* noundef %a, [4 x <8 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly16x8x4_t @test_vld4q_lane_p16(ptr noundef %a, [4 x <8 x i16>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly16x8x4_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.poly16x8x4_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly16x8x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly16x8x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <8 x i16>] [[B]].coerce, [4 x <8 x i16>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly16x8x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly16x8x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.poly16x8x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <8 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <8 x i16>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <8 x i16> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <8 x i16>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x i16> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <8 x i16>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i16> [[TMP8]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <8 x i16>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i16> [[TMP10]] to <16 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <8 x i16>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <8 x i16>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP9]] to <8 x i16>
 // CHECK:   [[TMP15:%.*]] = bitcast <16 x i8> [[TMP11]] to <8 x i16>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> } @llvm.aarch64.neon.ld4lane.v8i16.p0i8(<8 x i16> [[TMP12]], <8 x i16> [[TMP13]], <8 x i16> [[TMP14]], <8 x i16> [[TMP15]], i64 7, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> }*
-// CHECK:   store { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> } [[VLD4_LANE]], { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.poly16x8x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.poly16x8x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP17]], i8* align 16 [[TMP18]], i64 64, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[RETVAL]], align 16
+// CHECK:   [[VLD4_LANE:%.*]] = call { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> } @llvm.aarch64.neon.ld4lane.v8i16.p0(<8 x i16> [[TMP12]], <8 x i16> [[TMP13]], <8 x i16> [[TMP14]], <8 x i16> [[TMP15]], i64 7, ptr %a)
+// CHECK:   store { <8 x i16>, <8 x i16>, <8 x i16>, <8 x i16> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 64, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.poly16x8x4_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.poly16x8x4_t [[TMP19]]
 poly16x8x4_t test_vld4q_lane_p16(poly16_t  *a, poly16x8x4_t b) {
   return vld4q_lane_p16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly64x2x4_t @test_vld4q_lane_p64(i64* noundef %a, [4 x <2 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly64x2x4_t @test_vld4q_lane_p64(ptr noundef %a, [4 x <2 x i64>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly64x2x4_t, align 16
 // CHECK:   [[B:%.*]] = alloca %struct.poly64x2x4_t, align 16
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly64x2x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly64x2x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x2x4_t, %struct.poly64x2x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <2 x i64>] [[B]].coerce, [4 x <2 x i64>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly64x2x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly64x2x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.poly64x2x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x2x4_t, %struct.poly64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x2x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <2 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x i64>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x i64> [[TMP4]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x2x4_t, %struct.poly64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x i64>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x i64> [[TMP6]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly64x2x4_t, %struct.poly64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <2 x i64>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP9:%.*]] = bitcast <2 x i64> [[TMP8]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly64x2x4_t, %struct.poly64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <2 x i64>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP11:%.*]] = bitcast <2 x i64> [[TMP10]] to <16 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP5]] to <2 x i64>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP7]] to <2 x i64>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP9]] to <2 x i64>
 // CHECK:   [[TMP15:%.*]] = bitcast <16 x i8> [[TMP11]] to <2 x i64>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld4lane.v2i64.p0i8(<2 x i64> [[TMP12]], <2 x i64> [[TMP13]], <2 x i64> [[TMP14]], <2 x i64> [[TMP15]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> }*
-// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } [[VLD4_LANE]], { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.poly64x2x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.poly64x2x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP17]], i8* align 16 [[TMP18]], i64 64, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.poly64x2x4_t, %struct.poly64x2x4_t* [[RETVAL]], align 16
+// CHECK:   [[VLD4_LANE:%.*]] = call { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } @llvm.aarch64.neon.ld4lane.v2i64.p0(<2 x i64> [[TMP12]], <2 x i64> [[TMP13]], <2 x i64> [[TMP14]], <2 x i64> [[TMP15]], i64 1, ptr %a)
+// CHECK:   store { <2 x i64>, <2 x i64>, <2 x i64>, <2 x i64> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[RETVAL]], ptr align 16 [[__RET]], i64 64, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.poly64x2x4_t, ptr [[RETVAL]], align 16
 // CHECK:   ret %struct.poly64x2x4_t [[TMP19]]
 poly64x2x4_t test_vld4q_lane_p64(poly64_t  *a, poly64x2x4_t b) {
   return vld4q_lane_p64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint8x8x4_t @test_vld4_lane_u8(i8* noundef %a, [4 x <8 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint8x8x4_t @test_vld4_lane_u8(ptr noundef %a, [4 x <8 x i8>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint8x8x4_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.uint8x8x4_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint8x8x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint8x8x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x8x4_t, %struct.uint8x8x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <8 x i8>] [[B]].coerce, [4 x <8 x i8>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint8x8x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint8x8x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint8x8x4_t* [[__RET]] to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x8x4_t, %struct.uint8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX]], align 8
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x8x4_t, %struct.uint8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP4:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX2]], align 8
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint8x8x4_t, %struct.uint8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP5:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX4]], align 8
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint8x8x4_t, %struct.uint8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP6:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX6]], align 8
-// CHECK:   [[VLD4_LANE:%.*]] = call { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> } @llvm.aarch64.neon.ld4lane.v8i8.p0i8(<8 x i8> [[TMP3]], <8 x i8> [[TMP4]], <8 x i8> [[TMP5]], <8 x i8> [[TMP6]], i64 7, i8* %a)
-// CHECK:   [[TMP7:%.*]] = bitcast i8* [[TMP2]] to { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> }*
-// CHECK:   store { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> } [[VLD4_LANE]], { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> }* [[TMP7]]
-// CHECK:   [[TMP8:%.*]] = bitcast %struct.uint8x8x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP9:%.*]] = bitcast %struct.uint8x8x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP8]], i8* align 8 [[TMP9]], i64 32, i1 false)
-// CHECK:   [[TMP10:%.*]] = load %struct.uint8x8x4_t, %struct.uint8x8x4_t* [[RETVAL]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x8x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <8 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x i8>, ptr [[ARRAYIDX]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP4:%.*]] = load <8 x i8>, ptr [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP5:%.*]] = load <8 x i8>, ptr [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP6:%.*]] = load <8 x i8>, ptr [[ARRAYIDX6]], align 8
+// CHECK:   [[VLD4_LANE:%.*]] = call { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> } @llvm.aarch64.neon.ld4lane.v8i8.p0(<8 x i8> [[TMP3]], <8 x i8> [[TMP4]], <8 x i8> [[TMP5]], <8 x i8> [[TMP6]], i64 7, ptr %a)
+// CHECK:   store { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP10:%.*]] = load %struct.uint8x8x4_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.uint8x8x4_t [[TMP10]]
 uint8x8x4_t test_vld4_lane_u8(uint8_t  *a, uint8x8x4_t b) {
   return vld4_lane_u8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint16x4x4_t @test_vld4_lane_u16(i16* noundef %a, [4 x <4 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint16x4x4_t @test_vld4_lane_u16(ptr noundef %a, [4 x <4 x i16>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint16x4x4_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.uint16x4x4_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint16x4x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint16x4x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x4x4_t, %struct.uint16x4x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <4 x i16>] [[B]].coerce, [4 x <4 x i16>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint16x4x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint16x4x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint16x4x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x4x4_t, %struct.uint16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x4x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <4 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x i16>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x i16> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x4x4_t, %struct.uint16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x i16>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x i16> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x4x4_t, %struct.uint16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <4 x i16>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <4 x i16> [[TMP8]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint16x4x4_t, %struct.uint16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <4 x i16>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP11:%.*]] = bitcast <4 x i16> [[TMP10]] to <8 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP5]] to <4 x i16>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP7]] to <4 x i16>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP9]] to <4 x i16>
 // CHECK:   [[TMP15:%.*]] = bitcast <8 x i8> [[TMP11]] to <4 x i16>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> } @llvm.aarch64.neon.ld4lane.v4i16.p0i8(<4 x i16> [[TMP12]], <4 x i16> [[TMP13]], <4 x i16> [[TMP14]], <4 x i16> [[TMP15]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> }*
-// CHECK:   store { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> } [[VLD4_LANE]], { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.uint16x4x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.uint16x4x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP17]], i8* align 8 [[TMP18]], i64 32, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.uint16x4x4_t, %struct.uint16x4x4_t* [[RETVAL]], align 8
+// CHECK:   [[VLD4_LANE:%.*]] = call { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> } @llvm.aarch64.neon.ld4lane.v4i16.p0(<4 x i16> [[TMP12]], <4 x i16> [[TMP13]], <4 x i16> [[TMP14]], <4 x i16> [[TMP15]], i64 3, ptr %a)
+// CHECK:   store { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.uint16x4x4_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.uint16x4x4_t [[TMP19]]
 uint16x4x4_t test_vld4_lane_u16(uint16_t  *a, uint16x4x4_t b) {
   return vld4_lane_u16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint32x2x4_t @test_vld4_lane_u32(i32* noundef %a, [4 x <2 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint32x2x4_t @test_vld4_lane_u32(ptr noundef %a, [4 x <2 x i32>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint32x2x4_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.uint32x2x4_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint32x2x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint32x2x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x2x4_t, %struct.uint32x2x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <2 x i32>] [[B]].coerce, [4 x <2 x i32>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint32x2x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint32x2x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint32x2x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x2x4_t, %struct.uint32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x i32>], [4 x <2 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x2x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <2 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x i32>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x i32> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x2x4_t, %struct.uint32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x i32>], [4 x <2 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x i32>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x i32> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x2x4_t, %struct.uint32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x i32>], [4 x <2 x i32>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x i32>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <2 x i32>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <2 x i32> [[TMP8]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint32x2x4_t, %struct.uint32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x i32>], [4 x <2 x i32>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x i32>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <2 x i32>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP11:%.*]] = bitcast <2 x i32> [[TMP10]] to <8 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP5]] to <2 x i32>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP7]] to <2 x i32>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP9]] to <2 x i32>
 // CHECK:   [[TMP15:%.*]] = bitcast <8 x i8> [[TMP11]] to <2 x i32>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> } @llvm.aarch64.neon.ld4lane.v2i32.p0i8(<2 x i32> [[TMP12]], <2 x i32> [[TMP13]], <2 x i32> [[TMP14]], <2 x i32> [[TMP15]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> }*
-// CHECK:   store { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> } [[VLD4_LANE]], { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.uint32x2x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.uint32x2x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP17]], i8* align 8 [[TMP18]], i64 32, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.uint32x2x4_t, %struct.uint32x2x4_t* [[RETVAL]], align 8
+// CHECK:   [[VLD4_LANE:%.*]] = call { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> } @llvm.aarch64.neon.ld4lane.v2i32.p0(<2 x i32> [[TMP12]], <2 x i32> [[TMP13]], <2 x i32> [[TMP14]], <2 x i32> [[TMP15]], i64 1, ptr %a)
+// CHECK:   store { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.uint32x2x4_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.uint32x2x4_t [[TMP19]]
 uint32x2x4_t test_vld4_lane_u32(uint32_t  *a, uint32x2x4_t b) {
   return vld4_lane_u32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.uint64x1x4_t @test_vld4_lane_u64(i64* noundef %a, [4 x <1 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.uint64x1x4_t @test_vld4_lane_u64(ptr noundef %a, [4 x <1 x i64>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.uint64x1x4_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.uint64x1x4_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.uint64x1x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint64x1x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x1x4_t, %struct.uint64x1x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <1 x i64>] [[B]].coerce, [4 x <1 x i64>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint64x1x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint64x1x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.uint64x1x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x1x4_t, %struct.uint64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x1x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <1 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <1 x i64>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <1 x i64> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x1x4_t, %struct.uint64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <1 x i64>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <1 x i64> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint64x1x4_t, %struct.uint64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <1 x i64>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <1 x i64> [[TMP8]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint64x1x4_t, %struct.uint64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <1 x i64>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP11:%.*]] = bitcast <1 x i64> [[TMP10]] to <8 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP5]] to <1 x i64>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP7]] to <1 x i64>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x i64>
 // CHECK:   [[TMP15:%.*]] = bitcast <8 x i8> [[TMP11]] to <1 x i64>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld4lane.v1i64.p0i8(<1 x i64> [[TMP12]], <1 x i64> [[TMP13]], <1 x i64> [[TMP14]], <1 x i64> [[TMP15]], i64 0, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> }*
-// CHECK:   store { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> } [[VLD4_LANE]], { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.uint64x1x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.uint64x1x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP17]], i8* align 8 [[TMP18]], i64 32, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.uint64x1x4_t, %struct.uint64x1x4_t* [[RETVAL]], align 8
+// CHECK:   [[VLD4_LANE:%.*]] = call { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld4lane.v1i64.p0(<1 x i64> [[TMP12]], <1 x i64> [[TMP13]], <1 x i64> [[TMP14]], <1 x i64> [[TMP15]], i64 0, ptr %a)
+// CHECK:   store { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.uint64x1x4_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.uint64x1x4_t [[TMP19]]
 uint64x1x4_t test_vld4_lane_u64(uint64_t  *a, uint64x1x4_t b) {
   return vld4_lane_u64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int8x8x4_t @test_vld4_lane_s8(i8* noundef %a, [4 x <8 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int8x8x4_t @test_vld4_lane_s8(ptr noundef %a, [4 x <8 x i8>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int8x8x4_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.int8x8x4_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.int8x8x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int8x8x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x8x4_t, %struct.int8x8x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <8 x i8>] [[B]].coerce, [4 x <8 x i8>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int8x8x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int8x8x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int8x8x4_t* [[__RET]] to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x8x4_t, %struct.int8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX]], align 8
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x8x4_t, %struct.int8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP4:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX2]], align 8
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int8x8x4_t, %struct.int8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP5:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX4]], align 8
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int8x8x4_t, %struct.int8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP6:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX6]], align 8
-// CHECK:   [[VLD4_LANE:%.*]] = call { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> } @llvm.aarch64.neon.ld4lane.v8i8.p0i8(<8 x i8> [[TMP3]], <8 x i8> [[TMP4]], <8 x i8> [[TMP5]], <8 x i8> [[TMP6]], i64 7, i8* %a)
-// CHECK:   [[TMP7:%.*]] = bitcast i8* [[TMP2]] to { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> }*
-// CHECK:   store { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> } [[VLD4_LANE]], { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> }* [[TMP7]]
-// CHECK:   [[TMP8:%.*]] = bitcast %struct.int8x8x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP9:%.*]] = bitcast %struct.int8x8x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP8]], i8* align 8 [[TMP9]], i64 32, i1 false)
-// CHECK:   [[TMP10:%.*]] = load %struct.int8x8x4_t, %struct.int8x8x4_t* [[RETVAL]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x8x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <8 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x i8>, ptr [[ARRAYIDX]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP4:%.*]] = load <8 x i8>, ptr [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP5:%.*]] = load <8 x i8>, ptr [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP6:%.*]] = load <8 x i8>, ptr [[ARRAYIDX6]], align 8
+// CHECK:   [[VLD4_LANE:%.*]] = call { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> } @llvm.aarch64.neon.ld4lane.v8i8.p0(<8 x i8> [[TMP3]], <8 x i8> [[TMP4]], <8 x i8> [[TMP5]], <8 x i8> [[TMP6]], i64 7, ptr %a)
+// CHECK:   store { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP10:%.*]] = load %struct.int8x8x4_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.int8x8x4_t [[TMP10]]
 int8x8x4_t test_vld4_lane_s8(int8_t  *a, int8x8x4_t b) {
   return vld4_lane_s8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int16x4x4_t @test_vld4_lane_s16(i16* noundef %a, [4 x <4 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int16x4x4_t @test_vld4_lane_s16(ptr noundef %a, [4 x <4 x i16>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int16x4x4_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.int16x4x4_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.int16x4x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int16x4x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x4x4_t, %struct.int16x4x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <4 x i16>] [[B]].coerce, [4 x <4 x i16>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int16x4x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int16x4x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int16x4x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x4x4_t, %struct.int16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x4x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <4 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x i16>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x i16> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x4x4_t, %struct.int16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x i16>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x i16> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int16x4x4_t, %struct.int16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <4 x i16>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <4 x i16> [[TMP8]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int16x4x4_t, %struct.int16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <4 x i16>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP11:%.*]] = bitcast <4 x i16> [[TMP10]] to <8 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP5]] to <4 x i16>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP7]] to <4 x i16>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP9]] to <4 x i16>
 // CHECK:   [[TMP15:%.*]] = bitcast <8 x i8> [[TMP11]] to <4 x i16>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> } @llvm.aarch64.neon.ld4lane.v4i16.p0i8(<4 x i16> [[TMP12]], <4 x i16> [[TMP13]], <4 x i16> [[TMP14]], <4 x i16> [[TMP15]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> }*
-// CHECK:   store { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> } [[VLD4_LANE]], { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.int16x4x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.int16x4x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP17]], i8* align 8 [[TMP18]], i64 32, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.int16x4x4_t, %struct.int16x4x4_t* [[RETVAL]], align 8
+// CHECK:   [[VLD4_LANE:%.*]] = call { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> } @llvm.aarch64.neon.ld4lane.v4i16.p0(<4 x i16> [[TMP12]], <4 x i16> [[TMP13]], <4 x i16> [[TMP14]], <4 x i16> [[TMP15]], i64 3, ptr %a)
+// CHECK:   store { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.int16x4x4_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.int16x4x4_t [[TMP19]]
 int16x4x4_t test_vld4_lane_s16(int16_t  *a, int16x4x4_t b) {
   return vld4_lane_s16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int32x2x4_t @test_vld4_lane_s32(i32* noundef %a, [4 x <2 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int32x2x4_t @test_vld4_lane_s32(ptr noundef %a, [4 x <2 x i32>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int32x2x4_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.int32x2x4_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.int32x2x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int32x2x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x2x4_t, %struct.int32x2x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <2 x i32>] [[B]].coerce, [4 x <2 x i32>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int32x2x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int32x2x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int32x2x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x2x4_t, %struct.int32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x i32>], [4 x <2 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x2x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <2 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x i32>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x i32> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x2x4_t, %struct.int32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x i32>], [4 x <2 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x i32>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x i32> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int32x2x4_t, %struct.int32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x i32>], [4 x <2 x i32>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x i32>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <2 x i32>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <2 x i32> [[TMP8]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int32x2x4_t, %struct.int32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x i32>], [4 x <2 x i32>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x i32>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <2 x i32>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP11:%.*]] = bitcast <2 x i32> [[TMP10]] to <8 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP5]] to <2 x i32>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP7]] to <2 x i32>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP9]] to <2 x i32>
 // CHECK:   [[TMP15:%.*]] = bitcast <8 x i8> [[TMP11]] to <2 x i32>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> } @llvm.aarch64.neon.ld4lane.v2i32.p0i8(<2 x i32> [[TMP12]], <2 x i32> [[TMP13]], <2 x i32> [[TMP14]], <2 x i32> [[TMP15]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> }*
-// CHECK:   store { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> } [[VLD4_LANE]], { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.int32x2x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.int32x2x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP17]], i8* align 8 [[TMP18]], i64 32, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.int32x2x4_t, %struct.int32x2x4_t* [[RETVAL]], align 8
+// CHECK:   [[VLD4_LANE:%.*]] = call { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> } @llvm.aarch64.neon.ld4lane.v2i32.p0(<2 x i32> [[TMP12]], <2 x i32> [[TMP13]], <2 x i32> [[TMP14]], <2 x i32> [[TMP15]], i64 1, ptr %a)
+// CHECK:   store { <2 x i32>, <2 x i32>, <2 x i32>, <2 x i32> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.int32x2x4_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.int32x2x4_t [[TMP19]]
 int32x2x4_t test_vld4_lane_s32(int32_t  *a, int32x2x4_t b) {
   return vld4_lane_s32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.int64x1x4_t @test_vld4_lane_s64(i64* noundef %a, [4 x <1 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.int64x1x4_t @test_vld4_lane_s64(ptr noundef %a, [4 x <1 x i64>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.int64x1x4_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.int64x1x4_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.int64x1x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int64x1x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x1x4_t, %struct.int64x1x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <1 x i64>] [[B]].coerce, [4 x <1 x i64>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int64x1x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int64x1x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.int64x1x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x1x4_t, %struct.int64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x1x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <1 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <1 x i64>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <1 x i64> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x1x4_t, %struct.int64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <1 x i64>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <1 x i64> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int64x1x4_t, %struct.int64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <1 x i64>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <1 x i64> [[TMP8]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int64x1x4_t, %struct.int64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <1 x i64>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP11:%.*]] = bitcast <1 x i64> [[TMP10]] to <8 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP5]] to <1 x i64>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP7]] to <1 x i64>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x i64>
 // CHECK:   [[TMP15:%.*]] = bitcast <8 x i8> [[TMP11]] to <1 x i64>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld4lane.v1i64.p0i8(<1 x i64> [[TMP12]], <1 x i64> [[TMP13]], <1 x i64> [[TMP14]], <1 x i64> [[TMP15]], i64 0, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> }*
-// CHECK:   store { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> } [[VLD4_LANE]], { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.int64x1x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.int64x1x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP17]], i8* align 8 [[TMP18]], i64 32, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.int64x1x4_t, %struct.int64x1x4_t* [[RETVAL]], align 8
+// CHECK:   [[VLD4_LANE:%.*]] = call { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld4lane.v1i64.p0(<1 x i64> [[TMP12]], <1 x i64> [[TMP13]], <1 x i64> [[TMP14]], <1 x i64> [[TMP15]], i64 0, ptr %a)
+// CHECK:   store { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.int64x1x4_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.int64x1x4_t [[TMP19]]
 int64x1x4_t test_vld4_lane_s64(int64_t  *a, int64x1x4_t b) {
   return vld4_lane_s64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float16x4x4_t @test_vld4_lane_f16(half* noundef %a, [4 x <4 x half>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float16x4x4_t @test_vld4_lane_f16(ptr noundef %a, [4 x <4 x half>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float16x4x4_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.float16x4x4_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.float16x4x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.float16x4x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x4x4_t, %struct.float16x4x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <4 x half>] [[B]].coerce, [4 x <4 x half>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float16x4x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float16x4x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.float16x4x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast half* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x4x4_t, %struct.float16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x half>], [4 x <4 x half>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x half>, <4 x half>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x4x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <4 x half>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x half>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x half>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x half> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x4x4_t, %struct.float16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x half>], [4 x <4 x half>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x half>, <4 x half>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x half>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x half>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x half> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float16x4x4_t, %struct.float16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x half>], [4 x <4 x half>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <4 x half>, <4 x half>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x half>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <4 x half>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <4 x half> [[TMP8]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float16x4x4_t, %struct.float16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x half>], [4 x <4 x half>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <4 x half>, <4 x half>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x half>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <4 x half>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP11:%.*]] = bitcast <4 x half> [[TMP10]] to <8 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP5]] to <4 x half>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP7]] to <4 x half>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP9]] to <4 x half>
 // CHECK:   [[TMP15:%.*]] = bitcast <8 x i8> [[TMP11]] to <4 x half>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <4 x half>, <4 x half>, <4 x half>, <4 x half> } @llvm.aarch64.neon.ld4lane.v4f16.p0i8(<4 x half> [[TMP12]], <4 x half> [[TMP13]], <4 x half> [[TMP14]], <4 x half> [[TMP15]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <4 x half>, <4 x half>, <4 x half>, <4 x half> }*
-// CHECK:   store { <4 x half>, <4 x half>, <4 x half>, <4 x half> } [[VLD4_LANE]], { <4 x half>, <4 x half>, <4 x half>, <4 x half> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.float16x4x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.float16x4x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP17]], i8* align 8 [[TMP18]], i64 32, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.float16x4x4_t, %struct.float16x4x4_t* [[RETVAL]], align 8
+// CHECK:   [[VLD4_LANE:%.*]] = call { <4 x half>, <4 x half>, <4 x half>, <4 x half> } @llvm.aarch64.neon.ld4lane.v4f16.p0(<4 x half> [[TMP12]], <4 x half> [[TMP13]], <4 x half> [[TMP14]], <4 x half> [[TMP15]], i64 3, ptr %a)
+// CHECK:   store { <4 x half>, <4 x half>, <4 x half>, <4 x half> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.float16x4x4_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.float16x4x4_t [[TMP19]]
 float16x4x4_t test_vld4_lane_f16(float16_t  *a, float16x4x4_t b) {
   return vld4_lane_f16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float32x2x4_t @test_vld4_lane_f32(float* noundef %a, [4 x <2 x float>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float32x2x4_t @test_vld4_lane_f32(ptr noundef %a, [4 x <2 x float>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float32x2x4_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.float32x2x4_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.float32x2x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.float32x2x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x2x4_t, %struct.float32x2x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <2 x float>] [[B]].coerce, [4 x <2 x float>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float32x2x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float32x2x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.float32x2x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast float* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x2x4_t, %struct.float32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x float>], [4 x <2 x float>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <2 x float>, <2 x float>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x2x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <2 x float>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x float>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <2 x float>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <2 x float> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x2x4_t, %struct.float32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x float>], [4 x <2 x float>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <2 x float>, <2 x float>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x float>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <2 x float>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <2 x float> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float32x2x4_t, %struct.float32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x float>], [4 x <2 x float>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <2 x float>, <2 x float>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x float>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <2 x float>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <2 x float> [[TMP8]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float32x2x4_t, %struct.float32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x float>], [4 x <2 x float>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <2 x float>, <2 x float>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x float>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <2 x float>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP11:%.*]] = bitcast <2 x float> [[TMP10]] to <8 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP5]] to <2 x float>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP7]] to <2 x float>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP9]] to <2 x float>
 // CHECK:   [[TMP15:%.*]] = bitcast <8 x i8> [[TMP11]] to <2 x float>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <2 x float>, <2 x float>, <2 x float>, <2 x float> } @llvm.aarch64.neon.ld4lane.v2f32.p0i8(<2 x float> [[TMP12]], <2 x float> [[TMP13]], <2 x float> [[TMP14]], <2 x float> [[TMP15]], i64 1, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <2 x float>, <2 x float>, <2 x float>, <2 x float> }*
-// CHECK:   store { <2 x float>, <2 x float>, <2 x float>, <2 x float> } [[VLD4_LANE]], { <2 x float>, <2 x float>, <2 x float>, <2 x float> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.float32x2x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.float32x2x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP17]], i8* align 8 [[TMP18]], i64 32, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.float32x2x4_t, %struct.float32x2x4_t* [[RETVAL]], align 8
+// CHECK:   [[VLD4_LANE:%.*]] = call { <2 x float>, <2 x float>, <2 x float>, <2 x float> } @llvm.aarch64.neon.ld4lane.v2f32.p0(<2 x float> [[TMP12]], <2 x float> [[TMP13]], <2 x float> [[TMP14]], <2 x float> [[TMP15]], i64 1, ptr %a)
+// CHECK:   store { <2 x float>, <2 x float>, <2 x float>, <2 x float> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.float32x2x4_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.float32x2x4_t [[TMP19]]
 float32x2x4_t test_vld4_lane_f32(float32_t  *a, float32x2x4_t b) {
   return vld4_lane_f32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.float64x1x4_t @test_vld4_lane_f64(double* noundef %a, [4 x <1 x double>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.float64x1x4_t @test_vld4_lane_f64(ptr noundef %a, [4 x <1 x double>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.float64x1x4_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.float64x1x4_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.float64x1x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.float64x1x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x1x4_t, %struct.float64x1x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <1 x double>] [[B]].coerce, [4 x <1 x double>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float64x1x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float64x1x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.float64x1x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast double* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x1x4_t, %struct.float64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <1 x double>], [4 x <1 x double>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <1 x double>, <1 x double>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x1x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <1 x double>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <1 x double>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <1 x double>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <1 x double> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x1x4_t, %struct.float64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <1 x double>], [4 x <1 x double>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <1 x double>, <1 x double>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <1 x double>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <1 x double>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <1 x double> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float64x1x4_t, %struct.float64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <1 x double>], [4 x <1 x double>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <1 x double>, <1 x double>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <1 x double>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <1 x double>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <1 x double> [[TMP8]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float64x1x4_t, %struct.float64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <1 x double>], [4 x <1 x double>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <1 x double>, <1 x double>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <1 x double>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <1 x double>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP11:%.*]] = bitcast <1 x double> [[TMP10]] to <8 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP5]] to <1 x double>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP7]] to <1 x double>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x double>
 // CHECK:   [[TMP15:%.*]] = bitcast <8 x i8> [[TMP11]] to <1 x double>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <1 x double>, <1 x double>, <1 x double>, <1 x double> } @llvm.aarch64.neon.ld4lane.v1f64.p0i8(<1 x double> [[TMP12]], <1 x double> [[TMP13]], <1 x double> [[TMP14]], <1 x double> [[TMP15]], i64 0, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <1 x double>, <1 x double>, <1 x double>, <1 x double> }*
-// CHECK:   store { <1 x double>, <1 x double>, <1 x double>, <1 x double> } [[VLD4_LANE]], { <1 x double>, <1 x double>, <1 x double>, <1 x double> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.float64x1x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.float64x1x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP17]], i8* align 8 [[TMP18]], i64 32, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.float64x1x4_t, %struct.float64x1x4_t* [[RETVAL]], align 8
+// CHECK:   [[VLD4_LANE:%.*]] = call { <1 x double>, <1 x double>, <1 x double>, <1 x double> } @llvm.aarch64.neon.ld4lane.v1f64.p0(<1 x double> [[TMP12]], <1 x double> [[TMP13]], <1 x double> [[TMP14]], <1 x double> [[TMP15]], i64 0, ptr %a)
+// CHECK:   store { <1 x double>, <1 x double>, <1 x double>, <1 x double> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.float64x1x4_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.float64x1x4_t [[TMP19]]
 float64x1x4_t test_vld4_lane_f64(float64_t  *a, float64x1x4_t b) {
   return vld4_lane_f64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly8x8x4_t @test_vld4_lane_p8(i8* noundef %a, [4 x <8 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly8x8x4_t @test_vld4_lane_p8(ptr noundef %a, [4 x <8 x i8>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly8x8x4_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.poly8x8x4_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly8x8x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly8x8x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x8x4_t, %struct.poly8x8x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <8 x i8>] [[B]].coerce, [4 x <8 x i8>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly8x8x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly8x8x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.poly8x8x4_t* [[__RET]] to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x8x4_t, %struct.poly8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX]], align 8
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x8x4_t, %struct.poly8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP4:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX2]], align 8
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly8x8x4_t, %struct.poly8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP5:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX4]], align 8
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly8x8x4_t, %struct.poly8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP6:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX6]], align 8
-// CHECK:   [[VLD4_LANE:%.*]] = call { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> } @llvm.aarch64.neon.ld4lane.v8i8.p0i8(<8 x i8> [[TMP3]], <8 x i8> [[TMP4]], <8 x i8> [[TMP5]], <8 x i8> [[TMP6]], i64 7, i8* %a)
-// CHECK:   [[TMP7:%.*]] = bitcast i8* [[TMP2]] to { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> }*
-// CHECK:   store { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> } [[VLD4_LANE]], { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> }* [[TMP7]]
-// CHECK:   [[TMP8:%.*]] = bitcast %struct.poly8x8x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP9:%.*]] = bitcast %struct.poly8x8x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP8]], i8* align 8 [[TMP9]], i64 32, i1 false)
-// CHECK:   [[TMP10:%.*]] = load %struct.poly8x8x4_t, %struct.poly8x8x4_t* [[RETVAL]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x8x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <8 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x i8>, ptr [[ARRAYIDX]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP4:%.*]] = load <8 x i8>, ptr [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP5:%.*]] = load <8 x i8>, ptr [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP6:%.*]] = load <8 x i8>, ptr [[ARRAYIDX6]], align 8
+// CHECK:   [[VLD4_LANE:%.*]] = call { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> } @llvm.aarch64.neon.ld4lane.v8i8.p0(<8 x i8> [[TMP3]], <8 x i8> [[TMP4]], <8 x i8> [[TMP5]], <8 x i8> [[TMP6]], i64 7, ptr %a)
+// CHECK:   store { <8 x i8>, <8 x i8>, <8 x i8>, <8 x i8> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP10:%.*]] = load %struct.poly8x8x4_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.poly8x8x4_t [[TMP10]]
 poly8x8x4_t test_vld4_lane_p8(poly8_t  *a, poly8x8x4_t b) {
   return vld4_lane_p8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly16x4x4_t @test_vld4_lane_p16(i16* noundef %a, [4 x <4 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly16x4x4_t @test_vld4_lane_p16(ptr noundef %a, [4 x <4 x i16>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly16x4x4_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.poly16x4x4_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly16x4x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly16x4x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x4x4_t, %struct.poly16x4x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <4 x i16>] [[B]].coerce, [4 x <4 x i16>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly16x4x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly16x4x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.poly16x4x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x4x4_t, %struct.poly16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x4x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <4 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <4 x i16>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <4 x i16> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x4x4_t, %struct.poly16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <4 x i16>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <4 x i16> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x4x4_t, %struct.poly16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <4 x i16>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <4 x i16> [[TMP8]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly16x4x4_t, %struct.poly16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <4 x i16>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP11:%.*]] = bitcast <4 x i16> [[TMP10]] to <8 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP5]] to <4 x i16>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP7]] to <4 x i16>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP9]] to <4 x i16>
 // CHECK:   [[TMP15:%.*]] = bitcast <8 x i8> [[TMP11]] to <4 x i16>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> } @llvm.aarch64.neon.ld4lane.v4i16.p0i8(<4 x i16> [[TMP12]], <4 x i16> [[TMP13]], <4 x i16> [[TMP14]], <4 x i16> [[TMP15]], i64 3, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> }*
-// CHECK:   store { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> } [[VLD4_LANE]], { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.poly16x4x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.poly16x4x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP17]], i8* align 8 [[TMP18]], i64 32, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.poly16x4x4_t, %struct.poly16x4x4_t* [[RETVAL]], align 8
+// CHECK:   [[VLD4_LANE:%.*]] = call { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> } @llvm.aarch64.neon.ld4lane.v4i16.p0(<4 x i16> [[TMP12]], <4 x i16> [[TMP13]], <4 x i16> [[TMP14]], <4 x i16> [[TMP15]], i64 3, ptr %a)
+// CHECK:   store { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.poly16x4x4_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.poly16x4x4_t [[TMP19]]
 poly16x4x4_t test_vld4_lane_p16(poly16_t  *a, poly16x4x4_t b) {
   return vld4_lane_p16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} %struct.poly64x1x4_t @test_vld4_lane_p64(i64* noundef %a, [4 x <1 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} %struct.poly64x1x4_t @test_vld4_lane_p64(ptr noundef %a, [4 x <1 x i64>] %b.coerce) #2 {
 // CHECK:   [[RETVAL:%.*]] = alloca %struct.poly64x1x4_t, align 8
 // CHECK:   [[B:%.*]] = alloca %struct.poly64x1x4_t, align 8
 // CHECK:   [[__RET:%.*]] = alloca %struct.poly64x1x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly64x1x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x1x4_t, %struct.poly64x1x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <1 x i64>] [[B]].coerce, [4 x <1 x i64>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly64x1x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly64x1x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast %struct.poly64x1x4_t* [[__RET]] to i8*
-// CHECK:   [[TMP3:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x1x4_t, %struct.poly64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP4:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x1x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <1 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP4:%.*]] = load <1 x i64>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP5:%.*]] = bitcast <1 x i64> [[TMP4]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x1x4_t, %struct.poly64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP6:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP6:%.*]] = load <1 x i64>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP7:%.*]] = bitcast <1 x i64> [[TMP6]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly64x1x4_t, %struct.poly64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP8:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP8:%.*]] = load <1 x i64>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP9:%.*]] = bitcast <1 x i64> [[TMP8]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly64x1x4_t, %struct.poly64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP10:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP10:%.*]] = load <1 x i64>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP11:%.*]] = bitcast <1 x i64> [[TMP10]] to <8 x i8>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP5]] to <1 x i64>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP7]] to <1 x i64>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x i64>
 // CHECK:   [[TMP15:%.*]] = bitcast <8 x i8> [[TMP11]] to <1 x i64>
-// CHECK:   [[VLD4_LANE:%.*]] = call { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld4lane.v1i64.p0i8(<1 x i64> [[TMP12]], <1 x i64> [[TMP13]], <1 x i64> [[TMP14]], <1 x i64> [[TMP15]], i64 0, i8* [[TMP3]])
-// CHECK:   [[TMP16:%.*]] = bitcast i8* [[TMP2]] to { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> }*
-// CHECK:   store { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> } [[VLD4_LANE]], { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> }* [[TMP16]]
-// CHECK:   [[TMP17:%.*]] = bitcast %struct.poly64x1x4_t* [[RETVAL]] to i8*
-// CHECK:   [[TMP18:%.*]] = bitcast %struct.poly64x1x4_t* [[__RET]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP17]], i8* align 8 [[TMP18]], i64 32, i1 false)
-// CHECK:   [[TMP19:%.*]] = load %struct.poly64x1x4_t, %struct.poly64x1x4_t* [[RETVAL]], align 8
+// CHECK:   [[VLD4_LANE:%.*]] = call { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> } @llvm.aarch64.neon.ld4lane.v1i64.p0(<1 x i64> [[TMP12]], <1 x i64> [[TMP13]], <1 x i64> [[TMP14]], <1 x i64> [[TMP15]], i64 0, ptr %a)
+// CHECK:   store { <1 x i64>, <1 x i64>, <1 x i64>, <1 x i64> } [[VLD4_LANE]], ptr [[__RET]]
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[RETVAL]], ptr align 8 [[__RET]], i64 32, i1 false)
+// CHECK:   [[TMP19:%.*]] = load %struct.poly64x1x4_t, ptr [[RETVAL]], align 8
 // CHECK:   ret %struct.poly64x1x4_t [[TMP19]]
 poly64x1x4_t test_vld4_lane_p64(poly64_t  *a, poly64x1x4_t b) {
   return vld4_lane_p64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_u8(i8* noundef %a, <16 x i8> noundef %b) #0 {
+// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_u8(ptr noundef %a, <16 x i8> noundef %b) #0 {
 // CHECK:   [[TMP0:%.*]] = extractelement <16 x i8> %b, i32 15
-// CHECK:   store i8 [[TMP0]], i8* %a
+// CHECK:   store i8 [[TMP0]], ptr %a
 // CHECK:   ret void
 void test_vst1q_lane_u8(uint8_t  *a, uint8x16_t b) {
   vst1q_lane_u8(a, b, 15);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_u16(i16* noundef %a, <8 x i16> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i16* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_u16(ptr noundef %a, <8 x i16> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <8 x i16> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <8 x i16>
 // CHECK:   [[TMP3:%.*]] = extractelement <8 x i16> [[TMP2]], i32 7
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to i16*
-// CHECK:   store i16 [[TMP3]], i16* [[TMP4]]
+// CHECK:   store i16 [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1q_lane_u16(uint16_t  *a, uint16x8_t b) {
   vst1q_lane_u16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_u32(i32* noundef %a, <4 x i32> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i32* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_u32(ptr noundef %a, <4 x i32> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i32> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <4 x i32>
 // CHECK:   [[TMP3:%.*]] = extractelement <4 x i32> [[TMP2]], i32 3
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to i32*
-// CHECK:   store i32 [[TMP3]], i32* [[TMP4]]
+// CHECK:   store i32 [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1q_lane_u32(uint32_t  *a, uint32x4_t b) {
   vst1q_lane_u32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_u64(i64* noundef %a, <2 x i64> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i64* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_u64(ptr noundef %a, <2 x i64> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x i64> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <2 x i64>
 // CHECK:   [[TMP3:%.*]] = extractelement <2 x i64> [[TMP2]], i32 1
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to i64*
-// CHECK:   store i64 [[TMP3]], i64* [[TMP4]]
+// CHECK:   store i64 [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1q_lane_u64(uint64_t  *a, uint64x2_t b) {
   vst1q_lane_u64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_s8(i8* noundef %a, <16 x i8> noundef %b) #0 {
+// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_s8(ptr noundef %a, <16 x i8> noundef %b) #0 {
 // CHECK:   [[TMP0:%.*]] = extractelement <16 x i8> %b, i32 15
-// CHECK:   store i8 [[TMP0]], i8* %a
+// CHECK:   store i8 [[TMP0]], ptr %a
 // CHECK:   ret void
 void test_vst1q_lane_s8(int8_t  *a, int8x16_t b) {
   vst1q_lane_s8(a, b, 15);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_s16(i16* noundef %a, <8 x i16> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i16* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_s16(ptr noundef %a, <8 x i16> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <8 x i16> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <8 x i16>
 // CHECK:   [[TMP3:%.*]] = extractelement <8 x i16> [[TMP2]], i32 7
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to i16*
-// CHECK:   store i16 [[TMP3]], i16* [[TMP4]]
+// CHECK:   store i16 [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1q_lane_s16(int16_t  *a, int16x8_t b) {
   vst1q_lane_s16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_s32(i32* noundef %a, <4 x i32> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i32* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_s32(ptr noundef %a, <4 x i32> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i32> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <4 x i32>
 // CHECK:   [[TMP3:%.*]] = extractelement <4 x i32> [[TMP2]], i32 3
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to i32*
-// CHECK:   store i32 [[TMP3]], i32* [[TMP4]]
+// CHECK:   store i32 [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1q_lane_s32(int32_t  *a, int32x4_t b) {
   vst1q_lane_s32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_s64(i64* noundef %a, <2 x i64> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i64* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_s64(ptr noundef %a, <2 x i64> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x i64> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <2 x i64>
 // CHECK:   [[TMP3:%.*]] = extractelement <2 x i64> [[TMP2]], i32 1
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to i64*
-// CHECK:   store i64 [[TMP3]], i64* [[TMP4]]
+// CHECK:   store i64 [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1q_lane_s64(int64_t  *a, int64x2_t b) {
   vst1q_lane_s64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_f16(half* noundef %a, <8 x half> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast half* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_f16(ptr noundef %a, <8 x half> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <8 x half> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <8 x half>
 // CHECK:   [[TMP3:%.*]] = extractelement <8 x half> [[TMP2]], i32 7
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to half*
-// CHECK:   store half [[TMP3]], half* [[TMP4]]
+// CHECK:   store half [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1q_lane_f16(float16_t  *a, float16x8_t b) {
   vst1q_lane_f16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_f32(float* noundef %a, <4 x float> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast float* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_f32(ptr noundef %a, <4 x float> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x float> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <4 x float>
 // CHECK:   [[TMP3:%.*]] = extractelement <4 x float> [[TMP2]], i32 3
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to float*
-// CHECK:   store float [[TMP3]], float* [[TMP4]]
+// CHECK:   store float [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1q_lane_f32(float32_t  *a, float32x4_t b) {
   vst1q_lane_f32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_f64(double* noundef %a, <2 x double> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast double* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_f64(ptr noundef %a, <2 x double> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x double> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <2 x double>
 // CHECK:   [[TMP3:%.*]] = extractelement <2 x double> [[TMP2]], i32 1
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to double*
-// CHECK:   store double [[TMP3]], double* [[TMP4]]
+// CHECK:   store double [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1q_lane_f64(float64_t  *a, float64x2_t b) {
   vst1q_lane_f64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_p8(i8* noundef %a, <16 x i8> noundef %b) #0 {
+// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_p8(ptr noundef %a, <16 x i8> noundef %b) #0 {
 // CHECK:   [[TMP0:%.*]] = extractelement <16 x i8> %b, i32 15
-// CHECK:   store i8 [[TMP0]], i8* %a
+// CHECK:   store i8 [[TMP0]], ptr %a
 // CHECK:   ret void
 void test_vst1q_lane_p8(poly8_t  *a, poly8x16_t b) {
   vst1q_lane_p8(a, b, 15);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_p16(i16* noundef %a, <8 x i16> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i16* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_p16(ptr noundef %a, <8 x i16> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <8 x i16> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <8 x i16>
 // CHECK:   [[TMP3:%.*]] = extractelement <8 x i16> [[TMP2]], i32 7
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to i16*
-// CHECK:   store i16 [[TMP3]], i16* [[TMP4]]
+// CHECK:   store i16 [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1q_lane_p16(poly16_t  *a, poly16x8_t b) {
   vst1q_lane_p16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_p64(i64* noundef %a, <2 x i64> noundef %b) #0 {
-// CHECK:   [[TMP0:%.*]] = bitcast i64* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1q_lane_p64(ptr noundef %a, <2 x i64> noundef %b) #0 {
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x i64> %b to <16 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <2 x i64>
 // CHECK:   [[TMP3:%.*]] = extractelement <2 x i64> [[TMP2]], i32 1
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to i64*
-// CHECK:   store i64 [[TMP3]], i64* [[TMP4]]
+// CHECK:   store i64 [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1q_lane_p64(poly64_t  *a, poly64x2_t b) {
   vst1q_lane_p64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1_lane_u8(i8* noundef %a, <8 x i8> noundef %b) #1 {
+// CHECK-LABEL: define{{.*}} void @test_vst1_lane_u8(ptr noundef %a, <8 x i8> noundef %b) #1 {
 // CHECK:   [[TMP0:%.*]] = extractelement <8 x i8> %b, i32 7
-// CHECK:   store i8 [[TMP0]], i8* %a
+// CHECK:   store i8 [[TMP0]], ptr %a
 // CHECK:   ret void
 void test_vst1_lane_u8(uint8_t  *a, uint8x8_t b) {
   vst1_lane_u8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1_lane_u16(i16* noundef %a, <4 x i16> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i16* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1_lane_u16(ptr noundef %a, <4 x i16> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i16> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <4 x i16>
 // CHECK:   [[TMP3:%.*]] = extractelement <4 x i16> [[TMP2]], i32 3
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to i16*
-// CHECK:   store i16 [[TMP3]], i16* [[TMP4]]
+// CHECK:   store i16 [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1_lane_u16(uint16_t  *a, uint16x4_t b) {
   vst1_lane_u16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1_lane_u32(i32* noundef %a, <2 x i32> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i32* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1_lane_u32(ptr noundef %a, <2 x i32> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x i32> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <2 x i32>
 // CHECK:   [[TMP3:%.*]] = extractelement <2 x i32> [[TMP2]], i32 1
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to i32*
-// CHECK:   store i32 [[TMP3]], i32* [[TMP4]]
+// CHECK:   store i32 [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1_lane_u32(uint32_t  *a, uint32x2_t b) {
   vst1_lane_u32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1_lane_u64(i64* noundef %a, <1 x i64> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i64* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1_lane_u64(ptr noundef %a, <1 x i64> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <1 x i64> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <1 x i64>
 // CHECK:   [[TMP3:%.*]] = extractelement <1 x i64> [[TMP2]], i32 0
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to i64*
-// CHECK:   store i64 [[TMP3]], i64* [[TMP4]]
+// CHECK:   store i64 [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1_lane_u64(uint64_t  *a, uint64x1_t b) {
   vst1_lane_u64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1_lane_s8(i8* noundef %a, <8 x i8> noundef %b) #1 {
+// CHECK-LABEL: define{{.*}} void @test_vst1_lane_s8(ptr noundef %a, <8 x i8> noundef %b) #1 {
 // CHECK:   [[TMP0:%.*]] = extractelement <8 x i8> %b, i32 7
-// CHECK:   store i8 [[TMP0]], i8* %a
+// CHECK:   store i8 [[TMP0]], ptr %a
 // CHECK:   ret void
 void test_vst1_lane_s8(int8_t  *a, int8x8_t b) {
   vst1_lane_s8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1_lane_s16(i16* noundef %a, <4 x i16> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i16* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1_lane_s16(ptr noundef %a, <4 x i16> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i16> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <4 x i16>
 // CHECK:   [[TMP3:%.*]] = extractelement <4 x i16> [[TMP2]], i32 3
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to i16*
-// CHECK:   store i16 [[TMP3]], i16* [[TMP4]]
+// CHECK:   store i16 [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1_lane_s16(int16_t  *a, int16x4_t b) {
   vst1_lane_s16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1_lane_s32(i32* noundef %a, <2 x i32> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i32* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1_lane_s32(ptr noundef %a, <2 x i32> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x i32> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <2 x i32>
 // CHECK:   [[TMP3:%.*]] = extractelement <2 x i32> [[TMP2]], i32 1
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to i32*
-// CHECK:   store i32 [[TMP3]], i32* [[TMP4]]
+// CHECK:   store i32 [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1_lane_s32(int32_t  *a, int32x2_t b) {
   vst1_lane_s32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1_lane_s64(i64* noundef %a, <1 x i64> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i64* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1_lane_s64(ptr noundef %a, <1 x i64> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <1 x i64> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <1 x i64>
 // CHECK:   [[TMP3:%.*]] = extractelement <1 x i64> [[TMP2]], i32 0
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to i64*
-// CHECK:   store i64 [[TMP3]], i64* [[TMP4]]
+// CHECK:   store i64 [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1_lane_s64(int64_t  *a, int64x1_t b) {
   vst1_lane_s64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1_lane_f16(half* noundef %a, <4 x half> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast half* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1_lane_f16(ptr noundef %a, <4 x half> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x half> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <4 x half>
 // CHECK:   [[TMP3:%.*]] = extractelement <4 x half> [[TMP2]], i32 3
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to half*
-// CHECK:   store half [[TMP3]], half* [[TMP4]]
+// CHECK:   store half [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1_lane_f16(float16_t  *a, float16x4_t b) {
   vst1_lane_f16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1_lane_f32(float* noundef %a, <2 x float> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast float* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1_lane_f32(ptr noundef %a, <2 x float> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <2 x float> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <2 x float>
 // CHECK:   [[TMP3:%.*]] = extractelement <2 x float> [[TMP2]], i32 1
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to float*
-// CHECK:   store float [[TMP3]], float* [[TMP4]]
+// CHECK:   store float [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1_lane_f32(float32_t  *a, float32x2_t b) {
   vst1_lane_f32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1_lane_f64(double* noundef %a, <1 x double> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast double* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1_lane_f64(ptr noundef %a, <1 x double> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <1 x double> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <1 x double>
 // CHECK:   [[TMP3:%.*]] = extractelement <1 x double> [[TMP2]], i32 0
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to double*
-// CHECK:   store double [[TMP3]], double* [[TMP4]]
+// CHECK:   store double [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1_lane_f64(float64_t  *a, float64x1_t b) {
   vst1_lane_f64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1_lane_p8(i8* noundef %a, <8 x i8> noundef %b) #1 {
+// CHECK-LABEL: define{{.*}} void @test_vst1_lane_p8(ptr noundef %a, <8 x i8> noundef %b) #1 {
 // CHECK:   [[TMP0:%.*]] = extractelement <8 x i8> %b, i32 7
-// CHECK:   store i8 [[TMP0]], i8* %a
+// CHECK:   store i8 [[TMP0]], ptr %a
 // CHECK:   ret void
 void test_vst1_lane_p8(poly8_t  *a, poly8x8_t b) {
   vst1_lane_p8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1_lane_p16(i16* noundef %a, <4 x i16> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i16* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1_lane_p16(ptr noundef %a, <4 x i16> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <4 x i16> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <4 x i16>
 // CHECK:   [[TMP3:%.*]] = extractelement <4 x i16> [[TMP2]], i32 3
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to i16*
-// CHECK:   store i16 [[TMP3]], i16* [[TMP4]]
+// CHECK:   store i16 [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1_lane_p16(poly16_t  *a, poly16x4_t b) {
   vst1_lane_p16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst1_lane_p64(i64* noundef %a, <1 x i64> noundef %b) #1 {
-// CHECK:   [[TMP0:%.*]] = bitcast i64* %a to i8*
+// CHECK-LABEL: define{{.*}} void @test_vst1_lane_p64(ptr noundef %a, <1 x i64> noundef %b) #1 {
 // CHECK:   [[TMP1:%.*]] = bitcast <1 x i64> %b to <8 x i8>
 // CHECK:   [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to <1 x i64>
 // CHECK:   [[TMP3:%.*]] = extractelement <1 x i64> [[TMP2]], i32 0
-// CHECK:   [[TMP4:%.*]] = bitcast i8* [[TMP0]] to i64*
-// CHECK:   store i64 [[TMP3]], i64* [[TMP4]]
+// CHECK:   store i64 [[TMP3]], ptr %a
 // CHECK:   ret void
 void test_vst1_lane_p64(poly64_t  *a, poly64x1_t b) {
   vst1_lane_p64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_u8(i8* noundef %a, [2 x <16 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_u8(ptr noundef %a, [2 x <16 x i8>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint8x16x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint8x16x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x16x2_t, %struct.uint8x16x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <16 x i8>] [[B]].coerce, [2 x <16 x i8>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint8x16x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint8x16x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x16x2_t, %struct.uint8x16x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <16 x i8>], [2 x <16 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP2:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x16x2_t, %struct.uint8x16x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <16 x i8>], [2 x <16 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v16i8.p0i8(<16 x i8> [[TMP2]], <16 x i8> [[TMP3]], i64 15, i8* %a)
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x16x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <16 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x16x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <16 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP2:%.*]] = load <16 x i8>, ptr [[ARRAYIDX]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x16x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <16 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP3:%.*]] = load <16 x i8>, ptr [[ARRAYIDX2]], align 16
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v16i8.p0(<16 x i8> [[TMP2]], <16 x i8> [[TMP3]], i64 15, ptr %a)
 // CHECK:   ret void
 void test_vst2q_lane_u8(uint8_t  *a, uint8x16x2_t b) {
   vst2q_lane_u8(a, b, 15);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_u16(i16* noundef %a, [2 x <8 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_u16(ptr noundef %a, [2 x <8 x i16>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint16x8x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint16x8x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, %struct.uint16x8x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <8 x i16>] [[B]].coerce, [2 x <8 x i16>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint16x8x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint16x8x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, %struct.uint16x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <8 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x i16>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <8 x i16> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, %struct.uint16x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <8 x i16>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <16 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <16 x i8> [[TMP4]] to <8 x i16>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v8i16.p0i8(<8 x i16> [[TMP7]], <8 x i16> [[TMP8]], i64 7, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v8i16.p0(<8 x i16> [[TMP7]], <8 x i16> [[TMP8]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst2q_lane_u16(uint16_t  *a, uint16x8x2_t b) {
   vst2q_lane_u16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_u32(i32* noundef %a, [2 x <4 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_u32(ptr noundef %a, [2 x <4 x i32>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint32x4x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint32x4x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, %struct.uint32x4x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <4 x i32>] [[B]].coerce, [2 x <4 x i32>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint32x4x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint32x4x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, %struct.uint32x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <4 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x i32>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x i32> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, %struct.uint32x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x i32>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x i32> [[TMP5]] to <16 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <16 x i8> [[TMP4]] to <4 x i32>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP6]] to <4 x i32>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v4i32.p0i8(<4 x i32> [[TMP7]], <4 x i32> [[TMP8]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v4i32.p0(<4 x i32> [[TMP7]], <4 x i32> [[TMP8]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst2q_lane_u32(uint32_t  *a, uint32x4x2_t b) {
   vst2q_lane_u32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_u64(i64* noundef %a, [2 x <2 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_u64(ptr noundef %a, [2 x <2 x i64>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint64x2x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint64x2x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x2x2_t, %struct.uint64x2x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <2 x i64>] [[B]].coerce, [2 x <2 x i64>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint64x2x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint64x2x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x2x2_t, %struct.uint64x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x i64>], [2 x <2 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x2x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <2 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x i64>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x i64> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x2x2_t, %struct.uint64x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x i64>], [2 x <2 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x i64>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x i64> [[TMP5]] to <16 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <16 x i8> [[TMP4]] to <2 x i64>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP6]] to <2 x i64>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v2i64.p0i8(<2 x i64> [[TMP7]], <2 x i64> [[TMP8]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v2i64.p0(<2 x i64> [[TMP7]], <2 x i64> [[TMP8]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst2q_lane_u64(uint64_t  *a, uint64x2x2_t b) {
   vst2q_lane_u64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_s8(i8* noundef %a, [2 x <16 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_s8(ptr noundef %a, [2 x <16 x i8>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int8x16x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int8x16x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x16x2_t, %struct.int8x16x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <16 x i8>] [[B]].coerce, [2 x <16 x i8>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int8x16x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int8x16x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x16x2_t, %struct.int8x16x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <16 x i8>], [2 x <16 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP2:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x16x2_t, %struct.int8x16x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <16 x i8>], [2 x <16 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v16i8.p0i8(<16 x i8> [[TMP2]], <16 x i8> [[TMP3]], i64 15, i8* %a)
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x16x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <16 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x16x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <16 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP2:%.*]] = load <16 x i8>, ptr [[ARRAYIDX]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x16x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <16 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP3:%.*]] = load <16 x i8>, ptr [[ARRAYIDX2]], align 16
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v16i8.p0(<16 x i8> [[TMP2]], <16 x i8> [[TMP3]], i64 15, ptr %a)
 // CHECK:   ret void
 void test_vst2q_lane_s8(int8_t  *a, int8x16x2_t b) {
   vst2q_lane_s8(a, b, 15);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_s16(i16* noundef %a, [2 x <8 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_s16(ptr noundef %a, [2 x <8 x i16>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int16x8x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int16x8x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x2_t, %struct.int16x8x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <8 x i16>] [[B]].coerce, [2 x <8 x i16>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int16x8x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int16x8x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x2_t, %struct.int16x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <8 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x i16>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <8 x i16> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x2_t, %struct.int16x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <8 x i16>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <16 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <16 x i8> [[TMP4]] to <8 x i16>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v8i16.p0i8(<8 x i16> [[TMP7]], <8 x i16> [[TMP8]], i64 7, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v8i16.p0(<8 x i16> [[TMP7]], <8 x i16> [[TMP8]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst2q_lane_s16(int16_t  *a, int16x8x2_t b) {
   vst2q_lane_s16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_s32(i32* noundef %a, [2 x <4 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_s32(ptr noundef %a, [2 x <4 x i32>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int32x4x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int32x4x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x2_t, %struct.int32x4x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <4 x i32>] [[B]].coerce, [2 x <4 x i32>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int32x4x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int32x4x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x2_t, %struct.int32x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <4 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x i32>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x i32> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x2_t, %struct.int32x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x i32>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x i32> [[TMP5]] to <16 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <16 x i8> [[TMP4]] to <4 x i32>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP6]] to <4 x i32>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v4i32.p0i8(<4 x i32> [[TMP7]], <4 x i32> [[TMP8]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v4i32.p0(<4 x i32> [[TMP7]], <4 x i32> [[TMP8]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst2q_lane_s32(int32_t  *a, int32x4x2_t b) {
   vst2q_lane_s32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_s64(i64* noundef %a, [2 x <2 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_s64(ptr noundef %a, [2 x <2 x i64>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int64x2x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int64x2x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x2x2_t, %struct.int64x2x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <2 x i64>] [[B]].coerce, [2 x <2 x i64>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int64x2x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int64x2x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x2x2_t, %struct.int64x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x i64>], [2 x <2 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x2x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <2 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x i64>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x i64> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x2x2_t, %struct.int64x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x i64>], [2 x <2 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x i64>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x i64> [[TMP5]] to <16 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <16 x i8> [[TMP4]] to <2 x i64>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP6]] to <2 x i64>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v2i64.p0i8(<2 x i64> [[TMP7]], <2 x i64> [[TMP8]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v2i64.p0(<2 x i64> [[TMP7]], <2 x i64> [[TMP8]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst2q_lane_s64(int64_t  *a, int64x2x2_t b) {
   vst2q_lane_s64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_f16(half* noundef %a, [2 x <8 x half>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_f16(ptr noundef %a, [2 x <8 x half>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.float16x8x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.float16x8x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x2_t, %struct.float16x8x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <8 x half>] [[B]].coerce, [2 x <8 x half>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float16x8x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float16x8x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast half* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x2_t, %struct.float16x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x half>], [2 x <8 x half>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <8 x half>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x half>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x half>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <8 x half> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x2_t, %struct.float16x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x half>], [2 x <8 x half>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x half>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <8 x half>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <8 x half> [[TMP5]] to <16 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <16 x i8> [[TMP4]] to <8 x half>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x half>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v8f16.p0i8(<8 x half> [[TMP7]], <8 x half> [[TMP8]], i64 7, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v8f16.p0(<8 x half> [[TMP7]], <8 x half> [[TMP8]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst2q_lane_f16(float16_t  *a, float16x8x2_t b) {
   vst2q_lane_f16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_f32(float* noundef %a, [2 x <4 x float>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_f32(ptr noundef %a, [2 x <4 x float>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.float32x4x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.float32x4x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x2_t, %struct.float32x4x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <4 x float>] [[B]].coerce, [2 x <4 x float>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float32x4x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float32x4x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast float* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x2_t, %struct.float32x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x float>], [2 x <4 x float>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <4 x float>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x float>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x float>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x float> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x2_t, %struct.float32x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x float>], [2 x <4 x float>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x float>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x float>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x float> [[TMP5]] to <16 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <16 x i8> [[TMP4]] to <4 x float>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP6]] to <4 x float>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v4f32.p0i8(<4 x float> [[TMP7]], <4 x float> [[TMP8]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v4f32.p0(<4 x float> [[TMP7]], <4 x float> [[TMP8]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst2q_lane_f32(float32_t  *a, float32x4x2_t b) {
   vst2q_lane_f32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_f64(double* noundef %a, [2 x <2 x double>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_f64(ptr noundef %a, [2 x <2 x double>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.float64x2x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.float64x2x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x2x2_t, %struct.float64x2x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <2 x double>] [[B]].coerce, [2 x <2 x double>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float64x2x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float64x2x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast double* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x2x2_t, %struct.float64x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x double>], [2 x <2 x double>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x double>, <2 x double>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x2x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <2 x double>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x double>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x double>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x double> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x2x2_t, %struct.float64x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x double>], [2 x <2 x double>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x double>, <2 x double>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x double>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x double>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x double> [[TMP5]] to <16 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <16 x i8> [[TMP4]] to <2 x double>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP6]] to <2 x double>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v2f64.p0i8(<2 x double> [[TMP7]], <2 x double> [[TMP8]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v2f64.p0(<2 x double> [[TMP7]], <2 x double> [[TMP8]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst2q_lane_f64(float64_t  *a, float64x2x2_t b) {
   vst2q_lane_f64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_p8(i8* noundef %a, [2 x <16 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_p8(ptr noundef %a, [2 x <16 x i8>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.poly8x16x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly8x16x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x16x2_t, %struct.poly8x16x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <16 x i8>] [[B]].coerce, [2 x <16 x i8>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly8x16x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly8x16x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x16x2_t, %struct.poly8x16x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <16 x i8>], [2 x <16 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP2:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x16x2_t, %struct.poly8x16x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <16 x i8>], [2 x <16 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v16i8.p0i8(<16 x i8> [[TMP2]], <16 x i8> [[TMP3]], i64 15, i8* %a)
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x16x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <16 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x16x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <16 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP2:%.*]] = load <16 x i8>, ptr [[ARRAYIDX]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x16x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <16 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP3:%.*]] = load <16 x i8>, ptr [[ARRAYIDX2]], align 16
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v16i8.p0(<16 x i8> [[TMP2]], <16 x i8> [[TMP3]], i64 15, ptr %a)
 // CHECK:   ret void
 void test_vst2q_lane_p8(poly8_t  *a, poly8x16x2_t b) {
   vst2q_lane_p8(a, b, 15);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_p16(i16* noundef %a, [2 x <8 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_p16(ptr noundef %a, [2 x <8 x i16>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.poly16x8x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly16x8x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, %struct.poly16x8x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <8 x i16>] [[B]].coerce, [2 x <8 x i16>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly16x8x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly16x8x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, %struct.poly16x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <8 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x i16>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <8 x i16> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, %struct.poly16x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], [2 x <8 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <8 x i16>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <16 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <16 x i8> [[TMP4]] to <8 x i16>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v8i16.p0i8(<8 x i16> [[TMP7]], <8 x i16> [[TMP8]], i64 7, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v8i16.p0(<8 x i16> [[TMP7]], <8 x i16> [[TMP8]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst2q_lane_p16(poly16_t  *a, poly16x8x2_t b) {
   vst2q_lane_p16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_p64(i64* noundef %a, [2 x <2 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2q_lane_p64(ptr noundef %a, [2 x <2 x i64>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.poly64x2x2_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly64x2x2_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x2x2_t, %struct.poly64x2x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <2 x i64>] [[B]].coerce, [2 x <2 x i64>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly64x2x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly64x2x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x2x2_t, %struct.poly64x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x i64>], [2 x <2 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x2x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <2 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x i64>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x i64> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x2x2_t, %struct.poly64x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x i64>], [2 x <2 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x i64>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x i64> [[TMP5]] to <16 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <16 x i8> [[TMP4]] to <2 x i64>
 // CHECK:   [[TMP8:%.*]] = bitcast <16 x i8> [[TMP6]] to <2 x i64>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v2i64.p0i8(<2 x i64> [[TMP7]], <2 x i64> [[TMP8]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v2i64.p0(<2 x i64> [[TMP7]], <2 x i64> [[TMP8]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst2q_lane_p64(poly64_t  *a, poly64x2x2_t b) {
   vst2q_lane_p64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2_lane_u8(i8* noundef %a, [2 x <8 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2_lane_u8(ptr noundef %a, [2 x <8 x i8>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint8x8x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint8x8x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x8x2_t, %struct.uint8x8x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <8 x i8>] [[B]].coerce, [2 x <8 x i8>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint8x8x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint8x8x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x8x2_t, %struct.uint8x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i8>], [2 x <8 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP2:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX]], align 8
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x8x2_t, %struct.uint8x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i8>], [2 x <8 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP3:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX2]], align 8
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v8i8.p0i8(<8 x i8> [[TMP2]], <8 x i8> [[TMP3]], i64 7, i8* %a)
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x8x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <8 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP2:%.*]] = load <8 x i8>, ptr [[ARRAYIDX]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP3:%.*]] = load <8 x i8>, ptr [[ARRAYIDX2]], align 8
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v8i8.p0(<8 x i8> [[TMP2]], <8 x i8> [[TMP3]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst2_lane_u8(uint8_t  *a, uint8x8x2_t b) {
   vst2_lane_u8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2_lane_u16(i16* noundef %a, [2 x <4 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2_lane_u16(ptr noundef %a, [2 x <4 x i16>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint16x4x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint16x4x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x4x2_t, %struct.uint16x4x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <4 x i16>] [[B]].coerce, [2 x <4 x i16>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint16x4x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint16x4x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x4x2_t, %struct.uint16x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i16>], [2 x <4 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x4x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <4 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x i16>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x i16> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x4x2_t, %struct.uint16x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i16>], [2 x <4 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x i16>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x i16> [[TMP5]] to <8 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x i8> [[TMP4]] to <4 x i16>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP6]] to <4 x i16>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v4i16.p0i8(<4 x i16> [[TMP7]], <4 x i16> [[TMP8]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v4i16.p0(<4 x i16> [[TMP7]], <4 x i16> [[TMP8]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst2_lane_u16(uint16_t  *a, uint16x4x2_t b) {
   vst2_lane_u16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2_lane_u32(i32* noundef %a, [2 x <2 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2_lane_u32(ptr noundef %a, [2 x <2 x i32>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint32x2x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint32x2x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x2x2_t, %struct.uint32x2x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <2 x i32>] [[B]].coerce, [2 x <2 x i32>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint32x2x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint32x2x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x2x2_t, %struct.uint32x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x i32>], [2 x <2 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x2x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <2 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x i32>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x i32> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x2x2_t, %struct.uint32x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x i32>], [2 x <2 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x i32>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x i32> [[TMP5]] to <8 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x i8> [[TMP4]] to <2 x i32>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP6]] to <2 x i32>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v2i32.p0i8(<2 x i32> [[TMP7]], <2 x i32> [[TMP8]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v2i32.p0(<2 x i32> [[TMP7]], <2 x i32> [[TMP8]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst2_lane_u32(uint32_t  *a, uint32x2x2_t b) {
   vst2_lane_u32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2_lane_u64(i64* noundef %a, [2 x <1 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2_lane_u64(ptr noundef %a, [2 x <1 x i64>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint64x1x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint64x1x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x1x2_t, %struct.uint64x1x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <1 x i64>] [[B]].coerce, [2 x <1 x i64>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint64x1x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint64x1x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x1x2_t, %struct.uint64x1x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <1 x i64>], [2 x <1 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x1x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <1 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x1x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <1 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <1 x i64>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <1 x i64> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x1x2_t, %struct.uint64x1x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <1 x i64>], [2 x <1 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x1x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <1 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <1 x i64>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <1 x i64> [[TMP5]] to <8 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x i8> [[TMP4]] to <1 x i64>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP6]] to <1 x i64>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v1i64.p0i8(<1 x i64> [[TMP7]], <1 x i64> [[TMP8]], i64 0, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v1i64.p0(<1 x i64> [[TMP7]], <1 x i64> [[TMP8]], i64 0, ptr %a)
 // CHECK:   ret void
 void test_vst2_lane_u64(uint64_t  *a, uint64x1x2_t b) {
   vst2_lane_u64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2_lane_s8(i8* noundef %a, [2 x <8 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2_lane_s8(ptr noundef %a, [2 x <8 x i8>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int8x8x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int8x8x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x8x2_t, %struct.int8x8x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <8 x i8>] [[B]].coerce, [2 x <8 x i8>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int8x8x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int8x8x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x8x2_t, %struct.int8x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i8>], [2 x <8 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP2:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX]], align 8
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x8x2_t, %struct.int8x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i8>], [2 x <8 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP3:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX2]], align 8
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v8i8.p0i8(<8 x i8> [[TMP2]], <8 x i8> [[TMP3]], i64 7, i8* %a)
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x8x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <8 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP2:%.*]] = load <8 x i8>, ptr [[ARRAYIDX]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP3:%.*]] = load <8 x i8>, ptr [[ARRAYIDX2]], align 8
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v8i8.p0(<8 x i8> [[TMP2]], <8 x i8> [[TMP3]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst2_lane_s8(int8_t  *a, int8x8x2_t b) {
   vst2_lane_s8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2_lane_s16(i16* noundef %a, [2 x <4 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2_lane_s16(ptr noundef %a, [2 x <4 x i16>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int16x4x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int16x4x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x4x2_t, %struct.int16x4x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <4 x i16>] [[B]].coerce, [2 x <4 x i16>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int16x4x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int16x4x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x4x2_t, %struct.int16x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i16>], [2 x <4 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x4x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <4 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x i16>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x i16> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x4x2_t, %struct.int16x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i16>], [2 x <4 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x i16>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x i16> [[TMP5]] to <8 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x i8> [[TMP4]] to <4 x i16>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP6]] to <4 x i16>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v4i16.p0i8(<4 x i16> [[TMP7]], <4 x i16> [[TMP8]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v4i16.p0(<4 x i16> [[TMP7]], <4 x i16> [[TMP8]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst2_lane_s16(int16_t  *a, int16x4x2_t b) {
   vst2_lane_s16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2_lane_s32(i32* noundef %a, [2 x <2 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2_lane_s32(ptr noundef %a, [2 x <2 x i32>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int32x2x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int32x2x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x2x2_t, %struct.int32x2x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <2 x i32>] [[B]].coerce, [2 x <2 x i32>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int32x2x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int32x2x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x2x2_t, %struct.int32x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x i32>], [2 x <2 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x2x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <2 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x i32>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x i32> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x2x2_t, %struct.int32x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x i32>], [2 x <2 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x i32>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x i32> [[TMP5]] to <8 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x i8> [[TMP4]] to <2 x i32>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP6]] to <2 x i32>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v2i32.p0i8(<2 x i32> [[TMP7]], <2 x i32> [[TMP8]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v2i32.p0(<2 x i32> [[TMP7]], <2 x i32> [[TMP8]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst2_lane_s32(int32_t  *a, int32x2x2_t b) {
   vst2_lane_s32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2_lane_s64(i64* noundef %a, [2 x <1 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2_lane_s64(ptr noundef %a, [2 x <1 x i64>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int64x1x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int64x1x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x1x2_t, %struct.int64x1x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <1 x i64>] [[B]].coerce, [2 x <1 x i64>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int64x1x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int64x1x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x1x2_t, %struct.int64x1x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <1 x i64>], [2 x <1 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x1x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <1 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x1x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <1 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <1 x i64>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <1 x i64> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x1x2_t, %struct.int64x1x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <1 x i64>], [2 x <1 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x1x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <1 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <1 x i64>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <1 x i64> [[TMP5]] to <8 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x i8> [[TMP4]] to <1 x i64>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP6]] to <1 x i64>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v1i64.p0i8(<1 x i64> [[TMP7]], <1 x i64> [[TMP8]], i64 0, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v1i64.p0(<1 x i64> [[TMP7]], <1 x i64> [[TMP8]], i64 0, ptr %a)
 // CHECK:   ret void
 void test_vst2_lane_s64(int64_t  *a, int64x1x2_t b) {
   vst2_lane_s64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2_lane_f16(half* noundef %a, [2 x <4 x half>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2_lane_f16(ptr noundef %a, [2 x <4 x half>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.float16x4x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.float16x4x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x4x2_t, %struct.float16x4x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <4 x half>] [[B]].coerce, [2 x <4 x half>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float16x4x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float16x4x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast half* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x4x2_t, %struct.float16x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x half>], [2 x <4 x half>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x half>, <4 x half>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x4x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <4 x half>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x half>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x half>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x half> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x4x2_t, %struct.float16x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x half>], [2 x <4 x half>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x half>, <4 x half>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x half>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x half>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x half> [[TMP5]] to <8 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x i8> [[TMP4]] to <4 x half>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP6]] to <4 x half>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v4f16.p0i8(<4 x half> [[TMP7]], <4 x half> [[TMP8]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v4f16.p0(<4 x half> [[TMP7]], <4 x half> [[TMP8]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst2_lane_f16(float16_t  *a, float16x4x2_t b) {
   vst2_lane_f16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2_lane_f32(float* noundef %a, [2 x <2 x float>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2_lane_f32(ptr noundef %a, [2 x <2 x float>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.float32x2x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.float32x2x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x2x2_t, %struct.float32x2x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <2 x float>] [[B]].coerce, [2 x <2 x float>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float32x2x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float32x2x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast float* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x2x2_t, %struct.float32x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x float>], [2 x <2 x float>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x float>, <2 x float>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x2x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <2 x float>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <2 x float>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x float>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x float> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x2x2_t, %struct.float32x2x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x float>], [2 x <2 x float>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x float>, <2 x float>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x2x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <2 x float>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x float>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x float> [[TMP5]] to <8 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x i8> [[TMP4]] to <2 x float>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP6]] to <2 x float>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v2f32.p0i8(<2 x float> [[TMP7]], <2 x float> [[TMP8]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v2f32.p0(<2 x float> [[TMP7]], <2 x float> [[TMP8]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst2_lane_f32(float32_t  *a, float32x2x2_t b) {
   vst2_lane_f32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2_lane_f64(double* noundef %a, [2 x <1 x double>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2_lane_f64(ptr noundef %a, [2 x <1 x double>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.float64x1x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.float64x1x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x1x2_t, %struct.float64x1x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <1 x double>] [[B]].coerce, [2 x <1 x double>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float64x1x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float64x1x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast double* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x1x2_t, %struct.float64x1x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <1 x double>], [2 x <1 x double>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <1 x double>, <1 x double>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x1x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <1 x double>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x1x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <1 x double>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <1 x double>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <1 x double> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x1x2_t, %struct.float64x1x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <1 x double>], [2 x <1 x double>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <1 x double>, <1 x double>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x1x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <1 x double>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <1 x double>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <1 x double> [[TMP5]] to <8 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x i8> [[TMP4]] to <1 x double>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP6]] to <1 x double>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v1f64.p0i8(<1 x double> [[TMP7]], <1 x double> [[TMP8]], i64 0, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v1f64.p0(<1 x double> [[TMP7]], <1 x double> [[TMP8]], i64 0, ptr %a)
 // CHECK:   ret void
 void test_vst2_lane_f64(float64_t  *a, float64x1x2_t b) {
   vst2_lane_f64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2_lane_p8(i8* noundef %a, [2 x <8 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2_lane_p8(ptr noundef %a, [2 x <8 x i8>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.poly8x8x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly8x8x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x8x2_t, %struct.poly8x8x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <8 x i8>] [[B]].coerce, [2 x <8 x i8>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly8x8x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly8x8x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x8x2_t, %struct.poly8x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i8>], [2 x <8 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP2:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX]], align 8
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x8x2_t, %struct.poly8x8x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i8>], [2 x <8 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP3:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX2]], align 8
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v8i8.p0i8(<8 x i8> [[TMP2]], <8 x i8> [[TMP3]], i64 7, i8* %a)
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x8x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <8 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <8 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP2:%.*]] = load <8 x i8>, ptr [[ARRAYIDX]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x8x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <8 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP3:%.*]] = load <8 x i8>, ptr [[ARRAYIDX2]], align 8
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v8i8.p0(<8 x i8> [[TMP2]], <8 x i8> [[TMP3]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst2_lane_p8(poly8_t  *a, poly8x8x2_t b) {
   vst2_lane_p8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2_lane_p16(i16* noundef %a, [2 x <4 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2_lane_p16(ptr noundef %a, [2 x <4 x i16>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.poly16x4x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly16x4x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x4x2_t, %struct.poly16x4x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <4 x i16>] [[B]].coerce, [2 x <4 x i16>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly16x4x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly16x4x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x4x2_t, %struct.poly16x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i16>], [2 x <4 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x4x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <4 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <4 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x i16>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x i16> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x4x2_t, %struct.poly16x4x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i16>], [2 x <4 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x4x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <4 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x i16>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x i16> [[TMP5]] to <8 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x i8> [[TMP4]] to <4 x i16>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP6]] to <4 x i16>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v4i16.p0i8(<4 x i16> [[TMP7]], <4 x i16> [[TMP8]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v4i16.p0(<4 x i16> [[TMP7]], <4 x i16> [[TMP8]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst2_lane_p16(poly16_t  *a, poly16x4x2_t b) {
   vst2_lane_p16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst2_lane_p64(i64* noundef %a, [2 x <1 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst2_lane_p64(ptr noundef %a, [2 x <1 x i64>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.poly64x1x2_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly64x1x2_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x1x2_t, %struct.poly64x1x2_t* [[B]], i32 0, i32 0
-// CHECK:   store [2 x <1 x i64>] [[B]].coerce, [2 x <1 x i64>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly64x1x2_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly64x1x2_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 16, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x1x2_t, %struct.poly64x1x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <1 x i64>], [2 x <1 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x1x2_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [2 x <1 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 16, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x1x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x <1 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <1 x i64>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <1 x i64> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x1x2_t, %struct.poly64x1x2_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <1 x i64>], [2 x <1 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x1x2_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x <1 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <1 x i64>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <1 x i64> [[TMP5]] to <8 x i8>
 // CHECK:   [[TMP7:%.*]] = bitcast <8 x i8> [[TMP4]] to <1 x i64>
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i8> [[TMP6]] to <1 x i64>
-// CHECK:   call void @llvm.aarch64.neon.st2lane.v1i64.p0i8(<1 x i64> [[TMP7]], <1 x i64> [[TMP8]], i64 0, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st2lane.v1i64.p0(<1 x i64> [[TMP7]], <1 x i64> [[TMP8]], i64 0, ptr %a)
 // CHECK:   ret void
 void test_vst2_lane_p64(poly64_t  *a, poly64x1x2_t b) {
   vst2_lane_p64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_u8(i8* noundef %a, [3 x <16 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_u8(ptr noundef %a, [3 x <16 x i8>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint8x16x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint8x16x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x16x3_t, %struct.uint8x16x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <16 x i8>] [[B]].coerce, [3 x <16 x i8>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint8x16x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint8x16x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x16x3_t, %struct.uint8x16x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP2:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x16x3_t, %struct.uint8x16x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint8x16x3_t, %struct.uint8x16x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 16
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v16i8.p0i8(<16 x i8> [[TMP2]], <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], i64 15, i8* %a)
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x16x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <16 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x16x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <16 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP2:%.*]] = load <16 x i8>, ptr [[ARRAYIDX]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x16x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <16 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP3:%.*]] = load <16 x i8>, ptr [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint8x16x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <16 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP4:%.*]] = load <16 x i8>, ptr [[ARRAYIDX4]], align 16
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v16i8.p0(<16 x i8> [[TMP2]], <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], i64 15, ptr %a)
 // CHECK:   ret void
 void test_vst3q_lane_u8(uint8_t  *a, uint8x16x3_t b) {
   vst3q_lane_u8(a, b, 15);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_u16(i16* noundef %a, [3 x <8 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_u16(ptr noundef %a, [3 x <8 x i16>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint16x8x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint16x8x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <8 x i16>] [[B]].coerce, [3 x <8 x i16>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint16x8x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint16x8x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <8 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x i16>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <8 x i16> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <8 x i16>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, %struct.uint16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <8 x i16>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i16> [[TMP7]] to <16 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP4]] to <8 x i16>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP8]] to <8 x i16>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v8i16.p0i8(<8 x i16> [[TMP9]], <8 x i16> [[TMP10]], <8 x i16> [[TMP11]], i64 7, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v8i16.p0(<8 x i16> [[TMP9]], <8 x i16> [[TMP10]], <8 x i16> [[TMP11]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst3q_lane_u16(uint16_t  *a, uint16x8x3_t b) {
   vst3q_lane_u16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_u32(i32* noundef %a, [3 x <4 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_u32(ptr noundef %a, [3 x <4 x i32>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint32x4x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint32x4x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <4 x i32>] [[B]].coerce, [3 x <4 x i32>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint32x4x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint32x4x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <4 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x i32>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x i32> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x i32>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x i32> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, %struct.uint32x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i32>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <4 x i32>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <4 x i32> [[TMP7]] to <16 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP4]] to <4 x i32>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP6]] to <4 x i32>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP8]] to <4 x i32>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v4i32.p0i8(<4 x i32> [[TMP9]], <4 x i32> [[TMP10]], <4 x i32> [[TMP11]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v4i32.p0(<4 x i32> [[TMP9]], <4 x i32> [[TMP10]], <4 x i32> [[TMP11]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst3q_lane_u32(uint32_t  *a, uint32x4x3_t b) {
   vst3q_lane_u32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_u64(i64* noundef %a, [3 x <2 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_u64(ptr noundef %a, [3 x <2 x i64>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint64x2x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint64x2x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x2x3_t, %struct.uint64x2x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <2 x i64>] [[B]].coerce, [3 x <2 x i64>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint64x2x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint64x2x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x2x3_t, %struct.uint64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x i64>], [3 x <2 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x2x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <2 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x i64>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x i64> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x2x3_t, %struct.uint64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x i64>], [3 x <2 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x i64>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x i64> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint64x2x3_t, %struct.uint64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x i64>], [3 x <2 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <2 x i64>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <2 x i64> [[TMP7]] to <16 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP4]] to <2 x i64>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP6]] to <2 x i64>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP8]] to <2 x i64>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v2i64.p0i8(<2 x i64> [[TMP9]], <2 x i64> [[TMP10]], <2 x i64> [[TMP11]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v2i64.p0(<2 x i64> [[TMP9]], <2 x i64> [[TMP10]], <2 x i64> [[TMP11]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst3q_lane_u64(uint64_t  *a, uint64x2x3_t b) {
   vst3q_lane_u64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_s8(i8* noundef %a, [3 x <16 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_s8(ptr noundef %a, [3 x <16 x i8>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int8x16x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int8x16x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x16x3_t, %struct.int8x16x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <16 x i8>] [[B]].coerce, [3 x <16 x i8>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int8x16x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int8x16x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x16x3_t, %struct.int8x16x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP2:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x16x3_t, %struct.int8x16x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int8x16x3_t, %struct.int8x16x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 16
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v16i8.p0i8(<16 x i8> [[TMP2]], <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], i64 15, i8* %a)
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x16x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <16 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x16x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <16 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP2:%.*]] = load <16 x i8>, ptr [[ARRAYIDX]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x16x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <16 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP3:%.*]] = load <16 x i8>, ptr [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int8x16x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <16 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP4:%.*]] = load <16 x i8>, ptr [[ARRAYIDX4]], align 16
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v16i8.p0(<16 x i8> [[TMP2]], <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], i64 15, ptr %a)
 // CHECK:   ret void
 void test_vst3q_lane_s8(int8_t  *a, int8x16x3_t b) {
   vst3q_lane_s8(a, b, 15);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_s16(i16* noundef %a, [3 x <8 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_s16(ptr noundef %a, [3 x <8 x i16>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int16x8x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int16x8x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x3_t, %struct.int16x8x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <8 x i16>] [[B]].coerce, [3 x <8 x i16>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int16x8x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int16x8x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x3_t, %struct.int16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <8 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x i16>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <8 x i16> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x3_t, %struct.int16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <8 x i16>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int16x8x3_t, %struct.int16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <8 x i16>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i16> [[TMP7]] to <16 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP4]] to <8 x i16>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP8]] to <8 x i16>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v8i16.p0i8(<8 x i16> [[TMP9]], <8 x i16> [[TMP10]], <8 x i16> [[TMP11]], i64 7, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v8i16.p0(<8 x i16> [[TMP9]], <8 x i16> [[TMP10]], <8 x i16> [[TMP11]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst3q_lane_s16(int16_t  *a, int16x8x3_t b) {
   vst3q_lane_s16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_s32(i32* noundef %a, [3 x <4 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_s32(ptr noundef %a, [3 x <4 x i32>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int32x4x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int32x4x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x3_t, %struct.int32x4x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <4 x i32>] [[B]].coerce, [3 x <4 x i32>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int32x4x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int32x4x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x3_t, %struct.int32x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <4 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x i32>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x i32> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x3_t, %struct.int32x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x i32>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x i32> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int32x4x3_t, %struct.int32x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i32>], [3 x <4 x i32>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int32x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i32>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <4 x i32>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <4 x i32> [[TMP7]] to <16 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP4]] to <4 x i32>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP6]] to <4 x i32>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP8]] to <4 x i32>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v4i32.p0i8(<4 x i32> [[TMP9]], <4 x i32> [[TMP10]], <4 x i32> [[TMP11]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v4i32.p0(<4 x i32> [[TMP9]], <4 x i32> [[TMP10]], <4 x i32> [[TMP11]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst3q_lane_s32(int32_t  *a, int32x4x3_t b) {
   vst3q_lane_s32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_s64(i64* noundef %a, [3 x <2 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_s64(ptr noundef %a, [3 x <2 x i64>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int64x2x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int64x2x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x2x3_t, %struct.int64x2x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <2 x i64>] [[B]].coerce, [3 x <2 x i64>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int64x2x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int64x2x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x2x3_t, %struct.int64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x i64>], [3 x <2 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x2x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <2 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x i64>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x i64> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x2x3_t, %struct.int64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x i64>], [3 x <2 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x i64>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x i64> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int64x2x3_t, %struct.int64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x i64>], [3 x <2 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <2 x i64>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <2 x i64> [[TMP7]] to <16 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP4]] to <2 x i64>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP6]] to <2 x i64>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP8]] to <2 x i64>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v2i64.p0i8(<2 x i64> [[TMP9]], <2 x i64> [[TMP10]], <2 x i64> [[TMP11]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v2i64.p0(<2 x i64> [[TMP9]], <2 x i64> [[TMP10]], <2 x i64> [[TMP11]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst3q_lane_s64(int64_t  *a, int64x2x3_t b) {
   vst3q_lane_s64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_f16(half* noundef %a, [3 x <8 x half>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_f16(ptr noundef %a, [3 x <8 x half>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.float16x8x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.float16x8x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x3_t, %struct.float16x8x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <8 x half>] [[B]].coerce, [3 x <8 x half>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float16x8x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float16x8x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast half* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x3_t, %struct.float16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x half>], [3 x <8 x half>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <8 x half>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x half>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x half>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <8 x half> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x3_t, %struct.float16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x half>], [3 x <8 x half>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x half>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <8 x half>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <8 x half> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float16x8x3_t, %struct.float16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x half>], [3 x <8 x half>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x half>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <8 x half>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x half> [[TMP7]] to <16 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP4]] to <8 x half>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x half>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP8]] to <8 x half>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v8f16.p0i8(<8 x half> [[TMP9]], <8 x half> [[TMP10]], <8 x half> [[TMP11]], i64 7, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v8f16.p0(<8 x half> [[TMP9]], <8 x half> [[TMP10]], <8 x half> [[TMP11]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst3q_lane_f16(float16_t  *a, float16x8x3_t b) {
   vst3q_lane_f16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_f32(float* noundef %a, [3 x <4 x float>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_f32(ptr noundef %a, [3 x <4 x float>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.float32x4x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.float32x4x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x3_t, %struct.float32x4x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <4 x float>] [[B]].coerce, [3 x <4 x float>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float32x4x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float32x4x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast float* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x3_t, %struct.float32x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x float>], [3 x <4 x float>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <4 x float>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x float>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x float>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x float> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x3_t, %struct.float32x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x float>], [3 x <4 x float>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x float>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x float>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x float> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float32x4x3_t, %struct.float32x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x float>], [3 x <4 x float>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float32x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x float>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <4 x float>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <4 x float> [[TMP7]] to <16 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP4]] to <4 x float>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP6]] to <4 x float>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP8]] to <4 x float>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v4f32.p0i8(<4 x float> [[TMP9]], <4 x float> [[TMP10]], <4 x float> [[TMP11]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v4f32.p0(<4 x float> [[TMP9]], <4 x float> [[TMP10]], <4 x float> [[TMP11]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst3q_lane_f32(float32_t  *a, float32x4x3_t b) {
   vst3q_lane_f32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_f64(double* noundef %a, [3 x <2 x double>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_f64(ptr noundef %a, [3 x <2 x double>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.float64x2x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.float64x2x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x2x3_t, %struct.float64x2x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <2 x double>] [[B]].coerce, [3 x <2 x double>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float64x2x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float64x2x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast double* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x2x3_t, %struct.float64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x double>], [3 x <2 x double>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x double>, <2 x double>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x2x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <2 x double>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x double>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x double>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x double> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x2x3_t, %struct.float64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x double>], [3 x <2 x double>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x double>, <2 x double>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x double>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x double>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x double> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float64x2x3_t, %struct.float64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x double>], [3 x <2 x double>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <2 x double>, <2 x double>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x double>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <2 x double>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <2 x double> [[TMP7]] to <16 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP4]] to <2 x double>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP6]] to <2 x double>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP8]] to <2 x double>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v2f64.p0i8(<2 x double> [[TMP9]], <2 x double> [[TMP10]], <2 x double> [[TMP11]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v2f64.p0(<2 x double> [[TMP9]], <2 x double> [[TMP10]], <2 x double> [[TMP11]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst3q_lane_f64(float64_t  *a, float64x2x3_t b) {
   vst3q_lane_f64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_p8(i8* noundef %a, [3 x <16 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_p8(ptr noundef %a, [3 x <16 x i8>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.poly8x16x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly8x16x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x16x3_t, %struct.poly8x16x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <16 x i8>] [[B]].coerce, [3 x <16 x i8>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly8x16x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly8x16x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x16x3_t, %struct.poly8x16x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP2:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x16x3_t, %struct.poly8x16x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly8x16x3_t, %struct.poly8x16x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <16 x i8>], [3 x <16 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 16
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v16i8.p0i8(<16 x i8> [[TMP2]], <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], i64 15, i8* %a)
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x16x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <16 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x16x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <16 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP2:%.*]] = load <16 x i8>, ptr [[ARRAYIDX]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x16x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <16 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP3:%.*]] = load <16 x i8>, ptr [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly8x16x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <16 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP4:%.*]] = load <16 x i8>, ptr [[ARRAYIDX4]], align 16
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v16i8.p0(<16 x i8> [[TMP2]], <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], i64 15, ptr %a)
 // CHECK:   ret void
 void test_vst3q_lane_p8(poly8_t  *a, poly8x16x3_t b) {
   vst3q_lane_p8(a, b, 15);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_p16(i16* noundef %a, [3 x <8 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_p16(ptr noundef %a, [3 x <8 x i16>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.poly16x8x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly16x8x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <8 x i16>] [[B]].coerce, [3 x <8 x i16>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly16x8x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly16x8x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <8 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x i16>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <8 x i16> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <8 x i16>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, %struct.poly16x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], [3 x <8 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <8 x i16>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i16> [[TMP7]] to <16 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP4]] to <8 x i16>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP8]] to <8 x i16>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v8i16.p0i8(<8 x i16> [[TMP9]], <8 x i16> [[TMP10]], <8 x i16> [[TMP11]], i64 7, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v8i16.p0(<8 x i16> [[TMP9]], <8 x i16> [[TMP10]], <8 x i16> [[TMP11]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst3q_lane_p16(poly16_t  *a, poly16x8x3_t b) {
   vst3q_lane_p16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_p64(i64* noundef %a, [3 x <2 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3q_lane_p64(ptr noundef %a, [3 x <2 x i64>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.poly64x2x3_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly64x2x3_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x2x3_t, %struct.poly64x2x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <2 x i64>] [[B]].coerce, [3 x <2 x i64>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly64x2x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly64x2x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 48, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x2x3_t, %struct.poly64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x i64>], [3 x <2 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x2x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <2 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 48, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x i64>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x i64> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x2x3_t, %struct.poly64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x i64>], [3 x <2 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x i64>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x i64> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly64x2x3_t, %struct.poly64x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x i64>], [3 x <2 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly64x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <2 x i64>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <2 x i64> [[TMP7]] to <16 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <16 x i8> [[TMP4]] to <2 x i64>
 // CHECK:   [[TMP10:%.*]] = bitcast <16 x i8> [[TMP6]] to <2 x i64>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP8]] to <2 x i64>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v2i64.p0i8(<2 x i64> [[TMP9]], <2 x i64> [[TMP10]], <2 x i64> [[TMP11]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v2i64.p0(<2 x i64> [[TMP9]], <2 x i64> [[TMP10]], <2 x i64> [[TMP11]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst3q_lane_p64(poly64_t  *a, poly64x2x3_t b) {
   vst3q_lane_p64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3_lane_u8(i8* noundef %a, [3 x <8 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3_lane_u8(ptr noundef %a, [3 x <8 x i8>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint8x8x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint8x8x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x8x3_t, %struct.uint8x8x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <8 x i8>] [[B]].coerce, [3 x <8 x i8>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint8x8x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint8x8x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x8x3_t, %struct.uint8x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i8>], [3 x <8 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP2:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX]], align 8
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x8x3_t, %struct.uint8x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i8>], [3 x <8 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP3:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX2]], align 8
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint8x8x3_t, %struct.uint8x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i8>], [3 x <8 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP4:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX4]], align 8
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v8i8.p0i8(<8 x i8> [[TMP2]], <8 x i8> [[TMP3]], <8 x i8> [[TMP4]], i64 7, i8* %a)
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x8x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <8 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP2:%.*]] = load <8 x i8>, ptr [[ARRAYIDX]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP3:%.*]] = load <8 x i8>, ptr [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint8x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP4:%.*]] = load <8 x i8>, ptr [[ARRAYIDX4]], align 8
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v8i8.p0(<8 x i8> [[TMP2]], <8 x i8> [[TMP3]], <8 x i8> [[TMP4]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst3_lane_u8(uint8_t  *a, uint8x8x3_t b) {
   vst3_lane_u8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3_lane_u16(i16* noundef %a, [3 x <4 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3_lane_u16(ptr noundef %a, [3 x <4 x i16>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint16x4x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint16x4x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x4x3_t, %struct.uint16x4x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <4 x i16>] [[B]].coerce, [3 x <4 x i16>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint16x4x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint16x4x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x4x3_t, %struct.uint16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i16>], [3 x <4 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x4x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <4 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x i16>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x i16> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x4x3_t, %struct.uint16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i16>], [3 x <4 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x i16>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x i16> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x4x3_t, %struct.uint16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i16>], [3 x <4 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <4 x i16>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <4 x i16> [[TMP7]] to <8 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP4]] to <4 x i16>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP6]] to <4 x i16>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP8]] to <4 x i16>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v4i16.p0i8(<4 x i16> [[TMP9]], <4 x i16> [[TMP10]], <4 x i16> [[TMP11]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v4i16.p0(<4 x i16> [[TMP9]], <4 x i16> [[TMP10]], <4 x i16> [[TMP11]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst3_lane_u16(uint16_t  *a, uint16x4x3_t b) {
   vst3_lane_u16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3_lane_u32(i32* noundef %a, [3 x <2 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3_lane_u32(ptr noundef %a, [3 x <2 x i32>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint32x2x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint32x2x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x2x3_t, %struct.uint32x2x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <2 x i32>] [[B]].coerce, [3 x <2 x i32>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint32x2x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint32x2x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x2x3_t, %struct.uint32x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x i32>], [3 x <2 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x2x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <2 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x i32>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x i32> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x2x3_t, %struct.uint32x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x i32>], [3 x <2 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x i32>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x i32> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x2x3_t, %struct.uint32x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x i32>], [3 x <2 x i32>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x i32>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <2 x i32>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <2 x i32> [[TMP7]] to <8 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP4]] to <2 x i32>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP6]] to <2 x i32>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP8]] to <2 x i32>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v2i32.p0i8(<2 x i32> [[TMP9]], <2 x i32> [[TMP10]], <2 x i32> [[TMP11]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v2i32.p0(<2 x i32> [[TMP9]], <2 x i32> [[TMP10]], <2 x i32> [[TMP11]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst3_lane_u32(uint32_t  *a, uint32x2x3_t b) {
   vst3_lane_u32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3_lane_u64(i64* noundef %a, [3 x <1 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3_lane_u64(ptr noundef %a, [3 x <1 x i64>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint64x1x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint64x1x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x1x3_t, %struct.uint64x1x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <1 x i64>] [[B]].coerce, [3 x <1 x i64>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint64x1x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint64x1x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x1x3_t, %struct.uint64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <1 x i64>], [3 x <1 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x1x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <1 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <1 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <1 x i64>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <1 x i64> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x1x3_t, %struct.uint64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <1 x i64>], [3 x <1 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <1 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <1 x i64>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <1 x i64> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint64x1x3_t, %struct.uint64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <1 x i64>], [3 x <1 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <1 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <1 x i64>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <1 x i64> [[TMP7]] to <8 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP4]] to <1 x i64>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP6]] to <1 x i64>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP8]] to <1 x i64>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v1i64.p0i8(<1 x i64> [[TMP9]], <1 x i64> [[TMP10]], <1 x i64> [[TMP11]], i64 0, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v1i64.p0(<1 x i64> [[TMP9]], <1 x i64> [[TMP10]], <1 x i64> [[TMP11]], i64 0, ptr %a)
 // CHECK:   ret void
 void test_vst3_lane_u64(uint64_t  *a, uint64x1x3_t b) {
   vst3_lane_u64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3_lane_s8(i8* noundef %a, [3 x <8 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3_lane_s8(ptr noundef %a, [3 x <8 x i8>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int8x8x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int8x8x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x8x3_t, %struct.int8x8x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <8 x i8>] [[B]].coerce, [3 x <8 x i8>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int8x8x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int8x8x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x8x3_t, %struct.int8x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i8>], [3 x <8 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP2:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX]], align 8
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x8x3_t, %struct.int8x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i8>], [3 x <8 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP3:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX2]], align 8
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int8x8x3_t, %struct.int8x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i8>], [3 x <8 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP4:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX4]], align 8
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v8i8.p0i8(<8 x i8> [[TMP2]], <8 x i8> [[TMP3]], <8 x i8> [[TMP4]], i64 7, i8* %a)
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x8x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <8 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP2:%.*]] = load <8 x i8>, ptr [[ARRAYIDX]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP3:%.*]] = load <8 x i8>, ptr [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int8x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP4:%.*]] = load <8 x i8>, ptr [[ARRAYIDX4]], align 8
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v8i8.p0(<8 x i8> [[TMP2]], <8 x i8> [[TMP3]], <8 x i8> [[TMP4]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst3_lane_s8(int8_t  *a, int8x8x3_t b) {
   vst3_lane_s8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3_lane_s16(i16* noundef %a, [3 x <4 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3_lane_s16(ptr noundef %a, [3 x <4 x i16>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int16x4x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int16x4x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x4x3_t, %struct.int16x4x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <4 x i16>] [[B]].coerce, [3 x <4 x i16>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int16x4x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int16x4x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x4x3_t, %struct.int16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i16>], [3 x <4 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x4x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <4 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x i16>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x i16> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x4x3_t, %struct.int16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i16>], [3 x <4 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x i16>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x i16> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int16x4x3_t, %struct.int16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i16>], [3 x <4 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <4 x i16>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <4 x i16> [[TMP7]] to <8 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP4]] to <4 x i16>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP6]] to <4 x i16>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP8]] to <4 x i16>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v4i16.p0i8(<4 x i16> [[TMP9]], <4 x i16> [[TMP10]], <4 x i16> [[TMP11]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v4i16.p0(<4 x i16> [[TMP9]], <4 x i16> [[TMP10]], <4 x i16> [[TMP11]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst3_lane_s16(int16_t  *a, int16x4x3_t b) {
   vst3_lane_s16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3_lane_s32(i32* noundef %a, [3 x <2 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3_lane_s32(ptr noundef %a, [3 x <2 x i32>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int32x2x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int32x2x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x2x3_t, %struct.int32x2x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <2 x i32>] [[B]].coerce, [3 x <2 x i32>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int32x2x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int32x2x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x2x3_t, %struct.int32x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x i32>], [3 x <2 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x2x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <2 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x i32>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x i32> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x2x3_t, %struct.int32x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x i32>], [3 x <2 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x i32>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x i32> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int32x2x3_t, %struct.int32x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x i32>], [3 x <2 x i32>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int32x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x i32>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <2 x i32>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <2 x i32> [[TMP7]] to <8 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP4]] to <2 x i32>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP6]] to <2 x i32>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP8]] to <2 x i32>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v2i32.p0i8(<2 x i32> [[TMP9]], <2 x i32> [[TMP10]], <2 x i32> [[TMP11]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v2i32.p0(<2 x i32> [[TMP9]], <2 x i32> [[TMP10]], <2 x i32> [[TMP11]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst3_lane_s32(int32_t  *a, int32x2x3_t b) {
   vst3_lane_s32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3_lane_s64(i64* noundef %a, [3 x <1 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3_lane_s64(ptr noundef %a, [3 x <1 x i64>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int64x1x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int64x1x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x1x3_t, %struct.int64x1x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <1 x i64>] [[B]].coerce, [3 x <1 x i64>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int64x1x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int64x1x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x1x3_t, %struct.int64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <1 x i64>], [3 x <1 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x1x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <1 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <1 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <1 x i64>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <1 x i64> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x1x3_t, %struct.int64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <1 x i64>], [3 x <1 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <1 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <1 x i64>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <1 x i64> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int64x1x3_t, %struct.int64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <1 x i64>], [3 x <1 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <1 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <1 x i64>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <1 x i64> [[TMP7]] to <8 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP4]] to <1 x i64>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP6]] to <1 x i64>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP8]] to <1 x i64>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v1i64.p0i8(<1 x i64> [[TMP9]], <1 x i64> [[TMP10]], <1 x i64> [[TMP11]], i64 0, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v1i64.p0(<1 x i64> [[TMP9]], <1 x i64> [[TMP10]], <1 x i64> [[TMP11]], i64 0, ptr %a)
 // CHECK:   ret void
 void test_vst3_lane_s64(int64_t  *a, int64x1x3_t b) {
   vst3_lane_s64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3_lane_f16(half* noundef %a, [3 x <4 x half>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3_lane_f16(ptr noundef %a, [3 x <4 x half>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.float16x4x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.float16x4x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x4x3_t, %struct.float16x4x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <4 x half>] [[B]].coerce, [3 x <4 x half>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float16x4x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float16x4x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast half* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x4x3_t, %struct.float16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x half>], [3 x <4 x half>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x half>, <4 x half>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x4x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <4 x half>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x half>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x half>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x half> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x4x3_t, %struct.float16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x half>], [3 x <4 x half>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x half>, <4 x half>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x half>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x half>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x half> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float16x4x3_t, %struct.float16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x half>], [3 x <4 x half>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <4 x half>, <4 x half>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x half>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <4 x half>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <4 x half> [[TMP7]] to <8 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP4]] to <4 x half>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP6]] to <4 x half>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP8]] to <4 x half>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v4f16.p0i8(<4 x half> [[TMP9]], <4 x half> [[TMP10]], <4 x half> [[TMP11]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v4f16.p0(<4 x half> [[TMP9]], <4 x half> [[TMP10]], <4 x half> [[TMP11]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst3_lane_f16(float16_t  *a, float16x4x3_t b) {
   vst3_lane_f16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3_lane_f32(float* noundef %a, [3 x <2 x float>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3_lane_f32(ptr noundef %a, [3 x <2 x float>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.float32x2x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.float32x2x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x2x3_t, %struct.float32x2x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <2 x float>] [[B]].coerce, [3 x <2 x float>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float32x2x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float32x2x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast float* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x2x3_t, %struct.float32x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x float>], [3 x <2 x float>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x float>, <2 x float>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x2x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <2 x float>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <2 x float>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x float>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x float> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x2x3_t, %struct.float32x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x float>], [3 x <2 x float>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x float>, <2 x float>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <2 x float>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x float>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x float> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float32x2x3_t, %struct.float32x2x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x float>], [3 x <2 x float>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <2 x float>, <2 x float>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float32x2x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <2 x float>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <2 x float>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <2 x float> [[TMP7]] to <8 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP4]] to <2 x float>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP6]] to <2 x float>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP8]] to <2 x float>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v2f32.p0i8(<2 x float> [[TMP9]], <2 x float> [[TMP10]], <2 x float> [[TMP11]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v2f32.p0(<2 x float> [[TMP9]], <2 x float> [[TMP10]], <2 x float> [[TMP11]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst3_lane_f32(float32_t  *a, float32x2x3_t b) {
   vst3_lane_f32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3_lane_f64(double* noundef %a, [3 x <1 x double>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3_lane_f64(ptr noundef %a, [3 x <1 x double>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.float64x1x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.float64x1x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x1x3_t, %struct.float64x1x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <1 x double>] [[B]].coerce, [3 x <1 x double>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float64x1x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float64x1x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast double* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x1x3_t, %struct.float64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <1 x double>], [3 x <1 x double>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <1 x double>, <1 x double>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x1x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <1 x double>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <1 x double>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <1 x double>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <1 x double> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x1x3_t, %struct.float64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <1 x double>], [3 x <1 x double>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <1 x double>, <1 x double>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <1 x double>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <1 x double>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <1 x double> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float64x1x3_t, %struct.float64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <1 x double>], [3 x <1 x double>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <1 x double>, <1 x double>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <1 x double>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <1 x double>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <1 x double> [[TMP7]] to <8 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP4]] to <1 x double>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP6]] to <1 x double>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP8]] to <1 x double>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v1f64.p0i8(<1 x double> [[TMP9]], <1 x double> [[TMP10]], <1 x double> [[TMP11]], i64 0, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v1f64.p0(<1 x double> [[TMP9]], <1 x double> [[TMP10]], <1 x double> [[TMP11]], i64 0, ptr %a)
 // CHECK:   ret void
 void test_vst3_lane_f64(float64_t  *a, float64x1x3_t b) {
   vst3_lane_f64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3_lane_p8(i8* noundef %a, [3 x <8 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3_lane_p8(ptr noundef %a, [3 x <8 x i8>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.poly8x8x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly8x8x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x8x3_t, %struct.poly8x8x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <8 x i8>] [[B]].coerce, [3 x <8 x i8>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly8x8x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly8x8x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x8x3_t, %struct.poly8x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i8>], [3 x <8 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP2:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX]], align 8
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x8x3_t, %struct.poly8x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i8>], [3 x <8 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP3:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX2]], align 8
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly8x8x3_t, %struct.poly8x8x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i8>], [3 x <8 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP4:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX4]], align 8
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v8i8.p0i8(<8 x i8> [[TMP2]], <8 x i8> [[TMP3]], <8 x i8> [[TMP4]], i64 7, i8* %a)
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x8x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <8 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <8 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP2:%.*]] = load <8 x i8>, ptr [[ARRAYIDX]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <8 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP3:%.*]] = load <8 x i8>, ptr [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly8x8x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <8 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP4:%.*]] = load <8 x i8>, ptr [[ARRAYIDX4]], align 8
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v8i8.p0(<8 x i8> [[TMP2]], <8 x i8> [[TMP3]], <8 x i8> [[TMP4]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst3_lane_p8(poly8_t  *a, poly8x8x3_t b) {
   vst3_lane_p8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3_lane_p16(i16* noundef %a, [3 x <4 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3_lane_p16(ptr noundef %a, [3 x <4 x i16>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.poly16x4x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly16x4x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x4x3_t, %struct.poly16x4x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <4 x i16>] [[B]].coerce, [3 x <4 x i16>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly16x4x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly16x4x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x4x3_t, %struct.poly16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i16>], [3 x <4 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x4x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <4 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <4 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x i16>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x i16> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x4x3_t, %struct.poly16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i16>], [3 x <4 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <4 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x i16>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x i16> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x4x3_t, %struct.poly16x4x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i16>], [3 x <4 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x4x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <4 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <4 x i16>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <4 x i16> [[TMP7]] to <8 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP4]] to <4 x i16>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP6]] to <4 x i16>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP8]] to <4 x i16>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v4i16.p0i8(<4 x i16> [[TMP9]], <4 x i16> [[TMP10]], <4 x i16> [[TMP11]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v4i16.p0(<4 x i16> [[TMP9]], <4 x i16> [[TMP10]], <4 x i16> [[TMP11]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst3_lane_p16(poly16_t  *a, poly16x4x3_t b) {
   vst3_lane_p16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst3_lane_p64(i64* noundef %a, [3 x <1 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst3_lane_p64(ptr noundef %a, [3 x <1 x i64>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.poly64x1x3_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly64x1x3_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x1x3_t, %struct.poly64x1x3_t* [[B]], i32 0, i32 0
-// CHECK:   store [3 x <1 x i64>] [[B]].coerce, [3 x <1 x i64>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly64x1x3_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly64x1x3_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 24, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x1x3_t, %struct.poly64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <1 x i64>], [3 x <1 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x1x3_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [3 x <1 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 24, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [3 x <1 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <1 x i64>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <1 x i64> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x1x3_t, %struct.poly64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <1 x i64>], [3 x <1 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [3 x <1 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <1 x i64>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <1 x i64> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly64x1x3_t, %struct.poly64x1x3_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <1 x i64>], [3 x <1 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly64x1x3_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [3 x <1 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <1 x i64>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <1 x i64> [[TMP7]] to <8 x i8>
 // CHECK:   [[TMP9:%.*]] = bitcast <8 x i8> [[TMP4]] to <1 x i64>
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i8> [[TMP6]] to <1 x i64>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP8]] to <1 x i64>
-// CHECK:   call void @llvm.aarch64.neon.st3lane.v1i64.p0i8(<1 x i64> [[TMP9]], <1 x i64> [[TMP10]], <1 x i64> [[TMP11]], i64 0, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st3lane.v1i64.p0(<1 x i64> [[TMP9]], <1 x i64> [[TMP10]], <1 x i64> [[TMP11]], i64 0, ptr %a)
 // CHECK:   ret void
 void test_vst3_lane_p64(poly64_t  *a, poly64x1x3_t b) {
   vst3_lane_p64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_u8(i8* noundef %a, [4 x <16 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_u8(ptr noundef %a, [4 x <16 x i8>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint8x16x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint8x16x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, %struct.uint8x16x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <16 x i8>] [[B]].coerce, [4 x <16 x i8>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint8x16x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint8x16x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, %struct.uint8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP2:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, %struct.uint8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, %struct.uint8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 16
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, %struct.uint8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX6]], align 16
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v16i8.p0i8(<16 x i8> [[TMP2]], <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], i64 15, i8* %a)
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <16 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP2:%.*]] = load <16 x i8>, ptr [[ARRAYIDX]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP3:%.*]] = load <16 x i8>, ptr [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP4:%.*]] = load <16 x i8>, ptr [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP5:%.*]] = load <16 x i8>, ptr [[ARRAYIDX6]], align 16
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v16i8.p0(<16 x i8> [[TMP2]], <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], i64 15, ptr %a)
 // CHECK:   ret void
 void test_vst4q_lane_u8(uint8_t  *a, uint8x16x4_t b) {
   vst4q_lane_u8(a, b, 15);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_u16(i16* noundef %a, [4 x <8 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_u16(ptr noundef %a, [4 x <8 x i16>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint16x8x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint16x8x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <8 x i16>] [[B]].coerce, [4 x <8 x i16>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint16x8x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint16x8x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <8 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x i16>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <8 x i16> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <8 x i16>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <8 x i16>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i16> [[TMP7]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, %struct.uint16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <8 x i16>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i16> [[TMP9]] to <16 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP4]] to <8 x i16>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP8]] to <8 x i16>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP10]] to <8 x i16>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v8i16.p0i8(<8 x i16> [[TMP11]], <8 x i16> [[TMP12]], <8 x i16> [[TMP13]], <8 x i16> [[TMP14]], i64 7, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v8i16.p0(<8 x i16> [[TMP11]], <8 x i16> [[TMP12]], <8 x i16> [[TMP13]], <8 x i16> [[TMP14]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst4q_lane_u16(uint16_t  *a, uint16x8x4_t b) {
   vst4q_lane_u16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_u32(i32* noundef %a, [4 x <4 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_u32(ptr noundef %a, [4 x <4 x i32>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint32x4x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint32x4x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <4 x i32>] [[B]].coerce, [4 x <4 x i32>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint32x4x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint32x4x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <4 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x i32>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x i32> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x i32>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x i32> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i32>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <4 x i32>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <4 x i32> [[TMP7]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, %struct.uint32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i32>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <4 x i32>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP10:%.*]] = bitcast <4 x i32> [[TMP9]] to <16 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP4]] to <4 x i32>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP6]] to <4 x i32>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP8]] to <4 x i32>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP10]] to <4 x i32>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v4i32.p0i8(<4 x i32> [[TMP11]], <4 x i32> [[TMP12]], <4 x i32> [[TMP13]], <4 x i32> [[TMP14]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v4i32.p0(<4 x i32> [[TMP11]], <4 x i32> [[TMP12]], <4 x i32> [[TMP13]], <4 x i32> [[TMP14]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst4q_lane_u32(uint32_t  *a, uint32x4x4_t b) {
   vst4q_lane_u32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_u64(i64* noundef %a, [4 x <2 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_u64(ptr noundef %a, [4 x <2 x i64>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint64x2x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint64x2x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x2x4_t, %struct.uint64x2x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <2 x i64>] [[B]].coerce, [4 x <2 x i64>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint64x2x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint64x2x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x2x4_t, %struct.uint64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x2x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <2 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x i64>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x i64> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x2x4_t, %struct.uint64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x i64>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x i64> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint64x2x4_t, %struct.uint64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <2 x i64>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <2 x i64> [[TMP7]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint64x2x4_t, %struct.uint64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <2 x i64>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP10:%.*]] = bitcast <2 x i64> [[TMP9]] to <16 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP4]] to <2 x i64>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP6]] to <2 x i64>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP8]] to <2 x i64>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP10]] to <2 x i64>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v2i64.p0i8(<2 x i64> [[TMP11]], <2 x i64> [[TMP12]], <2 x i64> [[TMP13]], <2 x i64> [[TMP14]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v2i64.p0(<2 x i64> [[TMP11]], <2 x i64> [[TMP12]], <2 x i64> [[TMP13]], <2 x i64> [[TMP14]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst4q_lane_u64(uint64_t  *a, uint64x2x4_t b) {
   vst4q_lane_u64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_s8(i8* noundef %a, [4 x <16 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_s8(ptr noundef %a, [4 x <16 x i8>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int8x16x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int8x16x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x16x4_t, %struct.int8x16x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <16 x i8>] [[B]].coerce, [4 x <16 x i8>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int8x16x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int8x16x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x16x4_t, %struct.int8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP2:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x16x4_t, %struct.int8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int8x16x4_t, %struct.int8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 16
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int8x16x4_t, %struct.int8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX6]], align 16
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v16i8.p0i8(<16 x i8> [[TMP2]], <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], i64 15, i8* %a)
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x16x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <16 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP2:%.*]] = load <16 x i8>, ptr [[ARRAYIDX]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP3:%.*]] = load <16 x i8>, ptr [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP4:%.*]] = load <16 x i8>, ptr [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP5:%.*]] = load <16 x i8>, ptr [[ARRAYIDX6]], align 16
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v16i8.p0(<16 x i8> [[TMP2]], <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], i64 15, ptr %a)
 // CHECK:   ret void
 void test_vst4q_lane_s8(int8_t  *a, int8x16x4_t b) {
   vst4q_lane_s8(a, b, 15);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_s16(i16* noundef %a, [4 x <8 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_s16(ptr noundef %a, [4 x <8 x i16>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int16x8x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int16x8x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <8 x i16>] [[B]].coerce, [4 x <8 x i16>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int16x8x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int16x8x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x8x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <8 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x i16>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <8 x i16> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <8 x i16>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <8 x i16>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i16> [[TMP7]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int16x8x4_t, %struct.int16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <8 x i16>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i16> [[TMP9]] to <16 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP4]] to <8 x i16>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP8]] to <8 x i16>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP10]] to <8 x i16>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v8i16.p0i8(<8 x i16> [[TMP11]], <8 x i16> [[TMP12]], <8 x i16> [[TMP13]], <8 x i16> [[TMP14]], i64 7, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v8i16.p0(<8 x i16> [[TMP11]], <8 x i16> [[TMP12]], <8 x i16> [[TMP13]], <8 x i16> [[TMP14]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst4q_lane_s16(int16_t  *a, int16x8x4_t b) {
   vst4q_lane_s16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_s32(i32* noundef %a, [4 x <4 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_s32(ptr noundef %a, [4 x <4 x i32>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int32x4x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int32x4x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <4 x i32>] [[B]].coerce, [4 x <4 x i32>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int32x4x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int32x4x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x4x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <4 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x i32>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x i32> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x i32>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x i32> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i32>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <4 x i32>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <4 x i32> [[TMP7]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int32x4x4_t, %struct.int32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i32>], [4 x <4 x i32>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <4 x i32>, <4 x i32>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i32>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <4 x i32>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP10:%.*]] = bitcast <4 x i32> [[TMP9]] to <16 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP4]] to <4 x i32>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP6]] to <4 x i32>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP8]] to <4 x i32>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP10]] to <4 x i32>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v4i32.p0i8(<4 x i32> [[TMP11]], <4 x i32> [[TMP12]], <4 x i32> [[TMP13]], <4 x i32> [[TMP14]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v4i32.p0(<4 x i32> [[TMP11]], <4 x i32> [[TMP12]], <4 x i32> [[TMP13]], <4 x i32> [[TMP14]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst4q_lane_s32(int32_t  *a, int32x4x4_t b) {
   vst4q_lane_s32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_s64(i64* noundef %a, [4 x <2 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_s64(ptr noundef %a, [4 x <2 x i64>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int64x2x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.int64x2x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x2x4_t, %struct.int64x2x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <2 x i64>] [[B]].coerce, [4 x <2 x i64>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int64x2x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int64x2x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x2x4_t, %struct.int64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x2x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <2 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x i64>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x i64> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x2x4_t, %struct.int64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x i64>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x i64> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int64x2x4_t, %struct.int64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <2 x i64>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <2 x i64> [[TMP7]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int64x2x4_t, %struct.int64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <2 x i64>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP10:%.*]] = bitcast <2 x i64> [[TMP9]] to <16 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP4]] to <2 x i64>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP6]] to <2 x i64>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP8]] to <2 x i64>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP10]] to <2 x i64>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v2i64.p0i8(<2 x i64> [[TMP11]], <2 x i64> [[TMP12]], <2 x i64> [[TMP13]], <2 x i64> [[TMP14]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v2i64.p0(<2 x i64> [[TMP11]], <2 x i64> [[TMP12]], <2 x i64> [[TMP13]], <2 x i64> [[TMP14]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst4q_lane_s64(int64_t  *a, int64x2x4_t b) {
   vst4q_lane_s64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_f16(half* noundef %a, [4 x <8 x half>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_f16(ptr noundef %a, [4 x <8 x half>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.float16x8x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.float16x8x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <8 x half>] [[B]].coerce, [4 x <8 x half>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float16x8x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float16x8x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast half* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x half>], [4 x <8 x half>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x8x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <8 x half>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x half>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x half>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <8 x half> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x half>], [4 x <8 x half>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x half>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <8 x half>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <8 x half> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x half>], [4 x <8 x half>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x half>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <8 x half>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x half> [[TMP7]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float16x8x4_t, %struct.float16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x half>], [4 x <8 x half>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <8 x half>, <8 x half>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x half>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <8 x half>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x half> [[TMP9]] to <16 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP4]] to <8 x half>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x half>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP8]] to <8 x half>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP10]] to <8 x half>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v8f16.p0i8(<8 x half> [[TMP11]], <8 x half> [[TMP12]], <8 x half> [[TMP13]], <8 x half> [[TMP14]], i64 7, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v8f16.p0(<8 x half> [[TMP11]], <8 x half> [[TMP12]], <8 x half> [[TMP13]], <8 x half> [[TMP14]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst4q_lane_f16(float16_t  *a, float16x8x4_t b) {
   vst4q_lane_f16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_f32(float* noundef %a, [4 x <4 x float>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_f32(ptr noundef %a, [4 x <4 x float>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.float32x4x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.float32x4x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <4 x float>] [[B]].coerce, [4 x <4 x float>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float32x4x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float32x4x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast float* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x float>], [4 x <4 x float>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x4x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <4 x float>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x float>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x float>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x float> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x float>], [4 x <4 x float>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x float>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x float>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x float> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x float>], [4 x <4 x float>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x float>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <4 x float>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <4 x float> [[TMP7]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float32x4x4_t, %struct.float32x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x float>], [4 x <4 x float>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <4 x float>, <4 x float>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float32x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x float>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <4 x float>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP10:%.*]] = bitcast <4 x float> [[TMP9]] to <16 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP4]] to <4 x float>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP6]] to <4 x float>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP8]] to <4 x float>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP10]] to <4 x float>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v4f32.p0i8(<4 x float> [[TMP11]], <4 x float> [[TMP12]], <4 x float> [[TMP13]], <4 x float> [[TMP14]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v4f32.p0(<4 x float> [[TMP11]], <4 x float> [[TMP12]], <4 x float> [[TMP13]], <4 x float> [[TMP14]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst4q_lane_f32(float32_t  *a, float32x4x4_t b) {
   vst4q_lane_f32(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_f64(double* noundef %a, [4 x <2 x double>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_f64(ptr noundef %a, [4 x <2 x double>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.float64x2x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.float64x2x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x2x4_t, %struct.float64x2x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <2 x double>] [[B]].coerce, [4 x <2 x double>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float64x2x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float64x2x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast double* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x2x4_t, %struct.float64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x double>], [4 x <2 x double>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x double>, <2 x double>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x2x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <2 x double>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x double>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x double>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x double> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x2x4_t, %struct.float64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x double>], [4 x <2 x double>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x double>, <2 x double>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x double>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x double>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x double> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float64x2x4_t, %struct.float64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x double>], [4 x <2 x double>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <2 x double>, <2 x double>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x double>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <2 x double>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <2 x double> [[TMP7]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float64x2x4_t, %struct.float64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x double>], [4 x <2 x double>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <2 x double>, <2 x double>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x double>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <2 x double>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP10:%.*]] = bitcast <2 x double> [[TMP9]] to <16 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP4]] to <2 x double>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP6]] to <2 x double>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP8]] to <2 x double>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP10]] to <2 x double>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v2f64.p0i8(<2 x double> [[TMP11]], <2 x double> [[TMP12]], <2 x double> [[TMP13]], <2 x double> [[TMP14]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v2f64.p0(<2 x double> [[TMP11]], <2 x double> [[TMP12]], <2 x double> [[TMP13]], <2 x double> [[TMP14]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst4q_lane_f64(float64_t  *a, float64x2x4_t b) {
   vst4q_lane_f64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_p8(i8* noundef %a, [4 x <16 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_p8(ptr noundef %a, [4 x <16 x i8>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.poly8x16x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly8x16x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, %struct.poly8x16x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <16 x i8>] [[B]].coerce, [4 x <16 x i8>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly8x16x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly8x16x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, %struct.poly8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP2:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX]], align 16
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, %struct.poly8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX2]], align 16
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, %struct.poly8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP4:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX4]], align 16
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, %struct.poly8x16x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <16 x i8>], [4 x <16 x i8>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP5:%.*]] = load <16 x i8>, <16 x i8>* [[ARRAYIDX6]], align 16
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v16i8.p0i8(<16 x i8> [[TMP2]], <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], i64 15, i8* %a)
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <16 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP2:%.*]] = load <16 x i8>, ptr [[ARRAYIDX]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP3:%.*]] = load <16 x i8>, ptr [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP4:%.*]] = load <16 x i8>, ptr [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly8x16x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <16 x i8>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP5:%.*]] = load <16 x i8>, ptr [[ARRAYIDX6]], align 16
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v16i8.p0(<16 x i8> [[TMP2]], <16 x i8> [[TMP3]], <16 x i8> [[TMP4]], <16 x i8> [[TMP5]], i64 15, ptr %a)
 // CHECK:   ret void
 void test_vst4q_lane_p8(poly8_t  *a, poly8x16x4_t b) {
   vst4q_lane_p8(a, b, 15);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_p16(i16* noundef %a, [4 x <8 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_p16(ptr noundef %a, [4 x <8 x i16>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.poly16x8x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly16x8x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <8 x i16>] [[B]].coerce, [4 x <8 x i16>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly16x8x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly16x8x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <8 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <8 x i16>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <8 x i16> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <8 x i16>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <8 x i16> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <8 x i16>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <8 x i16> [[TMP7]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, %struct.poly16x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], [4 x <8 x i16>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <8 x i16>, <8 x i16>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly16x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i16>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <8 x i16>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP10:%.*]] = bitcast <8 x i16> [[TMP9]] to <16 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP4]] to <8 x i16>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP6]] to <8 x i16>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP8]] to <8 x i16>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP10]] to <8 x i16>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v8i16.p0i8(<8 x i16> [[TMP11]], <8 x i16> [[TMP12]], <8 x i16> [[TMP13]], <8 x i16> [[TMP14]], i64 7, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v8i16.p0(<8 x i16> [[TMP11]], <8 x i16> [[TMP12]], <8 x i16> [[TMP13]], <8 x i16> [[TMP14]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst4q_lane_p16(poly16_t  *a, poly16x8x4_t b) {
   vst4q_lane_p16(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_p64(i64* noundef %a, [4 x <2 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4q_lane_p64(ptr noundef %a, [4 x <2 x i64>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.poly64x2x4_t, align 16
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly64x2x4_t, align 16
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x2x4_t, %struct.poly64x2x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <2 x i64>] [[B]].coerce, [4 x <2 x i64>]* [[COERCE_DIVE]], align 16
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly64x2x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly64x2x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 16 [[TMP0]], i8* align 16 [[TMP1]], i64 64, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x2x4_t, %struct.poly64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX]], align 16
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x2x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <2 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 16
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 16 [[__S1]], ptr align 16 [[B]], i64 64, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x i64>, ptr [[ARRAYIDX]], align 16
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x i64> [[TMP3]] to <16 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x2x4_t, %struct.poly64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX2]], align 16
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x i64>, ptr [[ARRAYIDX2]], align 16
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x i64> [[TMP5]] to <16 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly64x2x4_t, %struct.poly64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX4]], align 16
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <2 x i64>, ptr [[ARRAYIDX4]], align 16
 // CHECK:   [[TMP8:%.*]] = bitcast <2 x i64> [[TMP7]] to <16 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly64x2x4_t, %struct.poly64x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x i64>], [4 x <2 x i64>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <2 x i64>, <2 x i64>* [[ARRAYIDX6]], align 16
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly64x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x i64>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <2 x i64>, ptr [[ARRAYIDX6]], align 16
 // CHECK:   [[TMP10:%.*]] = bitcast <2 x i64> [[TMP9]] to <16 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <16 x i8> [[TMP4]] to <2 x i64>
 // CHECK:   [[TMP12:%.*]] = bitcast <16 x i8> [[TMP6]] to <2 x i64>
 // CHECK:   [[TMP13:%.*]] = bitcast <16 x i8> [[TMP8]] to <2 x i64>
 // CHECK:   [[TMP14:%.*]] = bitcast <16 x i8> [[TMP10]] to <2 x i64>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v2i64.p0i8(<2 x i64> [[TMP11]], <2 x i64> [[TMP12]], <2 x i64> [[TMP13]], <2 x i64> [[TMP14]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v2i64.p0(<2 x i64> [[TMP11]], <2 x i64> [[TMP12]], <2 x i64> [[TMP13]], <2 x i64> [[TMP14]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst4q_lane_p64(poly64_t  *a, poly64x2x4_t b) {
   vst4q_lane_p64(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4_lane_u8(i8* noundef %a, [4 x <8 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4_lane_u8(ptr noundef %a, [4 x <8 x i8>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint8x8x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint8x8x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x8x4_t, %struct.uint8x8x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <8 x i8>] [[B]].coerce, [4 x <8 x i8>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint8x8x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint8x8x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x8x4_t, %struct.uint8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP2:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX]], align 8
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x8x4_t, %struct.uint8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP3:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX2]], align 8
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint8x8x4_t, %struct.uint8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP4:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX4]], align 8
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint8x8x4_t, %struct.uint8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP5:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX6]], align 8
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v8i8.p0i8(<8 x i8> [[TMP2]], <8 x i8> [[TMP3]], <8 x i8> [[TMP4]], <8 x i8> [[TMP5]], i64 7, i8* %a)
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint8x8x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <8 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP2:%.*]] = load <8 x i8>, ptr [[ARRAYIDX]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP3:%.*]] = load <8 x i8>, ptr [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP4:%.*]] = load <8 x i8>, ptr [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP5:%.*]] = load <8 x i8>, ptr [[ARRAYIDX6]], align 8
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v8i8.p0(<8 x i8> [[TMP2]], <8 x i8> [[TMP3]], <8 x i8> [[TMP4]], <8 x i8> [[TMP5]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst4_lane_u8(uint8_t  *a, uint8x8x4_t b) {
   vst4_lane_u8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4_lane_u16(i16* noundef %a, [4 x <4 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4_lane_u16(ptr noundef %a, [4 x <4 x i16>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint16x4x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint16x4x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x4x4_t, %struct.uint16x4x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <4 x i16>] [[B]].coerce, [4 x <4 x i16>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint16x4x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint16x4x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x4x4_t, %struct.uint16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint16x4x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <4 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x i16>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x i16> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x4x4_t, %struct.uint16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x i16>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x i16> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x4x4_t, %struct.uint16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <4 x i16>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <4 x i16> [[TMP7]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint16x4x4_t, %struct.uint16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <4 x i16>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP10:%.*]] = bitcast <4 x i16> [[TMP9]] to <8 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP4]] to <4 x i16>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP6]] to <4 x i16>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP8]] to <4 x i16>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP10]] to <4 x i16>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v4i16.p0i8(<4 x i16> [[TMP11]], <4 x i16> [[TMP12]], <4 x i16> [[TMP13]], <4 x i16> [[TMP14]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v4i16.p0(<4 x i16> [[TMP11]], <4 x i16> [[TMP12]], <4 x i16> [[TMP13]], <4 x i16> [[TMP14]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst4_lane_u16(uint16_t  *a, uint16x4x4_t b) {
   vst4_lane_u16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4_lane_u32(i32* noundef %a, [4 x <2 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4_lane_u32(ptr noundef %a, [4 x <2 x i32>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint32x2x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint32x2x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x2x4_t, %struct.uint32x2x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <2 x i32>] [[B]].coerce, [4 x <2 x i32>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint32x2x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint32x2x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x2x4_t, %struct.uint32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x i32>], [4 x <2 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint32x2x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <2 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x i32>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x i32> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x2x4_t, %struct.uint32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x i32>], [4 x <2 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x i32>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x i32> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x2x4_t, %struct.uint32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x i32>], [4 x <2 x i32>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x i32>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <2 x i32>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <2 x i32> [[TMP7]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint32x2x4_t, %struct.uint32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x i32>], [4 x <2 x i32>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x i32>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <2 x i32>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP10:%.*]] = bitcast <2 x i32> [[TMP9]] to <8 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP4]] to <2 x i32>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP6]] to <2 x i32>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP8]] to <2 x i32>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP10]] to <2 x i32>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v2i32.p0i8(<2 x i32> [[TMP11]], <2 x i32> [[TMP12]], <2 x i32> [[TMP13]], <2 x i32> [[TMP14]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v2i32.p0(<2 x i32> [[TMP11]], <2 x i32> [[TMP12]], <2 x i32> [[TMP13]], <2 x i32> [[TMP14]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst4_lane_u32(uint32_t  *a, uint32x2x4_t b) {
   vst4_lane_u32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4_lane_u64(i64* noundef %a, [4 x <1 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4_lane_u64(ptr noundef %a, [4 x <1 x i64>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.uint64x1x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.uint64x1x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x1x4_t, %struct.uint64x1x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <1 x i64>] [[B]].coerce, [4 x <1 x i64>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.uint64x1x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.uint64x1x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x1x4_t, %struct.uint64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.uint64x1x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <1 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.uint64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <1 x i64>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <1 x i64> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x1x4_t, %struct.uint64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.uint64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <1 x i64>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <1 x i64> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint64x1x4_t, %struct.uint64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.uint64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <1 x i64>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <1 x i64> [[TMP7]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint64x1x4_t, %struct.uint64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.uint64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <1 x i64>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP10:%.*]] = bitcast <1 x i64> [[TMP9]] to <8 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP4]] to <1 x i64>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP6]] to <1 x i64>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP8]] to <1 x i64>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP10]] to <1 x i64>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v1i64.p0i8(<1 x i64> [[TMP11]], <1 x i64> [[TMP12]], <1 x i64> [[TMP13]], <1 x i64> [[TMP14]], i64 0, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v1i64.p0(<1 x i64> [[TMP11]], <1 x i64> [[TMP12]], <1 x i64> [[TMP13]], <1 x i64> [[TMP14]], i64 0, ptr %a)
 // CHECK:   ret void
 void test_vst4_lane_u64(uint64_t  *a, uint64x1x4_t b) {
   vst4_lane_u64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4_lane_s8(i8* noundef %a, [4 x <8 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4_lane_s8(ptr noundef %a, [4 x <8 x i8>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int8x8x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int8x8x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x8x4_t, %struct.int8x8x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <8 x i8>] [[B]].coerce, [4 x <8 x i8>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int8x8x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int8x8x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x8x4_t, %struct.int8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP2:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX]], align 8
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x8x4_t, %struct.int8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP3:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX2]], align 8
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int8x8x4_t, %struct.int8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP4:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX4]], align 8
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int8x8x4_t, %struct.int8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP5:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX6]], align 8
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v8i8.p0i8(<8 x i8> [[TMP2]], <8 x i8> [[TMP3]], <8 x i8> [[TMP4]], <8 x i8> [[TMP5]], i64 7, i8* %a)
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int8x8x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <8 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP2:%.*]] = load <8 x i8>, ptr [[ARRAYIDX]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP3:%.*]] = load <8 x i8>, ptr [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP4:%.*]] = load <8 x i8>, ptr [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP5:%.*]] = load <8 x i8>, ptr [[ARRAYIDX6]], align 8
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v8i8.p0(<8 x i8> [[TMP2]], <8 x i8> [[TMP3]], <8 x i8> [[TMP4]], <8 x i8> [[TMP5]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst4_lane_s8(int8_t  *a, int8x8x4_t b) {
   vst4_lane_s8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4_lane_s16(i16* noundef %a, [4 x <4 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4_lane_s16(ptr noundef %a, [4 x <4 x i16>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int16x4x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int16x4x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x4x4_t, %struct.int16x4x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <4 x i16>] [[B]].coerce, [4 x <4 x i16>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int16x4x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int16x4x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x4x4_t, %struct.int16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int16x4x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <4 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x i16>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x i16> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x4x4_t, %struct.int16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x i16>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x i16> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int16x4x4_t, %struct.int16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <4 x i16>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <4 x i16> [[TMP7]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int16x4x4_t, %struct.int16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <4 x i16>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP10:%.*]] = bitcast <4 x i16> [[TMP9]] to <8 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP4]] to <4 x i16>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP6]] to <4 x i16>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP8]] to <4 x i16>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP10]] to <4 x i16>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v4i16.p0i8(<4 x i16> [[TMP11]], <4 x i16> [[TMP12]], <4 x i16> [[TMP13]], <4 x i16> [[TMP14]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v4i16.p0(<4 x i16> [[TMP11]], <4 x i16> [[TMP12]], <4 x i16> [[TMP13]], <4 x i16> [[TMP14]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst4_lane_s16(int16_t  *a, int16x4x4_t b) {
   vst4_lane_s16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4_lane_s32(i32* noundef %a, [4 x <2 x i32>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4_lane_s32(ptr noundef %a, [4 x <2 x i32>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int32x2x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int32x2x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x2x4_t, %struct.int32x2x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <2 x i32>] [[B]].coerce, [4 x <2 x i32>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int32x2x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int32x2x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i32* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x2x4_t, %struct.int32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x i32>], [4 x <2 x i32>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int32x2x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <2 x i32>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x i32>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x i32>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x i32> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x2x4_t, %struct.int32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x i32>], [4 x <2 x i32>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x i32>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x i32>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x i32> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int32x2x4_t, %struct.int32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x i32>], [4 x <2 x i32>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x i32>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <2 x i32>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <2 x i32> [[TMP7]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int32x2x4_t, %struct.int32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x i32>], [4 x <2 x i32>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <2 x i32>, <2 x i32>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x i32>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <2 x i32>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP10:%.*]] = bitcast <2 x i32> [[TMP9]] to <8 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP4]] to <2 x i32>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP6]] to <2 x i32>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP8]] to <2 x i32>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP10]] to <2 x i32>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v2i32.p0i8(<2 x i32> [[TMP11]], <2 x i32> [[TMP12]], <2 x i32> [[TMP13]], <2 x i32> [[TMP14]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v2i32.p0(<2 x i32> [[TMP11]], <2 x i32> [[TMP12]], <2 x i32> [[TMP13]], <2 x i32> [[TMP14]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst4_lane_s32(int32_t  *a, int32x2x4_t b) {
   vst4_lane_s32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4_lane_s64(i64* noundef %a, [4 x <1 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4_lane_s64(ptr noundef %a, [4 x <1 x i64>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.int64x1x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.int64x1x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x1x4_t, %struct.int64x1x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <1 x i64>] [[B]].coerce, [4 x <1 x i64>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.int64x1x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.int64x1x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x1x4_t, %struct.int64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.int64x1x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <1 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.int64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <1 x i64>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <1 x i64> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x1x4_t, %struct.int64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.int64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <1 x i64>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <1 x i64> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int64x1x4_t, %struct.int64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.int64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <1 x i64>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <1 x i64> [[TMP7]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int64x1x4_t, %struct.int64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.int64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <1 x i64>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP10:%.*]] = bitcast <1 x i64> [[TMP9]] to <8 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP4]] to <1 x i64>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP6]] to <1 x i64>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP8]] to <1 x i64>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP10]] to <1 x i64>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v1i64.p0i8(<1 x i64> [[TMP11]], <1 x i64> [[TMP12]], <1 x i64> [[TMP13]], <1 x i64> [[TMP14]], i64 0, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v1i64.p0(<1 x i64> [[TMP11]], <1 x i64> [[TMP12]], <1 x i64> [[TMP13]], <1 x i64> [[TMP14]], i64 0, ptr %a)
 // CHECK:   ret void
 void test_vst4_lane_s64(int64_t  *a, int64x1x4_t b) {
   vst4_lane_s64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4_lane_f16(half* noundef %a, [4 x <4 x half>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4_lane_f16(ptr noundef %a, [4 x <4 x half>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.float16x4x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.float16x4x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x4x4_t, %struct.float16x4x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <4 x half>] [[B]].coerce, [4 x <4 x half>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float16x4x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float16x4x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast half* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x4x4_t, %struct.float16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x half>], [4 x <4 x half>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x half>, <4 x half>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float16x4x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <4 x half>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x half>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x half>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x half> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x4x4_t, %struct.float16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x half>], [4 x <4 x half>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x half>, <4 x half>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x half>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x half>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x half> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float16x4x4_t, %struct.float16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x half>], [4 x <4 x half>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <4 x half>, <4 x half>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x half>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <4 x half>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <4 x half> [[TMP7]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float16x4x4_t, %struct.float16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x half>], [4 x <4 x half>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <4 x half>, <4 x half>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x half>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <4 x half>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP10:%.*]] = bitcast <4 x half> [[TMP9]] to <8 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP4]] to <4 x half>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP6]] to <4 x half>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP8]] to <4 x half>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP10]] to <4 x half>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v4f16.p0i8(<4 x half> [[TMP11]], <4 x half> [[TMP12]], <4 x half> [[TMP13]], <4 x half> [[TMP14]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v4f16.p0(<4 x half> [[TMP11]], <4 x half> [[TMP12]], <4 x half> [[TMP13]], <4 x half> [[TMP14]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst4_lane_f16(float16_t  *a, float16x4x4_t b) {
   vst4_lane_f16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4_lane_f32(float* noundef %a, [4 x <2 x float>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4_lane_f32(ptr noundef %a, [4 x <2 x float>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.float32x2x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.float32x2x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x2x4_t, %struct.float32x2x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <2 x float>] [[B]].coerce, [4 x <2 x float>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float32x2x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float32x2x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast float* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x2x4_t, %struct.float32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x float>], [4 x <2 x float>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <2 x float>, <2 x float>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float32x2x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <2 x float>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <2 x float>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <2 x float>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <2 x float> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x2x4_t, %struct.float32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x float>], [4 x <2 x float>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <2 x float>, <2 x float>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <2 x float>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <2 x float>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <2 x float> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float32x2x4_t, %struct.float32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x float>], [4 x <2 x float>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <2 x float>, <2 x float>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <2 x float>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <2 x float>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <2 x float> [[TMP7]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float32x2x4_t, %struct.float32x2x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x float>], [4 x <2 x float>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <2 x float>, <2 x float>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float32x2x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <2 x float>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <2 x float>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP10:%.*]] = bitcast <2 x float> [[TMP9]] to <8 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP4]] to <2 x float>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP6]] to <2 x float>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP8]] to <2 x float>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP10]] to <2 x float>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v2f32.p0i8(<2 x float> [[TMP11]], <2 x float> [[TMP12]], <2 x float> [[TMP13]], <2 x float> [[TMP14]], i64 1, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v2f32.p0(<2 x float> [[TMP11]], <2 x float> [[TMP12]], <2 x float> [[TMP13]], <2 x float> [[TMP14]], i64 1, ptr %a)
 // CHECK:   ret void
 void test_vst4_lane_f32(float32_t  *a, float32x2x4_t b) {
   vst4_lane_f32(a, b, 1);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4_lane_f64(double* noundef %a, [4 x <1 x double>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4_lane_f64(ptr noundef %a, [4 x <1 x double>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.float64x1x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.float64x1x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x1x4_t, %struct.float64x1x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <1 x double>] [[B]].coerce, [4 x <1 x double>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.float64x1x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.float64x1x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast double* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x1x4_t, %struct.float64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <1 x double>], [4 x <1 x double>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <1 x double>, <1 x double>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.float64x1x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <1 x double>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.float64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <1 x double>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <1 x double>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <1 x double> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x1x4_t, %struct.float64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <1 x double>], [4 x <1 x double>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <1 x double>, <1 x double>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.float64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <1 x double>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <1 x double>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <1 x double> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float64x1x4_t, %struct.float64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <1 x double>], [4 x <1 x double>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <1 x double>, <1 x double>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.float64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <1 x double>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <1 x double>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <1 x double> [[TMP7]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float64x1x4_t, %struct.float64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <1 x double>], [4 x <1 x double>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <1 x double>, <1 x double>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.float64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <1 x double>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <1 x double>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP10:%.*]] = bitcast <1 x double> [[TMP9]] to <8 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP4]] to <1 x double>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP6]] to <1 x double>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP8]] to <1 x double>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP10]] to <1 x double>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v1f64.p0i8(<1 x double> [[TMP11]], <1 x double> [[TMP12]], <1 x double> [[TMP13]], <1 x double> [[TMP14]], i64 0, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v1f64.p0(<1 x double> [[TMP11]], <1 x double> [[TMP12]], <1 x double> [[TMP13]], <1 x double> [[TMP14]], i64 0, ptr %a)
 // CHECK:   ret void
 void test_vst4_lane_f64(float64_t  *a, float64x1x4_t b) {
   vst4_lane_f64(a, b, 0);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4_lane_p8(i8* noundef %a, [4 x <8 x i8>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4_lane_p8(ptr noundef %a, [4 x <8 x i8>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.poly8x8x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly8x8x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x8x4_t, %struct.poly8x8x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <8 x i8>] [[B]].coerce, [4 x <8 x i8>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly8x8x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly8x8x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x8x4_t, %struct.poly8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP2:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX]], align 8
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x8x4_t, %struct.poly8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP3:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX2]], align 8
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly8x8x4_t, %struct.poly8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP4:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX4]], align 8
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly8x8x4_t, %struct.poly8x8x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i8>], [4 x <8 x i8>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP5:%.*]] = load <8 x i8>, <8 x i8>* [[ARRAYIDX6]], align 8
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v8i8.p0i8(<8 x i8> [[TMP2]], <8 x i8> [[TMP3]], <8 x i8> [[TMP4]], <8 x i8> [[TMP5]], i64 7, i8* %a)
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly8x8x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <8 x i8>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP2:%.*]] = load <8 x i8>, ptr [[ARRAYIDX]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP3:%.*]] = load <8 x i8>, ptr [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP4:%.*]] = load <8 x i8>, ptr [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly8x8x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <8 x i8>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP5:%.*]] = load <8 x i8>, ptr [[ARRAYIDX6]], align 8
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v8i8.p0(<8 x i8> [[TMP2]], <8 x i8> [[TMP3]], <8 x i8> [[TMP4]], <8 x i8> [[TMP5]], i64 7, ptr %a)
 // CHECK:   ret void
 void test_vst4_lane_p8(poly8_t  *a, poly8x8x4_t b) {
   vst4_lane_p8(a, b, 7);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4_lane_p16(i16* noundef %a, [4 x <4 x i16>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4_lane_p16(ptr noundef %a, [4 x <4 x i16>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.poly16x4x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly16x4x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x4x4_t, %struct.poly16x4x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <4 x i16>] [[B]].coerce, [4 x <4 x i16>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly16x4x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly16x4x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i16* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x4x4_t, %struct.poly16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly16x4x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <4 x i16>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <4 x i16>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <4 x i16> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x4x4_t, %struct.poly16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <4 x i16>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <4 x i16> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x4x4_t, %struct.poly16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <4 x i16>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <4 x i16> [[TMP7]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly16x4x4_t, %struct.poly16x4x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i16>], [4 x <4 x i16>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <4 x i16>, <4 x i16>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly16x4x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <4 x i16>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <4 x i16>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP10:%.*]] = bitcast <4 x i16> [[TMP9]] to <8 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP4]] to <4 x i16>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP6]] to <4 x i16>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP8]] to <4 x i16>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP10]] to <4 x i16>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v4i16.p0i8(<4 x i16> [[TMP11]], <4 x i16> [[TMP12]], <4 x i16> [[TMP13]], <4 x i16> [[TMP14]], i64 3, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v4i16.p0(<4 x i16> [[TMP11]], <4 x i16> [[TMP12]], <4 x i16> [[TMP13]], <4 x i16> [[TMP14]], i64 3, ptr %a)
 // CHECK:   ret void
 void test_vst4_lane_p16(poly16_t  *a, poly16x4x4_t b) {
   vst4_lane_p16(a, b, 3);
 }
 
-// CHECK-LABEL: define{{.*}} void @test_vst4_lane_p64(i64* noundef %a, [4 x <1 x i64>] %b.coerce) #2 {
+// CHECK-LABEL: define{{.*}} void @test_vst4_lane_p64(ptr noundef %a, [4 x <1 x i64>] %b.coerce) #2 {
 // CHECK:   [[B:%.*]] = alloca %struct.poly64x1x4_t, align 8
 // CHECK:   [[__S1:%.*]] = alloca %struct.poly64x1x4_t, align 8
-// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x1x4_t, %struct.poly64x1x4_t* [[B]], i32 0, i32 0
-// CHECK:   store [4 x <1 x i64>] [[B]].coerce, [4 x <1 x i64>]* [[COERCE_DIVE]], align 8
-// CHECK:   [[TMP0:%.*]] = bitcast %struct.poly64x1x4_t* [[__S1]] to i8*
-// CHECK:   [[TMP1:%.*]] = bitcast %struct.poly64x1x4_t* [[B]] to i8*
-// CHECK:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[TMP0]], i8* align 8 [[TMP1]], i64 32, i1 false)
-// CHECK:   [[TMP2:%.*]] = bitcast i64* %a to i8*
-// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x1x4_t, %struct.poly64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL]], i64 0, i64 0
-// CHECK:   [[TMP3:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX]], align 8
+// CHECK:   [[COERCE_DIVE:%.*]] = getelementptr inbounds %struct.poly64x1x4_t, ptr [[B]], i32 0, i32 0
+// CHECK:   store [4 x <1 x i64>] [[B]].coerce, ptr [[COERCE_DIVE]], align 8
+// CHECK:   call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[__S1]], ptr align 8 [[B]], i64 32, i1 false)
+// CHECK:   [[VAL:%.*]] = getelementptr inbounds %struct.poly64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL]], i64 0, i64 0
+// CHECK:   [[TMP3:%.*]] = load <1 x i64>, ptr [[ARRAYIDX]], align 8
 // CHECK:   [[TMP4:%.*]] = bitcast <1 x i64> [[TMP3]] to <8 x i8>
-// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x1x4_t, %struct.poly64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL1]], i64 0, i64 1
-// CHECK:   [[TMP5:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX2]], align 8
+// CHECK:   [[VAL1:%.*]] = getelementptr inbounds %struct.poly64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX2:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL1]], i64 0, i64 1
+// CHECK:   [[TMP5:%.*]] = load <1 x i64>, ptr [[ARRAYIDX2]], align 8
 // CHECK:   [[TMP6:%.*]] = bitcast <1 x i64> [[TMP5]] to <8 x i8>
-// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly64x1x4_t, %struct.poly64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL3]], i64 0, i64 2
-// CHECK:   [[TMP7:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX4]], align 8
+// CHECK:   [[VAL3:%.*]] = getelementptr inbounds %struct.poly64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX4:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL3]], i64 0, i64 2
+// CHECK:   [[TMP7:%.*]] = load <1 x i64>, ptr [[ARRAYIDX4]], align 8
 // CHECK:   [[TMP8:%.*]] = bitcast <1 x i64> [[TMP7]] to <8 x i8>
-// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly64x1x4_t, %struct.poly64x1x4_t* [[__S1]], i32 0, i32 0
-// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <1 x i64>], [4 x <1 x i64>]* [[VAL5]], i64 0, i64 3
-// CHECK:   [[TMP9:%.*]] = load <1 x i64>, <1 x i64>* [[ARRAYIDX6]], align 8
+// CHECK:   [[VAL5:%.*]] = getelementptr inbounds %struct.poly64x1x4_t, ptr [[__S1]], i32 0, i32 0
+// CHECK:   [[ARRAYIDX6:%.*]] = getelementptr inbounds [4 x <1 x i64>], ptr [[VAL5]], i64 0, i64 3
+// CHECK:   [[TMP9:%.*]] = load <1 x i64>, ptr [[ARRAYIDX6]], align 8
 // CHECK:   [[TMP10:%.*]] = bitcast <1 x i64> [[TMP9]] to <8 x i8>
 // CHECK:   [[TMP11:%.*]] = bitcast <8 x i8> [[TMP4]] to <1 x i64>
 // CHECK:   [[TMP12:%.*]] = bitcast <8 x i8> [[TMP6]] to <1 x i64>
 // CHECK:   [[TMP13:%.*]] = bitcast <8 x i8> [[TMP8]] to <1 x i64>
 // CHECK:   [[TMP14:%.*]] = bitcast <8 x i8> [[TMP10]] to <1 x i64>
-// CHECK:   call void @llvm.aarch64.neon.st4lane.v1i64.p0i8(<1 x i64> [[TMP11]], <1 x i64> [[TMP12]], <1 x i64> [[TMP13]], <1 x i64> [[TMP14]], i64 0, i8* [[TMP2]])
+// CHECK:   call void @llvm.aarch64.neon.st4lane.v1i64.p0(<1 x i64> [[TMP11]], <1 x i64> [[TMP12]], <1 x i64> [[TMP13]], <1 x i64> [[TMP14]], i64 0, ptr %a)
 // CHECK:   ret void
 void test_vst4_lane_p64(poly64_t  *a, poly64x1x4_t b) {
   vst4_lane_p64(a, b, 0);

diff  --git a/clang/test/CodeGen/arm-vaarg.c b/clang/test/CodeGen/arm-vaarg.c
index 4dab397a20de4..bed008039274a 100644
--- a/clang/test/CodeGen/arm-vaarg.c
+++ b/clang/test/CodeGen/arm-vaarg.c
@@ -1,4 +1,4 @@
-// RUN: %clang -Xclang -no-opaque-pointers -mfloat-abi=soft -target arm-linux-gnu -emit-llvm -S -o - %s | FileCheck %s
+// RUN: %clang -mfloat-abi=soft -target arm-linux-gnu -emit-llvm -S -o - %s | FileCheck %s
 
 struct Empty {};
 
@@ -7,15 +7,13 @@ struct Empty emptyvar;
 void take_args(int a, ...) {
 // CHECK: [[ALLOCA_VA_LIST:%[a-zA-Z0-9._]+]] = alloca %struct.__va_list, align 4
 // CHECK: call void @llvm.va_start
-// CHECK-NEXT: [[AP_ADDR:%[a-zA-Z0-9._]+]] = bitcast %struct.__va_list* [[ALLOCA_VA_LIST]] to i8**
-// CHECK-NEXT: [[LOAD_AP:%[a-zA-Z0-9._]+]] = load i8*, i8** [[AP_ADDR]], align 4
-// CHECK-NEXT: [[EMPTY_PTR:%[a-zA-Z0-9._]+]] = bitcast i8* [[LOAD_AP]] to %struct.Empty*
+// CHECK-NEXT: [[LOAD_AP:%[a-zA-Z0-9._]+]] = load ptr, ptr [[ALLOCA_VA_LIST]], align 4
 
   // It's conceivable that EMPTY_PTR may not actually be a valid pointer
   // (e.g. it's at the very bottom of the stack and the next page is
   // invalid). This doesn't matter provided it's never loaded (there's no
   // well-defined way to tell), but it becomes a problem if we do try to use it.
-// CHECK-NOT: load %struct.Empty, %struct.Empty* [[EMPTY_PTR]]
+// CHECK-NOT: load %struct.Empty, ptr [[LOAD_AP]]
   __builtin_va_list l;
   __builtin_va_start(l, a);
   emptyvar = __builtin_va_arg(l, struct Empty);

diff  --git a/clang/test/CodeGen/complex-convert.c b/clang/test/CodeGen/complex-convert.c
index 6041bdd00172e..23c5dde5c9c57 100644
--- a/clang/test/CodeGen/complex-convert.c
+++ b/clang/test/CodeGen/complex-convert.c
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -emit-llvm -o - %s | FileCheck %s
 
 // Test conversions between complex integer types and standard integer
 // types.  Tests binary operator conversion and assignment conversion
@@ -34,692 +34,692 @@ void foo(signed char sc, unsigned char uc, signed long long sll,
   // CHECK: store i64 %ull,
 
   sc1 = csc;
-  // CHECK: %[[VAR1:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]]  }, { i[[CHSIZE]], i[[CHSIZE]]  }* %[[CSC:[A-Za-z0-9.]+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR2:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR1]]
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR2]], i[[CHSIZE]]* %[[SC1:[A-Za-z0-9.]+]], align [[CHALIGN]]
+  // CHECK: %[[VAR1:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]]  }, ptr %[[CSC:[A-Za-z0-9.]+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR2:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR1]]
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR2]], ptr %[[SC1:[A-Za-z0-9.]+]], align [[CHALIGN]]
 
   sc1 = cuc;
-  // CHECK-NEXT: %[[VAR3:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]]  }, { i[[CHSIZE]], i[[CHSIZE]]  }* %[[CUC:[A-Za-z0-9.]+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR4:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR3]]
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR4]], i[[CHSIZE]]* %[[SC1]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR3:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]]  }, ptr %[[CUC:[A-Za-z0-9.]+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR4:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR3]]
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR4]], ptr %[[SC1]], align [[CHALIGN]]
 
   sc1 = csll;
-  // CHECK-NEXT: %[[VAR5:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]]  }, { i[[LLSIZE]], i[[LLSIZE]]  }* %[[CSLL:[A-Za-z0-9.]+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR6:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR5]]
+  // CHECK-NEXT: %[[VAR5:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]]  }, ptr %[[CSLL:[A-Za-z0-9.]+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR6:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR5]]
   // CHECK-NEXT: %[[VAR7:[A-Za-z0-9.]+]] = trunc i[[LLSIZE]] %[[VAR6]] to i[[CHSIZE]]
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR7]], i[[CHSIZE]]* %[[SC1]], align [[CHALIGN]]
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR7]], ptr %[[SC1]], align [[CHALIGN]]
 
   sc1 = cull;
-  // CHECK-NEXT: %[[VAR8:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]]  }, { i[[LLSIZE]], i[[LLSIZE]]  }* %[[CULL:[A-Za-z0-9.]+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR9:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR8]]
+  // CHECK-NEXT: %[[VAR8:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]]  }, ptr %[[CULL:[A-Za-z0-9.]+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR9:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR8]]
   // CHECK-NEXT: %[[VAR10:[A-Za-z0-9.]+]] = trunc i[[LLSIZE]] %[[VAR9]] to i[[CHSIZE]]
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR10]], i[[CHSIZE]]* %[[SC1]], align [[CHALIGN]]
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR10]], ptr %[[SC1]], align [[CHALIGN]]
   
   uc1 = csc;
-  // CHECK-NEXT: %[[VAR11:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]]  }, { i[[CHSIZE]], i[[CHSIZE]]  }* %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR12:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR11]]
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR12]], i[[CHSIZE]]* %[[UC1:[A-Za-z0-9.]+]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR11:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]]  }, ptr %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR12:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR11]]
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR12]], ptr %[[UC1:[A-Za-z0-9.]+]], align [[CHALIGN]]
 
   uc1 = cuc;
-  // CHECK-NEXT: %[[VAR13:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]]  }, { i[[CHSIZE]], i[[CHSIZE]]  }* %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR14:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR13]]
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR14]], i[[CHSIZE]]* %[[UC1]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR13:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]]  }, ptr %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR14:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR13]]
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR14]], ptr %[[UC1]], align [[CHALIGN]]
 
   uc1 = csll;
-  // CHECK-NEXT: %[[VAR15:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]]  }, { i[[LLSIZE]], i[[LLSIZE]]  }* %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR16:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR15]]
+  // CHECK-NEXT: %[[VAR15:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]]  }, ptr %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR16:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR15]]
   // CHECK-NEXT: %[[VAR17:[A-Za-z0-9.]+]] = trunc i[[LLSIZE]] %[[VAR16]] to i[[CHSIZE]]
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR17]], i[[CHSIZE]]* %[[UC1]], align [[CHALIGN]]
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR17]], ptr %[[UC1]], align [[CHALIGN]]
 
   uc1 = cull;
-  // CHECK-NEXT: %[[VAR18:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]]  }, { i[[LLSIZE]], i[[LLSIZE]]  }* %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR19:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR18]]
+  // CHECK-NEXT: %[[VAR18:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]]  }, ptr %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR19:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR18]]
   // CHECK-NEXT: %[[VAR20:[A-Za-z0-9.]+]] = trunc i[[LLSIZE]] %[[VAR19]] to i[[CHSIZE]]
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR20]], i[[CHSIZE]]* %[[UC1]], align [[CHALIGN]]
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR20]], ptr %[[UC1]], align [[CHALIGN]]
 
   sll1 = csc;
-  // CHECK-NEXT: %[[VAR21:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]]  }, { i[[CHSIZE]], i[[CHSIZE]]  }* %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR22:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR21]]
+  // CHECK-NEXT: %[[VAR21:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]]  }, ptr %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR22:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR21]]
   // CHECK-NEXT: %[[VAR23:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR22]] to i[[LLSIZE]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR23]], i[[LLSIZE]]* %[[SLL1:[A-Za-z0-9]+]], align [[LLALIGN]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR23]], ptr %[[SLL1:[A-Za-z0-9]+]], align [[LLALIGN]]
 
   sll1 = cuc;
-  // CHECK-NEXT: %[[VAR24:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]]  }, { i[[CHSIZE]], i[[CHSIZE]]  }* %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR25:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR24]]
+  // CHECK-NEXT: %[[VAR24:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]]  }, ptr %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR25:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR24]]
   // CHECK-NEXT: %[[VAR26:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR25]] to i[[LLSIZE]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR26]], i[[LLSIZE]]* %[[SLL1]], align [[LLALIGN]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR26]], ptr %[[SLL1]], align [[LLALIGN]]
 
   sll1 = csll;
-  // CHECK-NEXT: %[[VAR27:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR28:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR27]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR28]], i[[LLSIZE]]* %[[SLL1]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR27:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR28:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR27]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR28]], ptr %[[SLL1]], align [[LLALIGN]]
 
   sll1 = cull;
-  // CHECK-NEXT: %[[VAR29:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR30:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR29]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR30]], i[[LLSIZE]]* %[[SLL1]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR29:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR30:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR29]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR30]], ptr %[[SLL1]], align [[LLALIGN]]
   
   ull1 = csc;
-  // CHECK-NEXT: %[[VAR31:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR32:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR31]]
+  // CHECK-NEXT: %[[VAR31:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR32:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR31]]
   // CHECK-NEXT: %[[VAR33:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR32]] to i[[LLSIZE]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR33]], i[[LLSIZE]]* %[[ULL1:[A-Za-z0-9]+]], align [[LLALIGN]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR33]], ptr %[[ULL1:[A-Za-z0-9]+]], align [[LLALIGN]]
 
   ull1 = cuc;
-  // CHECK-NEXT: %[[VAR34:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR35:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR34]]
+  // CHECK-NEXT: %[[VAR34:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR35:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR34]]
   // CHECK-NEXT: %[[VAR36:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR35]] to i[[LLSIZE]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR36]], i[[LLSIZE]]* %[[ULL1]], align [[LLALIGN]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR36]], ptr %[[ULL1]], align [[LLALIGN]]
 
   ull1 = csll;
-  // CHECK-NEXT: %[[VAR37:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR38:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR37]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR38]], i[[LLSIZE]]* %[[ULL1]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR37:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR38:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR37]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR38]], ptr %[[ULL1]], align [[LLALIGN]]
 
   ull1 = cull;
-  // CHECK-NEXT: %[[VAR39:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR40:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR39]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR40]], i[[LLSIZE]]* %[[ULL1]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR39:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR40:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR39]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR40]], ptr %[[ULL1]], align [[LLALIGN]]
 
   csc1 = sc;
-  // CHECK-NEXT: %[[VAR41:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[SCADDR:[A-Za-z0-9.]+]], align [[CHALIGN]]
-  // CHECK-NEXT: %[[VAR42:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC1:[A-Za-z0-9.]+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR43:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR41]], i[[CHSIZE]]* %[[VAR42]]
-  // CHECK-NEXT: store i[[CHSIZE]] 0, i[[CHSIZE]]* %[[VAR43]]
+  // CHECK-NEXT: %[[VAR41:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[SCADDR:[A-Za-z0-9.]+]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR42:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC1:[A-Za-z0-9.]+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR43:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR41]], ptr %[[VAR42]]
+  // CHECK-NEXT: store i[[CHSIZE]] 0, ptr %[[VAR43]]
 
   csc1 = uc;
-  // CHECK-NEXT: %[[VAR44:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[UCADDR:[A-Za-z0-9.]+]], align [[CHALIGN]]
-  // CHECK-NEXT: %[[VAR45:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR46:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR44]], i[[CHSIZE]]* %[[VAR45]]
-  // CHECK-NEXT: store i[[CHSIZE]] 0, i[[CHSIZE]]* %[[VAR46]]
+  // CHECK-NEXT: %[[VAR44:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[UCADDR:[A-Za-z0-9.]+]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR45:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR46:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR44]], ptr %[[VAR45]]
+  // CHECK-NEXT: store i[[CHSIZE]] 0, ptr %[[VAR46]]
 
   csc1 = sll;
-  // CHECK-NEXT: %[[VAR47:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[SLLADDR:[A-Za-z0-9.]+]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR47:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[SLLADDR:[A-Za-z0-9.]+]], align [[LLALIGN]]
   // CHECK-NEXT: %[[VAR48:[A-Za-z0-9.]+]] = trunc i[[LLSIZE]] %[[VAR47]] to i[[CHSIZE]]
-  // CHECK-NEXT: %[[VAR49:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR50:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR48]], i[[CHSIZE]]* %[[VAR49]]
-  // CHECK-NEXT: store i[[CHSIZE]] 0, i[[CHSIZE]]* %[[VAR50]]
+  // CHECK-NEXT: %[[VAR49:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR50:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR48]], ptr %[[VAR49]]
+  // CHECK-NEXT: store i[[CHSIZE]] 0, ptr %[[VAR50]]
 
   csc1 = ull;
-  // CHECK-NEXT: %[[VAR51:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[ULLADDR:[A-Za-z0-9.]+]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR51:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[ULLADDR:[A-Za-z0-9.]+]], align [[LLALIGN]]
   // CHECK-NEXT: %[[VAR52:[A-Za-z0-9.]+]] = trunc i[[LLSIZE]] %[[VAR51]] to i[[CHSIZE]]
-  // CHECK-NEXT: %[[VAR53:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR54:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR52]], i[[CHSIZE]]* %[[VAR53]]
-  // CHECK-NEXT: store i[[CHSIZE]] 0, i[[CHSIZE]]* %[[VAR54]]
+  // CHECK-NEXT: %[[VAR53:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR54:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR52]], ptr %[[VAR53]]
+  // CHECK-NEXT: store i[[CHSIZE]] 0, ptr %[[VAR54]]
   
   cuc1 = sc;
-  // CHECK-NEXT: %[[VAR55:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[SCADDR]], align [[CHALIGN]]
-  // CHECK-NEXT: %[[VAR56:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC1:[A-Za-z0-9.]+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR57:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR55]], i[[CHSIZE]]* %[[VAR56]]
-  // CHECK-NEXT: store i[[CHSIZE]] 0, i[[CHSIZE]]* %[[VAR57]]
+  // CHECK-NEXT: %[[VAR55:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[SCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR56:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC1:[A-Za-z0-9.]+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR57:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR55]], ptr %[[VAR56]]
+  // CHECK-NEXT: store i[[CHSIZE]] 0, ptr %[[VAR57]]
 
   cuc1 = uc;
-  // CHECK-NEXT: %[[VAR58:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[UCADDR]], align [[CHALIGN]]
-  // CHECK-NEXT: %[[VAR59:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR60:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR58]], i[[CHSIZE]]* %[[VAR59]]
-  // CHECK-NEXT: store i[[CHSIZE]] 0, i[[CHSIZE]]* %[[VAR60]]
+  // CHECK-NEXT: %[[VAR58:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[UCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR59:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR60:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR58]], ptr %[[VAR59]]
+  // CHECK-NEXT: store i[[CHSIZE]] 0, ptr %[[VAR60]]
 
   cuc1 = sll;
-  // CHECK-NEXT: %[[VAR61:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[SLLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR61:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[SLLADDR]], align [[LLALIGN]]
   // CHECK-NEXT: %[[VAR62:[A-Za-z0-9.]+]] = trunc i[[LLSIZE]] %[[VAR61]] to i[[CHSIZE]]
-  // CHECK-NEXT: %[[VAR63:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR64:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR62]], i[[CHSIZE]]* %[[VAR63]]
-  // CHECK-NEXT: store i[[CHSIZE]] 0, i[[CHSIZE]]* %[[VAR64]]
+  // CHECK-NEXT: %[[VAR63:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR64:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR62]], ptr %[[VAR63]]
+  // CHECK-NEXT: store i[[CHSIZE]] 0, ptr %[[VAR64]]
 
   cuc1 = ull;
-  // CHECK-NEXT: %[[VAR65:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[ULLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR65:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[ULLADDR]], align [[LLALIGN]]
   // CHECK-NEXT: %[[VAR66:[A-Za-z0-9.]+]] = trunc i[[LLSIZE]] %[[VAR65]] to i[[CHSIZE]]
-  // CHECK-NEXT: %[[VAR67:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR68:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR66]], i[[CHSIZE]]* %[[VAR67]]
-  // CHECK-NEXT: store i[[CHSIZE]] 0, i[[CHSIZE]]* %[[VAR68]]
+  // CHECK-NEXT: %[[VAR67:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR68:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR66]], ptr %[[VAR67]]
+  // CHECK-NEXT: store i[[CHSIZE]] 0, ptr %[[VAR68]]
 
   csll1 = sc;
-  // CHECK-NEXT: %[[VAR69:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[SCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR69:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[SCADDR]], align [[CHALIGN]]
   // CHECK-NEXT: %[[VAR70:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR69]] to i[[LLSIZE]]
-  // CHECK-NEXT: %[[VAR71:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1:[A-Za-z0-9.]+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR72:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR70]], i[[LLSIZE]]* %[[VAR71]]
-  // CHECK-NEXT: store i[[LLSIZE]] 0, i[[LLSIZE]]* %[[VAR72]]
+  // CHECK-NEXT: %[[VAR71:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1:[A-Za-z0-9.]+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR72:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR70]], ptr %[[VAR71]]
+  // CHECK-NEXT: store i[[LLSIZE]] 0, ptr %[[VAR72]]
 
   csll1 = uc;
-  // CHECK-NEXT: %[[VAR73:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[UCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR73:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[UCADDR]], align [[CHALIGN]]
   // CHECK-NEXT: %[[VAR74:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR73]] to i[[LLSIZE]]
-  // CHECK-NEXT: %[[VAR75:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR76:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR74]], i[[LLSIZE]]* %[[VAR75]]
-  // CHECK-NEXT: store i[[LLSIZE]] 0, i[[LLSIZE]]* %[[VAR76]]
+  // CHECK-NEXT: %[[VAR75:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR76:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR74]], ptr %[[VAR75]]
+  // CHECK-NEXT: store i[[LLSIZE]] 0, ptr %[[VAR76]]
 
   csll1 = sll;
-  // CHECK-NEXT: %[[VAR77:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[SLLADDR]], align [[LLALIGN]]
-  // CHECK-NEXT: %[[VAR78:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR79:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR77]], i[[LLSIZE]]* %[[VAR78]]
-  // CHECK-NEXT: store i[[LLSIZE]] 0, i[[LLSIZE]]* %[[VAR79]]
+  // CHECK-NEXT: %[[VAR77:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[SLLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR78:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR79:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR77]], ptr %[[VAR78]]
+  // CHECK-NEXT: store i[[LLSIZE]] 0, ptr %[[VAR79]]
 
   csll1 = ull;
-  // CHECK-NEXT: %[[VAR77:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[ULLADDR]], align [[LLALIGN]]
-  // CHECK-NEXT: %[[VAR78:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR79:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR77]], i[[LLSIZE]]* %[[VAR78]]
-  // CHECK-NEXT: store i[[LLSIZE]] 0, i[[LLSIZE]]* %[[VAR79]]
+  // CHECK-NEXT: %[[VAR77:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[ULLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR78:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR79:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR77]], ptr %[[VAR78]]
+  // CHECK-NEXT: store i[[LLSIZE]] 0, ptr %[[VAR79]]
 
   cull1 = sc;
-  // CHECK-NEXT: %[[VAR80:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[SCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR80:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[SCADDR]], align [[CHALIGN]]
   // CHECK-NEXT: %[[VAR81:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR80]] to i[[LLSIZE]]
-  // CHECK-NEXT: %[[VAR82:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1:[A-Za-z0-9.]+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR83:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR81]], i[[LLSIZE]]* %[[VAR82]]
-  // CHECK-NEXT: store i[[LLSIZE]] 0, i[[LLSIZE]]* %[[VAR83]]
+  // CHECK-NEXT: %[[VAR82:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1:[A-Za-z0-9.]+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR83:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR81]], ptr %[[VAR82]]
+  // CHECK-NEXT: store i[[LLSIZE]] 0, ptr %[[VAR83]]
 
   cull1 = uc;
-  // CHECK-NEXT: %[[VAR84:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[UCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR84:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[UCADDR]], align [[CHALIGN]]
   // CHECK-NEXT: %[[VAR85:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR84]] to i[[LLSIZE]]
-  // CHECK-NEXT: %[[VAR86:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR87:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR85]], i[[LLSIZE]]* %[[VAR86]]
-  // CHECK-NEXT: store i[[LLSIZE]] 0, i[[LLSIZE]]* %[[VAR87]]
+  // CHECK-NEXT: %[[VAR86:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR87:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR85]], ptr %[[VAR86]]
+  // CHECK-NEXT: store i[[LLSIZE]] 0, ptr %[[VAR87]]
 
   cull1 = sll;
-  // CHECK-NEXT: %[[VAR88:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[SLLADDR]], align [[LLALIGN]]
-  // CHECK-NEXT: %[[VAR89:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR90:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR88]], i[[LLSIZE]]* %[[VAR89]]
-  // CHECK-NEXT: store i[[LLSIZE]] 0, i[[LLSIZE]]* %[[VAR90]]
+  // CHECK-NEXT: %[[VAR88:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[SLLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR89:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR90:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR88]], ptr %[[VAR89]]
+  // CHECK-NEXT: store i[[LLSIZE]] 0, ptr %[[VAR90]]
 
   cull1 = ull;
-  // CHECK-NEXT: %[[VAR91:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[ULLADDR]], align [[LLALIGN]]
-  // CHECK-NEXT: %[[VAR92:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR93:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR91]], i[[LLSIZE]]* %[[VAR92]]
-  // CHECK-NEXT: store i[[LLSIZE]] 0, i[[LLSIZE]]* %[[VAR93]]
+  // CHECK-NEXT: %[[VAR91:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[ULLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR92:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR93:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR91]], ptr %[[VAR92]]
+  // CHECK-NEXT: store i[[LLSIZE]] 0, ptr %[[VAR93]]
 
   csc1 = sc + csc;
-  // CHECK-NEXT: %[[VAR94:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[SCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR94:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[SCADDR]], align [[CHALIGN]]
   // CHECK-NEXT: %[[VAR95:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR94]] to i[[ARSIZE:[0-9]+]]
-  // CHECK-NEXT: %[[VAR96:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR97:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR96]]
-  // CHECK-NEXT: %[[VAR98:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR99:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR98]]
+  // CHECK-NEXT: %[[VAR96:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR97:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR96]]
+  // CHECK-NEXT: %[[VAR98:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR99:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR98]]
   // CHECK-NEXT: %[[VAR100:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR97]] to i[[ARSIZE]]
   // CHECK-NEXT: %[[VAR101:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR99]] to i[[ARSIZE]]
   // CHECK-NEXT: %[[VAR102:[A-Za-z0-9.]+]] = add i[[ARSIZE]] %[[VAR95]], %[[VAR100]]
   // CHECK-NEXT: %[[VAR103:[A-Za-z0-9.]+]] = add i[[ARSIZE]] 0, %[[VAR101]]
   // CHECK-NEXT: %[[VAR104:[A-Za-z0-9.]+]] = trunc i[[ARSIZE]] %[[VAR102]] to i[[CHSIZE]]
   // CHECK-NEXT: %[[VAR105:[A-Za-z0-9.]+]] = trunc i[[ARSIZE]] %[[VAR103]] to i[[CHSIZE]]
-  // CHECK-NEXT: %[[VAR106:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR107:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR104]], i[[CHSIZE]]* %[[VAR106]]
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR105]], i[[CHSIZE]]* %[[VAR107]]
+  // CHECK-NEXT: %[[VAR106:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR107:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR104]], ptr %[[VAR106]]
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR105]], ptr %[[VAR107]]
 
   cuc1 = sc + cuc;
-  // CHECK-NEXT: %[[VAR108:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[SCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR108:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[SCADDR]], align [[CHALIGN]]
   // CHECK-NEXT: %[[VAR109:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR108]] to i[[ARSIZE]]
-  // CHECK-NEXT: %[[VAR110:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR111:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR110]]
-  // CHECK-NEXT: %[[VAR112:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR113:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR112]]
+  // CHECK-NEXT: %[[VAR110:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR111:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR110]]
+  // CHECK-NEXT: %[[VAR112:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR113:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR112]]
   // CHECK-NEXT: %[[VAR114:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR111]] to i[[ARSIZE]]
   // CHECK-NEXT: %[[VAR115:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR113]] to i[[ARSIZE]]
   // CHECK-NEXT: %[[VAR116:[A-Za-z0-9.]+]] = add i[[ARSIZE]] %[[VAR109]], %[[VAR114]]
   // CHECK-NEXT: %[[VAR117:[A-Za-z0-9.]+]] = add i[[ARSIZE]] 0, %[[VAR115]]
   // CHECK-NEXT: %[[VAR118:[A-Za-z0-9.]+]] = trunc i[[ARSIZE]] %[[VAR116]] to i[[CHSIZE]]
   // CHECK-NEXT: %[[VAR119:[A-Za-z0-9.]+]] = trunc i[[ARSIZE]] %[[VAR117]] to i[[CHSIZE]]
-  // CHECK-NEXT: %[[VAR120:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR121:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR118]], i[[CHSIZE]]* %[[VAR120]]
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR119]], i[[CHSIZE]]* %[[VAR121]]
+  // CHECK-NEXT: %[[VAR120:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR121:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR118]], ptr %[[VAR120]]
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR119]], ptr %[[VAR121]]
 
   csll1 = sc + csll;
-  // CHECK-NEXT: %[[VAR122:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[SCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR122:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[SCADDR]], align [[CHALIGN]]
   // CHECK-NEXT: %[[VAR123:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR122]] to i[[LLSIZE]]
-  // CHECK-NEXT: %[[VAR124:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR125:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR124]]
-  // CHECK-NEXT: %[[VAR126:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR127:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR126]]
+  // CHECK-NEXT: %[[VAR124:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR125:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR124]]
+  // CHECK-NEXT: %[[VAR126:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR127:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR126]]
   // CHECK-NEXT: %[[VAR128:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR123]], %[[VAR125]]
   // CHECK-NEXT: %[[VAR129:[A-Za-z0-9.]+]] = add i[[LLSIZE]] 0, %[[VAR127]]
-  // CHECK-NEXT: %[[VAR130:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR131:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR128]], i[[LLSIZE]]* %[[VAR130]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR129]], i[[LLSIZE]]* %[[VAR131]]
+  // CHECK-NEXT: %[[VAR130:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR131:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR128]], ptr %[[VAR130]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR129]], ptr %[[VAR131]]
 
   cull1 = sc + cull;
-  // CHECK-NEXT: %[[VAR132:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[SCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR132:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[SCADDR]], align [[CHALIGN]]
   // CHECK-NEXT: %[[VAR133:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR132]] to i[[LLSIZE]]
-  // CHECK-NEXT: %[[VAR134:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR135:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR134]]
-  // CHECK-NEXT: %[[VAR136:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR137:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR136]]
+  // CHECK-NEXT: %[[VAR134:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR135:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR134]]
+  // CHECK-NEXT: %[[VAR136:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR137:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR136]]
   // CHECK-NEXT: %[[VAR138:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR133]], %[[VAR135]]
   // CHECK-NEXT: %[[VAR139:[A-Za-z0-9.]+]] = add i[[LLSIZE]] 0, %[[VAR137]]
-  // CHECK-NEXT: %[[VAR140:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR141:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR138]], i[[LLSIZE]]* %[[VAR140]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR139]], i[[LLSIZE]]* %[[VAR141]]
+  // CHECK-NEXT: %[[VAR140:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR141:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR138]], ptr %[[VAR140]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR139]], ptr %[[VAR141]]
   
   csc1 = uc + csc;
-  // CHECK-NEXT: %[[VAR142:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[UCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR142:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[UCADDR]], align [[CHALIGN]]
   // CHECK-NEXT: %[[VAR143:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR142]] to i[[ARSIZE]]
-  // CHECK-NEXT: %[[VAR144:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR145:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR144]]
-  // CHECK-NEXT: %[[VAR146:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR147:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR146]]
+  // CHECK-NEXT: %[[VAR144:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR145:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR144]]
+  // CHECK-NEXT: %[[VAR146:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR147:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR146]]
   // CHECK-NEXT: %[[VAR148:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR145]] to i[[ARSIZE]]
   // CHECK-NEXT: %[[VAR149:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR147]] to i[[ARSIZE]]
   // CHECK-NEXT: %[[VAR150:[A-Za-z0-9.]+]] = add i[[ARSIZE]] %[[VAR143]], %[[VAR148]]
   // CHECK-NEXT: %[[VAR151:[A-Za-z0-9.]+]] = add i[[ARSIZE]] 0, %[[VAR149]]
   // CHECK-NEXT: %[[VAR152:[A-Za-z0-9.]+]] = trunc i[[ARSIZE]] %[[VAR150]] to i[[CHSIZE]]
   // CHECK-NEXT: %[[VAR153:[A-Za-z0-9.]+]] = trunc i[[ARSIZE]] %[[VAR151]] to i[[CHSIZE]]
-  // CHECK-NEXT: %[[VAR154:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR155:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR152]], i[[CHSIZE]]* %[[VAR154]]
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR153]], i[[CHSIZE]]* %[[VAR155]]
+  // CHECK-NEXT: %[[VAR154:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR155:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR152]], ptr %[[VAR154]]
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR153]], ptr %[[VAR155]]
 
   cuc1 = uc + cuc;
-  // CHECK-NEXT: %[[VAR156:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[UCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR156:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[UCADDR]], align [[CHALIGN]]
   // CHECK-NEXT: %[[VAR157:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR156]] to i[[ARSIZE]]
-  // CHECK-NEXT: %[[VAR158:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR159:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR158]]
-  // CHECK-NEXT: %[[VAR160:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR161:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR160]]
+  // CHECK-NEXT: %[[VAR158:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR159:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR158]]
+  // CHECK-NEXT: %[[VAR160:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR161:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR160]]
   // CHECK-NEXT: %[[VAR162:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR159]] to i[[ARSIZE]]
   // CHECK-NEXT: %[[VAR163:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR161]] to i[[ARSIZE]]
   // CHECK-NEXT: %[[VAR164:[A-Za-z0-9.]+]] = add i[[ARSIZE]] %[[VAR157]], %[[VAR162]]
   // CHECK-NEXT: %[[VAR165:[A-Za-z0-9.]+]] = add i[[ARSIZE]] 0, %[[VAR163]]
   // CHECK-NEXT: %[[VAR166:[A-Za-z0-9.]+]] = trunc i[[ARSIZE]] %[[VAR164]] to i[[CHSIZE]]
   // CHECK-NEXT: %[[VAR167:[A-Za-z0-9.]+]] = trunc i[[ARSIZE]] %[[VAR165]] to i[[CHSIZE]]
-  // CHECK-NEXT: %[[VAR168:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR169:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR166]], i[[CHSIZE]]* %[[VAR168]]
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR167]], i[[CHSIZE]]* %[[VAR169]]
+  // CHECK-NEXT: %[[VAR168:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR169:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR166]], ptr %[[VAR168]]
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR167]], ptr %[[VAR169]]
 
   csll1 = uc + csll;
-  // CHECK-NEXT: %[[VAR170:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[UCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR170:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[UCADDR]], align [[CHALIGN]]
   // CHECK-NEXT: %[[VAR171:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR170]] to i[[LLSIZE]]
-  // CHECK-NEXT: %[[VAR172:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR173:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR172]]
-  // CHECK-NEXT: %[[VAR174:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR175:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR174]]
+  // CHECK-NEXT: %[[VAR172:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR173:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR172]]
+  // CHECK-NEXT: %[[VAR174:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR175:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR174]]
   // CHECK-NEXT: %[[VAR176:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR171]], %[[VAR173]]
   // CHECK-NEXT: %[[VAR177:[A-Za-z0-9.]+]] = add i[[LLSIZE]] 0, %[[VAR175]]
-  // CHECK-NEXT: %[[VAR178:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR179:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR176]], i[[LLSIZE]]* %[[VAR178]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR177]], i[[LLSIZE]]* %[[VAR179]]
+  // CHECK-NEXT: %[[VAR178:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR179:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR176]], ptr %[[VAR178]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR177]], ptr %[[VAR179]]
 
   cull1 = uc + cull;
-  // CHECK-NEXT: %[[VAR180:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[UCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR180:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[UCADDR]], align [[CHALIGN]]
   // CHECK-NEXT: %[[VAR181:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR180]] to i[[LLSIZE]]
-  // CHECK-NEXT: %[[VAR182:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR183:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR182]]
-  // CHECK-NEXT: %[[VAR184:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR185:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR184]]
+  // CHECK-NEXT: %[[VAR182:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR183:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR182]]
+  // CHECK-NEXT: %[[VAR184:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR185:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR184]]
   // CHECK-NEXT: %[[VAR186:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR181]], %[[VAR183]]
   // CHECK-NEXT: %[[VAR187:[A-Za-z0-9.]+]] = add i[[LLSIZE]] 0, %[[VAR185]]
-  // CHECK-NEXT: %[[VAR188:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR189:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR186]], i[[LLSIZE]]* %[[VAR188]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR187]], i[[LLSIZE]]* %[[VAR189]]
+  // CHECK-NEXT: %[[VAR188:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR189:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR186]], ptr %[[VAR188]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR187]], ptr %[[VAR189]]
 
   csll1 = sll + csc;
-  // CHECK-NEXT: %[[VAR190:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[SLLADDR]], align [[LLALIGN]]
-  // CHECK-NEXT: %[[VAR191:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR192:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR191]]
-  // CHECK-NEXT: %[[VAR193:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR194:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR193]]
+  // CHECK-NEXT: %[[VAR190:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[SLLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR191:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR192:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR191]]
+  // CHECK-NEXT: %[[VAR193:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR194:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR193]]
   // CHECK-NEXT: %[[VAR195:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR192]] to i[[LLSIZE]]
   // CHECK-NEXT: %[[VAR196:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR194]] to i[[LLSIZE]]
   // CHECK-NEXT: %[[VAR197:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR190]], %[[VAR195]]
   // CHECK-NEXT: %[[VAR198:[A-Za-z0-9.]+]] = add i[[LLSIZE]] 0, %[[VAR196]]
-  // CHECK-NEXT: %[[VAR199:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR200:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR197]], i[[LLSIZE]]* %[[VAR199]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR198]], i[[LLSIZE]]* %[[VAR200]]
+  // CHECK-NEXT: %[[VAR199:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR200:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR197]], ptr %[[VAR199]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR198]], ptr %[[VAR200]]
 
   csll1 = sll + cuc;
-  // CHECK-NEXT: %[[VAR201:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[SLLADDR]], align [[LLALIGN]]
-  // CHECK-NEXT: %[[VAR202:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR203:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR202]]
-  // CHECK-NEXT: %[[VAR204:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR205:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR204]]
+  // CHECK-NEXT: %[[VAR201:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[SLLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR202:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR203:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR202]]
+  // CHECK-NEXT: %[[VAR204:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR205:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR204]]
   // CHECK-NEXT: %[[VAR206:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR203]] to i[[LLSIZE]]
   // CHECK-NEXT: %[[VAR207:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR205]] to i[[LLSIZE]]
   // CHECK-NEXT: %[[VAR208:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR201]], %[[VAR206]]
   // CHECK-NEXT: %[[VAR209:[A-Za-z0-9.]+]] = add i[[LLSIZE]] 0, %[[VAR207]]
-  // CHECK-NEXT: %[[VAR210:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR211:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR208]], i[[LLSIZE]]* %[[VAR210]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR209]], i[[LLSIZE]]* %[[VAR211]]
+  // CHECK-NEXT: %[[VAR210:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR211:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR208]], ptr %[[VAR210]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR209]], ptr %[[VAR211]]
 
   csll1 = sll + csll;
-  // CHECK-NEXT: %[[VAR212:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[SLLADDR]], align [[LLALIGN]]
-  // CHECK-NEXT: %[[VAR213:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR214:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR213]]
-  // CHECK-NEXT: %[[VAR215:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR216:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR215]]
+  // CHECK-NEXT: %[[VAR212:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[SLLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR213:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR214:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR213]]
+  // CHECK-NEXT: %[[VAR215:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR216:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR215]]
   // CHECK-NEXT: %[[VAR217:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR212]], %[[VAR214]]
   // CHECK-NEXT: %[[VAR218:[A-Za-z0-9.]+]] = add i[[LLSIZE]] 0, %[[VAR216]]
-  // CHECK-NEXT: %[[VAR219:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR220:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR217]], i[[LLSIZE]]* %[[VAR219]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR218]], i[[LLSIZE]]* %[[VAR220]]
+  // CHECK-NEXT: %[[VAR219:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR220:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR217]], ptr %[[VAR219]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR218]], ptr %[[VAR220]]
 
   csll1 = sll + cull;
-  // CHECK-NEXT: %[[VAR221:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[SLLADDR]], align [[LLALIGN]]
-  // CHECK-NEXT: %[[VAR222:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR223:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR222]]
-  // CHECK-NEXT: %[[VAR224:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR225:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR224]]
+  // CHECK-NEXT: %[[VAR221:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[SLLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR222:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR223:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR222]]
+  // CHECK-NEXT: %[[VAR224:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR225:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR224]]
   // CHECK-NEXT: %[[VAR226:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR221]], %[[VAR223]]
   // CHECK-NEXT: %[[VAR227:[A-Za-z0-9.]+]] = add i[[LLSIZE]] 0, %[[VAR225]]
-  // CHECK-NEXT: %[[VAR228:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR229:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR226]], i[[LLSIZE]]* %[[VAR228]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR227]], i[[LLSIZE]]* %[[VAR229]]
+  // CHECK-NEXT: %[[VAR228:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR229:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR226]], ptr %[[VAR228]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR227]], ptr %[[VAR229]]
   
   csll1 = ull + csc;
-  // CHECK-NEXT: %[[VAR230:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[ULLADDR]], align [[LLALIGN]]
-  // CHECK-NEXT: %[[VAR231:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR232:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR231]]
-  // CHECK-NEXT: %[[VAR233:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR234:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR233]]
+  // CHECK-NEXT: %[[VAR230:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[ULLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR231:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR232:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR231]]
+  // CHECK-NEXT: %[[VAR233:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR234:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR233]]
   // CHECK-NEXT: %[[VAR235:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR232]] to i[[LLSIZE]]
   // CHECK-NEXT: %[[VAR236:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR234]] to i[[LLSIZE]]
   // CHECK-NEXT: %[[VAR237:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR230]], %[[VAR235]]
   // CHECK-NEXT: %[[VAR238:[A-Za-z0-9.]+]] = add i[[LLSIZE]] 0, %[[VAR236]]
-  // CHECK-NEXT: %[[VAR239:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR240:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR237]], i[[LLSIZE]]* %[[VAR239]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR238]], i[[LLSIZE]]* %[[VAR240]]
+  // CHECK-NEXT: %[[VAR239:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR240:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR237]], ptr %[[VAR239]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR238]], ptr %[[VAR240]]
 
   cull1 = ull + cuc;
-  // CHECK-NEXT: %[[VAR241:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[ULLADDR]], align [[LLALIGN]]
-  // CHECK-NEXT: %[[VAR242:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR243:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR242]]
-  // CHECK-NEXT: %[[VAR244:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR245:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR244]]
+  // CHECK-NEXT: %[[VAR241:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[ULLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR242:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR243:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR242]]
+  // CHECK-NEXT: %[[VAR244:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR245:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR244]]
   // CHECK-NEXT: %[[VAR246:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR243]] to i[[LLSIZE]]
   // CHECK-NEXT: %[[VAR247:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR245]] to i[[LLSIZE]]
   // CHECK-NEXT: %[[VAR248:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR241]], %[[VAR246]]
   // CHECK-NEXT: %[[VAR249:[A-Za-z0-9.]+]] = add i[[LLSIZE]] 0, %[[VAR247]]
-  // CHECK-NEXT: %[[VAR250:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR251:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR248]], i[[LLSIZE]]* %[[VAR250]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR249]], i[[LLSIZE]]* %[[VAR251]]
+  // CHECK-NEXT: %[[VAR250:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR251:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR248]], ptr %[[VAR250]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR249]], ptr %[[VAR251]]
 
   csll1 = ull + csll;
-  // CHECK-NEXT: %[[VAR252:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[ULLADDR]], align [[LLALIGN]]
-  // CHECK-NEXT: %[[VAR253:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR254:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR253]]
-  // CHECK-NEXT: %[[VAR255:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR256:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR255]]
+  // CHECK-NEXT: %[[VAR252:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[ULLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR253:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR254:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR253]]
+  // CHECK-NEXT: %[[VAR255:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR256:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR255]]
   // CHECK-NEXT: %[[VAR257:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR252]], %[[VAR254]]
   // CHECK-NEXT: %[[VAR258:[A-Za-z0-9.]+]] = add i[[LLSIZE]] 0, %[[VAR256]]
-  // CHECK-NEXT: %[[VAR259:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR260:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR257]], i[[LLSIZE]]* %[[VAR259]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR258]], i[[LLSIZE]]* %[[VAR260]]
+  // CHECK-NEXT: %[[VAR259:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR260:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR257]], ptr %[[VAR259]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR258]], ptr %[[VAR260]]
 
   cull1 = ull + cull;
-  // CHECK-NEXT: %[[VAR261:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[ULLADDR]], align [[LLALIGN]]
-  // CHECK-NEXT: %[[VAR262:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR263:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR262]]
-  // CHECK-NEXT: %[[VAR264:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR265:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR264]]
+  // CHECK-NEXT: %[[VAR261:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[ULLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR262:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR263:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR262]]
+  // CHECK-NEXT: %[[VAR264:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR265:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR264]]
   // CHECK-NEXT: %[[VAR266:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR261]], %[[VAR263]]
   // CHECK-NEXT: %[[VAR267:[A-Za-z0-9.]+]] = add i[[LLSIZE]] 0, %[[VAR265]]
-  // CHECK-NEXT: %[[VAR268:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR269:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR266]], i[[LLSIZE]]* %[[VAR268]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR267]], i[[LLSIZE]]* %[[VAR269]]
+  // CHECK-NEXT: %[[VAR268:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR269:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR266]], ptr %[[VAR268]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR267]], ptr %[[VAR269]]
 
   csc1 = csc + sc;
-  // CHECK-NEXT: %[[VAR270:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR271:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR270]]
-  // CHECK-NEXT: %[[VAR272:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR273:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR272]]
+  // CHECK-NEXT: %[[VAR270:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR271:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR270]]
+  // CHECK-NEXT: %[[VAR272:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR273:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR272]]
   // CHECK-NEXT: %[[VAR274:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR271]] to i[[ARSIZE]]
   // CHECK-NEXT: %[[VAR275:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR273]] to i[[ARSIZE]]
-  // CHECK-NEXT: %[[VAR276:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[SCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR276:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[SCADDR]], align [[CHALIGN]]
   // CHECK-NEXT: %[[VAR277:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR276]] to i[[ARSIZE]]
   // CHECK-NEXT: %[[VAR278:[A-Za-z0-9.]+]] = add i[[ARSIZE]] %[[VAR274]], %[[VAR277]]
   // CHECK-NEXT: %[[VAR279:[A-Za-z0-9.]+]] = add i[[ARSIZE]] %[[VAR275]], 0
   // CHECK-NEXT: %[[VAR280:[A-Za-z0-9.]+]] = trunc i[[ARSIZE]] %[[VAR278]] to i[[CHSIZE]]
   // CHECK-NEXT: %[[VAR281:[A-Za-z0-9.]+]] = trunc i[[ARSIZE]] %[[VAR279]] to i[[CHSIZE]]
-  // CHECK-NEXT: %[[VAR282:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR283:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR280]], i[[CHSIZE]]* %[[VAR282]]
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR281]], i[[CHSIZE]]* %[[VAR283]]
+  // CHECK-NEXT: %[[VAR282:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR283:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR280]], ptr %[[VAR282]]
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR281]], ptr %[[VAR283]]
 
   csc1 = csc + uc;
-  // CHECK-NEXT: %[[VAR284:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR285:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR284]]
-  // CHECK-NEXT: %[[VAR286:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR287:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR286]]
+  // CHECK-NEXT: %[[VAR284:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR285:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR284]]
+  // CHECK-NEXT: %[[VAR286:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR287:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR286]]
   // CHECK-NEXT: %[[VAR288:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR285]] to i[[ARSIZE]]
   // CHECK-NEXT: %[[VAR289:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR287]] to i[[ARSIZE]]
-  // CHECK-NEXT: %[[VAR290:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[UCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR290:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[UCADDR]], align [[CHALIGN]]
   // CHECK-NEXT: %[[VAR291:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR290]] to i[[ARSIZE]]
   // CHECK-NEXT: %[[VAR292:[A-Za-z0-9.]+]] = add i[[ARSIZE]] %[[VAR288]], %[[VAR291]]
   // CHECK-NEXT: %[[VAR293:[A-Za-z0-9.]+]] = add i[[ARSIZE]] %[[VAR289]], 0
   // CHECK-NEXT: %[[VAR294:[A-Za-z0-9.]+]] = trunc i[[ARSIZE]] %[[VAR292]] to i[[CHSIZE]]
   // CHECK-NEXT: %[[VAR295:[A-Za-z0-9.]+]] = trunc i[[ARSIZE]] %[[VAR293]] to i[[CHSIZE]]
-  // CHECK-NEXT: %[[VAR296:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR297:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR294]], i[[CHSIZE]]* %[[VAR296]]
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR295]], i[[CHSIZE]]* %[[VAR297]]
+  // CHECK-NEXT: %[[VAR296:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR297:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR294]], ptr %[[VAR296]]
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR295]], ptr %[[VAR297]]
 
   csll1 = csc + sll;
-  // CHECK-NEXT: %[[VAR298:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR299:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR298]]
-  // CHECK-NEXT: %[[VAR300:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR301:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR300]]
+  // CHECK-NEXT: %[[VAR298:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR299:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR298]]
+  // CHECK-NEXT: %[[VAR300:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR301:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR300]]
   // CHECK-NEXT: %[[VAR302:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR299]] to i[[LLSIZE]]
   // CHECK-NEXT: %[[VAR303:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR301]] to i[[LLSIZE]]
-  // CHECK-NEXT: %[[VAR304:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[SLLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR304:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[SLLADDR]], align [[LLALIGN]]
   // CHECK-NEXT: %[[VAR305:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR302]], %[[VAR304]]
   // CHECK-NEXT: %[[VAR306:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR303]], 0
-  // CHECK-NEXT: %[[VAR307:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR308:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR305]], i[[LLSIZE]]* %[[VAR307]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR306]], i[[LLSIZE]]* %[[VAR308]]
+  // CHECK-NEXT: %[[VAR307:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR308:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR305]], ptr %[[VAR307]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR306]], ptr %[[VAR308]]
 
   csll1 = csc + ull;
-  // CHECK-NEXT: %[[VAR309:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR310:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR309]]
-  // CHECK-NEXT: %[[VAR311:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR312:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR311]]
+  // CHECK-NEXT: %[[VAR309:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR310:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR309]]
+  // CHECK-NEXT: %[[VAR311:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR312:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR311]]
   // CHECK-NEXT: %[[VAR313:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR310]] to i[[LLSIZE]]
   // CHECK-NEXT: %[[VAR314:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR312]] to i[[LLSIZE]]
-  // CHECK-NEXT: %[[VAR315:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[ULLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR315:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[ULLADDR]], align [[LLALIGN]]
   // CHECK-NEXT: %[[VAR316:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR313]], %[[VAR315]]
   // CHECK-NEXT: %[[VAR317:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR314]], 0
-  // CHECK-NEXT: %[[VAR318:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR319:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR316]], i[[LLSIZE]]* %[[VAR318]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR317]], i[[LLSIZE]]* %[[VAR319]]
+  // CHECK-NEXT: %[[VAR318:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR319:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR316]], ptr %[[VAR318]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR317]], ptr %[[VAR319]]
   
   csc1 = cuc + sc;
-  // CHECK-NEXT: %[[VAR320:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR321:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR320]]
-  // CHECK-NEXT: %[[VAR322:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR323:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR322]]
+  // CHECK-NEXT: %[[VAR320:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR321:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR320]]
+  // CHECK-NEXT: %[[VAR322:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR323:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR322]]
   // CHECK-NEXT: %[[VAR324:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR321]] to i[[ARSIZE]]
   // CHECK-NEXT: %[[VAR325:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR323]] to i[[ARSIZE]]
-  // CHECK-NEXT: %[[VAR326:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[SCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR326:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[SCADDR]], align [[CHALIGN]]
   // CHECK-NEXT: %[[VAR327:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR326]] to i[[ARSIZE]]
   // CHECK-NEXT: %[[VAR328:[A-Za-z0-9.]+]] = add i[[ARSIZE]] %[[VAR324]], %[[VAR327]]
   // CHECK-NEXT: %[[VAR329:[A-Za-z0-9.]+]] = add i[[ARSIZE]] %[[VAR325]], 0
   // CHECK-NEXT: %[[VAR330:[A-Za-z0-9.]+]] = trunc i[[ARSIZE]] %[[VAR328]] to i[[CHSIZE]]
   // CHECK-NEXT: %[[VAR331:[A-Za-z0-9.]+]] = trunc i[[ARSIZE]] %[[VAR329]] to i[[CHSIZE]]
-  // CHECK-NEXT: %[[VAR332:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR333:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR330]], i[[CHSIZE]]* %[[VAR332]]
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR331]], i[[CHSIZE]]* %[[VAR333]]
+  // CHECK-NEXT: %[[VAR332:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR333:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CSC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR330]], ptr %[[VAR332]]
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR331]], ptr %[[VAR333]]
 
   cuc1 = cuc + uc;
-  // CHECK-NEXT: %[[VAR334:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR335:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR334]]
-  // CHECK-NEXT: %[[VAR336:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR337:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR336]]
+  // CHECK-NEXT: %[[VAR334:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR335:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR334]]
+  // CHECK-NEXT: %[[VAR336:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR337:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR336]]
   // CHECK-NEXT: %[[VAR338:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR335]] to i[[ARSIZE]]
   // CHECK-NEXT: %[[VAR339:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR337]] to i[[ARSIZE]]
-  // CHECK-NEXT: %[[VAR340:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[UCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR340:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[UCADDR]], align [[CHALIGN]]
   // CHECK-NEXT: %[[VAR341:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR340]] to i[[ARSIZE]]
   // CHECK-NEXT: %[[VAR342:[A-Za-z0-9.]+]] = add i[[ARSIZE]] %[[VAR338]], %[[VAR341]]
   // CHECK-NEXT: %[[VAR343:[A-Za-z0-9.]+]] = add i[[ARSIZE]] %[[VAR339]], 0
   // CHECK-NEXT: %[[VAR344:[A-Za-z0-9.]+]] = trunc i[[ARSIZE]] %[[VAR342]] to i[[CHSIZE]]
   // CHECK-NEXT: %[[VAR345:[A-Za-z0-9.]+]] = trunc i[[ARSIZE]] %[[VAR343]] to i[[CHSIZE]]
-  // CHECK-NEXT: %[[VAR346:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR347:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR344]], i[[CHSIZE]]* %[[VAR346]]
-  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR345]], i[[CHSIZE]]* %[[VAR347]]
+  // CHECK-NEXT: %[[VAR346:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR347:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR344]], ptr %[[VAR346]]
+  // CHECK-NEXT: store i[[CHSIZE]] %[[VAR345]], ptr %[[VAR347]]
 
   csll1 = cuc + sll;
-  // CHECK-NEXT: %[[VAR348:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR349:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR348]]
-  // CHECK-NEXT: %[[VAR350:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR351:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR350]]
+  // CHECK-NEXT: %[[VAR348:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR349:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR348]]
+  // CHECK-NEXT: %[[VAR350:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR351:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR350]]
   // CHECK-NEXT: %[[VAR352:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR349]] to i[[LLSIZE]]
   // CHECK-NEXT: %[[VAR353:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR351]] to i[[LLSIZE]]
-  // CHECK-NEXT: %[[VAR354:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[SLLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR354:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[SLLADDR]], align [[LLALIGN]]
   // CHECK-NEXT: %[[VAR355:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR352]], %[[VAR354]]
   // CHECK-NEXT: %[[VAR356:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR353]], 0
-  // CHECK-NEXT: %[[VAR357:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR358:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR355]], i[[LLSIZE]]* %[[VAR357]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR356]], i[[LLSIZE]]* %[[VAR358]]
+  // CHECK-NEXT: %[[VAR357:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR358:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR355]], ptr %[[VAR357]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR356]], ptr %[[VAR358]]
 
   cull1 = cuc + ull;
-  // CHECK-NEXT: %[[VAR357:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR358:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR357]]
-  // CHECK-NEXT: %[[VAR359:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, { i[[CHSIZE]], i[[CHSIZE]] }* %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR360:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[VAR359]]
+  // CHECK-NEXT: %[[VAR357:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR358:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR357]]
+  // CHECK-NEXT: %[[VAR359:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[CHSIZE]], i[[CHSIZE]] }, ptr %[[CUC]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR360:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[VAR359]]
   // CHECK-NEXT: %[[VAR361:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR358]] to i[[LLSIZE]]
   // CHECK-NEXT: %[[VAR362:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR360]] to i[[LLSIZE]]
-  // CHECK-NEXT: %[[VAR363:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[ULLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR363:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[ULLADDR]], align [[LLALIGN]]
   // CHECK-NEXT: %[[VAR364:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR361]], %[[VAR363]]
   // CHECK-NEXT: %[[VAR365:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR362]], 0
-  // CHECK-NEXT: %[[VAR366:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR367:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR364]], i[[LLSIZE]]* %[[VAR366]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR365]], i[[LLSIZE]]* %[[VAR367]]
+  // CHECK-NEXT: %[[VAR366:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR367:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR364]], ptr %[[VAR366]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR365]], ptr %[[VAR367]]
 
   csll1 = csll + sc;
-  // CHECK-NEXT: %[[VAR368:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR369:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR368]]
-  // CHECK-NEXT: %[[VAR370:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR371:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR370]]
-  // CHECK-NEXT: %[[VAR372:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[SCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR368:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR369:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR368]]
+  // CHECK-NEXT: %[[VAR370:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR371:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR370]]
+  // CHECK-NEXT: %[[VAR372:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[SCADDR]], align [[CHALIGN]]
   // CHECK-NEXT: %[[VAR373:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR372]] to i[[LLSIZE]]
   // CHECK-NEXT: %[[VAR374:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR369]], %[[VAR373]]
   // CHECK-NEXT: %[[VAR375:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR371]], 0
-  // CHECK-NEXT: %[[VAR376:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR377:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR374]], i[[LLSIZE]]* %[[VAR376]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR375]], i[[LLSIZE]]* %[[VAR377]]
+  // CHECK-NEXT: %[[VAR376:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR377:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR374]], ptr %[[VAR376]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR375]], ptr %[[VAR377]]
 
   csll1 = csll + uc;
-  // CHECK-NEXT: %[[VAR378:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR379:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR378]]
-  // CHECK-NEXT: %[[VAR380:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR381:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR380]]
-  // CHECK-NEXT: %[[VAR382:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[UCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR378:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR379:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR378]]
+  // CHECK-NEXT: %[[VAR380:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR381:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR380]]
+  // CHECK-NEXT: %[[VAR382:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[UCADDR]], align [[CHALIGN]]
   // CHECK-NEXT: %[[VAR383:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR382]] to i[[LLSIZE]]
   // CHECK-NEXT: %[[VAR384:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR379]], %[[VAR383]]
   // CHECK-NEXT: %[[VAR385:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR381]], 0
-  // CHECK-NEXT: %[[VAR386:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR387:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR384]], i[[LLSIZE]]* %[[VAR386]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR385]], i[[LLSIZE]]* %[[VAR387]]
+  // CHECK-NEXT: %[[VAR386:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR387:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR384]], ptr %[[VAR386]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR385]], ptr %[[VAR387]]
 
   csll1 = csll + sll;
-  // CHECK-NEXT: %[[VAR388:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR389:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR388]]
-  // CHECK-NEXT: %[[VAR390:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR391:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR390]]
-  // CHECK-NEXT: %[[VAR392:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[SLLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR388:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR389:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR388]]
+  // CHECK-NEXT: %[[VAR390:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR391:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR390]]
+  // CHECK-NEXT: %[[VAR392:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[SLLADDR]], align [[LLALIGN]]
   // CHECK-NEXT: %[[VAR393:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR389]], %[[VAR392]]
   // CHECK-NEXT: %[[VAR394:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR391]], 0
-  // CHECK-NEXT: %[[VAR395:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR396:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR393]], i[[LLSIZE]]* %[[VAR395]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR394]], i[[LLSIZE]]* %[[VAR396]]
+  // CHECK-NEXT: %[[VAR395:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR396:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR393]], ptr %[[VAR395]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR394]], ptr %[[VAR396]]
 
   csll1 = csll + ull;
-  // CHECK-NEXT: %[[VAR397:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR398:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR397]]
-  // CHECK-NEXT: %[[VAR399:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR400:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR399]]
-  // CHECK-NEXT: %[[VAR401:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[ULLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR397:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR398:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR397]]
+  // CHECK-NEXT: %[[VAR399:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR400:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR399]]
+  // CHECK-NEXT: %[[VAR401:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[ULLADDR]], align [[LLALIGN]]
   // CHECK-NEXT: %[[VAR402:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR398]], %[[VAR401]]
   // CHECK-NEXT: %[[VAR403:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR400]], 0
-  // CHECK-NEXT: %[[VAR404:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR405:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR402]], i[[LLSIZE]]* %[[VAR404]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR403]], i[[LLSIZE]]* %[[VAR405]]
+  // CHECK-NEXT: %[[VAR404:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR405:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR402]], ptr %[[VAR404]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR403]], ptr %[[VAR405]]
   
   csll1 = cull + sc;
-  // CHECK-NEXT: %[[VAR406:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR407:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR406]]
-  // CHECK-NEXT: %[[VAR408:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR409:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR408]]
-  // CHECK-NEXT: %[[VAR410:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[SCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR406:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR407:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR406]]
+  // CHECK-NEXT: %[[VAR408:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR409:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR408]]
+  // CHECK-NEXT: %[[VAR410:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[SCADDR]], align [[CHALIGN]]
   // CHECK-NEXT: %[[VAR411:[A-Za-z0-9.]+]] = sext i[[CHSIZE]] %[[VAR410]] to i[[LLSIZE]]
   // CHECK-NEXT: %[[VAR412:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR407]], %[[VAR411]]
   // CHECK-NEXT: %[[VAR413:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR409]], 0
-  // CHECK-NEXT: %[[VAR414:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR415:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR412]], i[[LLSIZE]]* %[[VAR414]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR413]], i[[LLSIZE]]* %[[VAR415]]
+  // CHECK-NEXT: %[[VAR414:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR415:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR412]], ptr %[[VAR414]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR413]], ptr %[[VAR415]]
 
   cull1 = cull + uc;
-  // CHECK-NEXT: %[[VAR416:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR417:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR416]]
-  // CHECK-NEXT: %[[VAR418:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR419:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR418]]
-  // CHECK-NEXT: %[[VAR420:[A-Za-z0-9.]+]] = load i[[CHSIZE]], i[[CHSIZE]]* %[[UCADDR]], align [[CHALIGN]]
+  // CHECK-NEXT: %[[VAR416:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR417:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR416]]
+  // CHECK-NEXT: %[[VAR418:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR419:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR418]]
+  // CHECK-NEXT: %[[VAR420:[A-Za-z0-9.]+]] = load i[[CHSIZE]], ptr %[[UCADDR]], align [[CHALIGN]]
   // CHECK-NEXT: %[[VAR421:[A-Za-z0-9.]+]] = zext i[[CHSIZE]] %[[VAR420]] to i[[LLSIZE]]
   // CHECK-NEXT: %[[VAR422:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR417]], %[[VAR421]]
   // CHECK-NEXT: %[[VAR423:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR419]], 0
-  // CHECK-NEXT: %[[VAR424:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR425:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR422]], i[[LLSIZE]]* %[[VAR424]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR423]], i[[LLSIZE]]* %[[VAR425]]
+  // CHECK-NEXT: %[[VAR424:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR425:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR422]], ptr %[[VAR424]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR423]], ptr %[[VAR425]]
 
   csll1 = cull + sll;
-  // CHECK-NEXT: %[[VAR426:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR427:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR426]]
-  // CHECK-NEXT: %[[VAR428:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR429:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR428]]
-  // CHECK-NEXT: %[[VAR430:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[SLLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR426:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR427:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR426]]
+  // CHECK-NEXT: %[[VAR428:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR429:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR428]]
+  // CHECK-NEXT: %[[VAR430:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[SLLADDR]], align [[LLALIGN]]
   // CHECK-NEXT: %[[VAR431:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR427]], %[[VAR430]]
   // CHECK-NEXT: %[[VAR432:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR429]], 0
-  // CHECK-NEXT: %[[VAR433:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR434:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR431]], i[[LLSIZE]]* %[[VAR433]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR432]], i[[LLSIZE]]* %[[VAR434]]
+  // CHECK-NEXT: %[[VAR433:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR434:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CSLL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR431]], ptr %[[VAR433]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR432]], ptr %[[VAR434]]
 
   cull1 = cull + ull;
-  // CHECK-NEXT: %[[VAR435:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR436:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR435]]
-  // CHECK-NEXT: %[[VAR437:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: %[[VAR438:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[VAR437]]
-  // CHECK-NEXT: %[[VAR439:[A-Za-z0-9.]+]] = load i[[LLSIZE]], i[[LLSIZE]]* %[[ULLADDR]], align [[LLALIGN]]
+  // CHECK-NEXT: %[[VAR435:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR436:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR435]]
+  // CHECK-NEXT: %[[VAR437:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: %[[VAR438:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[VAR437]]
+  // CHECK-NEXT: %[[VAR439:[A-Za-z0-9.]+]] = load i[[LLSIZE]], ptr %[[ULLADDR]], align [[LLALIGN]]
   // CHECK-NEXT: %[[VAR440:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR436]], %[[VAR439]]
   // CHECK-NEXT: %[[VAR441:[A-Za-z0-9.]+]] = add i[[LLSIZE]] %[[VAR438]], 0
-  // CHECK-NEXT: %[[VAR442:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // CHECK-NEXT: %[[VAR443:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, { i[[LLSIZE]], i[[LLSIZE]] }* %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR440]], i[[LLSIZE]]* %[[VAR442]]
-  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR441]], i[[LLSIZE]]* %[[VAR443]]
+  // CHECK-NEXT: %[[VAR442:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
+  // CHECK-NEXT: %[[VAR443:[A-Za-z0-9.]+]] = getelementptr inbounds { i[[LLSIZE]], i[[LLSIZE]] }, ptr %[[CULL1]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR440]], ptr %[[VAR442]]
+  // CHECK-NEXT: store i[[LLSIZE]] %[[VAR441]], ptr %[[VAR443]]
 }
 
 // This code used to cause a crash; test that it no longer does so.

diff  --git a/clang/test/CodeGen/instrument-objc-method.m b/clang/test/CodeGen/instrument-objc-method.m
index ea7a3cfa82a52..82457beb94a32 100644
--- a/clang/test/CodeGen/instrument-objc-method.m
+++ b/clang/test/CodeGen/instrument-objc-method.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -S -triple x86_64-apple-darwin10 -debug-info-kind=standalone -emit-llvm -o - %s -finstrument-functions | FileCheck %s
-// RUN: %clang_cc1 -no-opaque-pointers -S -triple x86_64-apple-darwin10 -debug-info-kind=standalone -emit-llvm -o - %s -finstrument-function-entry-bare | FileCheck -check-prefix=BARE %s
+// RUN: %clang_cc1 -S -triple x86_64-apple-darwin10 -debug-info-kind=standalone -emit-llvm -o - %s -finstrument-functions | FileCheck %s
+// RUN: %clang_cc1 -S -triple x86_64-apple-darwin10 -debug-info-kind=standalone -emit-llvm -o - %s -finstrument-function-entry-bare | FileCheck -check-prefix=BARE %s
 
 @interface ObjCClass
 @end
@@ -28,7 +28,7 @@ + (void)load __attribute__((no_instrument_function)) {
 - (void)dealloc __attribute__((no_instrument_function)) {
 }
 
-// CHECK: declare void @__cyg_profile_func_enter(i8*, i8*)
-// CHECK: declare void @__cyg_profile_func_exit(i8*, i8*)
+// CHECK: declare void @__cyg_profile_func_enter(ptr, ptr)
+// CHECK: declare void @__cyg_profile_func_exit(ptr, ptr)
 // BARE: declare void @__cyg_profile_func_enter_bare
 @end

diff  --git a/clang/test/CodeGen/named_reg_global.c b/clang/test/CodeGen/named_reg_global.c
index 0fbae7adf7e6e..eab98797feb17 100644
--- a/clang/test/CodeGen/named_reg_global.c
+++ b/clang/test/CodeGen/named_reg_global.c
@@ -1,6 +1,6 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK-X86-64
-// RUN: %clang_cc1 -no-opaque-pointers -triple arm64-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK-ARM
-// RUN: %clang_cc1 -no-opaque-pointers -triple armv7-linux-gnu -target-abi apcs-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK-ARM
+// RUN: %clang_cc1 -triple x86_64-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK-X86-64
+// RUN: %clang_cc1 -triple arm64-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK-ARM
+// RUN: %clang_cc1 -triple armv7-linux-gnu -target-abi apcs-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK-ARM
 
 // CHECK-NOT: @sp = common global
 
@@ -31,7 +31,7 @@ unsigned long get_stack_pointer_addr(void) {
 // CHECK: declare{{.*}} i[[bits]] @llvm.read_register.i[[bits]](metadata)
 
 // CHECK: define{{.*}} void @set_stack_pointer_addr(i[[bits]] noundef %addr) #0 {
-// CHECK: [[sto:%[0-9]+]] = load i[[bits]], i[[bits]]* %
+// CHECK: [[sto:%[0-9]+]] = load i[[bits]], ptr %
 // CHECK: call void @llvm.write_register.i[[bits]](metadata !0, i[[bits]] [[sto]])
 // CHECK: ret void
 void set_stack_pointer_addr(unsigned long addr) {
@@ -44,13 +44,13 @@ int fn1(void) {
   return (*p4TH).word.len;
 }
 // CHECK: %[[regr:[0-9]+]] = call i[[bits]] @llvm.read_register.i[[bits]](metadata !0)
-// CHECK: inttoptr i[[bits]] %[[regr]] to %struct.p4_Thread*
+// CHECK: inttoptr i[[bits]] %[[regr]] to ptr
 
 // CHECK: define {{.*}}@fn2
 void fn2(struct p4_Thread *val) {
   p4TH = val;
 }
-// CHECK: %[[regw:[0-9]+]] = ptrtoint %struct.p4_Thread* %{{.*}} to i[[bits]]
+// CHECK: %[[regw:[0-9]+]] = ptrtoint ptr %{{.*}} to i[[bits]]
 // CHECK: call void @llvm.write_register.i[[bits]](metadata !0, i[[bits]] %[[regw]])
 
 // CHECK-X86-64: !llvm.named.register.rsp = !{!0}

diff  --git a/clang/test/CodeGen/sanitize-thread-no-checking-at-run-time.m b/clang/test/CodeGen/sanitize-thread-no-checking-at-run-time.m
index 28312c674e937..ec3726bb18c53 100644
--- a/clang/test/CodeGen/sanitize-thread-no-checking-at-run-time.m
+++ b/clang/test/CodeGen/sanitize-thread-no-checking-at-run-time.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin -x objective-c++ -fblocks -emit-llvm -o - %s | FileCheck -check-prefix=WITHOUT %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin -x objective-c++ -fblocks -emit-llvm -o - %s -fsanitize=thread | FileCheck -check-prefix=TSAN %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin -x objective-c++ -fblocks -emit-llvm -o - %s | FileCheck -check-prefix=WITHOUT %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin -x objective-c++ -fblocks -emit-llvm -o - %s -fsanitize=thread | FileCheck -check-prefix=TSAN %s
 
 // WITHOUT-NOT: "sanitize_thread_no_checking_at_run_time"
 
@@ -35,7 +35,7 @@ - (void)dealloc {
 void test2(id x) {
   extern void test2_helper(id (^)(void));
   test2_helper(^{ return x; });
-// TSAN: define linkonce_odr hidden void @__destroy_helper_block_8_32o(i8* noundef %0) unnamed_addr [[ATTR:#[0-9]+]]
+// TSAN: define linkonce_odr hidden void @__destroy_helper_block_8_32o(ptr noundef %0) unnamed_addr [[ATTR:#[0-9]+]]
 }
 
 // TSAN: attributes [[ATTR]] = { noinline nounwind {{.*}} "sanitize_thread_no_checking_at_run_time" {{.*}} }

diff  --git a/clang/test/CodeGen/thinlto-opaque-typed-mix.ll b/clang/test/CodeGen/thinlto-opaque-typed-mix.ll
index 1cd301f290e9b..c146e1a0128b8 100644
--- a/clang/test/CodeGen/thinlto-opaque-typed-mix.ll
+++ b/clang/test/CodeGen/thinlto-opaque-typed-mix.ll
@@ -15,9 +15,9 @@
 target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
 target triple = "x86_64--"
 
-declare i8* @f2()
+declare ptr @f2()
 
 define i32 @main() {
-  call i8* @f2()
+  call ptr @f2()
   ret i32 0
 }

diff  --git a/clang/test/CodeGen/ubsan-pointer-overflow.m b/clang/test/CodeGen/ubsan-pointer-overflow.m
index d3da6942bd72e..9192598da92fc 100644
--- a/clang/test/CodeGen/ubsan-pointer-overflow.m
+++ b/clang/test/CodeGen/ubsan-pointer-overflow.m
@@ -1,17 +1,17 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -w -emit-llvm -o - %s -fsanitize=pointer-overflow | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -w -emit-llvm -o - %s -fsanitize=pointer-overflow | FileCheck %s
 
 // CHECK-LABEL: define{{.*}} void @variable_len_array_arith
 void variable_len_array_arith(int n, int k) {
   int vla[n];
   int (*p)[n] = &vla;
 
-  // CHECK: getelementptr inbounds i32, i32* {{.*}}, i64 [[INC:%.*]]
+  // CHECK: getelementptr inbounds i32, ptr {{.*}}, i64 [[INC:%.*]]
   // CHECK: @llvm.smul.with.overflow.i64(i64 4, i64 [[INC]]), !nosanitize
   // CHECK-NOT: select
   // CHECK: call void @__ubsan_handle_pointer_overflow{{.*}}
   ++p;
 
-  // CHECK: getelementptr inbounds i32, i32* {{.*}}, i64 [[IDXPROM:%.*]]
+  // CHECK: getelementptr inbounds i32, ptr {{.*}}, i64 [[IDXPROM:%.*]]
   // CHECK: @llvm.smul.with.overflow.i64(i64 4, i64 [[IDXPROM]]), !nosanitize
   // CHECK: select
   // CHECK: call void @__ubsan_handle_pointer_overflow{{.*}}

diff  --git a/clang/test/CodeGenCUDA/Inputs/device-code.ll b/clang/test/CodeGenCUDA/Inputs/device-code.ll
index 5943a000c1d8d..f63a5a0bbf10d 100644
--- a/clang/test/CodeGenCUDA/Inputs/device-code.ll
+++ b/clang/test/CodeGenCUDA/Inputs/device-code.ll
@@ -4,7 +4,7 @@
 
 target triple = "nvptx-unknown-cuda"
 
-declare i32 @__nvvm_reflect(i8*)
+declare i32 @__nvvm_reflect(ptr)
 
 @"$str" = private addrspace(1) constant [8 x i8] c"USE_MUL\00"
 
@@ -17,7 +17,7 @@ define void @used_subfunc(float %a) {
 }
 
 define float @_Z17device_mul_or_addff(float %a, float %b) {
-  %reflect = call i32 @__nvvm_reflect(i8* addrspacecast (i8 addrspace(1)* getelementptr inbounds ([8 x i8], [8 x i8] addrspace(1)* @"$str", i32 0, i32 0) to i8*))
+  %reflect = call i32 @__nvvm_reflect(ptr addrspacecast (ptr addrspace(1) @"$str" to ptr))
   %cmp = icmp ne i32 %reflect, 0
   br i1 %cmp, label %use_mul, label %use_add
 

diff  --git a/clang/test/CodeGenCUDA/cxx-call-kernel.cpp b/clang/test/CodeGenCUDA/cxx-call-kernel.cpp
index def6dbabfc814..653e03c153cbe 100644
--- a/clang/test/CodeGenCUDA/cxx-call-kernel.cpp
+++ b/clang/test/CodeGenCUDA/cxx-call-kernel.cpp
@@ -1,13 +1,13 @@
 // REQUIRES: x86-registered-target
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-linux-gnu \
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu \
 // RUN:   -x hip -emit-llvm-bc %s -o %t.hip.bc
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-linux-gnu \
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu \
 // RUN:   -mlink-bitcode-file %t.hip.bc -DHIP_PLATFORM -emit-llvm \
 // RUN:   %s -o - | FileCheck %s
 
 #include "Inputs/cuda.h"
 
-// CHECK: @_Z2g1i = constant void (i32)* @_Z17__device_stub__g1i, align 8
+// CHECK: @_Z2g1i = constant ptr @_Z17__device_stub__g1i, align 8
 #if __HIP__
 __global__ void g1(int x) {}
 #else

diff  --git a/clang/test/CodeGenCXX/amdgcn-automatic-variable.cpp b/clang/test/CodeGenCXX/amdgcn-automatic-variable.cpp
index a3108e48b6887..c1f9310141000 100644
--- a/clang/test/CodeGenCXX/amdgcn-automatic-variable.cpp
+++ b/clang/test/CodeGenCXX/amdgcn-automatic-variable.cpp
@@ -1,13 +1,13 @@
 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
-// RUN: %clang_cc1 -no-opaque-pointers -O0 -triple amdgcn---amdgiz -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -O0 -triple amdgcn---amdgiz -emit-llvm %s -o - | FileCheck %s
 
 // CHECK-LABEL: @_Z5func1Pi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[X_ADDR:%.*]] = alloca i32*, align 8, addrspace(5)
-// CHECK-NEXT:    [[X_ADDR_ASCAST:%.*]] = addrspacecast i32* addrspace(5)* [[X_ADDR]] to i32**
-// CHECK-NEXT:    store i32* [[X:%.*]], i32** [[X_ADDR_ASCAST]], align 8
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32*, i32** [[X_ADDR_ASCAST]], align 8
-// CHECK-NEXT:    store i32 1, i32* [[TMP0]], align 4
+// CHECK-NEXT:    [[X_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
+// CHECK-NEXT:    [[X_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[X_ADDR]] to ptr
+// CHECK-NEXT:    store ptr [[X:%.*]], ptr [[X_ADDR_ASCAST]], align 8
+// CHECK-NEXT:    [[TMP0:%.*]] = load ptr, ptr [[X_ADDR_ASCAST]], align 8
+// CHECK-NEXT:    store i32 1, ptr [[TMP0]], align 4
 // CHECK-NEXT:    ret void
 //
 void func1(int *x) {
@@ -19,25 +19,25 @@ void func1(int *x) {
 // CHECK-NEXT:    [[LV1:%.*]] = alloca i32, align 4, addrspace(5)
 // CHECK-NEXT:    [[LV2:%.*]] = alloca i32, align 4, addrspace(5)
 // CHECK-NEXT:    [[LA:%.*]] = alloca [100 x i32], align 4, addrspace(5)
-// CHECK-NEXT:    [[LP1:%.*]] = alloca i32*, align 8, addrspace(5)
-// CHECK-NEXT:    [[LP2:%.*]] = alloca i32*, align 8, addrspace(5)
+// CHECK-NEXT:    [[LP1:%.*]] = alloca ptr, align 8, addrspace(5)
+// CHECK-NEXT:    [[LP2:%.*]] = alloca ptr, align 8, addrspace(5)
 // CHECK-NEXT:    [[LVC:%.*]] = alloca i32, align 4, addrspace(5)
-// CHECK-NEXT:    [[LV1_ASCAST:%.*]] = addrspacecast i32 addrspace(5)* [[LV1]] to i32*
-// CHECK-NEXT:    [[LV2_ASCAST:%.*]] = addrspacecast i32 addrspace(5)* [[LV2]] to i32*
-// CHECK-NEXT:    [[LA_ASCAST:%.*]] = addrspacecast [100 x i32] addrspace(5)* [[LA]] to [100 x i32]*
-// CHECK-NEXT:    [[LP1_ASCAST:%.*]] = addrspacecast i32* addrspace(5)* [[LP1]] to i32**
-// CHECK-NEXT:    [[LP2_ASCAST:%.*]] = addrspacecast i32* addrspace(5)* [[LP2]] to i32**
-// CHECK-NEXT:    [[LVC_ASCAST:%.*]] = addrspacecast i32 addrspace(5)* [[LVC]] to i32*
-// CHECK-NEXT:    store i32 1, i32* [[LV1_ASCAST]], align 4
-// CHECK-NEXT:    store i32 2, i32* [[LV2_ASCAST]], align 4
-// CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds [100 x i32], [100 x i32]* [[LA_ASCAST]], i64 0, i64 0
-// CHECK-NEXT:    store i32 3, i32* [[ARRAYIDX]], align 4
-// CHECK-NEXT:    store i32* [[LV1_ASCAST]], i32** [[LP1_ASCAST]], align 8
-// CHECK-NEXT:    [[ARRAYDECAY:%.*]] = getelementptr inbounds [100 x i32], [100 x i32]* [[LA_ASCAST]], i64 0, i64 0
-// CHECK-NEXT:    store i32* [[ARRAYDECAY]], i32** [[LP2_ASCAST]], align 8
-// CHECK-NEXT:    call void @_Z5func1Pi(i32* noundef [[LV1_ASCAST]])
-// CHECK-NEXT:    store i32 4, i32* [[LVC_ASCAST]], align 4
-// CHECK-NEXT:    store i32 4, i32* [[LV1_ASCAST]], align 4
+// CHECK-NEXT:    [[LV1_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[LV1]] to ptr
+// CHECK-NEXT:    [[LV2_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[LV2]] to ptr
+// CHECK-NEXT:    [[LA_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[LA]] to ptr
+// CHECK-NEXT:    [[LP1_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[LP1]] to ptr
+// CHECK-NEXT:    [[LP2_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[LP2]] to ptr
+// CHECK-NEXT:    [[LVC_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[LVC]] to ptr
+// CHECK-NEXT:    store i32 1, ptr [[LV1_ASCAST]], align 4
+// CHECK-NEXT:    store i32 2, ptr [[LV2_ASCAST]], align 4
+// CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds [100 x i32], ptr [[LA_ASCAST]], i64 0, i64 0
+// CHECK-NEXT:    store i32 3, ptr [[ARRAYIDX]], align 4
+// CHECK-NEXT:    store ptr [[LV1_ASCAST]], ptr [[LP1_ASCAST]], align 8
+// CHECK-NEXT:    [[ARRAYDECAY:%.*]] = getelementptr inbounds [100 x i32], ptr [[LA_ASCAST]], i64 0, i64 0
+// CHECK-NEXT:    store ptr [[ARRAYDECAY]], ptr [[LP2_ASCAST]], align 8
+// CHECK-NEXT:    call void @_Z5func1Pi(ptr noundef [[LV1_ASCAST]])
+// CHECK-NEXT:    store i32 4, ptr [[LVC_ASCAST]], align 4
+// CHECK-NEXT:    store i32 4, ptr [[LV1_ASCAST]], align 4
 // CHECK-NEXT:    ret void
 //
 void func2(void) {
@@ -73,9 +73,9 @@ int x;
 // CHECK-LABEL: @_Z5func3v(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[A:%.*]] = alloca [[CLASS_A:%.*]], align 4, addrspace(5)
-// CHECK-NEXT:    [[A_ASCAST:%.*]] = addrspacecast [[CLASS_A]] addrspace(5)* [[A]] to %class.A*
-// CHECK-NEXT:    call void @_ZN1AC1Ev(%class.A* noundef nonnull align 4 dereferenceable(4) [[A_ASCAST]])
-// CHECK-NEXT:    call void @_ZN1AD1Ev(%class.A* noundef nonnull align 4 dereferenceable(4) [[A_ASCAST]])
+// CHECK-NEXT:    [[A_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[A]] to ptr
+// CHECK-NEXT:    call void @_ZN1AC1Ev(ptr noundef nonnull align 4 dereferenceable(4) [[A_ASCAST]])
+// CHECK-NEXT:    call void @_ZN1AD1Ev(ptr noundef nonnull align 4 dereferenceable(4) [[A_ASCAST]])
 // CHECK-NEXT:    ret void
 //
 void func3() {
@@ -85,9 +85,9 @@ void func3() {
 // CHECK-LABEL: @_Z5func4i(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[X_ADDR:%.*]] = alloca i32, align 4, addrspace(5)
-// CHECK-NEXT:    [[X_ADDR_ASCAST:%.*]] = addrspacecast i32 addrspace(5)* [[X_ADDR]] to i32*
-// CHECK-NEXT:    store i32 [[X:%.*]], i32* [[X_ADDR_ASCAST]], align 4
-// CHECK-NEXT:    call void @_Z5func1Pi(i32* noundef [[X_ADDR_ASCAST]])
+// CHECK-NEXT:    [[X_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[X_ADDR]] to ptr
+// CHECK-NEXT:    store i32 [[X:%.*]], ptr [[X_ADDR_ASCAST]], align 4
+// CHECK-NEXT:    call void @_Z5func1Pi(ptr noundef [[X_ADDR_ASCAST]])
 // CHECK-NEXT:    ret void
 //
 void func4(int x) {
@@ -97,7 +97,7 @@ void func4(int x) {
 // CHECK-LABEL: @_Z5func5v(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[X:%.*]] = alloca i32, align 4, addrspace(5)
-// CHECK-NEXT:    [[X_ASCAST:%.*]] = addrspacecast i32 addrspace(5)* [[X]] to i32*
+// CHECK-NEXT:    [[X_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[X]] to ptr
 // CHECK-NEXT:    ret void
 //
 void func5() {
@@ -108,7 +108,7 @@ void func5() {
 // CHECK-LABEL: @_Z5func6v(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[X:%.*]] = alloca i32, align 4, addrspace(5)
-// CHECK-NEXT:    [[X_ASCAST:%.*]] = addrspacecast i32 addrspace(5)* [[X]] to i32*
+// CHECK-NEXT:    [[X_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[X]] to ptr
 // CHECK-NEXT:    ret void
 //
 void func6() {
@@ -120,10 +120,10 @@ extern void use(int *);
 // CHECK-LABEL: @_Z5func7v(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[X:%.*]] = alloca i32, align 4, addrspace(5)
-// CHECK-NEXT:    [[X_ASCAST:%.*]] = addrspacecast i32 addrspace(5)* [[X]] to i32*
+// CHECK-NEXT:    [[X_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[X]] to ptr
 // CHECK-NEXT:    br label [[LATER:%.*]]
 // CHECK:       later:
-// CHECK-NEXT:    call void @_Z3usePi(i32* noundef [[X_ASCAST]])
+// CHECK-NEXT:    call void @_Z3usePi(ptr noundef [[X_ASCAST]])
 // CHECK-NEXT:    ret void
 //
 void func7() {

diff  --git a/clang/test/CodeGenCXX/amdgcn-func-arg.cpp b/clang/test/CodeGenCXX/amdgcn-func-arg.cpp
index e8b0975b1f521..f6b75d589dd84 100644
--- a/clang/test/CodeGenCXX/amdgcn-func-arg.cpp
+++ b/clang/test/CodeGenCXX/amdgcn-func-arg.cpp
@@ -1,5 +1,5 @@
 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
-// RUN: %clang_cc1 -no-opaque-pointers -O0 -triple amdgcn -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -O0 -triple amdgcn -emit-llvm %s -o - | FileCheck %s
 
 class A {
 public:
@@ -20,10 +20,10 @@ void func_with_ref_arg(B &b);
 
 // CHECK-LABEL: @_Z22func_with_indirect_arg1A(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[P:%.*]] = alloca %class.A*, align 8, addrspace(5)
-// CHECK-NEXT:    [[P_ASCAST:%.*]] = addrspacecast %class.A* addrspace(5)* [[P]] to %class.A**
-// CHECK-NEXT:    [[A_ASCAST:%.*]] = addrspacecast [[CLASS_A:%.*]] addrspace(5)* [[A:%.*]] to %class.A*
-// CHECK-NEXT:    store %class.A* [[A_ASCAST]], %class.A** [[P_ASCAST]], align 8
+// CHECK-NEXT:    [[P:%.*]] = alloca ptr, align 8, addrspace(5)
+// CHECK-NEXT:    [[P_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[P]] to ptr
+// CHECK-NEXT:    [[A_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[A:%.*]] to ptr
+// CHECK-NEXT:    store ptr [[A_ASCAST]], ptr [[P_ASCAST]], align 8
 // CHECK-NEXT:    ret void
 //
 void func_with_indirect_arg(A a) {
@@ -34,17 +34,15 @@ void func_with_indirect_arg(A a) {
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[A:%.*]] = alloca [[CLASS_A:%.*]], align 4, addrspace(5)
 // CHECK-NEXT:    [[AGG_TMP:%.*]] = alloca [[CLASS_A]], align 4, addrspace(5)
-// CHECK-NEXT:    [[A_ASCAST:%.*]] = addrspacecast [[CLASS_A]] addrspace(5)* [[A]] to %class.A*
-// CHECK-NEXT:    [[AGG_TMP_ASCAST:%.*]] = addrspacecast [[CLASS_A]] addrspace(5)* [[AGG_TMP]] to %class.A*
-// CHECK-NEXT:    call void @_ZN1AC1Ev(%class.A* noundef nonnull align 4 dereferenceable(4) [[A_ASCAST]])
-// CHECK-NEXT:    [[TMP0:%.*]] = bitcast %class.A* [[AGG_TMP_ASCAST]] to i8*
-// CHECK-NEXT:    [[TMP1:%.*]] = bitcast %class.A* [[A_ASCAST]] to i8*
-// CHECK-NEXT:    call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[TMP0]], i8* align 4 [[TMP1]], i64 4, i1 false)
-// CHECK-NEXT:    [[AGG_TMP_ASCAST_ASCAST:%.*]] = addrspacecast %class.A* [[AGG_TMP_ASCAST]] to [[CLASS_A]] addrspace(5)*
-// CHECK-NEXT:    call void @_Z22func_with_indirect_arg1A([[CLASS_A]] addrspace(5)* noundef [[AGG_TMP_ASCAST_ASCAST]])
-// CHECK-NEXT:    call void @_ZN1AD1Ev(%class.A* noundef nonnull align 4 dereferenceable(4) [[AGG_TMP_ASCAST]])
-// CHECK-NEXT:    call void @_Z17func_with_ref_argR1A(%class.A* noundef nonnull align 4 dereferenceable(4) [[A_ASCAST]])
-// CHECK-NEXT:    call void @_ZN1AD1Ev(%class.A* noundef nonnull align 4 dereferenceable(4) [[A_ASCAST]])
+// CHECK-NEXT:    [[A_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[A]] to ptr
+// CHECK-NEXT:    [[AGG_TMP_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[AGG_TMP]] to ptr
+// CHECK-NEXT:    call void @_ZN1AC1Ev(ptr noundef nonnull align 4 dereferenceable(4) [[A_ASCAST]])
+// CHECK-NEXT:    call void @llvm.memcpy.p0.p0.i64(ptr align 4 [[AGG_TMP_ASCAST]], ptr align 4 [[A_ASCAST]], i64 4, i1 false)
+// CHECK-NEXT:    [[AGG_TMP_ASCAST_ASCAST:%.*]] = addrspacecast ptr [[AGG_TMP_ASCAST]] to ptr addrspace(5)
+// CHECK-NEXT:    call void @_Z22func_with_indirect_arg1A(ptr addrspace(5) noundef [[AGG_TMP_ASCAST_ASCAST]])
+// CHECK-NEXT:    call void @_ZN1AD1Ev(ptr noundef nonnull align 4 dereferenceable(4) [[AGG_TMP_ASCAST]])
+// CHECK-NEXT:    call void @_Z17func_with_ref_argR1A(ptr noundef nonnull align 4 dereferenceable(4) [[A_ASCAST]])
+// CHECK-NEXT:    call void @_ZN1AD1Ev(ptr noundef nonnull align 4 dereferenceable(4) [[A_ASCAST]])
 // CHECK-NEXT:    ret void
 //
 void test_indirect_arg_auto() {
@@ -56,13 +54,12 @@ void test_indirect_arg_auto() {
 // CHECK-LABEL: @_Z24test_indirect_arg_globalv(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[AGG_TMP:%.*]] = alloca [[CLASS_A:%.*]], align 4, addrspace(5)
-// CHECK-NEXT:    [[AGG_TMP_ASCAST:%.*]] = addrspacecast [[CLASS_A]] addrspace(5)* [[AGG_TMP]] to %class.A*
-// CHECK-NEXT:    [[TMP0:%.*]] = bitcast %class.A* [[AGG_TMP_ASCAST]] to i8*
-// CHECK-NEXT:    call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[TMP0]], i8* align 4 addrspacecast (i8 addrspace(1)* bitcast ([[CLASS_A]] addrspace(1)* @g_a to i8 addrspace(1)*) to i8*), i64 4, i1 false)
-// CHECK-NEXT:    [[AGG_TMP_ASCAST_ASCAST:%.*]] = addrspacecast %class.A* [[AGG_TMP_ASCAST]] to [[CLASS_A]] addrspace(5)*
-// CHECK-NEXT:    call void @_Z22func_with_indirect_arg1A([[CLASS_A]] addrspace(5)* noundef [[AGG_TMP_ASCAST_ASCAST]])
-// CHECK-NEXT:    call void @_ZN1AD1Ev(%class.A* noundef nonnull align 4 dereferenceable(4) [[AGG_TMP_ASCAST]])
-// CHECK-NEXT:    call void @_Z17func_with_ref_argR1A(%class.A* noundef nonnull align 4 dereferenceable(4) addrspacecast ([[CLASS_A]] addrspace(1)* @g_a to %class.A*))
+// CHECK-NEXT:    [[AGG_TMP_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[AGG_TMP]] to ptr
+// CHECK-NEXT:    call void @llvm.memcpy.p0.p0.i64(ptr align 4 [[AGG_TMP_ASCAST]], ptr align 4 addrspacecast (ptr addrspace(1) @g_a to ptr), i64 4, i1 false)
+// CHECK-NEXT:    [[AGG_TMP_ASCAST_ASCAST:%.*]] = addrspacecast ptr [[AGG_TMP_ASCAST]] to ptr addrspace(5)
+// CHECK-NEXT:    call void @_Z22func_with_indirect_arg1A(ptr addrspace(5) noundef [[AGG_TMP_ASCAST_ASCAST]])
+// CHECK-NEXT:    call void @_ZN1AD1Ev(ptr noundef nonnull align 4 dereferenceable(4) [[AGG_TMP_ASCAST]])
+// CHECK-NEXT:    call void @_Z17func_with_ref_argR1A(ptr noundef nonnull align 4 dereferenceable(4) addrspacecast (ptr addrspace(1) @g_a to ptr))
 // CHECK-NEXT:    ret void
 //
 void test_indirect_arg_global() {
@@ -72,10 +69,10 @@ void test_indirect_arg_global() {
 
 // CHECK-LABEL: @_Z19func_with_byval_arg1B(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[P:%.*]] = alloca %class.B*, align 8, addrspace(5)
-// CHECK-NEXT:    [[P_ASCAST:%.*]] = addrspacecast %class.B* addrspace(5)* [[P]] to %class.B**
-// CHECK-NEXT:    [[B_ASCAST:%.*]] = addrspacecast [[CLASS_B:%.*]] addrspace(5)* [[B:%.*]] to %class.B*
-// CHECK-NEXT:    store %class.B* [[B_ASCAST]], %class.B** [[P_ASCAST]], align 8
+// CHECK-NEXT:    [[P:%.*]] = alloca ptr, align 8, addrspace(5)
+// CHECK-NEXT:    [[P_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[P]] to ptr
+// CHECK-NEXT:    [[B_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[B:%.*]] to ptr
+// CHECK-NEXT:    store ptr [[B_ASCAST]], ptr [[P_ASCAST]], align 8
 // CHECK-NEXT:    ret void
 //
 void func_with_byval_arg(B b) {
@@ -86,14 +83,12 @@ void func_with_byval_arg(B b) {
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[B:%.*]] = alloca [[CLASS_B:%.*]], align 4, addrspace(5)
 // CHECK-NEXT:    [[AGG_TMP:%.*]] = alloca [[CLASS_B]], align 4, addrspace(5)
-// CHECK-NEXT:    [[B_ASCAST:%.*]] = addrspacecast [[CLASS_B]] addrspace(5)* [[B]] to %class.B*
-// CHECK-NEXT:    [[AGG_TMP_ASCAST:%.*]] = addrspacecast [[CLASS_B]] addrspace(5)* [[AGG_TMP]] to %class.B*
-// CHECK-NEXT:    [[TMP0:%.*]] = bitcast %class.B* [[AGG_TMP_ASCAST]] to i8*
-// CHECK-NEXT:    [[TMP1:%.*]] = bitcast %class.B* [[B_ASCAST]] to i8*
-// CHECK-NEXT:    call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[TMP0]], i8* align 4 [[TMP1]], i64 400, i1 false)
-// CHECK-NEXT:    [[AGG_TMP_ASCAST_ASCAST:%.*]] = addrspacecast %class.B* [[AGG_TMP_ASCAST]] to [[CLASS_B]] addrspace(5)*
-// CHECK-NEXT:    call void @_Z19func_with_byval_arg1B([[CLASS_B]] addrspace(5)* noundef byval([[CLASS_B]]) align 4 [[AGG_TMP_ASCAST_ASCAST]])
-// CHECK-NEXT:    call void @_Z17func_with_ref_argR1B(%class.B* noundef nonnull align 4 dereferenceable(400) [[B_ASCAST]])
+// CHECK-NEXT:    [[B_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[B]] to ptr
+// CHECK-NEXT:    [[AGG_TMP_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[AGG_TMP]] to ptr
+// CHECK-NEXT:    call void @llvm.memcpy.p0.p0.i64(ptr align 4 [[AGG_TMP_ASCAST]], ptr align 4 [[B_ASCAST]], i64 400, i1 false)
+// CHECK-NEXT:    [[AGG_TMP_ASCAST_ASCAST:%.*]] = addrspacecast ptr [[AGG_TMP_ASCAST]] to ptr addrspace(5)
+// CHECK-NEXT:    call void @_Z19func_with_byval_arg1B(ptr addrspace(5) noundef byval([[CLASS_B]]) align 4 [[AGG_TMP_ASCAST_ASCAST]])
+// CHECK-NEXT:    call void @_Z17func_with_ref_argR1B(ptr noundef nonnull align 4 dereferenceable(400) [[B_ASCAST]])
 // CHECK-NEXT:    ret void
 //
 void test_byval_arg_auto() {
@@ -105,12 +100,11 @@ void test_byval_arg_auto() {
 // CHECK-LABEL: @_Z21test_byval_arg_globalv(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[AGG_TMP:%.*]] = alloca [[CLASS_B:%.*]], align 4, addrspace(5)
-// CHECK-NEXT:    [[AGG_TMP_ASCAST:%.*]] = addrspacecast [[CLASS_B]] addrspace(5)* [[AGG_TMP]] to %class.B*
-// CHECK-NEXT:    [[TMP0:%.*]] = bitcast %class.B* [[AGG_TMP_ASCAST]] to i8*
-// CHECK-NEXT:    call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[TMP0]], i8* align 4 addrspacecast (i8 addrspace(1)* bitcast ([[CLASS_B]] addrspace(1)* @g_b to i8 addrspace(1)*) to i8*), i64 400, i1 false)
-// CHECK-NEXT:    [[AGG_TMP_ASCAST_ASCAST:%.*]] = addrspacecast %class.B* [[AGG_TMP_ASCAST]] to [[CLASS_B]] addrspace(5)*
-// CHECK-NEXT:    call void @_Z19func_with_byval_arg1B([[CLASS_B]] addrspace(5)* noundef byval([[CLASS_B]]) align 4 [[AGG_TMP_ASCAST_ASCAST]])
-// CHECK-NEXT:    call void @_Z17func_with_ref_argR1B(%class.B* noundef nonnull align 4 dereferenceable(400) addrspacecast ([[CLASS_B]] addrspace(1)* @g_b to %class.B*))
+// CHECK-NEXT:    [[AGG_TMP_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[AGG_TMP]] to ptr
+// CHECK-NEXT:    call void @llvm.memcpy.p0.p0.i64(ptr align 4 [[AGG_TMP_ASCAST]], ptr align 4 addrspacecast (ptr addrspace(1) @g_b to ptr), i64 400, i1 false)
+// CHECK-NEXT:    [[AGG_TMP_ASCAST_ASCAST:%.*]] = addrspacecast ptr [[AGG_TMP_ASCAST]] to ptr addrspace(5)
+// CHECK-NEXT:    call void @_Z19func_with_byval_arg1B(ptr addrspace(5) noundef byval([[CLASS_B]]) align 4 [[AGG_TMP_ASCAST_ASCAST]])
+// CHECK-NEXT:    call void @_Z17func_with_ref_argR1B(ptr noundef nonnull align 4 dereferenceable(400) addrspacecast (ptr addrspace(1) @g_b to ptr))
 // CHECK-NEXT:    ret void
 //
 void test_byval_arg_global() {

diff  --git a/clang/test/CodeGenCoroutines/coro-alloc-exp-namespace.cpp b/clang/test/CodeGenCoroutines/coro-alloc-exp-namespace.cpp
index a93a85adfbb37..9377fd9e0e8c7 100644
--- a/clang/test/CodeGenCoroutines/coro-alloc-exp-namespace.cpp
+++ b/clang/test/CodeGenCoroutines/coro-alloc-exp-namespace.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-linux-gnu -fcoroutines-ts -std=c++14 \
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -fcoroutines-ts -std=c++14 \
 // RUN:    -Wno-coroutine-missing-unhandled-exception -emit-llvm %s -o - -disable-llvm-passes \
 // RUN:   | FileCheck %s
 
@@ -60,19 +60,19 @@ extern "C" void f0(global_new_delete_tag) {
 
   // CHECK: [[AllocBB]]:
   // CHECK: %[[SIZE:.+]] = call i64 @llvm.coro.size.i64()
-  // CHECK: %[[MEM:.+]] = call noalias noundef nonnull i8* @_Znwm(i64 noundef %[[SIZE]])
+  // CHECK: %[[MEM:.+]] = call noalias noundef nonnull ptr @_Znwm(i64 noundef %[[SIZE]])
   // CHECK: br label %[[InitBB]]
 
   // CHECK: [[InitBB]]:
-  // CHECK: %[[PHI:.+]] = phi i8* [ null, %{{.+}} ], [ %call, %[[AllocBB]] ]
-  // CHECK: %[[FRAME:.+]] = call i8* @llvm.coro.begin(token %[[ID]], i8* %[[PHI]])
+  // CHECK: %[[PHI:.+]] = phi ptr [ null, %{{.+}} ], [ %call, %[[AllocBB]] ]
+  // CHECK: %[[FRAME:.+]] = call ptr @llvm.coro.begin(token %[[ID]], ptr %[[PHI]])
 
-  // CHECK: %[[MEM:.+]] = call i8* @llvm.coro.free(token %[[ID]], i8* %[[FRAME]])
-  // CHECK: %[[NeedDealloc:.+]] = icmp ne i8* %[[MEM]], null
+  // CHECK: %[[MEM:.+]] = call ptr @llvm.coro.free(token %[[ID]], ptr %[[FRAME]])
+  // CHECK: %[[NeedDealloc:.+]] = icmp ne ptr %[[MEM]], null
   // CHECK: br i1 %[[NeedDealloc]], label %[[FreeBB:.+]], label %[[Afterwards:.+]]
 
   // CHECK: [[FreeBB]]:
-  // CHECK: call void @_ZdlPv(i8* noundef %[[MEM]])
+  // CHECK: call void @_ZdlPv(ptr noundef %[[MEM]])
   // CHECK: br label %[[Afterwards]]
 
   // CHECK: [[Afterwards]]:
@@ -97,11 +97,11 @@ struct std::experimental::coroutine_traits<void, promise_new_tag> {
 extern "C" void f1(promise_new_tag) {
   // CHECK: %[[ID:.+]] = call token @llvm.coro.id(i32 16
   // CHECK: %[[SIZE:.+]] = call i64 @llvm.coro.size.i64()
-  // CHECK: call noundef i8* @_ZNSt12experimental16coroutine_traitsIJv15promise_new_tagEE12promise_typenwEm(i64 noundef %[[SIZE]])
+  // CHECK: call noundef ptr @_ZNSt12experimental16coroutine_traitsIJv15promise_new_tagEE12promise_typenwEm(i64 noundef %[[SIZE]])
 
-  // CHECK: %[[FRAME:.+]] = call i8* @llvm.coro.begin(
-  // CHECK: %[[MEM:.+]] = call i8* @llvm.coro.free(token %[[ID]], i8* %[[FRAME]])
-  // CHECK: call void @_ZdlPv(i8* noundef %[[MEM]])
+  // CHECK: %[[FRAME:.+]] = call ptr @llvm.coro.begin(
+  // CHECK: %[[MEM:.+]] = call ptr @llvm.coro.free(token %[[ID]], ptr %[[FRAME]])
+  // CHECK: call void @_ZdlPv(ptr noundef %[[MEM]])
   co_return;
 }
 
@@ -121,20 +121,20 @@ struct std::experimental::coroutine_traits<void, promise_matching_placement_new_
 
 // CHECK-LABEL: f1a(
 extern "C" void f1a(promise_matching_placement_new_tag, int x, float y, double z) {
-  // CHECK: store i32 %x, i32* %x.addr, align 4
-  // CHECK: store float %y, float* %y.addr, align 4
-  // CHECK: store double %z, double* %z.addr, align 8
+  // CHECK: store i32 %x, ptr %x.addr, align 4
+  // CHECK: store float %y, ptr %y.addr, align 4
+  // CHECK: store double %z, ptr %z.addr, align 8
   // CHECK: %[[ID:.+]] = call token @llvm.coro.id(i32 16
   // CHECK: %[[SIZE:.+]] = call i64 @llvm.coro.size.i64()
-  // CHECK: %[[INT:.+]] = load i32, i32* %x.addr, align 4
-  // CHECK: %[[FLOAT:.+]] = load float, float* %y.addr, align 4
-  // CHECK: %[[DOUBLE:.+]] = load double, double* %z.addr, align 8
-  // CHECK: call noundef i8* @_ZNSt12experimental16coroutine_traitsIJv34promise_matching_placement_new_tagifdEE12promise_typenwEmS1_ifd(i64 noundef %[[SIZE]], i32 noundef %[[INT]], float noundef %[[FLOAT]], double noundef %[[DOUBLE]])
+  // CHECK: %[[INT:.+]] = load i32, ptr %x.addr, align 4
+  // CHECK: %[[FLOAT:.+]] = load float, ptr %y.addr, align 4
+  // CHECK: %[[DOUBLE:.+]] = load double, ptr %z.addr, align 8
+  // CHECK: call noundef ptr @_ZNSt12experimental16coroutine_traitsIJv34promise_matching_placement_new_tagifdEE12promise_typenwEmS1_ifd(i64 noundef %[[SIZE]], i32 noundef %[[INT]], float noundef %[[FLOAT]], double noundef %[[DOUBLE]])
   co_return;
 }
 
 // Declare a placement form operator new, such as the one described in
-// C++ 18.6.1.3.1, which takes a void* argument.
+// C++ 18.6.1.3.1, which takes a ptr argument.
 void *operator new(SizeT __sz, void *__p) noexcept;
 
 struct promise_matching_global_placement_new_tag {};
@@ -155,7 +155,7 @@ struct std::experimental::coroutine_traits<void, promise_matching_global_placeme
 // within the scope of the promise type's class.
 // CHECK-LABEL: f1b(
 extern "C" void f1b(promise_matching_global_placement_new_tag, dummy *) {
-  // CHECK: call noalias noundef nonnull i8* @_Znwm(i64
+  // CHECK: call noalias noundef nonnull ptr @_Znwm(i64
   co_return;
 }
 
@@ -176,11 +176,11 @@ struct std::experimental::coroutine_traits<void, promise_delete_tag> {
 extern "C" void f2(promise_delete_tag) {
   // CHECK: %[[ID:.+]] = call token @llvm.coro.id(i32 16
   // CHECK: %[[SIZE:.+]] = call i64 @llvm.coro.size.i64()
-  // CHECK: call noalias noundef nonnull i8* @_Znwm(i64 noundef %[[SIZE]])
+  // CHECK: call noalias noundef nonnull ptr @_Znwm(i64 noundef %[[SIZE]])
 
-  // CHECK: %[[FRAME:.+]] = call i8* @llvm.coro.begin(
-  // CHECK: %[[MEM:.+]] = call i8* @llvm.coro.free(token %[[ID]], i8* %[[FRAME]])
-  // CHECK: call void @_ZNSt12experimental16coroutine_traitsIJv18promise_delete_tagEE12promise_typedlEPv(i8* noundef %[[MEM]])
+  // CHECK: %[[FRAME:.+]] = call ptr @llvm.coro.begin(
+  // CHECK: %[[MEM:.+]] = call ptr @llvm.coro.free(token %[[ID]], ptr %[[FRAME]])
+  // CHECK: call void @_ZNSt12experimental16coroutine_traitsIJv18promise_delete_tagEE12promise_typedlEPv(ptr noundef %[[MEM]])
   co_return;
 }
 
@@ -201,12 +201,12 @@ struct std::experimental::coroutine_traits<void, promise_sized_delete_tag> {
 extern "C" void f3(promise_sized_delete_tag) {
   // CHECK: %[[ID:.+]] = call token @llvm.coro.id(i32 16
   // CHECK: %[[SIZE:.+]] = call i64 @llvm.coro.size.i64()
-  // CHECK: call noalias noundef nonnull i8* @_Znwm(i64 noundef %[[SIZE]])
+  // CHECK: call noalias noundef nonnull ptr @_Znwm(i64 noundef %[[SIZE]])
 
-  // CHECK: %[[FRAME:.+]] = call i8* @llvm.coro.begin(
-  // CHECK: %[[MEM:.+]] = call i8* @llvm.coro.free(token %[[ID]], i8* %[[FRAME]])
+  // CHECK: %[[FRAME:.+]] = call ptr @llvm.coro.begin(
+  // CHECK: %[[MEM:.+]] = call ptr @llvm.coro.free(token %[[ID]], ptr %[[FRAME]])
   // CHECK: %[[SIZE2:.+]] = call i64 @llvm.coro.size.i64()
-  // CHECK: call void @_ZNSt12experimental16coroutine_traitsIJv24promise_sized_delete_tagEE12promise_typedlEPvm(i8* noundef %[[MEM]], i64 noundef %[[SIZE2]])
+  // CHECK: call void @_ZNSt12experimental16coroutine_traitsIJv24promise_sized_delete_tagEE12promise_typedlEPvm(ptr noundef %[[MEM]], i64 noundef %[[SIZE2]])
   co_return;
 }
 
@@ -228,20 +228,20 @@ extern "C" int f4(promise_on_alloc_failure_tag) {
   // CHECK: %[[RetVal:.+]] = alloca i32
   // CHECK: %[[ID:.+]] = call token @llvm.coro.id(i32 16
   // CHECK: %[[SIZE:.+]] = call i64 @llvm.coro.size.i64()
-  // CHECK: %[[MEM:.+]] = call noalias noundef i8* @_ZnwmRKSt9nothrow_t(i64 noundef %[[SIZE]], %"struct.std::nothrow_t"* noundef nonnull align 1 dereferenceable(1) @_ZStL7nothrow)
-  // CHECK: %[[OK:.+]] = icmp ne i8* %[[MEM]], null
+  // CHECK: %[[MEM:.+]] = call noalias noundef ptr @_ZnwmRKSt9nothrow_t(i64 noundef %[[SIZE]], ptr noundef nonnull align 1 dereferenceable(1) @_ZStL7nothrow)
+  // CHECK: %[[OK:.+]] = icmp ne ptr %[[MEM]], null
   // CHECK: br i1 %[[OK]], label %[[OKBB:.+]], label %[[ERRBB:.+]]
 
   // CHECK: [[ERRBB]]:
   // CHECK:   %[[FailRet:.+]] = call noundef i32 @_ZNSt12experimental16coroutine_traitsIJi28promise_on_alloc_failure_tagEE12promise_type39get_return_object_on_allocation_failureEv(
-  // CHECK:   store i32 %[[FailRet]], i32* %[[RetVal]]
+  // CHECK:   store i32 %[[FailRet]], ptr %[[RetVal]]
   // CHECK:   br label %[[RetBB:.+]]
 
   // CHECK: [[OKBB]]:
   // CHECK:   %[[OkRet:.+]] = call noundef i32 @_ZNSt12experimental16coroutine_traitsIJi28promise_on_alloc_failure_tagEE12promise_type17get_return_objectEv(
 
   // CHECK: [[RetBB]]:
-  // CHECK:   %[[LoadRet:.+]] = load i32, i32* %[[RetVal]], align 4
+  // CHECK:   %[[LoadRet:.+]] = load i32, ptr %[[RetVal]], align 4
   // CHECK:   ret i32 %[[LoadRet]]
   co_return;
 }

diff  --git a/clang/test/CodeGenCoroutines/coro-alloc.cpp b/clang/test/CodeGenCoroutines/coro-alloc.cpp
index c16af68664e72..05b3d56483d0e 100644
--- a/clang/test/CodeGenCoroutines/coro-alloc.cpp
+++ b/clang/test/CodeGenCoroutines/coro-alloc.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-linux-gnu -std=c++20 \
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -std=c++20 \
 // RUN:    -Wno-coroutine-missing-unhandled-exception -emit-llvm %s -o - -disable-llvm-passes \
 // RUN:   | FileCheck %s
 
@@ -58,19 +58,19 @@ extern "C" void f0(global_new_delete_tag) {
 
   // CHECK: [[AllocBB]]:
   // CHECK: %[[SIZE:.+]] = call i64 @llvm.coro.size.i64()
-  // CHECK: %[[MEM:.+]] = call noalias noundef nonnull i8* @_Znwm(i64 noundef %[[SIZE]])
+  // CHECK: %[[MEM:.+]] = call noalias noundef nonnull ptr @_Znwm(i64 noundef %[[SIZE]])
   // CHECK: br label %[[InitBB]]
 
   // CHECK: [[InitBB]]:
-  // CHECK: %[[PHI:.+]] = phi i8* [ null, %{{.+}} ], [ %call, %[[AllocBB]] ]
-  // CHECK: %[[FRAME:.+]] = call i8* @llvm.coro.begin(token %[[ID]], i8* %[[PHI]])
+  // CHECK: %[[PHI:.+]] = phi ptr [ null, %{{.+}} ], [ %call, %[[AllocBB]] ]
+  // CHECK: %[[FRAME:.+]] = call ptr @llvm.coro.begin(token %[[ID]], ptr %[[PHI]])
 
-  // CHECK: %[[MEM:.+]] = call i8* @llvm.coro.free(token %[[ID]], i8* %[[FRAME]])
-  // CHECK: %[[NeedDealloc:.+]] = icmp ne i8* %[[MEM]], null
+  // CHECK: %[[MEM:.+]] = call ptr @llvm.coro.free(token %[[ID]], ptr %[[FRAME]])
+  // CHECK: %[[NeedDealloc:.+]] = icmp ne ptr %[[MEM]], null
   // CHECK: br i1 %[[NeedDealloc]], label %[[FreeBB:.+]], label %[[Afterwards:.+]]
 
   // CHECK: [[FreeBB]]:
-  // CHECK: call void @_ZdlPv(i8* noundef %[[MEM]])
+  // CHECK: call void @_ZdlPv(ptr noundef %[[MEM]])
   // CHECK: br label %[[Afterwards]]
 
   // CHECK: [[Afterwards]]:
@@ -95,11 +95,11 @@ struct std::coroutine_traits<void, promise_new_tag> {
 extern "C" void f1(promise_new_tag ) {
   // CHECK: %[[ID:.+]] = call token @llvm.coro.id(i32 16
   // CHECK: %[[SIZE:.+]] = call i64 @llvm.coro.size.i64()
-  // CHECK: call noundef i8* @_ZNSt16coroutine_traitsIJv15promise_new_tagEE12promise_typenwEm(i64 noundef %[[SIZE]])
+  // CHECK: call noundef ptr @_ZNSt16coroutine_traitsIJv15promise_new_tagEE12promise_typenwEm(i64 noundef %[[SIZE]])
 
-  // CHECK: %[[FRAME:.+]] = call i8* @llvm.coro.begin(
-  // CHECK: %[[MEM:.+]] = call i8* @llvm.coro.free(token %[[ID]], i8* %[[FRAME]])
-  // CHECK: call void @_ZdlPv(i8* noundef %[[MEM]])
+  // CHECK: %[[FRAME:.+]] = call ptr @llvm.coro.begin(
+  // CHECK: %[[MEM:.+]] = call ptr @llvm.coro.free(token %[[ID]], ptr %[[FRAME]])
+  // CHECK: call void @_ZdlPv(ptr noundef %[[MEM]])
   co_return;
 }
 
@@ -119,20 +119,20 @@ struct std::coroutine_traits<void, promise_matching_placement_new_tag, int, floa
 
 // CHECK-LABEL: f1a(
 extern "C" void f1a(promise_matching_placement_new_tag, int x, float y , double z) {
-  // CHECK: store i32 %x, i32* %x.addr, align 4
-  // CHECK: store float %y, float* %y.addr, align 4
-  // CHECK: store double %z, double* %z.addr, align 8
+  // CHECK: store i32 %x, ptr %x.addr, align 4
+  // CHECK: store float %y, ptr %y.addr, align 4
+  // CHECK: store double %z, ptr %z.addr, align 8
   // CHECK: %[[ID:.+]] = call token @llvm.coro.id(i32 16
   // CHECK: %[[SIZE:.+]] = call i64 @llvm.coro.size.i64()
-  // CHECK: %[[INT:.+]] = load i32, i32* %x.addr, align 4
-  // CHECK: %[[FLOAT:.+]] = load float, float* %y.addr, align 4
-  // CHECK: %[[DOUBLE:.+]] = load double, double* %z.addr, align 8
-  // CHECK: call noundef i8* @_ZNSt16coroutine_traitsIJv34promise_matching_placement_new_tagifdEE12promise_typenwEmS0_ifd(i64 noundef %[[SIZE]], i32 noundef %[[INT]], float noundef %[[FLOAT]], double noundef %[[DOUBLE]])
+  // CHECK: %[[INT:.+]] = load i32, ptr %x.addr, align 4
+  // CHECK: %[[FLOAT:.+]] = load float, ptr %y.addr, align 4
+  // CHECK: %[[DOUBLE:.+]] = load double, ptr %z.addr, align 8
+  // CHECK: call noundef ptr @_ZNSt16coroutine_traitsIJv34promise_matching_placement_new_tagifdEE12promise_typenwEmS0_ifd(i64 noundef %[[SIZE]], i32 noundef %[[INT]], float noundef %[[FLOAT]], double noundef %[[DOUBLE]])
   co_return;
 }
 
 // Declare a placement form operator new, such as the one described in
-// C++ 18.6.1.3.1, which takes a void* argument.
+// C++ 18.6.1.3.1, which takes a ptr argument.
 void* operator new(SizeT __sz, void *__p) noexcept;
 
 struct promise_matching_global_placement_new_tag {};
@@ -153,7 +153,7 @@ struct std::coroutine_traits<void, promise_matching_global_placement_new_tag, du
 // within the scope of the promise type's class.
 // CHECK-LABEL: f1b(
 extern "C" void f1b(promise_matching_global_placement_new_tag, dummy *) {
-  // CHECK: call noalias noundef nonnull i8* @_Znwm(i64
+  // CHECK: call noalias noundef nonnull ptr @_Znwm(i64
   co_return;
 }
 
@@ -174,11 +174,11 @@ struct std::coroutine_traits<void, promise_delete_tag> {
 extern "C" void f2(promise_delete_tag) {
   // CHECK: %[[ID:.+]] = call token @llvm.coro.id(i32 16
   // CHECK: %[[SIZE:.+]] = call i64 @llvm.coro.size.i64()
-  // CHECK: call noalias noundef nonnull i8* @_Znwm(i64 noundef %[[SIZE]])
+  // CHECK: call noalias noundef nonnull ptr @_Znwm(i64 noundef %[[SIZE]])
 
-  // CHECK: %[[FRAME:.+]] = call i8* @llvm.coro.begin(
-  // CHECK: %[[MEM:.+]] = call i8* @llvm.coro.free(token %[[ID]], i8* %[[FRAME]])
-  // CHECK: call void @_ZNSt16coroutine_traitsIJv18promise_delete_tagEE12promise_typedlEPv(i8* noundef %[[MEM]])
+  // CHECK: %[[FRAME:.+]] = call ptr @llvm.coro.begin(
+  // CHECK: %[[MEM:.+]] = call ptr @llvm.coro.free(token %[[ID]], ptr %[[FRAME]])
+  // CHECK: call void @_ZNSt16coroutine_traitsIJv18promise_delete_tagEE12promise_typedlEPv(ptr noundef %[[MEM]])
   co_return;
 }
 
@@ -199,12 +199,12 @@ struct std::coroutine_traits<void, promise_sized_delete_tag> {
 extern "C" void f3(promise_sized_delete_tag) {
   // CHECK: %[[ID:.+]] = call token @llvm.coro.id(i32 16
   // CHECK: %[[SIZE:.+]] = call i64 @llvm.coro.size.i64()
-  // CHECK: call noalias noundef nonnull i8* @_Znwm(i64 noundef %[[SIZE]])
+  // CHECK: call noalias noundef nonnull ptr @_Znwm(i64 noundef %[[SIZE]])
 
-  // CHECK: %[[FRAME:.+]] = call i8* @llvm.coro.begin(
-  // CHECK: %[[MEM:.+]] = call i8* @llvm.coro.free(token %[[ID]], i8* %[[FRAME]])
+  // CHECK: %[[FRAME:.+]] = call ptr @llvm.coro.begin(
+  // CHECK: %[[MEM:.+]] = call ptr @llvm.coro.free(token %[[ID]], ptr %[[FRAME]])
   // CHECK: %[[SIZE2:.+]] = call i64 @llvm.coro.size.i64()
-  // CHECK: call void @_ZNSt16coroutine_traitsIJv24promise_sized_delete_tagEE12promise_typedlEPvm(i8* noundef %[[MEM]], i64 noundef %[[SIZE2]])
+  // CHECK: call void @_ZNSt16coroutine_traitsIJv24promise_sized_delete_tagEE12promise_typedlEPvm(ptr noundef %[[MEM]], i64 noundef %[[SIZE2]])
   co_return;
 }
 
@@ -226,20 +226,20 @@ extern "C" int f4(promise_on_alloc_failure_tag) {
   // CHECK: %[[RetVal:.+]] = alloca i32
   // CHECK: %[[ID:.+]] = call token @llvm.coro.id(i32 16
   // CHECK: %[[SIZE:.+]] = call i64 @llvm.coro.size.i64()
-  // CHECK: %[[MEM:.+]] = call noalias noundef i8* @_ZnwmRKSt9nothrow_t(i64 noundef %[[SIZE]], %"struct.std::nothrow_t"* noundef nonnull align 1 dereferenceable(1) @_ZStL7nothrow)
-  // CHECK: %[[OK:.+]] = icmp ne i8* %[[MEM]], null
+  // CHECK: %[[MEM:.+]] = call noalias noundef ptr @_ZnwmRKSt9nothrow_t(i64 noundef %[[SIZE]], ptr noundef nonnull align 1 dereferenceable(1) @_ZStL7nothrow)
+  // CHECK: %[[OK:.+]] = icmp ne ptr %[[MEM]], null
   // CHECK: br i1 %[[OK]], label %[[OKBB:.+]], label %[[ERRBB:.+]]
 
   // CHECK: [[ERRBB]]:
   // CHECK:   %[[FailRet:.+]] = call noundef i32 @_ZNSt16coroutine_traitsIJi28promise_on_alloc_failure_tagEE12promise_type39get_return_object_on_allocation_failureEv(
-  // CHECK:   store i32 %[[FailRet]], i32* %[[RetVal]]
+  // CHECK:   store i32 %[[FailRet]], ptr %[[RetVal]]
   // CHECK:   br label %[[RetBB:.+]]
 
   // CHECK: [[OKBB]]:
   // CHECK:   %[[OkRet:.+]] = call noundef i32 @_ZNSt16coroutine_traitsIJi28promise_on_alloc_failure_tagEE12promise_type17get_return_objectEv(
 
   // CHECK: [[RetBB]]:
-  // CHECK:   %[[LoadRet:.+]] = load i32, i32* %[[RetVal]], align 4
+  // CHECK:   %[[LoadRet:.+]] = load i32, ptr %[[RetVal]], align 4
   // CHECK:   ret i32 %[[LoadRet]]
   co_return;
 }
@@ -264,11 +264,11 @@ struct std::coroutine_traits<void, promise_sized_delete_tag2> {
 extern "C" void f5(promise_sized_delete_tag2) {
   // CHECK: %[[ID:.+]] = call token @llvm.coro.id(i32 16
   // CHECK: %[[SIZE:.+]] = call i64 @llvm.coro.size.i64()
-  // CHECK: call noalias noundef nonnull i8* @_Znwm(i64 noundef %[[SIZE]])
+  // CHECK: call noalias noundef nonnull ptr @_Znwm(i64 noundef %[[SIZE]])
 
-  // CHECK: %[[FRAME:.+]] = call i8* @llvm.coro.begin(
-  // CHECK: %[[MEM:.+]] = call i8* @llvm.coro.free(token %[[ID]], i8* %[[FRAME]])
+  // CHECK: %[[FRAME:.+]] = call ptr @llvm.coro.begin(
+  // CHECK: %[[MEM:.+]] = call ptr @llvm.coro.free(token %[[ID]], ptr %[[FRAME]])
   // CHECK: %[[SIZE2:.+]] = call i64 @llvm.coro.size.i64()
-  // CHECK: call void @_ZNSt16coroutine_traitsIJv25promise_sized_delete_tag2EE12promise_typedlEPvm(i8* noundef %[[MEM]], i64 noundef %[[SIZE2]])
+  // CHECK: call void @_ZNSt16coroutine_traitsIJv25promise_sized_delete_tag2EE12promise_typedlEPvm(ptr noundef %[[MEM]], i64 noundef %[[SIZE2]])
   co_return;
 }

diff  --git a/clang/test/CodeGenCoroutines/coro-await-exp-namespace.cpp b/clang/test/CodeGenCoroutines/coro-await-exp-namespace.cpp
index faa23dd5cde53..6efbaa3e01739 100644
--- a/clang/test/CodeGenCoroutines/coro-await-exp-namespace.cpp
+++ b/clang/test/CodeGenCoroutines/coro-await-exp-namespace.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -no-enable-noundef-analysis -triple x86_64-unknown-linux-gnu -fcoroutines-ts -std=c++14 \
+// RUN: %clang_cc1 -no-enable-noundef-analysis -triple x86_64-unknown-linux-gnu -fcoroutines-ts -std=c++14 \
 // RUN:   -emit-llvm %s -o - -disable-llvm-passes -Wno-coroutine -Wno-unused | FileCheck %s
 
 namespace std {
@@ -53,19 +53,19 @@ struct std::experimental::coroutine_traits<void> {
 
 // CHECK-LABEL: f0(
 extern "C" void f0() {
-  // CHECK: %[[FRAME:.+]] = call i8* @llvm.coro.begin(
+  // CHECK: %[[FRAME:.+]] = call ptr @llvm.coro.begin(
 
   // See if initial_suspend was issued:
   // ----------------------------------
   // CHECK: call void @_ZNSt12experimental16coroutine_traitsIJvEE12promise_type15initial_suspendEv(
-  // CHECK-NEXT: call zeroext i1 @_ZN9init_susp11await_readyEv(%struct.init_susp*
+  // CHECK-NEXT: call zeroext i1 @_ZN9init_susp11await_readyEv(ptr
   // CHECK: %[[INITSP_ID:.+]] = call token @llvm.coro.save(
   // CHECK: call i8 @llvm.coro.suspend(token %[[INITSP_ID]], i1 false)
 
   co_await suspend_always{};
   // See if we need to suspend:
   // --------------------------
-  // CHECK: %[[READY:.+]] = call zeroext i1 @_ZN14suspend_always11await_readyEv(%struct.suspend_always* {{[^,]*}} %[[AWAITABLE:.+]])
+  // CHECK: %[[READY:.+]] = call zeroext i1 @_ZN14suspend_always11await_readyEv(ptr {{[^,]*}} %[[AWAITABLE:.+]])
   // CHECK: br i1 %[[READY]], label %[[READY_BB:.+]], label %[[SUSPEND_BB:.+]]
 
   // If we are suspending:
@@ -75,10 +75,10 @@ extern "C" void f0() {
   // ---------------------------
   // Build the coroutine handle and pass it to await_suspend
   // ---------------------------
-  // CHECK: call i8* @_ZNSt12experimental16coroutine_handleINS_16coroutine_traitsIJvEE12promise_typeEE12from_addressEPv(i8* %[[FRAME]])
-  //   ... many lines of code to coerce coroutine_handle into an i8* scalar
-  // CHECK: %[[CH:.+]] = load i8*, i8** %{{.+}}
-  // CHECK: call void @_ZN14suspend_always13await_suspendENSt12experimental16coroutine_handleIvEE(%struct.suspend_always* {{[^,]*}} %[[AWAITABLE]], i8* %[[CH]])
+  // CHECK: call ptr @_ZNSt12experimental16coroutine_handleINS_16coroutine_traitsIJvEE12promise_typeEE12from_addressEPv(ptr %[[FRAME]])
+  //   ... many lines of code to coerce coroutine_handle into an ptr scalar
+  // CHECK: %[[CH:.+]] = load ptr, ptr %{{.+}}
+  // CHECK: call void @_ZN14suspend_always13await_suspendENSt12experimental16coroutine_handleIvEE(ptr {{[^,]*}} %[[AWAITABLE]], ptr %[[CH]])
   // -------------------------
   // Generate a suspend point:
   // -------------------------
@@ -95,12 +95,12 @@ extern "C" void f0() {
   // When coroutine is resumed, call await_resume
   // --------------------------
   // CHECK: [[READY_BB]]:
-  // CHECK:  call void @_ZN14suspend_always12await_resumeEv(%struct.suspend_always* {{[^,]*}} %[[AWAITABLE]])
+  // CHECK:  call void @_ZN14suspend_always12await_resumeEv(ptr {{[^,]*}} %[[AWAITABLE]])
 
   // See if final_suspend was issued:
   // ----------------------------------
   // CHECK: call void @_ZNSt12experimental16coroutine_traitsIJvEE12promise_type13final_suspendEv(
-  // CHECK-NEXT: call zeroext i1 @_ZN10final_susp11await_readyEv(%struct.final_susp*
+  // CHECK-NEXT: call zeroext i1 @_ZN10final_susp11await_readyEv(ptr
   // CHECK: %[[FINALSP_ID:.+]] = call token @llvm.coro.save(
   // CHECK: call i8 @llvm.coro.suspend(token %[[FINALSP_ID]], i1 true)
 }
@@ -127,13 +127,13 @@ struct std::experimental::coroutine_traits<void, int> {
 // CHECK-LABEL: f1(
 extern "C" void f1(int) {
   // CHECK: %[[PROMISE:.+]] = alloca %"struct.std::experimental::coroutine_traits<void, int>::promise_type"
-  // CHECK: %[[FRAME:.+]] = call i8* @llvm.coro.begin(
+  // CHECK: %[[FRAME:.+]] = call ptr @llvm.coro.begin(
   co_yield 42;
-  // CHECK: call void @_ZNSt12experimental16coroutine_traitsIJviEE12promise_type11yield_valueEi(%struct.suspend_maybe* sret(%struct.suspend_maybe) align 4 %[[AWAITER:.+]], %"struct.std::experimental::coroutine_traits<void, int>::promise_type"* {{[^,]*}} %[[PROMISE]], i32 42)
+  // CHECK: call void @_ZNSt12experimental16coroutine_traitsIJviEE12promise_type11yield_valueEi(ptr sret(%struct.suspend_maybe) align 4 %[[AWAITER:.+]], ptr {{[^,]*}} %[[PROMISE]], i32 42)
 
   // See if we need to suspend:
   // --------------------------
-  // CHECK: %[[READY:.+]] = call zeroext i1 @_ZN13suspend_maybe11await_readyEv(%struct.suspend_maybe* {{[^,]*}} %[[AWAITABLE]])
+  // CHECK: %[[READY:.+]] = call zeroext i1 @_ZN13suspend_maybe11await_readyEv(ptr {{[^,]*}} %[[AWAITABLE]])
   // CHECK: br i1 %[[READY]], label %[[READY_BB:.+]], label %[[SUSPEND_BB:.+]]
 
   // If we are suspending:
@@ -143,10 +143,10 @@ extern "C" void f1(int) {
   // ---------------------------
   // Build the coroutine handle and pass it to await_suspend
   // ---------------------------
-  // CHECK: call i8* @_ZNSt12experimental16coroutine_handleINS_16coroutine_traitsIJviEE12promise_typeEE12from_addressEPv(i8* %[[FRAME]])
-  //   ... many lines of code to coerce coroutine_handle into an i8* scalar
-  // CHECK: %[[CH:.+]] = load i8*, i8** %{{.+}}
-  // CHECK: %[[YES:.+]] = call zeroext i1 @_ZN13suspend_maybe13await_suspendENSt12experimental16coroutine_handleIvEE(%struct.suspend_maybe* {{[^,]*}} %[[AWAITABLE]], i8* %[[CH]])
+  // CHECK: call ptr @_ZNSt12experimental16coroutine_handleINS_16coroutine_traitsIJviEE12promise_typeEE12from_addressEPv(ptr %[[FRAME]])
+  //   ... many lines of code to coerce coroutine_handle into an ptr scalar
+  // CHECK: %[[CH:.+]] = load ptr, ptr %{{.+}}
+  // CHECK: %[[YES:.+]] = call zeroext i1 @_ZN13suspend_maybe13await_suspendENSt12experimental16coroutine_handleIvEE(ptr {{[^,]*}} %[[AWAITABLE]], ptr %[[CH]])
   // -------------------------------------------
   // See if await_suspend decided not to suspend
   // -------------------------------------------
@@ -156,7 +156,7 @@ extern "C" void f1(int) {
   // CHECK:    call i8 @llvm.coro.suspend(token %[[SUSPEND_ID]], i1 false)
 
   // CHECK: [[READY_BB]]:
-  // CHECK:     call void @_ZN13suspend_maybe12await_resumeEv(%struct.suspend_maybe* {{[^,]*}} %[[AWAITABLE]])
+  // CHECK:     call void @_ZN13suspend_maybe12await_resumeEv(ptr {{[^,]*}} %[[AWAITABLE]])
 }
 
 struct ComplexAwaiter {
@@ -169,14 +169,14 @@ extern "C" void UseComplex(_Complex float);
 // CHECK-LABEL: @TestComplex(
 extern "C" void TestComplex() {
   UseComplex(co_await ComplexAwaiter{});
-  // CHECK: call <2 x float> @_ZN14ComplexAwaiter12await_resumeEv(%struct.ComplexAwaiter*
+  // CHECK: call <2 x float> @_ZN14ComplexAwaiter12await_resumeEv(ptr
   // CHECK: call void @UseComplex(<2 x float> %{{.+}})
 
   co_await ComplexAwaiter{};
-  // CHECK: call <2 x float> @_ZN14ComplexAwaiter12await_resumeEv(%struct.ComplexAwaiter*
+  // CHECK: call <2 x float> @_ZN14ComplexAwaiter12await_resumeEv(ptr
 
   _Complex float Val = co_await ComplexAwaiter{};
-  // CHECK: call <2 x float> @_ZN14ComplexAwaiter12await_resumeEv(%struct.ComplexAwaiter*
+  // CHECK: call <2 x float> @_ZN14ComplexAwaiter12await_resumeEv(ptr
 }
 
 struct Aggr {
@@ -199,22 +199,22 @@ extern "C" void UseAggr(Aggr &&);
 extern "C" void TestAggr() {
   UseAggr(co_await AggrAwaiter{});
   Whatever();
-  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(%struct.Aggr* sret(%struct.Aggr) align 4 %[[AwaitResume:.+]],
-  // CHECK: call void @UseAggr(%struct.Aggr* nonnull align 4 dereferenceable(12) %[[AwaitResume]])
-  // CHECK: call void @_ZN4AggrD1Ev(%struct.Aggr* {{[^,]*}} %[[AwaitResume]])
+  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(ptr sret(%struct.Aggr) align 4 %[[AwaitResume:.+]],
+  // CHECK: call void @UseAggr(ptr nonnull align 4 dereferenceable(12) %[[AwaitResume]])
+  // CHECK: call void @_ZN4AggrD1Ev(ptr {{[^,]*}} %[[AwaitResume]])
   // CHECK: call void @Whatever()
 
   co_await AggrAwaiter{};
   Whatever();
-  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(%struct.Aggr* sret(%struct.Aggr) align 4 %[[AwaitResume2:.+]],
-  // CHECK: call void @_ZN4AggrD1Ev(%struct.Aggr* {{[^,]*}} %[[AwaitResume2]])
+  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(ptr sret(%struct.Aggr) align 4 %[[AwaitResume2:.+]],
+  // CHECK: call void @_ZN4AggrD1Ev(ptr {{[^,]*}} %[[AwaitResume2]])
   // CHECK: call void @Whatever()
 
   Aggr Val = co_await AggrAwaiter{};
   Whatever();
-  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(%struct.Aggr* sret(%struct.Aggr) align 4 %[[AwaitResume3:.+]],
+  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(ptr sret(%struct.Aggr) align 4 %[[AwaitResume3:.+]],
   // CHECK: call void @Whatever()
-  // CHECK: call void @_ZN4AggrD1Ev(%struct.Aggr* {{[^,]*}} %[[AwaitResume3]])
+  // CHECK: call void @_ZN4AggrD1Ev(ptr {{[^,]*}} %[[AwaitResume3]])
 }
 
 struct ScalarAwaiter {
@@ -228,17 +228,17 @@ extern "C" void UseScalar(int);
 // CHECK-LABEL: @TestScalar(
 extern "C" void TestScalar() {
   UseScalar(co_await ScalarAwaiter{});
-  // CHECK: %[[Result:.+]] = call i32 @_ZN13ScalarAwaiter12await_resumeEv(%struct.ScalarAwaiter*
+  // CHECK: %[[Result:.+]] = call i32 @_ZN13ScalarAwaiter12await_resumeEv(ptr
   // CHECK: call void @UseScalar(i32 %[[Result]])
 
   int Val = co_await ScalarAwaiter{};
-  // CHECK: %[[Result2:.+]] = call i32 @_ZN13ScalarAwaiter12await_resumeEv(%struct.ScalarAwaiter*
-  // CHECK: store i32 %[[Result2]], i32* %[[TMP_EXPRCLEANUP:.+]],
-  // CHECK: %[[TMP:.+]] = load i32, i32* %[[TMP_EXPRCLEANUP]],
-  // CHECK: store i32 %[[TMP]], i32* %Val,
+  // CHECK: %[[Result2:.+]] = call i32 @_ZN13ScalarAwaiter12await_resumeEv(ptr
+  // CHECK: store i32 %[[Result2]], ptr %[[TMP_EXPRCLEANUP:.+]],
+  // CHECK: %[[TMP:.+]] = load i32, ptr %[[TMP_EXPRCLEANUP]],
+  // CHECK: store i32 %[[TMP]], ptr %Val,
 
   co_await ScalarAwaiter{};
-  // CHECK: call i32 @_ZN13ScalarAwaiter12await_resumeEv(%struct.ScalarAwaiter*
+  // CHECK: call i32 @_ZN13ScalarAwaiter12await_resumeEv(ptr
 }
 
 // Test operator co_await codegen.
@@ -253,21 +253,21 @@ struct MyAgg {
 extern "C" void TestOpAwait() {
   co_await MyInt(42);
   // CHECK: call void @_Zaw5MyInt(i32 42)
-  // CHECK: call i32 @_ZN13ScalarAwaiter12await_resumeEv(%struct.ScalarAwaiter* {{[^,]*}} %
+  // CHECK: call i32 @_ZN13ScalarAwaiter12await_resumeEv(ptr {{[^,]*}} %
 
   co_await MyAgg{};
-  // CHECK: call void @_ZN5MyAggawEv(%struct.MyAgg* {{[^,]*}} %
-  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(%struct.Aggr* sret(%struct.Aggr) align 4 %
+  // CHECK: call void @_ZN5MyAggawEv(ptr {{[^,]*}} %
+  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(ptr sret(%struct.Aggr) align 4 %
 }
 
 // CHECK-LABEL: EndlessLoop(
 extern "C" void EndlessLoop() {
-  // CHECK: %[[FRAME:.+]] = call i8* @llvm.coro.begin(
+  // CHECK: %[[FRAME:.+]] = call ptr @llvm.coro.begin(
 
   // See if initial_suspend was issued:
   // ----------------------------------
   // CHECK: call void @_ZNSt12experimental16coroutine_traitsIJvEE12promise_type15initial_suspendEv(
-  // CHECK-NEXT: call zeroext i1 @_ZN9init_susp11await_readyEv(%struct.init_susp*
+  // CHECK-NEXT: call zeroext i1 @_ZN9init_susp11await_readyEv(ptr
 
   for (;;)
     co_await suspend_always{};
@@ -275,7 +275,7 @@ extern "C" void EndlessLoop() {
   // Verify that final_suspend was NOT issued:
   // ----------------------------------
   // CHECK-NOT: call void @_ZNSt12experimental16coroutine_traitsIJvEE12promise_type13final_suspendEv(
-  // CHECK-NOT: call zeroext i1 @_ZN10final_susp11await_readyEv(%struct.final_susp*
+  // CHECK-NOT: call zeroext i1 @_ZN10final_susp11await_readyEv(ptr
 }
 
 // Verifies that we don't crash when awaiting on an lvalue.
@@ -310,30 +310,30 @@ struct std::experimental::coroutine_traits<void, double> {
 void AwaitReturnsLValue(double) {
   AwaitResumeReturnsLValue a;
   // CHECK: %[[AVAR:.+]] = alloca %struct.AwaitResumeReturnsLValue,
-  // CHECK: %[[XVAR:.+]] = alloca %struct.RefTag*,
+  // CHECK: %[[XVAR:.+]] = alloca ptr,
 
-  // CHECK: %[[YVAR:.+]] = alloca %struct.RefTag*,
+  // CHECK: %[[YVAR:.+]] = alloca ptr,
   // CHECK-NEXT: %[[TMP1:.+]] = alloca %struct.AwaitResumeReturnsLValue,
 
-  // CHECK: %[[TMP_EXPRCLEANUP1:.+]] = alloca %struct.RefTag*,
-  // CHECK: %[[ZVAR:.+]] = alloca %struct.RefTag*,
+  // CHECK: %[[TMP_EXPRCLEANUP1:.+]] = alloca ptr,
+  // CHECK: %[[ZVAR:.+]] = alloca ptr,
   // CHECK-NEXT: %[[TMP2:.+]] = alloca %struct.AwaitResumeReturnsLValue,
-  // CHECK: %[[TMP_EXPRCLEANUP2:.+]] = alloca %struct.RefTag*,
+  // CHECK: %[[TMP_EXPRCLEANUP2:.+]] = alloca ptr,
 
-  // CHECK: %[[RES1:.+]] = call nonnull align 1 dereferenceable({{.*}}) %struct.RefTag* @_ZN24AwaitResumeReturnsLValue12await_resumeEv(%struct.AwaitResumeReturnsLValue* {{[^,]*}} %[[AVAR]])
-  // CHECK-NEXT: store %struct.RefTag* %[[RES1]], %struct.RefTag** %[[XVAR]],
+  // CHECK: %[[RES1:.+]] = call nonnull align 1 dereferenceable({{.*}}) ptr @_ZN24AwaitResumeReturnsLValue12await_resumeEv(ptr {{[^,]*}} %[[AVAR]])
+  // CHECK-NEXT: store ptr %[[RES1]], ptr %[[XVAR]],
   RefTag &x = co_await a;
 
-  // CHECK: %[[RES2:.+]] = call nonnull align 1 dereferenceable({{.*}}) %struct.RefTag* @_ZN24AwaitResumeReturnsLValue12await_resumeEv(%struct.AwaitResumeReturnsLValue* {{[^,]*}} %[[TMP1]])
-  // CHECK-NEXT: store %struct.RefTag* %[[RES2]], %struct.RefTag** %[[TMP_EXPRCLEANUP1]],
-  // CHECK: %[[LOAD_TMP1:.+]] = load %struct.RefTag*, %struct.RefTag** %[[TMP_EXPRCLEANUP1]],
-  // CHECK: store %struct.RefTag* %[[LOAD_TMP1]], %struct.RefTag** %[[YVAR]],
+  // CHECK: %[[RES2:.+]] = call nonnull align 1 dereferenceable({{.*}}) ptr @_ZN24AwaitResumeReturnsLValue12await_resumeEv(ptr {{[^,]*}} %[[TMP1]])
+  // CHECK-NEXT: store ptr %[[RES2]], ptr %[[TMP_EXPRCLEANUP1]],
+  // CHECK: %[[LOAD_TMP1:.+]] = load ptr, ptr %[[TMP_EXPRCLEANUP1]],
+  // CHECK: store ptr %[[LOAD_TMP1]], ptr %[[YVAR]],
 
   RefTag &y = co_await AwaitResumeReturnsLValue{};
-  // CHECK: %[[RES3:.+]] = call nonnull align 1 dereferenceable({{.*}}) %struct.RefTag* @_ZN24AwaitResumeReturnsLValue12await_resumeEv(%struct.AwaitResumeReturnsLValue* {{[^,]*}} %[[TMP2]])
-  // CHECK-NEXT: store %struct.RefTag* %[[RES3]], %struct.RefTag** %[[TMP_EXPRCLEANUP2]],
-  // CHECK: %[[LOAD_TMP2:.+]] = load %struct.RefTag*, %struct.RefTag** %[[TMP_EXPRCLEANUP2]],
-  // CHECK: store %struct.RefTag* %[[LOAD_TMP2]], %struct.RefTag** %[[ZVAR]],
+  // CHECK: %[[RES3:.+]] = call nonnull align 1 dereferenceable({{.*}}) ptr @_ZN24AwaitResumeReturnsLValue12await_resumeEv(ptr {{[^,]*}} %[[TMP2]])
+  // CHECK-NEXT: store ptr %[[RES3]], ptr %[[TMP_EXPRCLEANUP2]],
+  // CHECK: %[[LOAD_TMP2:.+]] = load ptr, ptr %[[TMP_EXPRCLEANUP2]],
+  // CHECK: store ptr %[[LOAD_TMP2]], ptr %[[ZVAR]],
   RefTag &z = co_yield 42;
 }
 
@@ -347,9 +347,9 @@ struct TailCallAwait {
 extern "C" void TestTailcall() {
   co_await TailCallAwait{};
 
-  // CHECK: %[[RESULT:.+]] = call i8* @_ZN13TailCallAwait13await_suspendENSt12experimental16coroutine_handleIvEE(%struct.TailCallAwait*
-  // CHECK: %[[COERCE:.+]] = getelementptr inbounds %"struct.std::experimental::coroutine_handle", %"struct.std::experimental::coroutine_handle"* %[[TMP:.+]], i32 0, i32 0
-  // CHECK: store i8* %[[RESULT]], i8** %[[COERCE]]
-  // CHECK: %[[ADDR:.+]] = call i8* @_ZNSt12experimental16coroutine_handleIvE7addressEv(%"struct.std::experimental::coroutine_handle"* {{[^,]*}} %[[TMP]])
-  // CHECK: call void @llvm.coro.resume(i8* %[[ADDR]])
+  // CHECK: %[[RESULT:.+]] = call ptr @_ZN13TailCallAwait13await_suspendENSt12experimental16coroutine_handleIvEE(ptr
+  // CHECK: %[[COERCE:.+]] = getelementptr inbounds %"struct.std::experimental::coroutine_handle", ptr %[[TMP:.+]], i32 0, i32 0
+  // CHECK: store ptr %[[RESULT]], ptr %[[COERCE]]
+  // CHECK: %[[ADDR:.+]] = call ptr @_ZNSt12experimental16coroutine_handleIvE7addressEv(ptr {{[^,]*}} %[[TMP]])
+  // CHECK: call void @llvm.coro.resume(ptr %[[ADDR]])
 }

diff  --git a/clang/test/CodeGenCoroutines/coro-await-resume-eh-exp-namespace.cpp b/clang/test/CodeGenCoroutines/coro-await-resume-eh-exp-namespace.cpp
index aab9fc100e6dc..380ce751e6c81 100644
--- a/clang/test/CodeGenCoroutines/coro-await-resume-eh-exp-namespace.cpp
+++ b/clang/test/CodeGenCoroutines/coro-await-resume-eh-exp-namespace.cpp
@@ -3,7 +3,7 @@
 // executing the handler specified by the promise type's 'unhandled_exception'
 // member function.
 //
-// RUN: %clang_cc1 -no-opaque-pointers -std=c++14 -fcoroutines-ts \
+// RUN: %clang_cc1 -std=c++14 -fcoroutines-ts \
 // RUN:   -triple=x86_64-unknown-linux-gnu -emit-llvm -o - %s \
 // RUN:   -fexceptions -fcxx-exceptions -disable-llvm-passes \
 // RUN:   | FileCheck %s
@@ -34,14 +34,14 @@ throwing_task f() {
   // of 'await_resume' threw an exception. Exceptions thrown in
   // 'await_resume' are unwound to RESUMELPAD.
   // CHECK: init.ready:
-  // CHECK-NEXT: store i1 true, i1* %[[RESUMETHREW:.+]], align 1
+  // CHECK-NEXT: store i1 true, ptr %[[RESUMETHREW:.+]], align 1
   // CHECK-NEXT: invoke void @_ZN18throwing_awaitable12await_resumeEv
   // CHECK-NEXT: to label %[[RESUMECONT:.+]] unwind label %[[RESUMELPAD:.+]]
 
   // If 'await_resume' does not throw an exception, 'false' is stored in
   // variable RESUMETHREW.
   // CHECK: [[RESUMECONT]]:
-  // CHECK-NEXT: store i1 false, i1* %[[RESUMETHREW]]
+  // CHECK-NEXT: store i1 false, ptr %[[RESUMETHREW]]
   // CHECK-NEXT: br label %[[RESUMETRYCONT:.+]]
 
   // 'unhandled_exception' is called for the exception thrown in
@@ -69,7 +69,7 @@ throwing_task f() {
   // suspend is executed immediately. Otherwise, the coroutine body is
   // executed, and then the final suspend.
   // CHECK: [[CLEANUPCONT]]:
-  // CHECK-NEXT: %[[RESUMETHREWLOAD:.+]] = load i1, i1* %[[RESUMETHREW]]
+  // CHECK-NEXT: %[[RESUMETHREWLOAD:.+]] = load i1, ptr %[[RESUMETHREW]]
   // CHECK-NEXT: br i1 %[[RESUMETHREWLOAD]], label %[[RESUMEDCONT:.+]], label %[[RESUMEDBODY:.+]]
 
   // CHECK: [[RESUMEDBODY]]:
@@ -109,6 +109,6 @@ noexcept_task g() {
   // This means that no i1 are stored before or after calling await_resume:
   // CHECK: init.ready:
   // CHECK-NEXT: call void @_ZN18noexcept_awaitable12await_resumeEv
-  // CHECK-NOT: store i1 false, i1*
+  // CHECK-NOT: store i1 false, ptr
   co_return;
 }

diff  --git a/clang/test/CodeGenCoroutines/coro-await-resume-eh.cpp b/clang/test/CodeGenCoroutines/coro-await-resume-eh.cpp
index aa14028244624..8253337ba7390 100644
--- a/clang/test/CodeGenCoroutines/coro-await-resume-eh.cpp
+++ b/clang/test/CodeGenCoroutines/coro-await-resume-eh.cpp
@@ -3,7 +3,7 @@
 // executing the handler specified by the promise type's 'unhandled_exception'
 // member function.
 //
-// RUN: %clang_cc1 -no-opaque-pointers -std=c++20 \
+// RUN: %clang_cc1 -std=c++20 \
 // RUN:   -triple=x86_64-unknown-linux-gnu -emit-llvm -o - %s \
 // RUN:   -fexceptions -fcxx-exceptions -disable-llvm-passes \
 // RUN:   | FileCheck %s
@@ -32,14 +32,14 @@ throwing_task f() {
   // of 'await_resume' threw an exception. Exceptions thrown in
   // 'await_resume' are unwound to RESUMELPAD.
   // CHECK: init.ready:
-  // CHECK-NEXT: store i1 true, i1* %[[RESUMETHREW:.+]], align 1
+  // CHECK-NEXT: store i1 true, ptr %[[RESUMETHREW:.+]], align 1
   // CHECK-NEXT: invoke void @_ZN18throwing_awaitable12await_resumeEv
   // CHECK-NEXT: to label %[[RESUMECONT:.+]] unwind label %[[RESUMELPAD:.+]]
 
   // If 'await_resume' does not throw an exception, 'false' is stored in
   // variable RESUMETHREW.
   // CHECK: [[RESUMECONT]]:
-  // CHECK-NEXT: store i1 false, i1* %[[RESUMETHREW]]
+  // CHECK-NEXT: store i1 false, ptr %[[RESUMETHREW]]
   // CHECK-NEXT: br label %[[RESUMETRYCONT:.+]]
 
   // 'unhandled_exception' is called for the exception thrown in
@@ -67,7 +67,7 @@ throwing_task f() {
   // suspend is executed immediately. Otherwise, the coroutine body is
   // executed, and then the final suspend.
   // CHECK: [[CLEANUPCONT]]:
-  // CHECK-NEXT: %[[RESUMETHREWLOAD:.+]] = load i1, i1* %[[RESUMETHREW]]
+  // CHECK-NEXT: %[[RESUMETHREWLOAD:.+]] = load i1, ptr %[[RESUMETHREW]]
   // CHECK-NEXT: br i1 %[[RESUMETHREWLOAD]], label %[[RESUMEDCONT:.+]], label %[[RESUMEDBODY:.+]]
 
   // CHECK: [[RESUMEDBODY]]:
@@ -107,6 +107,6 @@ noexcept_task g() {
   // This means that no i1 are stored before or after calling await_resume:
   // CHECK: init.ready:
   // CHECK-NEXT: call void @_ZN18noexcept_awaitable12await_resumeEv
-  // CHECK-NOT: store i1 false, i1*
+  // CHECK-NOT: store i1 false, ptr
   co_return;
 }

diff  --git a/clang/test/CodeGenCoroutines/coro-await.cpp b/clang/test/CodeGenCoroutines/coro-await.cpp
index 7685376a9a6df..e1c7039469afc 100644
--- a/clang/test/CodeGenCoroutines/coro-await.cpp
+++ b/clang/test/CodeGenCoroutines/coro-await.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -no-enable-noundef-analysis -triple x86_64-unknown-linux-gnu -std=c++20 \
+// RUN: %clang_cc1 -no-enable-noundef-analysis -triple x86_64-unknown-linux-gnu -std=c++20 \
 // RUN:   -emit-llvm %s -o - -disable-llvm-passes -Wno-coroutine -Wno-unused | FileCheck %s
 
 namespace std {
@@ -51,19 +51,19 @@ struct std::coroutine_traits<void> {
 
 // CHECK-LABEL: f0(
 extern "C" void f0() {
-  // CHECK: %[[FRAME:.+]] = call i8* @llvm.coro.begin(
+  // CHECK: %[[FRAME:.+]] = call ptr @llvm.coro.begin(
 
   // See if initial_suspend was issued:
   // ----------------------------------
   // CHECK: call void @_ZNSt16coroutine_traitsIJvEE12promise_type15initial_suspendEv(
-  // CHECK-NEXT: call zeroext i1 @_ZN9init_susp11await_readyEv(%struct.init_susp*
+  // CHECK-NEXT: call zeroext i1 @_ZN9init_susp11await_readyEv(ptr
   // CHECK: %[[INITSP_ID:.+]] = call token @llvm.coro.save(
   // CHECK: call i8 @llvm.coro.suspend(token %[[INITSP_ID]], i1 false)
 
   co_await suspend_always{};
   // See if we need to suspend:
   // --------------------------
-  // CHECK: %[[READY:.+]] = call zeroext i1 @_ZN14suspend_always11await_readyEv(%struct.suspend_always* {{[^,]*}} %[[AWAITABLE:.+]])
+  // CHECK: %[[READY:.+]] = call zeroext i1 @_ZN14suspend_always11await_readyEv(ptr {{[^,]*}} %[[AWAITABLE:.+]])
   // CHECK: br i1 %[[READY]], label %[[READY_BB:.+]], label %[[SUSPEND_BB:.+]]
 
   // If we are suspending:
@@ -73,10 +73,10 @@ extern "C" void f0() {
   // ---------------------------
   // Build the coroutine handle and pass it to await_suspend
   // ---------------------------
-  // CHECK: call i8* @_ZNSt16coroutine_handleINSt16coroutine_traitsIJvEE12promise_typeEE12from_addressEPv(i8* %[[FRAME]])
-  //   ... many lines of code to coerce coroutine_handle into an i8* scalar
-  // CHECK: %[[CH:.+]] = load i8*, i8** %{{.+}}
-  // CHECK: call void @_ZN14suspend_always13await_suspendESt16coroutine_handleIvE(%struct.suspend_always* {{[^,]*}} %[[AWAITABLE]], i8* %[[CH]])
+  // CHECK: call ptr @_ZNSt16coroutine_handleINSt16coroutine_traitsIJvEE12promise_typeEE12from_addressEPv(ptr %[[FRAME]])
+  //   ... many lines of code to coerce coroutine_handle into an ptr scalar
+  // CHECK: %[[CH:.+]] = load ptr, ptr %{{.+}}
+  // CHECK: call void @_ZN14suspend_always13await_suspendESt16coroutine_handleIvE(ptr {{[^,]*}} %[[AWAITABLE]], ptr %[[CH]])
   // -------------------------
   // Generate a suspend point:
   // -------------------------
@@ -93,12 +93,12 @@ extern "C" void f0() {
   // When coroutine is resumed, call await_resume
   // --------------------------
   // CHECK: [[READY_BB]]:
-  // CHECK:  call void @_ZN14suspend_always12await_resumeEv(%struct.suspend_always* {{[^,]*}} %[[AWAITABLE]])
+  // CHECK:  call void @_ZN14suspend_always12await_resumeEv(ptr {{[^,]*}} %[[AWAITABLE]])
 
   // See if final_suspend was issued:
   // ----------------------------------
   // CHECK: call void @_ZNSt16coroutine_traitsIJvEE12promise_type13final_suspendEv(
-  // CHECK-NEXT: call zeroext i1 @_ZN10final_susp11await_readyEv(%struct.final_susp*
+  // CHECK-NEXT: call zeroext i1 @_ZN10final_susp11await_readyEv(ptr
   // CHECK: %[[FINALSP_ID:.+]] = call token @llvm.coro.save(
   // CHECK: call i8 @llvm.coro.suspend(token %[[FINALSP_ID]], i1 true)
 }
@@ -125,13 +125,13 @@ struct std::coroutine_traits<void, int> {
 // CHECK-LABEL: f1(
 extern "C" void f1(int) {
   // CHECK: %[[PROMISE:.+]] = alloca %"struct.std::coroutine_traits<void, int>::promise_type"
-  // CHECK: %[[FRAME:.+]] = call i8* @llvm.coro.begin(
+  // CHECK: %[[FRAME:.+]] = call ptr @llvm.coro.begin(
   co_yield 42;
-  // CHECK: call void @_ZNSt16coroutine_traitsIJviEE12promise_type11yield_valueEi(%struct.suspend_maybe* sret(%struct.suspend_maybe) align 4 %[[AWAITER:.+]], %"struct.std::coroutine_traits<void, int>::promise_type"* {{[^,]*}} %[[PROMISE]], i32 42)
+  // CHECK: call void @_ZNSt16coroutine_traitsIJviEE12promise_type11yield_valueEi(ptr sret(%struct.suspend_maybe) align 4 %[[AWAITER:.+]], ptr {{[^,]*}} %[[PROMISE]], i32 42)
 
   // See if we need to suspend:
   // --------------------------
-  // CHECK: %[[READY:.+]] = call zeroext i1 @_ZN13suspend_maybe11await_readyEv(%struct.suspend_maybe* {{[^,]*}} %[[AWAITABLE]])
+  // CHECK: %[[READY:.+]] = call zeroext i1 @_ZN13suspend_maybe11await_readyEv(ptr {{[^,]*}} %[[AWAITABLE]])
   // CHECK: br i1 %[[READY]], label %[[READY_BB:.+]], label %[[SUSPEND_BB:.+]]
 
   // If we are suspending:
@@ -141,10 +141,10 @@ extern "C" void f1(int) {
   // ---------------------------
   // Build the coroutine handle and pass it to await_suspend
   // ---------------------------
-  // CHECK: call i8* @_ZNSt16coroutine_handleINSt16coroutine_traitsIJviEE12promise_typeEE12from_addressEPv(i8* %[[FRAME]])
-  //   ... many lines of code to coerce coroutine_handle into an i8* scalar
-  // CHECK: %[[CH:.+]] = load i8*, i8** %{{.+}}
-  // CHECK: %[[YES:.+]] = call zeroext i1 @_ZN13suspend_maybe13await_suspendESt16coroutine_handleIvE(%struct.suspend_maybe* {{[^,]*}} %[[AWAITABLE]], i8* %[[CH]])
+  // CHECK: call ptr @_ZNSt16coroutine_handleINSt16coroutine_traitsIJviEE12promise_typeEE12from_addressEPv(ptr %[[FRAME]])
+  //   ... many lines of code to coerce coroutine_handle into an ptr scalar
+  // CHECK: %[[CH:.+]] = load ptr, ptr %{{.+}}
+  // CHECK: %[[YES:.+]] = call zeroext i1 @_ZN13suspend_maybe13await_suspendESt16coroutine_handleIvE(ptr {{[^,]*}} %[[AWAITABLE]], ptr %[[CH]])
   // -------------------------------------------
   // See if await_suspend decided not to suspend
   // -------------------------------------------
@@ -154,7 +154,7 @@ extern "C" void f1(int) {
   // CHECK:    call i8 @llvm.coro.suspend(token %[[SUSPEND_ID]], i1 false)
 
   // CHECK: [[READY_BB]]:
-  // CHECK:     call void @_ZN13suspend_maybe12await_resumeEv(%struct.suspend_maybe* {{[^,]*}} %[[AWAITABLE]])
+  // CHECK:     call void @_ZN13suspend_maybe12await_resumeEv(ptr {{[^,]*}} %[[AWAITABLE]])
 }
 
 struct ComplexAwaiter {
@@ -167,14 +167,14 @@ extern "C" void UseComplex(_Complex float);
 // CHECK-LABEL: @TestComplex(
 extern "C" void TestComplex() {
   UseComplex(co_await ComplexAwaiter{});
-  // CHECK: call <2 x float> @_ZN14ComplexAwaiter12await_resumeEv(%struct.ComplexAwaiter*
+  // CHECK: call <2 x float> @_ZN14ComplexAwaiter12await_resumeEv(ptr
   // CHECK: call void @UseComplex(<2 x float> %{{.+}})
 
   co_await ComplexAwaiter{};
-  // CHECK: call <2 x float> @_ZN14ComplexAwaiter12await_resumeEv(%struct.ComplexAwaiter*
+  // CHECK: call <2 x float> @_ZN14ComplexAwaiter12await_resumeEv(ptr
 
   _Complex float Val = co_await ComplexAwaiter{};
-  // CHECK: call <2 x float> @_ZN14ComplexAwaiter12await_resumeEv(%struct.ComplexAwaiter*
+  // CHECK: call <2 x float> @_ZN14ComplexAwaiter12await_resumeEv(ptr
 }
 
 struct Aggr { int X, Y, Z; ~Aggr(); };
@@ -194,22 +194,22 @@ extern "C" void UseAggr(Aggr&&);
 extern "C" void TestAggr() {
   UseAggr(co_await AggrAwaiter{});
   Whatever();
-  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(%struct.Aggr* sret(%struct.Aggr) align 4 %[[AwaitResume:.+]],
-  // CHECK: call void @UseAggr(%struct.Aggr* nonnull align 4 dereferenceable(12) %[[AwaitResume]])
-  // CHECK: call void @_ZN4AggrD1Ev(%struct.Aggr* {{[^,]*}} %[[AwaitResume]])
+  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(ptr sret(%struct.Aggr) align 4 %[[AwaitResume:.+]],
+  // CHECK: call void @UseAggr(ptr nonnull align 4 dereferenceable(12) %[[AwaitResume]])
+  // CHECK: call void @_ZN4AggrD1Ev(ptr {{[^,]*}} %[[AwaitResume]])
   // CHECK: call void @Whatever()
 
   co_await AggrAwaiter{};
   Whatever();
-  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(%struct.Aggr* sret(%struct.Aggr) align 4 %[[AwaitResume2:.+]],
-  // CHECK: call void @_ZN4AggrD1Ev(%struct.Aggr* {{[^,]*}} %[[AwaitResume2]])
+  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(ptr sret(%struct.Aggr) align 4 %[[AwaitResume2:.+]],
+  // CHECK: call void @_ZN4AggrD1Ev(ptr {{[^,]*}} %[[AwaitResume2]])
   // CHECK: call void @Whatever()
 
   Aggr Val = co_await AggrAwaiter{};
   Whatever();
-  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(%struct.Aggr* sret(%struct.Aggr) align 4 %[[AwaitResume3:.+]],
+  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(ptr sret(%struct.Aggr) align 4 %[[AwaitResume3:.+]],
   // CHECK: call void @Whatever()
-  // CHECK: call void @_ZN4AggrD1Ev(%struct.Aggr* {{[^,]*}} %[[AwaitResume3]])
+  // CHECK: call void @_ZN4AggrD1Ev(ptr {{[^,]*}} %[[AwaitResume3]])
 }
 
 struct ScalarAwaiter {
@@ -223,17 +223,17 @@ extern "C" void UseScalar(int);
 // CHECK-LABEL: @TestScalar(
 extern "C" void TestScalar() {
   UseScalar(co_await ScalarAwaiter{});
-  // CHECK: %[[Result:.+]] = call i32 @_ZN13ScalarAwaiter12await_resumeEv(%struct.ScalarAwaiter*
+  // CHECK: %[[Result:.+]] = call i32 @_ZN13ScalarAwaiter12await_resumeEv(ptr
   // CHECK: call void @UseScalar(i32 %[[Result]])
 
   int Val = co_await ScalarAwaiter{};
-  // CHECK: %[[Result2:.+]] = call i32 @_ZN13ScalarAwaiter12await_resumeEv(%struct.ScalarAwaiter*
-  // CHECK: store i32 %[[Result2]], i32* %[[TMP_EXPRCLEANUP:.+]],
-  // CHECK: %[[TMP:.+]] = load i32, i32* %[[TMP_EXPRCLEANUP]],
-  // CHECK: store i32 %[[TMP]], i32* %Val,
+  // CHECK: %[[Result2:.+]] = call i32 @_ZN13ScalarAwaiter12await_resumeEv(ptr
+  // CHECK: store i32 %[[Result2]], ptr %[[TMP_EXPRCLEANUP:.+]],
+  // CHECK: %[[TMP:.+]] = load i32, ptr %[[TMP_EXPRCLEANUP]],
+  // CHECK: store i32 %[[TMP]], ptr %Val,
 
   co_await ScalarAwaiter{};
-  // CHECK: call i32 @_ZN13ScalarAwaiter12await_resumeEv(%struct.ScalarAwaiter*
+  // CHECK: call i32 @_ZN13ScalarAwaiter12await_resumeEv(ptr
 }
 
 // Test operator co_await codegen.
@@ -248,21 +248,21 @@ struct MyAgg {
 extern "C" void TestOpAwait() {
   co_await MyInt(42);
   // CHECK: call void @_Zaw5MyInt(i32 42)
-  // CHECK: call i32 @_ZN13ScalarAwaiter12await_resumeEv(%struct.ScalarAwaiter* {{[^,]*}} %
+  // CHECK: call i32 @_ZN13ScalarAwaiter12await_resumeEv(ptr {{[^,]*}} %
 
   co_await MyAgg{};
-  // CHECK: call void @_ZN5MyAggawEv(%struct.MyAgg* {{[^,]*}} %
-  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(%struct.Aggr* sret(%struct.Aggr) align 4 %
+  // CHECK: call void @_ZN5MyAggawEv(ptr {{[^,]*}} %
+  // CHECK: call void @_ZN11AggrAwaiter12await_resumeEv(ptr sret(%struct.Aggr) align 4 %
 }
 
 // CHECK-LABEL: EndlessLoop(
 extern "C" void EndlessLoop() {
-  // CHECK: %[[FRAME:.+]] = call i8* @llvm.coro.begin(
+  // CHECK: %[[FRAME:.+]] = call ptr @llvm.coro.begin(
 
   // See if initial_suspend was issued:
   // ----------------------------------
   // CHECK: call void @_ZNSt16coroutine_traitsIJvEE12promise_type15initial_suspendEv(
-  // CHECK-NEXT: call zeroext i1 @_ZN9init_susp11await_readyEv(%struct.init_susp*
+  // CHECK-NEXT: call zeroext i1 @_ZN9init_susp11await_readyEv(ptr
 
   for (;;)
     co_await suspend_always{};
@@ -270,7 +270,7 @@ extern "C" void EndlessLoop() {
   // Verify that final_suspend was NOT issued:
   // ----------------------------------
   // CHECK-NOT: call void @_ZNSt16coroutine_traitsIJvEE12promise_type13final_suspendEv(
-  // CHECK-NOT: call zeroext i1 @_ZN10final_susp11await_readyEv(%struct.final_susp*
+  // CHECK-NOT: call zeroext i1 @_ZN10final_susp11await_readyEv(ptr
 }
 
 // Verifies that we don't crash when awaiting on an lvalue.
@@ -305,30 +305,30 @@ struct std::coroutine_traits<void, double> {
 void AwaitReturnsLValue(double) {
   AwaitResumeReturnsLValue a;
   // CHECK: %[[AVAR:.+]] = alloca %struct.AwaitResumeReturnsLValue,
-  // CHECK: %[[XVAR:.+]] = alloca %struct.RefTag*,
+  // CHECK: %[[XVAR:.+]] = alloca ptr,
 
-  // CHECK: %[[YVAR:.+]] = alloca %struct.RefTag*,
+  // CHECK: %[[YVAR:.+]] = alloca ptr,
   // CHECK-NEXT: %[[TMP1:.+]] = alloca %struct.AwaitResumeReturnsLValue,
 
-  // CHECK: %[[TMP_EXPRCLEANUP1:.+]] = alloca %struct.RefTag*,
-  // CHECK: %[[ZVAR:.+]] = alloca %struct.RefTag*,
+  // CHECK: %[[TMP_EXPRCLEANUP1:.+]] = alloca ptr,
+  // CHECK: %[[ZVAR:.+]] = alloca ptr,
   // CHECK-NEXT: %[[TMP2:.+]] = alloca %struct.AwaitResumeReturnsLValue,
-  // CHECK: %[[TMP_EXPRCLEANUP2:.+]] = alloca %struct.RefTag*,
+  // CHECK: %[[TMP_EXPRCLEANUP2:.+]] = alloca ptr,
 
-  // CHECK: %[[RES1:.+]] = call nonnull align 1 dereferenceable({{.*}}) %struct.RefTag* @_ZN24AwaitResumeReturnsLValue12await_resumeEv(%struct.AwaitResumeReturnsLValue* {{[^,]*}} %[[AVAR]])
-  // CHECK-NEXT: store %struct.RefTag* %[[RES1]], %struct.RefTag** %[[XVAR]],
+  // CHECK: %[[RES1:.+]] = call nonnull align 1 dereferenceable({{.*}}) ptr @_ZN24AwaitResumeReturnsLValue12await_resumeEv(ptr {{[^,]*}} %[[AVAR]])
+  // CHECK-NEXT: store ptr %[[RES1]], ptr %[[XVAR]],
   RefTag& x = co_await a;
 
-  // CHECK: %[[RES2:.+]] = call nonnull align 1 dereferenceable({{.*}}) %struct.RefTag* @_ZN24AwaitResumeReturnsLValue12await_resumeEv(%struct.AwaitResumeReturnsLValue* {{[^,]*}} %[[TMP1]])
-  // CHECK-NEXT: store %struct.RefTag* %[[RES2]], %struct.RefTag** %[[TMP_EXPRCLEANUP1]],
-  // CHECK: %[[LOAD_TMP1:.+]] = load %struct.RefTag*, %struct.RefTag** %[[TMP_EXPRCLEANUP1]],
-  // CHECK: store %struct.RefTag* %[[LOAD_TMP1]], %struct.RefTag** %[[YVAR]],
+  // CHECK: %[[RES2:.+]] = call nonnull align 1 dereferenceable({{.*}}) ptr @_ZN24AwaitResumeReturnsLValue12await_resumeEv(ptr {{[^,]*}} %[[TMP1]])
+  // CHECK-NEXT: store ptr %[[RES2]], ptr %[[TMP_EXPRCLEANUP1]],
+  // CHECK: %[[LOAD_TMP1:.+]] = load ptr, ptr %[[TMP_EXPRCLEANUP1]],
+  // CHECK: store ptr %[[LOAD_TMP1]], ptr %[[YVAR]],
 
   RefTag& y = co_await AwaitResumeReturnsLValue{};
-  // CHECK: %[[RES3:.+]] = call nonnull align 1 dereferenceable({{.*}}) %struct.RefTag* @_ZN24AwaitResumeReturnsLValue12await_resumeEv(%struct.AwaitResumeReturnsLValue* {{[^,]*}} %[[TMP2]])
-  // CHECK-NEXT: store %struct.RefTag* %[[RES3]], %struct.RefTag** %[[TMP_EXPRCLEANUP2]],
-  // CHECK: %[[LOAD_TMP2:.+]] = load %struct.RefTag*, %struct.RefTag** %[[TMP_EXPRCLEANUP2]],
-  // CHECK: store %struct.RefTag* %[[LOAD_TMP2]], %struct.RefTag** %[[ZVAR]],
+  // CHECK: %[[RES3:.+]] = call nonnull align 1 dereferenceable({{.*}}) ptr @_ZN24AwaitResumeReturnsLValue12await_resumeEv(ptr {{[^,]*}} %[[TMP2]])
+  // CHECK-NEXT: store ptr %[[RES3]], ptr %[[TMP_EXPRCLEANUP2]],
+  // CHECK: %[[LOAD_TMP2:.+]] = load ptr, ptr %[[TMP_EXPRCLEANUP2]],
+  // CHECK: store ptr %[[LOAD_TMP2]], ptr %[[ZVAR]],
   RefTag& z = co_yield 42;
 }
 
@@ -342,9 +342,9 @@ struct TailCallAwait {
 extern "C" void TestTailcall() {
   co_await TailCallAwait{};
 
-  // CHECK: %[[RESULT:.+]] = call i8* @_ZN13TailCallAwait13await_suspendESt16coroutine_handleIvE(%struct.TailCallAwait*
-  // CHECK: %[[COERCE:.+]] = getelementptr inbounds %"struct.std::coroutine_handle", %"struct.std::coroutine_handle"* %[[TMP:.+]], i32 0, i32 0
-  // CHECK: store i8* %[[RESULT]], i8** %[[COERCE]]
-  // CHECK: %[[ADDR:.+]] = call i8* @_ZNSt16coroutine_handleIvE7addressEv(%"struct.std::coroutine_handle"* {{[^,]*}} %[[TMP]])
-  // CHECK: call void @llvm.coro.resume(i8* %[[ADDR]])
+  // CHECK: %[[RESULT:.+]] = call ptr @_ZN13TailCallAwait13await_suspendESt16coroutine_handleIvE(ptr
+  // CHECK: %[[COERCE:.+]] = getelementptr inbounds %"struct.std::coroutine_handle", ptr %[[TMP:.+]], i32 0, i32 0
+  // CHECK: store ptr %[[RESULT]], ptr %[[COERCE]]
+  // CHECK: %[[ADDR:.+]] = call ptr @_ZNSt16coroutine_handleIvE7addressEv(ptr {{[^,]*}} %[[TMP]])
+  // CHECK: call void @llvm.coro.resume(ptr %[[ADDR]])
 }

diff  --git a/clang/test/CodeGenCoroutines/coro-cleanup-exp-namespace.cpp b/clang/test/CodeGenCoroutines/coro-cleanup-exp-namespace.cpp
index 1233790abfa58..03023cc0694d2 100644
--- a/clang/test/CodeGenCoroutines/coro-cleanup-exp-namespace.cpp
+++ b/clang/test/CodeGenCoroutines/coro-cleanup-exp-namespace.cpp
@@ -1,5 +1,5 @@
 // Verify that coroutine promise and allocated memory are freed up on exception.
-// RUN: %clang_cc1 -no-opaque-pointers -std=c++1z -fcoroutines-ts -triple=x86_64-unknown-linux-gnu -emit-llvm -o - %s -fexceptions -fcxx-exceptions -disable-llvm-passes | FileCheck %s
+// RUN: %clang_cc1 -std=c++1z -fcoroutines-ts -triple=x86_64-unknown-linux-gnu -emit-llvm -o - %s -fexceptions -fcxx-exceptions -disable-llvm-passes | FileCheck %s
 
 namespace std::experimental {
 template <typename... T> struct coroutine_traits;
@@ -41,7 +41,7 @@ void may_throw();
 
 // CHECK-LABEL: define{{.*}} void @_Z1fv(
 void f() {
-  // CHECK: call noalias noundef nonnull i8* @_Znwm(i64
+  // CHECK: call noalias noundef nonnull ptr @_Znwm(i64
 
   // If promise constructor throws, check that we free the memory.
 
@@ -63,12 +63,12 @@ void f() {
 
   // CHECK: [[CatchPad]]:
   // CHECK-NEXT:  landingpad
-  // CHECK-NEXT:       catch i8* null
+  // CHECK-NEXT:       catch ptr null
   // CHECK:  call void @_ZN7CleanupD1Ev(
   // CHECK:  br label %[[Catch:.+]]
 
   // CHECK: [[Catch]]:
-  // CHECK:    call i8* @__cxa_begin_catch(
+  // CHECK:    call ptr @__cxa_begin_catch(
   // CHECK:    call void @_ZNSt12experimental16coroutine_traitsIJvEE12promise_type19unhandled_exceptionEv(
   // CHECK:    invoke void @__cxa_end_catch()
   // CHECK-NEXT:    to label %[[Cont:.+]] unwind
@@ -80,12 +80,12 @@ void f() {
 
   // CHECK: [[Cleanup]]:
   // CHECK: call void @_ZNSt12experimental16coroutine_traitsIJvEE12promise_typeD1Ev(
-  // CHECK: %[[Mem0:.+]] = call i8* @llvm.coro.free(
-  // CHECK: call void @_ZdlPv(i8* noundef %[[Mem0]]
+  // CHECK: %[[Mem0:.+]] = call ptr @llvm.coro.free(
+  // CHECK: call void @_ZdlPv(ptr noundef %[[Mem0]]
 
   // CHECK: [[Dealloc]]:
-  // CHECK:   %[[Mem:.+]] = call i8* @llvm.coro.free(
-  // CHECK:   call void @_ZdlPv(i8* noundef %[[Mem]])
+  // CHECK:   %[[Mem:.+]] = call ptr @llvm.coro.free(
+  // CHECK:   call void @_ZdlPv(ptr noundef %[[Mem]])
 
   co_return;
 }

diff  --git a/clang/test/CodeGenCoroutines/coro-cleanup.cpp b/clang/test/CodeGenCoroutines/coro-cleanup.cpp
index 06b70853a02a2..e07876305dabf 100644
--- a/clang/test/CodeGenCoroutines/coro-cleanup.cpp
+++ b/clang/test/CodeGenCoroutines/coro-cleanup.cpp
@@ -1,5 +1,5 @@
 // Verify that coroutine promise and allocated memory are freed up on exception.
-// RUN: %clang_cc1 -no-opaque-pointers -std=c++20 -triple=x86_64-unknown-linux-gnu -emit-llvm -o - %s -fexceptions -fcxx-exceptions -disable-llvm-passes | FileCheck %s
+// RUN: %clang_cc1 -std=c++20 -triple=x86_64-unknown-linux-gnu -emit-llvm -o - %s -fexceptions -fcxx-exceptions -disable-llvm-passes | FileCheck %s
 
 namespace std {
 template <typename... T> struct coroutine_traits;
@@ -39,7 +39,7 @@ void may_throw();
 
 // CHECK-LABEL: define{{.*}} void @_Z1fv(
 void f() {
-  // CHECK: call noalias noundef nonnull i8* @_Znwm(i64
+  // CHECK: call noalias noundef nonnull ptr @_Znwm(i64
 
   // If promise constructor throws, check that we free the memory.
 
@@ -61,12 +61,12 @@ void f() {
 
   // CHECK: [[CatchPad]]:
   // CHECK-NEXT:  landingpad
-  // CHECK-NEXT:       catch i8* null
+  // CHECK-NEXT:       catch ptr null
   // CHECK:  call void @_ZN7CleanupD1Ev(
   // CHECK:  br label %[[Catch:.+]]
 
   // CHECK: [[Catch]]:
-  // CHECK:    call i8* @__cxa_begin_catch(
+  // CHECK:    call ptr @__cxa_begin_catch(
   // CHECK:    call void @_ZNSt16coroutine_traitsIJvEE12promise_type19unhandled_exceptionEv(
   // CHECK:    invoke void @__cxa_end_catch()
   // CHECK-NEXT:    to label %[[Cont:.+]] unwind
@@ -78,12 +78,12 @@ void f() {
 
   // CHECK: [[Cleanup]]:
   // CHECK: call void @_ZNSt16coroutine_traitsIJvEE12promise_typeD1Ev(
-  // CHECK: %[[Mem0:.+]] = call i8* @llvm.coro.free(
-  // CHECK: call void @_ZdlPv(i8* noundef %[[Mem0]]
+  // CHECK: %[[Mem0:.+]] = call ptr @llvm.coro.free(
+  // CHECK: call void @_ZdlPv(ptr noundef %[[Mem0]]
 
   // CHECK: [[Dealloc]]:
-  // CHECK:   %[[Mem:.+]] = call i8* @llvm.coro.free(
-  // CHECK:   call void @_ZdlPv(i8* noundef %[[Mem]])
+  // CHECK:   %[[Mem:.+]] = call ptr @llvm.coro.free(
+  // CHECK:   call void @_ZdlPv(ptr noundef %[[Mem]])
 
   co_return;
 }

diff  --git a/clang/test/CodeGenCoroutines/coro-dest-slot-exp-namespace.cpp b/clang/test/CodeGenCoroutines/coro-dest-slot-exp-namespace.cpp
index 43862fa130437..14be9a1d03601 100644
--- a/clang/test/CodeGenCoroutines/coro-dest-slot-exp-namespace.cpp
+++ b/clang/test/CodeGenCoroutines/coro-dest-slot-exp-namespace.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-linux-gnu -fcoroutines-ts -std=c++14 -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -fcoroutines-ts -std=c++14 -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
 
 #include "Inputs/coroutine-exp-namespace.h"
 
@@ -33,7 +33,7 @@ extern "C" coro f(int) { co_return; }
 // CHECK: call void @_ZNSt12experimental13coroutines_v113suspend_never12await_resumeEv(
 // CHECK: %[[CLEANUP_DEST1:.+]] = phi i32 [ 0, %[[FINAL_READY]] ], [ 2, %[[FINAL_CLEANUP]] ]
 // CHECK: %[[CLEANUP_DEST2:.+]] = phi i32 [ %[[CLEANUP_DEST0]], %{{.+}} ], [ %[[CLEANUP_DEST1]], %{{.+}} ], [ 0, %{{.+}} ]
-// CHECK: call i8* @llvm.coro.free(
+// CHECK: call ptr @llvm.coro.free(
 // CHECK: switch i32 %[[CLEANUP_DEST2]], label %{{.+}} [
 // CHECK-NEXT: i32 0
 // CHECK-NEXT: i32 2

diff  --git a/clang/test/CodeGenCoroutines/coro-dest-slot.cpp b/clang/test/CodeGenCoroutines/coro-dest-slot.cpp
index abdedfb9f68f4..d794c74cd52d6 100644
--- a/clang/test/CodeGenCoroutines/coro-dest-slot.cpp
+++ b/clang/test/CodeGenCoroutines/coro-dest-slot.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-linux-gnu -std=c++20 -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -std=c++20 -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
 
 #include "Inputs/coroutine.h"
 
@@ -33,7 +33,7 @@ extern "C" coro f(int) { co_return; }
 // CHECK: call void @_ZNSt13suspend_never12await_resumeEv(
 // CHECK: %[[CLEANUP_DEST1:.+]] = phi i32 [ 0, %[[FINAL_READY]] ], [ 2, %[[FINAL_CLEANUP]] ]
 // CHECK: %[[CLEANUP_DEST2:.+]] = phi i32 [ %[[CLEANUP_DEST0]], %{{.+}} ], [ %[[CLEANUP_DEST1]], %{{.+}} ], [ 0, %{{.+}} ]
-// CHECK: call i8* @llvm.coro.free(
+// CHECK: call ptr @llvm.coro.free(
 // CHECK: switch i32 %[[CLEANUP_DEST2]], label %{{.+}} [
 // CHECK-NEXT: i32 0
 // CHECK-NEXT: i32 2

diff  --git a/clang/test/CodeGenCoroutines/coro-eh-cleanup-exp-namespace.cpp b/clang/test/CodeGenCoroutines/coro-eh-cleanup-exp-namespace.cpp
index feebef3006d1a..9ba91e77004fa 100644
--- a/clang/test/CodeGenCoroutines/coro-eh-cleanup-exp-namespace.cpp
+++ b/clang/test/CodeGenCoroutines/coro-eh-cleanup-exp-namespace.cpp
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -std=c++1z -fcoroutines-ts -triple=x86_64-pc-windows-msvc18.0.0 -emit-llvm %s -o - -fexceptions -fcxx-exceptions -disable-llvm-passes | FileCheck %s
-// RUN: %clang_cc1 -no-opaque-pointers -std=c++1z -fcoroutines-ts -triple=x86_64-unknown-linux-gnu -emit-llvm -o - %s -fexceptions -fcxx-exceptions -disable-llvm-passes | FileCheck --check-prefix=CHECK-LPAD %s
+// RUN: %clang_cc1 -std=c++1z -fcoroutines-ts -triple=x86_64-pc-windows-msvc18.0.0 -emit-llvm %s -o - -fexceptions -fcxx-exceptions -disable-llvm-passes | FileCheck %s
+// RUN: %clang_cc1 -std=c++1z -fcoroutines-ts -triple=x86_64-unknown-linux-gnu -emit-llvm -o - %s -fexceptions -fcxx-exceptions -disable-llvm-passes | FileCheck --check-prefix=CHECK-LPAD %s
 
 namespace std::experimental {
 template <typename R, typename... T> struct coroutine_traits {
@@ -62,27 +62,27 @@ coro_t f() {
 
 // CHECK: [[COROENDBB]]:
 // CHECK-NEXT: %[[CLPAD:.+]] = cleanuppad within none
-// CHECK-NEXT: call i1 @llvm.coro.end(i8* null, i1 true) [ "funclet"(token %[[CLPAD]]) ]
+// CHECK-NEXT: call i1 @llvm.coro.end(ptr null, i1 true) [ "funclet"(token %[[CLPAD]]) ]
 // CHECK-NEXT: cleanupret from %[[CLPAD]] unwind label
 
 // CHECK-LPAD: @_Z1fv(
 // CHECK-LPAD:   invoke void @_Z9may_throwv()
 // CHECK-LPAD:       to label %[[CONT:.+]] unwind label %[[EHCLEANUP:.+]]
 // CHECK-LPAD: [[EHCLEANUP]]:
-// CHECK-LPAD:    landingpad { i8*, i32 }
+// CHECK-LPAD:    landingpad { ptr, i32 }
 // CHECK-LPAD:          catch
 // CHECK-LPAD:   call void @_ZN7CleanupD1Ev(
-// CHECK-LPAD:   call i8* @__cxa_begin_catch
+// CHECK-LPAD:   call ptr @__cxa_begin_catch
 // CHECK-LPAD:   call void @_ZN6coro_t12promise_type19unhandled_exceptionEv
 // CHECK-LPAD:   invoke void @__cxa_end_catch()
 // CHECK-LPAD:             to label %{{.+}} unwind label %[[UNWINDBB:.+]]
 
 // CHECK-LPAD: [[UNWINDBB]]:
-// CHECK-LPAD:   %[[I1RESUME:.+]] = call i1 @llvm.coro.end(i8* null, i1 true)
+// CHECK-LPAD:   %[[I1RESUME:.+]] = call i1 @llvm.coro.end(ptr null, i1 true)
 // CHECK-LPAD:   br i1  %[[I1RESUME]], label %[[EHRESUME:.+]], label
 // CHECK-LPAD: [[EHRESUME]]:
-// CHECK-LPAD-NEXT:  %[[exn:.+]] = load i8*, i8** %exn.slot, align 8
-// CHECK-LPAD-NEXT:  %[[sel:.+]] = load i32, i32* %ehselector.slot, align 4
-// CHECK-LPAD-NEXT:  %[[val1:.+]] = insertvalue { i8*, i32 } poison, i8* %[[exn]], 0
-// CHECK-LPAD-NEXT:  %[[val2:.+]] = insertvalue { i8*, i32 } %[[val1]], i32 %[[sel]], 1
-// CHECK-LPAD-NEXT:  resume { i8*, i32 } %[[val2]]
+// CHECK-LPAD-NEXT:  %[[exn:.+]] = load ptr, ptr %exn.slot, align 8
+// CHECK-LPAD-NEXT:  %[[sel:.+]] = load i32, ptr %ehselector.slot, align 4
+// CHECK-LPAD-NEXT:  %[[val1:.+]] = insertvalue { ptr, i32 } poison, ptr %[[exn]], 0
+// CHECK-LPAD-NEXT:  %[[val2:.+]] = insertvalue { ptr, i32 } %[[val1]], i32 %[[sel]], 1
+// CHECK-LPAD-NEXT:  resume { ptr, i32 } %[[val2]]

diff  --git a/clang/test/CodeGenCoroutines/coro-eh-cleanup.cpp b/clang/test/CodeGenCoroutines/coro-eh-cleanup.cpp
index 756aa453fc315..c4a6ae96f551e 100644
--- a/clang/test/CodeGenCoroutines/coro-eh-cleanup.cpp
+++ b/clang/test/CodeGenCoroutines/coro-eh-cleanup.cpp
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -std=c++20 -triple=x86_64-pc-windows-msvc18.0.0 -emit-llvm %s -o - -fexceptions -fcxx-exceptions -disable-llvm-passes | FileCheck %s
-// RUN: %clang_cc1 -no-opaque-pointers -std=c++20 -triple=x86_64-unknown-linux-gnu -emit-llvm -o - %s -fexceptions -fcxx-exceptions -disable-llvm-passes | FileCheck --check-prefix=CHECK-LPAD %s
+// RUN: %clang_cc1 -std=c++20 -triple=x86_64-pc-windows-msvc18.0.0 -emit-llvm %s -o - -fexceptions -fcxx-exceptions -disable-llvm-passes | FileCheck %s
+// RUN: %clang_cc1 -std=c++20 -triple=x86_64-unknown-linux-gnu -emit-llvm -o - %s -fexceptions -fcxx-exceptions -disable-llvm-passes | FileCheck --check-prefix=CHECK-LPAD %s
 
 namespace std {
 template <typename R, typename... T> struct coroutine_traits {
@@ -60,27 +60,27 @@ coro_t f() {
 
 // CHECK: [[COROENDBB]]:
 // CHECK-NEXT: %[[CLPAD:.+]] = cleanuppad within none
-// CHECK-NEXT: call i1 @llvm.coro.end(i8* null, i1 true) [ "funclet"(token %[[CLPAD]]) ]
+// CHECK-NEXT: call i1 @llvm.coro.end(ptr null, i1 true) [ "funclet"(token %[[CLPAD]]) ]
 // CHECK-NEXT: cleanupret from %[[CLPAD]] unwind label
 
 // CHECK-LPAD: @_Z1fv(
 // CHECK-LPAD:   invoke void @_Z9may_throwv()
 // CHECK-LPAD:       to label %[[CONT:.+]] unwind label %[[EHCLEANUP:.+]]
 // CHECK-LPAD: [[EHCLEANUP]]:
-// CHECK-LPAD:    landingpad { i8*, i32 }
+// CHECK-LPAD:    landingpad { ptr, i32 }
 // CHECK-LPAD:          catch
 // CHECK-LPAD:   call void @_ZN7CleanupD1Ev(
-// CHECK-LPAD:   call i8* @__cxa_begin_catch
+// CHECK-LPAD:   call ptr @__cxa_begin_catch
 // CHECK-LPAD:   call void @_ZN6coro_t12promise_type19unhandled_exceptionEv
 // CHECK-LPAD:   invoke void @__cxa_end_catch()
 // CHECK-LPAD:             to label %{{.+}} unwind label %[[UNWINDBB:.+]]
 
 // CHECK-LPAD: [[UNWINDBB]]:
-// CHECK-LPAD:   %[[I1RESUME:.+]] = call i1 @llvm.coro.end(i8* null, i1 true)
+// CHECK-LPAD:   %[[I1RESUME:.+]] = call i1 @llvm.coro.end(ptr null, i1 true)
 // CHECK-LPAD:   br i1  %[[I1RESUME]], label %[[EHRESUME:.+]], label
 // CHECK-LPAD: [[EHRESUME]]:
-// CHECK-LPAD-NEXT:  %[[exn:.+]] = load i8*, i8** %exn.slot, align 8
-// CHECK-LPAD-NEXT:  %[[sel:.+]] = load i32, i32* %ehselector.slot, align 4
-// CHECK-LPAD-NEXT:  %[[val1:.+]] = insertvalue { i8*, i32 } poison, i8* %[[exn]], 0
-// CHECK-LPAD-NEXT:  %[[val2:.+]] = insertvalue { i8*, i32 } %[[val1]], i32 %[[sel]], 1
-// CHECK-LPAD-NEXT:  resume { i8*, i32 } %[[val2]]
+// CHECK-LPAD-NEXT:  %[[exn:.+]] = load ptr, ptr %exn.slot, align 8
+// CHECK-LPAD-NEXT:  %[[sel:.+]] = load i32, ptr %ehselector.slot, align 4
+// CHECK-LPAD-NEXT:  %[[val1:.+]] = insertvalue { ptr, i32 } poison, ptr %[[exn]], 0
+// CHECK-LPAD-NEXT:  %[[val2:.+]] = insertvalue { ptr, i32 } %[[val1]], i32 %[[sel]], 1
+// CHECK-LPAD-NEXT:  resume { ptr, i32 } %[[val2]]

diff  --git a/clang/test/CodeGenCoroutines/coro-gro-exp-namespace.cpp b/clang/test/CodeGenCoroutines/coro-gro-exp-namespace.cpp
index b58828bfdd553..bdd2e1149bf0a 100644
--- a/clang/test/CodeGenCoroutines/coro-gro-exp-namespace.cpp
+++ b/clang/test/CodeGenCoroutines/coro-gro-exp-namespace.cpp
@@ -1,6 +1,6 @@
 // Verifies lifetime of __gro local variable
 // Verify that coroutine promise and allocated memory are freed up on exception.
-// RUN: %clang_cc1 -no-opaque-pointers -std=c++1z -fcoroutines-ts -triple=x86_64-unknown-linux-gnu -emit-llvm -o - %s -disable-llvm-passes | FileCheck %s
+// RUN: %clang_cc1 -std=c++1z -fcoroutines-ts -triple=x86_64-unknown-linux-gnu -emit-llvm -o - %s -disable-llvm-passes | FileCheck %s
 
 namespace std::experimental {
 template <typename... T> struct coroutine_traits;
@@ -50,11 +50,11 @@ int f() {
   // CHECK: %[[RetVal:.+]] = alloca i32
 
   // CHECK: %[[Size:.+]] = call i64 @llvm.coro.size.i64()
-  // CHECK: call noalias noundef nonnull i8* @_Znwm(i64 noundef %[[Size]])
+  // CHECK: call noalias noundef nonnull ptr @_Znwm(i64 noundef %[[Size]])
   // CHECK: call void @_ZNSt12experimental16coroutine_traitsIJiEE12promise_typeC1Ev(
-  // CHECK: call void @_ZNSt12experimental16coroutine_traitsIJiEE12promise_type17get_return_objectEv(%struct.GroType* sret(%struct.GroType) align {{[0-9]+}} %[[GRO:.+]],
+  // CHECK: call void @_ZNSt12experimental16coroutine_traitsIJiEE12promise_type17get_return_objectEv(ptr sret(%struct.GroType) align {{[0-9]+}} %[[GRO:.+]],
   // CHECK: %[[Conv:.+]] = call noundef i32 @_ZN7GroTypecviEv({{.*}}[[GRO]]
-  // CHECK: store i32 %[[Conv]], i32* %[[RetVal]]
+  // CHECK: store i32 %[[Conv]], ptr %[[RetVal]]
 
   Cleanup cleanup;
   doSomething();
@@ -67,10 +67,10 @@ int f() {
   // Destroy promise and free the memory.
 
   // CHECK: call void @_ZNSt12experimental16coroutine_traitsIJiEE12promise_typeD1Ev(
-  // CHECK: %[[Mem:.+]] = call i8* @llvm.coro.free(
-  // CHECK: call void @_ZdlPv(i8* noundef %[[Mem]])
+  // CHECK: %[[Mem:.+]] = call ptr @llvm.coro.free(
+  // CHECK: call void @_ZdlPv(ptr noundef %[[Mem]])
 
   // CHECK: coro.ret:
-  // CHECK:   %[[LoadRet:.+]] = load i32, i32* %[[RetVal]]
+  // CHECK:   %[[LoadRet:.+]] = load i32, ptr %[[RetVal]]
   // CHECK:   ret i32 %[[LoadRet]]
 }

diff  --git a/clang/test/CodeGenCoroutines/coro-gro.cpp b/clang/test/CodeGenCoroutines/coro-gro.cpp
index 40ce751ff098c..fad75c9433076 100644
--- a/clang/test/CodeGenCoroutines/coro-gro.cpp
+++ b/clang/test/CodeGenCoroutines/coro-gro.cpp
@@ -1,6 +1,6 @@
 // Verifies lifetime of __gro local variable
 // Verify that coroutine promise and allocated memory are freed up on exception.
-// RUN: %clang_cc1 -no-opaque-pointers -std=c++20 -triple=x86_64-unknown-linux-gnu -emit-llvm -o - %s -disable-llvm-passes | FileCheck %s
+// RUN: %clang_cc1 -std=c++20 -triple=x86_64-unknown-linux-gnu -emit-llvm -o - %s -disable-llvm-passes | FileCheck %s
 
 namespace std {
 template <typename... T> struct coroutine_traits;
@@ -48,11 +48,11 @@ int f() {
   // CHECK: %[[RetVal:.+]] = alloca i32
 
   // CHECK: %[[Size:.+]] = call i64 @llvm.coro.size.i64()
-  // CHECK: call noalias noundef nonnull i8* @_Znwm(i64 noundef %[[Size]])
+  // CHECK: call noalias noundef nonnull ptr @_Znwm(i64 noundef %[[Size]])
   // CHECK: call void @_ZNSt16coroutine_traitsIJiEE12promise_typeC1Ev(
-  // CHECK: call void @_ZNSt16coroutine_traitsIJiEE12promise_type17get_return_objectEv(%struct.GroType* sret(%struct.GroType) align {{[0-9]+}} %[[GRO:.+]],
+  // CHECK: call void @_ZNSt16coroutine_traitsIJiEE12promise_type17get_return_objectEv(ptr sret(%struct.GroType) align {{[0-9]+}} %[[GRO:.+]],
   // CHECK: %[[Conv:.+]] = call noundef i32 @_ZN7GroTypecviEv({{.*}}[[GRO]]
-  // CHECK: store i32 %[[Conv]], i32* %[[RetVal]]
+  // CHECK: store i32 %[[Conv]], ptr %[[RetVal]]
 
   Cleanup cleanup;
   doSomething();
@@ -65,10 +65,10 @@ int f() {
   // Destroy promise and free the memory.
 
   // CHECK: call void @_ZNSt16coroutine_traitsIJiEE12promise_typeD1Ev(
-  // CHECK: %[[Mem:.+]] = call i8* @llvm.coro.free(
-  // CHECK: call void @_ZdlPv(i8* noundef %[[Mem]])
+  // CHECK: %[[Mem:.+]] = call ptr @llvm.coro.free(
+  // CHECK: call void @_ZdlPv(ptr noundef %[[Mem]])
 
   // CHECK: coro.ret:
-  // CHECK:   %[[LoadRet:.+]] = load i32, i32* %[[RetVal]]
+  // CHECK:   %[[LoadRet:.+]] = load i32, ptr %[[RetVal]]
   // CHECK:   ret i32 %[[LoadRet]]
 }

diff  --git a/clang/test/CodeGenCoroutines/coro-gro2-exp-namespace.cpp b/clang/test/CodeGenCoroutines/coro-gro2-exp-namespace.cpp
index 5b73ec083ff73..10c606450d5bf 100644
--- a/clang/test/CodeGenCoroutines/coro-gro2-exp-namespace.cpp
+++ b/clang/test/CodeGenCoroutines/coro-gro2-exp-namespace.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-linux-gnu -fcoroutines-ts -std=c++14 -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -fcoroutines-ts -std=c++14 -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
 
 #include "Inputs/coroutine-exp-namespace.h"
 
@@ -33,13 +33,13 @@ struct coro {
 };
 
 // Verify that the RVO is applied.
-// CHECK-LABEL: define{{.*}} void @_Z1fi(%struct.coro* noalias sret(%struct.coro) align 8 %agg.result, i32 noundef %0)
+// CHECK-LABEL: define{{.*}} void @_Z1fi(ptr noalias sret(%struct.coro) align 8 %agg.result, i32 noundef %0)
 coro f(int) {
-  // CHECK: %call = call noalias noundef nonnull i8* @_Znwm(
+  // CHECK: %call = call noalias noundef nonnull ptr @_Znwm(
   // CHECK-NEXT: br label %[[CoroInit:.*]]
 
   // CHECK: {{.*}}[[CoroInit]]:
-  // CHECK: call void @{{.*get_return_objectEv}}(%struct.coro* sret(%struct.coro) align 8 %agg.result
+  // CHECK: call void @{{.*get_return_objectEv}}(ptr sret(%struct.coro) align 8 %agg.result
   co_return;
 }
 
@@ -61,19 +61,19 @@ struct coro_two {
 };
 
 // Verify that the NRVO is applied to the Gro object.
-// CHECK-LABEL: define{{.*}} void @_Z1hi(%struct.coro_two* noalias sret(%struct.coro_two) align 8 %agg.result, i32 noundef %0)
+// CHECK-LABEL: define{{.*}} void @_Z1hi(ptr noalias sret(%struct.coro_two) align 8 %agg.result, i32 noundef %0)
 coro_two h(int) {
 
-  // CHECK: %call = call noalias noundef i8* @_ZnwmRKSt9nothrow_t
-  // CHECK-NEXT: %[[CheckNull:.*]] = icmp ne i8* %call, null
+  // CHECK: %call = call noalias noundef ptr @_ZnwmRKSt9nothrow_t
+  // CHECK-NEXT: %[[CheckNull:.*]] = icmp ne ptr %call, null
   // CHECK-NEXT: br i1 %[[CheckNull]], label %[[InitOnSuccess:.*]], label %[[InitOnFailure:.*]]
 
   // CHECK: {{.*}}[[InitOnFailure]]:
-  // CHECK-NEXT: call void @{{.*get_return_object_on_allocation_failureEv}}(%struct.coro_two* sret(%struct.coro_two) align 8 %agg.result
+  // CHECK-NEXT: call void @{{.*get_return_object_on_allocation_failureEv}}(ptr sret(%struct.coro_two) align 8 %agg.result
   // CHECK-NEXT: br label %[[RetLabel:.*]]
 
   // CHECK: {{.*}}[[InitOnSuccess]]:
-  // CHECK: call void @{{.*get_return_objectEv}}(%struct.coro_two* sret(%struct.coro_two) align 8 %agg.result
+  // CHECK: call void @{{.*get_return_objectEv}}(ptr sret(%struct.coro_two) align 8 %agg.result
 
   // CHECK: [[RetLabel]]:
   // CHECK-NEXT: ret void

diff  --git a/clang/test/CodeGenCoroutines/coro-gro2.cpp b/clang/test/CodeGenCoroutines/coro-gro2.cpp
index ba3fc9b884d81..bb52d21af746c 100644
--- a/clang/test/CodeGenCoroutines/coro-gro2.cpp
+++ b/clang/test/CodeGenCoroutines/coro-gro2.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-linux-gnu -fcoroutines-ts -std=c++20 -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -fcoroutines-ts -std=c++20 -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
 
 #include "Inputs/coroutine.h"
 
@@ -34,13 +34,13 @@ struct coro {
 };
 
 // Verify that the RVO is applied.
-// CHECK-LABEL: define{{.*}} void @_Z1fi(%struct.coro* noalias sret(%struct.coro) align 8 %agg.result, i32 noundef %0)
+// CHECK-LABEL: define{{.*}} void @_Z1fi(ptr noalias sret(%struct.coro) align 8 %agg.result, i32 noundef %0)
 coro f(int) {
-// CHECK: %call = call noalias noundef nonnull i8* @_Znwm(
+// CHECK: %call = call noalias noundef nonnull ptr @_Znwm(
 // CHECK-NEXT: br label %[[CoroInit:.*]]
 
 // CHECK: {{.*}}[[CoroInit]]:
-// CHECK: call void @{{.*get_return_objectEv}}(%struct.coro* sret(%struct.coro) align 8 %agg.result
+// CHECK: call void @{{.*get_return_objectEv}}(ptr sret(%struct.coro) align 8 %agg.result
 co_return;
 }
 
@@ -63,19 +63,19 @@ struct coro_two {
 };
 
 // Verify that the RVO is applied.
-// CHECK-LABEL: define{{.*}} void @_Z1hi(%struct.coro_two* noalias sret(%struct.coro_two) align 8 %agg.result, i32 noundef %0)
+// CHECK-LABEL: define{{.*}} void @_Z1hi(ptr noalias sret(%struct.coro_two) align 8 %agg.result, i32 noundef %0)
 coro_two h(int) {
 
-  // CHECK: %call = call noalias noundef i8* @_ZnwmRKSt9nothrow_t
-  // CHECK-NEXT: %[[CheckNull:.*]] = icmp ne i8* %call, null
+  // CHECK: %call = call noalias noundef ptr @_ZnwmRKSt9nothrow_t
+  // CHECK-NEXT: %[[CheckNull:.*]] = icmp ne ptr %call, null
   // CHECK-NEXT: br i1 %[[CheckNull]], label %[[InitOnSuccess:.*]], label %[[InitOnFailure:.*]]
 
   // CHECK: {{.*}}[[InitOnFailure]]:
-  // CHECK-NEXT: call void @{{.*get_return_object_on_allocation_failureEv}}(%struct.coro_two* sret(%struct.coro_two) align 8 %agg.result
+  // CHECK-NEXT: call void @{{.*get_return_object_on_allocation_failureEv}}(ptr sret(%struct.coro_two) align 8 %agg.result
   // CHECK-NEXT: br label %[[RetLabel:.*]]
 
   // CHECK: {{.*}}[[InitOnSuccess]]:
-  // CHECK: call void @{{.*get_return_objectEv}}(%struct.coro_two* sret(%struct.coro_two) align 8 %agg.result
+  // CHECK: call void @{{.*get_return_objectEv}}(ptr sret(%struct.coro_two) align 8 %agg.result
 
   // CHECK: [[RetLabel]]:
   // CHECK-NEXT: ret void

diff  --git a/clang/test/CodeGenCoroutines/coro-halo.cpp b/clang/test/CodeGenCoroutines/coro-halo.cpp
index 14ae53c41aebe..6244f130b7be2 100644
--- a/clang/test/CodeGenCoroutines/coro-halo.cpp
+++ b/clang/test/CodeGenCoroutines/coro-halo.cpp
@@ -1,5 +1,5 @@
 // This tests that the coroutine heap allocation elision optimization could happen succesfully.
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-linux-gnu -std=c++20 -O2 -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -std=c++20 -O2 -emit-llvm %s -o - | FileCheck %s
 
 #include "Inputs/coroutine.h"
 #include "Inputs/numeric.h"

diff  --git a/clang/test/CodeGenCoroutines/coro-promise-dtor-exp-namespace.cpp b/clang/test/CodeGenCoroutines/coro-promise-dtor-exp-namespace.cpp
index 75c0e29c3eaf3..30b655c03bb2b 100644
--- a/clang/test/CodeGenCoroutines/coro-promise-dtor-exp-namespace.cpp
+++ b/clang/test/CodeGenCoroutines/coro-promise-dtor-exp-namespace.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -std=c++14 -fcoroutines-ts -triple=x86_64-pc-windows-msvc18.0.0 -emit-llvm -o - %s -fexceptions -fcxx-exceptions -disable-llvm-passes | FileCheck %s
+// RUN: %clang_cc1 -std=c++14 -fcoroutines-ts -triple=x86_64-pc-windows-msvc18.0.0 -emit-llvm -o - %s -fexceptions -fcxx-exceptions -disable-llvm-passes | FileCheck %s
 // -triple=x86_64-unknown-linux-gnu
 
 #include "Inputs/coroutine-exp-namespace.h"
@@ -32,7 +32,7 @@ coro_t f() {
 
 // CHECK-LABEL: define dso_local void @"?f@@YA?AUcoro_t@@XZ"(
 
-// CHECK:  invoke noundef %"struct.coro_t::promise_type"* @"??0promise_type at coro_t@@QEAA at XZ"(
+// CHECK:  invoke noundef ptr @"??0promise_type at coro_t@@QEAA at XZ"(
 // CHECK:  invoke void @"?get_return_object at promise_type@coro_t@@QEAA?AU2 at XZ"(
 
 // CHECK:  call void @"??1promise_type at coro_t@@QEAA at XZ"

diff  --git a/clang/test/CodeGenCoroutines/coro-promise-dtor.cpp b/clang/test/CodeGenCoroutines/coro-promise-dtor.cpp
index 5371b7d9772b9..afaaf876a8103 100644
--- a/clang/test/CodeGenCoroutines/coro-promise-dtor.cpp
+++ b/clang/test/CodeGenCoroutines/coro-promise-dtor.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -std=c++20 -triple=x86_64-pc-windows-msvc18.0.0 -emit-llvm -o - %s -fexceptions -fcxx-exceptions -disable-llvm-passes | FileCheck %s
+// RUN: %clang_cc1 -std=c++20 -triple=x86_64-pc-windows-msvc18.0.0 -emit-llvm -o - %s -fexceptions -fcxx-exceptions -disable-llvm-passes | FileCheck %s
 // -triple=x86_64-unknown-linux-gnu
 
 #include "Inputs/coroutine.h"
@@ -28,7 +28,7 @@ coro_t f() {
 
 // CHECK-LABEL: define dso_local void @"?f@@YA?AUcoro_t@@XZ"(
 
-// CHECK:  invoke noundef %"struct.coro_t::promise_type"* @"??0promise_type at coro_t@@QEAA at XZ"(
+// CHECK:  invoke noundef ptr @"??0promise_type at coro_t@@QEAA at XZ"(
 // CHECK:  invoke void @"?get_return_object at promise_type@coro_t@@QEAA?AU2 at XZ"(
 
 // CHECK:  call void @"??1promise_type at coro_t@@QEAA at XZ"

diff  --git a/clang/test/CodeGenCoroutines/coro-ret-void-exp-namespace.cpp b/clang/test/CodeGenCoroutines/coro-ret-void-exp-namespace.cpp
index 9968aa03eaa17..44c09616097f8 100644
--- a/clang/test/CodeGenCoroutines/coro-ret-void-exp-namespace.cpp
+++ b/clang/test/CodeGenCoroutines/coro-ret-void-exp-namespace.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -std=c++14 -fcoroutines-ts -triple=x86_64-unknown-linux-gnu -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
+// RUN: %clang_cc1 -std=c++14 -fcoroutines-ts -triple=x86_64-unknown-linux-gnu -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
 
 #include "Inputs/coroutine-exp-namespace.h"
 
@@ -18,8 +18,8 @@ coro1 f() {
 }
 
 // CHECK-LABEL: define{{.*}} void @_Z1fv(
-// CHECK: call void @_ZNSt12experimental13coroutines_v113suspend_never12await_resumeEv(%"struct.std::experimental::coroutines_v1::suspend_never"*
-// CHECK: call void @_ZN5coro112promise_type11return_voidEv(%"struct.coro1::promise_type"* {{[^,]*}} %__promise)
+// CHECK: call void @_ZNSt12experimental13coroutines_v113suspend_never12await_resumeEv(ptr
+// CHECK: call void @_ZN5coro112promise_type11return_voidEv(ptr {{[^,]*}} %__promise)
 
 struct A {
   A();
@@ -31,9 +31,9 @@ coro1 f2() {
 }
 
 // CHECK-LABEL: define{{.*}} void @_Z2f2v(
-// CHECK: call void @_ZN1AC1Ev(%struct.A* {{[^,]*}} %[[AVar:.*]])
-// CHECK-NEXT: call void @_ZN1AD1Ev(%struct.A* {{[^,]*}} %[[AVar]])
-// CHECK-NEXT: call void @_ZN5coro112promise_type11return_voidEv(%"struct.coro1::promise_type"*
+// CHECK: call void @_ZN1AC1Ev(ptr {{[^,]*}} %[[AVar:.*]])
+// CHECK-NEXT: call void @_ZN1AD1Ev(ptr {{[^,]*}} %[[AVar]])
+// CHECK-NEXT: call void @_ZN5coro112promise_type11return_voidEv(ptr
 
 struct coro2 {
   struct promise_type {
@@ -49,5 +49,5 @@ coro2 g() {
 }
 
 // CHECK-LABEL: define{{.*}} void @_Z1gv(
-// CHECK: call void @_ZNSt12experimental13coroutines_v113suspend_never12await_resumeEv(%"struct.std::experimental::coroutines_v1::suspend_never"*
-// CHECK: call void @_ZN5coro212promise_type12return_valueEi(%"struct.coro2::promise_type"* {{[^,]*}} %__promise, i32 noundef 42)
+// CHECK: call void @_ZNSt12experimental13coroutines_v113suspend_never12await_resumeEv(ptr
+// CHECK: call void @_ZN5coro212promise_type12return_valueEi(ptr {{[^,]*}} %__promise, i32 noundef 42)

diff  --git a/clang/test/CodeGenCoroutines/coro-ret-void.cpp b/clang/test/CodeGenCoroutines/coro-ret-void.cpp
index ae139afabc4d4..73caba114cc60 100644
--- a/clang/test/CodeGenCoroutines/coro-ret-void.cpp
+++ b/clang/test/CodeGenCoroutines/coro-ret-void.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -std=c++20 -triple=x86_64-unknown-linux-gnu -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
+// RUN: %clang_cc1 -std=c++20 -triple=x86_64-unknown-linux-gnu -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
 
 #include "Inputs/coroutine.h"
 
@@ -17,8 +17,8 @@ coro1 f() {
 }
 
 // CHECK-LABEL: define{{.*}} void @_Z1fv(
-// CHECK: call void @_ZNSt13suspend_never12await_resumeEv(%"struct.std::suspend_never"*
-// CHECK: call void @_ZN5coro112promise_type11return_voidEv(%"struct.coro1::promise_type"* {{[^,]*}} %__promise)
+// CHECK: call void @_ZNSt13suspend_never12await_resumeEv(ptr
+// CHECK: call void @_ZN5coro112promise_type11return_voidEv(ptr {{[^,]*}} %__promise)
 
 struct A {
   A();
@@ -30,9 +30,9 @@ coro1 f2() {
 }
 
 // CHECK-LABEL: define{{.*}} void @_Z2f2v(
-// CHECK: call void @_ZN1AC1Ev(%struct.A* {{[^,]*}} %[[AVar:.*]])
-// CHECK-NEXT: call void @_ZN1AD1Ev(%struct.A* {{[^,]*}} %[[AVar]])
-// CHECK-NEXT: call void @_ZN5coro112promise_type11return_voidEv(%"struct.coro1::promise_type"*
+// CHECK: call void @_ZN1AC1Ev(ptr {{[^,]*}} %[[AVar:.*]])
+// CHECK-NEXT: call void @_ZN1AD1Ev(ptr {{[^,]*}} %[[AVar]])
+// CHECK-NEXT: call void @_ZN5coro112promise_type11return_voidEv(ptr
 
 struct coro2 {
   struct promise_type {
@@ -49,5 +49,5 @@ coro2 g() {
 }
 
 // CHECK-LABEL: define{{.*}} void @_Z1gv(
-// CHECK: call void @_ZNSt13suspend_never12await_resumeEv(%"struct.std::suspend_never"*
-// CHECK: call void @_ZN5coro212promise_type12return_valueEi(%"struct.coro2::promise_type"* {{[^,]*}} %__promise, i32 noundef 42)
+// CHECK: call void @_ZNSt13suspend_never12await_resumeEv(ptr
+// CHECK: call void @_ZN5coro212promise_type12return_valueEi(ptr {{[^,]*}} %__promise, i32 noundef 42)

diff  --git a/clang/test/CodeGenCoroutines/coro-return-exp-namespace.cpp b/clang/test/CodeGenCoroutines/coro-return-exp-namespace.cpp
index 7ab8d2c3dd4e2..39de61067e2c1 100644
--- a/clang/test/CodeGenCoroutines/coro-return-exp-namespace.cpp
+++ b/clang/test/CodeGenCoroutines/coro-return-exp-namespace.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-linux-gnu -fcoroutines-ts -std=c++1z -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -fcoroutines-ts -std=c++1z -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
 
 namespace std::experimental {
 template <typename... T> struct coroutine_traits;
@@ -33,8 +33,8 @@ template <> struct std::experimental::coroutine_traits<void> {
 // CHECK-LABEL: f0(
 extern "C" void f0() {
   // CHECK: %__promise = alloca %"struct.std::experimental::coroutine_traits<void>::promise_type"
-  // CHECK: %call = call noalias noundef nonnull i8* @_Znwm(
-  // CHECK: call void @_ZNSt12experimental16coroutine_traitsIJvEE12promise_type11return_voidEv(%"struct.std::experimental::coroutine_traits<void>::promise_type"* {{[^,]*}} %__promise)
+  // CHECK: %call = call noalias noundef nonnull ptr @_Znwm(
+  // CHECK: call void @_ZNSt12experimental16coroutine_traitsIJvEE12promise_type11return_voidEv(ptr {{[^,]*}} %__promise)
   // CHECK: call void @_ZdlPv
   co_return;
 }
@@ -52,8 +52,8 @@ struct std::experimental::coroutine_traits<int> {
 // CHECK-LABEL: f1(
 extern "C" int f1() {
   // CHECK: %__promise = alloca %"struct.std::experimental::coroutine_traits<int>::promise_type"
-  // CHECK: %call = call noalias noundef nonnull i8* @_Znwm(
-  // CHECK: call void @_ZNSt12experimental16coroutine_traitsIJiEE12promise_type12return_valueEi(%"struct.std::experimental::coroutine_traits<int>::promise_type"* {{[^,]*}} %__promise, i32 noundef 42)
+  // CHECK: %call = call noalias noundef nonnull ptr @_Znwm(
+  // CHECK: call void @_ZNSt12experimental16coroutine_traitsIJiEE12promise_type12return_valueEi(ptr {{[^,]*}} %__promise, i32 noundef 42)
   // CHECK: call void @_ZdlPv
   co_return 42;
 }

diff  --git a/clang/test/CodeGenCoroutines/coro-return-voidtype-initlist-exp-namespace.cpp b/clang/test/CodeGenCoroutines/coro-return-voidtype-initlist-exp-namespace.cpp
index 2751baebf2625..fbcab6386103c 100644
--- a/clang/test/CodeGenCoroutines/coro-return-voidtype-initlist-exp-namespace.cpp
+++ b/clang/test/CodeGenCoroutines/coro-return-voidtype-initlist-exp-namespace.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-linux-gnu -fcoroutines-ts -std=c++1z -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -fcoroutines-ts -std=c++1z -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
 
 namespace std {
 template <typename a>
@@ -72,7 +72,7 @@ template <typename bc>
 f<std::b<std::c<int, int>>> J<bc>::bo() {
   std::c<int> bu;
   int bw(0);
-  // CHECK: void @_ZN1j12return_valueESt1bISt1cIiiEE(%struct.j* {{[^,]*}} %__promise)
+  // CHECK: void @_ZN1j12return_valueESt1bISt1cIiiEE(ptr {{[^,]*}} %__promise)
   co_return {0, co_await ax(bu, bw)};
 }
 void bh() {

diff  --git a/clang/test/CodeGenCoroutines/coro-return-voidtype-initlist.cpp b/clang/test/CodeGenCoroutines/coro-return-voidtype-initlist.cpp
index b6cac61031598..66dfdedf958e3 100644
--- a/clang/test/CodeGenCoroutines/coro-return-voidtype-initlist.cpp
+++ b/clang/test/CodeGenCoroutines/coro-return-voidtype-initlist.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-linux-gnu -std=c++20 -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -std=c++20 -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
 
 namespace std {
 template <typename a>
@@ -70,7 +70,7 @@ template <typename bc>
 f<std::b<std::c<int, int>>> J<bc>::bo() {
   std::c<int> bu;
   int bw(0);
-  // CHECK: void @_ZN1j12return_valueESt1bISt1cIiiEE(%struct.j* {{[^,]*}} %__promise)
+  // CHECK: void @_ZN1j12return_valueESt1bISt1cIiiEE(ptr {{[^,]*}} %__promise)
   co_return{0, co_await ax(bu, bw)};
 }
 void bh() {

diff  --git a/clang/test/CodeGenCoroutines/coro-return.cpp b/clang/test/CodeGenCoroutines/coro-return.cpp
index b6260bbcc6c46..6f6cd712aa6fb 100644
--- a/clang/test/CodeGenCoroutines/coro-return.cpp
+++ b/clang/test/CodeGenCoroutines/coro-return.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-linux-gnu -std=c++20 -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -std=c++20 -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
 
 namespace std {
 template <typename... T> struct coroutine_traits;
@@ -33,8 +33,8 @@ template <> struct std::coroutine_traits<void> {
 // CHECK-LABEL: f0(
 extern "C" void f0() {
   // CHECK: %__promise = alloca %"struct.std::coroutine_traits<void>::promise_type"
-  // CHECK: %call = call noalias noundef nonnull i8* @_Znwm(
-  // CHECK: call void @_ZNSt16coroutine_traitsIJvEE12promise_type11return_voidEv(%"struct.std::coroutine_traits<void>::promise_type"* {{[^,]*}} %__promise)
+  // CHECK: %call = call noalias noundef nonnull ptr @_Znwm(
+  // CHECK: call void @_ZNSt16coroutine_traitsIJvEE12promise_type11return_voidEv(ptr {{[^,]*}} %__promise)
   // CHECK: call void @_ZdlPv
   co_return;
 }
@@ -52,8 +52,8 @@ struct std::coroutine_traits<int> {
 // CHECK-LABEL: f1(
 extern "C" int f1() {
   // CHECK: %__promise = alloca %"struct.std::coroutine_traits<int>::promise_type"
-  // CHECK: %call = call noalias noundef nonnull i8* @_Znwm(
-  // CHECK: call void @_ZNSt16coroutine_traitsIJiEE12promise_type12return_valueEi(%"struct.std::coroutine_traits<int>::promise_type"* {{[^,]*}} %__promise, i32 noundef 42)
+  // CHECK: %call = call noalias noundef nonnull ptr @_Znwm(
+  // CHECK: call void @_ZNSt16coroutine_traitsIJiEE12promise_type12return_valueEi(ptr {{[^,]*}} %__promise, i32 noundef 42)
   // CHECK: call void @_ZdlPv
   co_return 42;
 }

diff  --git a/clang/test/CodeGenCoroutines/coro-symmetric-transfer-01-exp-namespace.cpp b/clang/test/CodeGenCoroutines/coro-symmetric-transfer-01-exp-namespace.cpp
index 77da8f7720ab6..c60c2e0e271a7 100644
--- a/clang/test/CodeGenCoroutines/coro-symmetric-transfer-01-exp-namespace.cpp
+++ b/clang/test/CodeGenCoroutines/coro-symmetric-transfer-01-exp-namespace.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -no-enable-noundef-analysis -triple x86_64-unknown-linux-gnu -fcoroutines-ts -std=c++14 -O0 -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
+// RUN: %clang_cc1 -no-enable-noundef-analysis -triple x86_64-unknown-linux-gnu -fcoroutines-ts -std=c++14 -O0 -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
 // RUN: %clang -fcoroutines-ts -std=c++14 -O0 -emit-llvm -c  %s -o %t -Xclang -disable-llvm-passes && %clang -c %t
 
 #include "Inputs/coroutine-exp-namespace.h"
@@ -51,13 +51,11 @@ detached_task foo() {
 
 // check that the lifetime of the coroutine handle used to obtain the address is contained within single basic block, and hence does not live across suspension points.
 // CHECK-LABEL: final.suspend:
-// CHECK:         %{{.+}} = call token @llvm.coro.save(i8* null)
-// CHECK:         %[[HDL_CAST1:.+]] = bitcast %"struct.std::experimental::coroutines_v1::coroutine_handle.0"* %[[HDL:.+]] to i8*
-// CHECK:         call void @llvm.lifetime.start.p0i8(i64 8, i8* %[[HDL_CAST1]])
-// CHECK:         %[[CALL:.+]] = call i8* @_ZN13detached_task12promise_type13final_awaiter13await_suspendENSt12experimental13coroutines_v116coroutine_handleIS0_EE(
-// CHECK:         %[[HDL_CAST2:.+]] = getelementptr inbounds %"struct.std::experimental::coroutines_v1::coroutine_handle.0", %"struct.std::experimental::coroutines_v1::coroutine_handle.0"* %[[HDL]], i32 0, i32 0
-// CHECK:         store i8* %[[CALL]], i8** %[[HDL_CAST2]], align 8
-// CHECK:         %[[HDL_TRANSFER:.+]] = call i8* @_ZNKSt12experimental13coroutines_v116coroutine_handleIvE7addressEv(%"struct.std::experimental::coroutines_v1::coroutine_handle.0"* nonnull align 8 dereferenceable(8) %[[HDL]])
-// CHECK:         %[[HDL_CAST3:.+]] = bitcast %"struct.std::experimental::coroutines_v1::coroutine_handle.0"* %[[HDL]] to i8*
-// CHECK:         call void @llvm.lifetime.end.p0i8(i64 8, i8* %[[HDL_CAST3]])
-// CHECK:         call void @llvm.coro.resume(i8* %[[HDL_TRANSFER]])
+// CHECK:         %{{.+}} = call token @llvm.coro.save(ptr null)
+// CHECK:         call void @llvm.lifetime.start.p0(i64 8, ptr %[[HDL:.+]])
+// CHECK:         %[[CALL:.+]] = call ptr @_ZN13detached_task12promise_type13final_awaiter13await_suspendENSt12experimental13coroutines_v116coroutine_handleIS0_EE(
+// CHECK:         %[[HDL_CAST2:.+]] = getelementptr inbounds %"struct.std::experimental::coroutines_v1::coroutine_handle.0", ptr %[[HDL]], i32 0, i32 0
+// CHECK:         store ptr %[[CALL]], ptr %[[HDL_CAST2]], align 8
+// CHECK:         %[[HDL_TRANSFER:.+]] = call ptr @_ZNKSt12experimental13coroutines_v116coroutine_handleIvE7addressEv(ptr nonnull align 8 dereferenceable(8) %[[HDL]])
+// CHECK:         call void @llvm.lifetime.end.p0(i64 8, ptr %[[HDL]])
+// CHECK:         call void @llvm.coro.resume(ptr %[[HDL_TRANSFER]])

diff  --git a/clang/test/CodeGenCoroutines/coro-symmetric-transfer-01.cpp b/clang/test/CodeGenCoroutines/coro-symmetric-transfer-01.cpp
index 79c8d48c3c478..c0b9e9ee2c558 100644
--- a/clang/test/CodeGenCoroutines/coro-symmetric-transfer-01.cpp
+++ b/clang/test/CodeGenCoroutines/coro-symmetric-transfer-01.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-linux-gnu -std=c++20 -O0 -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -std=c++20 -O0 -emit-llvm %s -o - -disable-llvm-passes | FileCheck %s
 // RUN: %clang -std=c++20 -O0 -emit-llvm -c  %s -o %t -Xclang -disable-llvm-passes && %clang -c %t
 
 #include "Inputs/coroutine.h"
@@ -49,13 +49,11 @@ detached_task foo() {
 
 // check that the lifetime of the coroutine handle used to obtain the address is contained within single basic block, and hence does not live across suspension points.
 // CHECK-LABEL: final.suspend:
-// CHECK:         %{{.+}} = call token @llvm.coro.save(i8* null)
-// CHECK:         %[[HDL_CAST1:.+]] = bitcast %"struct.std::coroutine_handle.0"* %[[HDL:.+]] to i8*
-// CHECK:         call void @llvm.lifetime.start.p0i8(i64 8, i8* %[[HDL_CAST1]])
-// CHECK:         %[[CALL:.+]] = call i8* @_ZN13detached_task12promise_type13final_awaiter13await_suspendESt16coroutine_handleIS0_E(
-// CHECK:         %[[HDL_CAST2:.+]] = getelementptr inbounds %"struct.std::coroutine_handle.0", %"struct.std::coroutine_handle.0"* %[[HDL]], i32 0, i32 0
-// CHECK:         store i8* %[[CALL]], i8** %[[HDL_CAST2]], align 8
-// CHECK:         %[[HDL_TRANSFER:.+]] = call noundef i8* @_ZNKSt16coroutine_handleIvE7addressEv(%"struct.std::coroutine_handle.0"* noundef {{.*}}%[[HDL]])
-// CHECK:         %[[HDL_CAST3:.+]] = bitcast %"struct.std::coroutine_handle.0"* %[[HDL]] to i8*
-// CHECK:         call void @llvm.lifetime.end.p0i8(i64 8, i8* %[[HDL_CAST3]])
-// CHECK:         call void @llvm.coro.resume(i8* %[[HDL_TRANSFER]])
+// CHECK:         %{{.+}} = call token @llvm.coro.save(ptr null)
+// CHECK:         call void @llvm.lifetime.start.p0(i64 8, ptr %[[HDL:.+]])
+// CHECK:         %[[CALL:.+]] = call ptr @_ZN13detached_task12promise_type13final_awaiter13await_suspendESt16coroutine_handleIS0_E(
+// CHECK:         %[[HDL_CAST2:.+]] = getelementptr inbounds %"struct.std::coroutine_handle.0", ptr %[[HDL]], i32 0, i32 0
+// CHECK:         store ptr %[[CALL]], ptr %[[HDL_CAST2]], align 8
+// CHECK:         %[[HDL_TRANSFER:.+]] = call noundef ptr @_ZNKSt16coroutine_handleIvE7addressEv(ptr noundef {{.*}}%[[HDL]])
+// CHECK:         call void @llvm.lifetime.end.p0(i64 8, ptr %[[HDL]])
+// CHECK:         call void @llvm.coro.resume(ptr %[[HDL_TRANSFER]])

diff  --git a/clang/test/CodeGenHIP/hipspv-addr-spaces.cpp b/clang/test/CodeGenHIP/hipspv-addr-spaces.cpp
index 486e1606e7ba3..9fcdc460482e7 100644
--- a/clang/test/CodeGenHIP/hipspv-addr-spaces.cpp
+++ b/clang/test/CodeGenHIP/hipspv-addr-spaces.cpp
@@ -1,11 +1,11 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple spirv64 -x hip -emit-llvm -fcuda-is-device \
+// RUN: %clang_cc1 -triple spirv64 -x hip -emit-llvm -fcuda-is-device \
 // RUN:   -o - %s | FileCheck %s
 
 #define __device__ __attribute__((device))
 #define __shared__ __attribute__((shared))
 #define __constant__ __attribute__((constant))
 
-// CHECK: %struct.foo_t = type { i32, i32 addrspace(4)* }
+// CHECK: %struct.foo_t = type { i32, ptr addrspace(4) }
 
 // CHECK: @d ={{.*}} addrspace(1) externally_initialized global
 __device__ int d;
@@ -25,30 +25,30 @@ __device__ struct foo_t {
 // Check literals are placed in address space 1 (CrossWorkGroup/__global).
 // CHECK: @.str ={{.*}} unnamed_addr addrspace(1) constant
 
-// CHECK: define{{.*}} spir_func noundef i32 addrspace(4)* @_Z3barPi(i32 addrspace(4)*
+// CHECK: define{{.*}} spir_func noundef ptr addrspace(4) @_Z3barPi(ptr addrspace(4)
 __device__ int* bar(int *x) {
   return x;
 }
 
-// CHECK: define{{.*}} spir_func noundef i32 addrspace(4)* @_Z5baz_dv()
+// CHECK: define{{.*}} spir_func noundef ptr addrspace(4) @_Z5baz_dv()
 __device__ int* baz_d() {
-  // CHECK: ret i32 addrspace(4)* addrspacecast (i32 addrspace(1)* @d to i32 addrspace(4)*
+  // CHECK: ret ptr addrspace(4) addrspacecast (ptr addrspace(1) @d to ptr addrspace(4)
   return &d;
 }
 
-// CHECK: define{{.*}} spir_func noundef i32 addrspace(4)* @_Z5baz_cv()
+// CHECK: define{{.*}} spir_func noundef ptr addrspace(4) @_Z5baz_cv()
 __device__ int* baz_c() {
-  // CHECK: ret i32 addrspace(4)* addrspacecast (i32 addrspace(1)* @c to i32 addrspace(4)*
+  // CHECK: ret ptr addrspace(4) addrspacecast (ptr addrspace(1) @c to ptr addrspace(4)
   return &c;
 }
 
-// CHECK: define{{.*}} spir_func noundef i32 addrspace(4)* @_Z5baz_sv()
+// CHECK: define{{.*}} spir_func noundef ptr addrspace(4) @_Z5baz_sv()
 __device__ int* baz_s() {
-  // CHECK: ret i32 addrspace(4)* addrspacecast (i32 addrspace(3)* @s to i32 addrspace(4)*
+  // CHECK: ret ptr addrspace(4) addrspacecast (ptr addrspace(3) @s to ptr addrspace(4)
   return &s;
 }
 
-// CHECK: define{{.*}} spir_func noundef i8 addrspace(4)* @_Z3quzv()
+// CHECK: define{{.*}} spir_func noundef ptr addrspace(4) @_Z3quzv()
 __device__ const char* quz() {
   return "abc";
 }

diff  --git a/clang/test/CodeGenHIP/hipspv-kernel.cpp b/clang/test/CodeGenHIP/hipspv-kernel.cpp
index cc4fd30a0fbd2..b9a6df12c7968 100644
--- a/clang/test/CodeGenHIP/hipspv-kernel.cpp
+++ b/clang/test/CodeGenHIP/hipspv-kernel.cpp
@@ -1,9 +1,9 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple spirv64 -x hip -emit-llvm -fcuda-is-device \
+// RUN: %clang_cc1 -triple spirv64 -x hip -emit-llvm -fcuda-is-device \
 // RUN:   -o - %s | FileCheck %s
 
 #define __global__ __attribute__((global))
 
-// CHECK: define {{.*}}spir_kernel void @_Z3fooPff(float addrspace(1)* {{.*}}, float {{.*}})
+// CHECK: define {{.*}}spir_kernel void @_Z3fooPff(ptr addrspace(1) {{.*}}, float {{.*}})
 __global__ void foo(float *a, float b) {
   *a = b;
 }

diff  --git a/clang/test/CodeGenHIP/printf.cpp b/clang/test/CodeGenHIP/printf.cpp
index b9146310ddd0d..8c8b801dbbf56 100644
--- a/clang/test/CodeGenHIP/printf.cpp
+++ b/clang/test/CodeGenHIP/printf.cpp
@@ -1,5 +1,5 @@
 // REQUIRES: amdgpu-registered-target
-// RUN: %clang_cc1 -no-opaque-pointers -triple amdgcn-amd-amdhsa -x hip -emit-llvm -fcuda-is-device \
+// RUN: %clang_cc1 -triple amdgcn-amd-amdhsa -x hip -emit-llvm -fcuda-is-device \
 // RUN:   -o - %s | FileCheck --enable-var-scope %s
 
 #define __device__ __attribute__((device))
@@ -21,7 +21,7 @@ __device__ int foo1() {
 // CHECK: [[APPEND5:%.*]] = call i64 @__ockl_printf_append_args(i64 [[APPEND4]], i32 1, i64 4, i64 0, i64 0, i64 0, i64 0, i64 0, i64 0, i32 0)
 // CHECK: [[STRLEN2:%.*]] = phi i64 [ %{{[^,]*}}, %{{[^ ]*}} ], [ 0, %{{[^ ]*}} ]
 // CHECK: [[APPEND6:%.*]] = call i64 @__ockl_printf_append_string_n(i64 [[APPEND5]], {{.*}}, i64 [[STRLEN2]], i32 0)
-// CHECK: [[PTR2INT:%.*]] = ptrtoint i8* %{{.*}} to i64
+// CHECK: [[PTR2INT:%.*]] = ptrtoint ptr %{{.*}} to i64
 // CHECK: [[APPEND7:%.*]] = call i64 @__ockl_printf_append_args(i64 [[APPEND6]], i32 1, i64 [[PTR2INT]], i64 0, i64 0, i64 0, i64 0, i64 0, i64 0, i32 1)
 // CHECK: [[RETURN:%.*]]  = trunc i64 [[APPEND7]] to i32
 // CHECK: ret i32 [[RETURN]]
@@ -38,7 +38,7 @@ __device__ int foo2() {
 // CHECK: [[APPEND1:%.*]] = call i64 @__ockl_printf_append_string_n(i64 [[BEGIN]], {{.*}}, i64 [[STRLEN1]], i32 0)
 // CHECK: [[STRLEN2:%.*]] = phi i64 [ %{{[^,]*}}, %{{[^ ]*}} ], [ 0, %{{[^ ]*}} ]
 // CHECK: [[APPEND2:%.*]] = call i64 @__ockl_printf_append_string_n(i64 [[APPEND1]], {{.*}}, i64 [[STRLEN2]], i32 0)
-// CHECK: [[PTR2INT:%.*]] = ptrtoint i8* %{{.*}} to i64
+// CHECK: [[PTR2INT:%.*]] = ptrtoint ptr %{{.*}} to i64
 // CHECK: [[APPEND3:%.*]] = call i64 @__ockl_printf_append_args(i64 [[APPEND2]], i32 1, i64 [[PTR2INT]], i64 0, i64 0, i64 0, i64 0, i64 0, i64 0, i32 1)
 // CHECK: [[RETURN:%.*]]  = trunc i64 [[APPEND3]] to i32
 // CHECK: ret i32 [[RETURN]]

diff  --git a/clang/test/CodeGenObjC/2010-02-01-utf16-with-null.m b/clang/test/CodeGenObjC/2010-02-01-utf16-with-null.m
index 290b7a3eac0c3..811e8a3a4ea11 100644
--- a/clang/test/CodeGenObjC/2010-02-01-utf16-with-null.m
+++ b/clang/test/CodeGenObjC/2010-02-01-utf16-with-null.m
@@ -1,7 +1,7 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-apple-darwin -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -triple i686-apple-darwin -emit-llvm %s -o - | FileCheck %s
 // rdar://7589850
 
 // CHECK: @.str = private unnamed_addr constant [9 x i16] [i16 103, i16 111, i16 111, i16 100, i16 0, i16 98, i16 121, i16 101, i16 0], section "__TEXT,__ustring", align 2
-// CHECK: @_unnamed_cfstring_ = private global %struct.__NSConstantString_tag { i32* getelementptr inbounds ([0 x i32], [0 x i32]* @__CFConstantStringClassReference, i32 0, i32 0), i32 2000, i8* bitcast ([9 x i16]* @.str to i8*), i32 8 }, section "__DATA,__cfstring"
-// CHECK: @P ={{.*}} global i8* bitcast (%struct.__NSConstantString_tag* @_unnamed_cfstring_ to i8*), align 4
+// CHECK: @_unnamed_cfstring_ = private global %struct.__NSConstantString_tag { ptr @__CFConstantStringClassReference, i32 2000, ptr @.str, i32 8 }, section "__DATA,__cfstring"
+// CHECK: @P ={{.*}} global ptr @_unnamed_cfstring_, align 4
 void *P = @"good\0bye";

diff  --git a/clang/test/CodeGenObjC/2010-02-09-DbgSelf.m b/clang/test/CodeGenObjC/2010-02-09-DbgSelf.m
index 0f8ad9d9c7b1d..df2f8bec20f05 100644
--- a/clang/test/CodeGenObjC/2010-02-09-DbgSelf.m
+++ b/clang/test/CodeGenObjC/2010-02-09-DbgSelf.m
@@ -1,6 +1,6 @@
-// RUN: %clang_cc1 -no-opaque-pointers -x objective-c -emit-llvm -debug-info-kind=limited < %s | FileCheck %s
+// RUN: %clang_cc1 -x objective-c -emit-llvm -debug-info-kind=limited < %s | FileCheck %s
 // Test to check that "self" argument is assigned a location.
-// CHECK: call void @llvm.dbg.declare(metadata %0** %{{[^,]+}}, metadata [[SELF:![0-9]*]], metadata !{{.*}})
+// CHECK: call void @llvm.dbg.declare(metadata ptr %{{[^,]+}}, metadata [[SELF:![0-9]*]], metadata !{{.*}})
 // CHECK: [[SELF]] = !DILocalVariable(name: "self", arg: 1,
 
 @interface Foo 

diff  --git a/clang/test/CodeGenObjC/NSFastEnumeration.m b/clang/test/CodeGenObjC/NSFastEnumeration.m
index 757921b1ec79c..564a8b1228f77 100644
--- a/clang/test/CodeGenObjC/NSFastEnumeration.m
+++ b/clang/test/CodeGenObjC/NSFastEnumeration.m
@@ -1,16 +1,16 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-apple-ios10.3 -fobjc-runtime=ios-6.0 -Os -emit-llvm -o - %s | FileCheck %s -check-prefix CHECK32
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686--windows-msvc -fobjc-runtime=ios-6.0 -Os -emit-llvm -o - %s | FileCheck %s -check-prefix CHECK32
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-ios10.3 -fobjc-runtime=ios-6.0 -Os -emit-llvm -o - %s | FileCheck %s -check-prefix CHECK64
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64--windows-msvc -fobjc-runtime=ios-6.0 -Os -emit-llvm -o - %s | FileCheck %s -check-prefix CHECK64
+// RUN: %clang_cc1 -triple i686-apple-ios10.3 -fobjc-runtime=ios-6.0 -Os -emit-llvm -o - %s | FileCheck %s -check-prefix CHECK32
+// RUN: %clang_cc1 -triple i686--windows-msvc -fobjc-runtime=ios-6.0 -Os -emit-llvm -o - %s | FileCheck %s -check-prefix CHECK32
+// RUN: %clang_cc1 -triple x86_64-apple-ios10.3 -fobjc-runtime=ios-6.0 -Os -emit-llvm -o - %s | FileCheck %s -check-prefix CHECK64
+// RUN: %clang_cc1 -triple x86_64--windows-msvc -fobjc-runtime=ios-6.0 -Os -emit-llvm -o - %s | FileCheck %s -check-prefix CHECK64
 
 void f(id a) {
   for (id i in a)
     (void)i;
 }
 
-// CHECK32: call i32 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i32 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i32)*)
-// CHECK32: call i32 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i32 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i32)*)
+// CHECK32: call i32 @objc_msgSend
+// CHECK32: call i32 @objc_msgSend
 
-// CHECK64: call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)
-// CHECK64: call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)
+// CHECK64: call i64 @objc_msgSend
+// CHECK64: call i64 @objc_msgSend
 

diff  --git a/clang/test/CodeGenObjC/arc-arm.m b/clang/test/CodeGenObjC/arc-arm.m
index 8f90900788793..a605fab33e3a2 100644
--- a/clang/test/CodeGenObjC/arc-arm.m
+++ b/clang/test/CodeGenObjC/arc-arm.m
@@ -1,21 +1,21 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple armv7-apple-darwin10 -emit-llvm -fblocks -fobjc-arc -o - %s | FileCheck %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple arm64-apple-ios -emit-llvm -fblocks -fobjc-arc -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple armv7-apple-darwin10 -emit-llvm -fblocks -fobjc-arc -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple arm64-apple-ios -emit-llvm -fblocks -fobjc-arc -o - %s | FileCheck %s
 
 // <rdar://12438598>: use an autorelease marker on ARM64.
 
 id test0(void) {
   extern id test0_helper(void);
-  // CHECK:      [[T0:%.*]] = call [[CC:(arm_aapcscc )?]]i8* @test0_helper()
-  // CHECK-NEXT: ret i8* [[T0]]
+  // CHECK:      [[T0:%.*]] = call [[CC:(arm_aapcscc )?]]ptr @test0_helper()
+  // CHECK-NEXT: ret ptr [[T0]]
   return test0_helper();
 }
 
 void test1(void) {
   extern id test1_helper(void);
-  // CHECK:      [[T0:%.*]] = call [[CC]]i8* @test1_helper()
+  // CHECK:      [[T0:%.*]] = call [[CC]]ptr @test1_helper()
   // CHECK-NEXT: call void asm sideeffect "mov\09{{fp, fp|r7, r7}}\09\09// marker for objc_retainAutoreleaseReturnValue"
-  // CHECK-NEXT: [[T1:%.*]] = call [[CC]]i8* @llvm.objc.retainAutoreleasedReturnValue(i8* [[T0]])
-  // CHECK-NEXT: store i8* [[T1]],
+  // CHECK-NEXT: [[T1:%.*]] = call [[CC]]ptr @llvm.objc.retainAutoreleasedReturnValue(ptr [[T0]])
+  // CHECK-NEXT: store ptr [[T1]],
   // CHECK-NEXT: call [[CC]]void @llvm.objc.storeStrong(
   // CHECK-NEXT: ret void
   id x = test1_helper();
@@ -25,15 +25,14 @@ void test1(void) {
 @class A;
 A *test2(void) {
   extern A *test2_helper(void);
-  // CHECK:      [[T0:%.*]] = call [[CC]][[A:%.*]]* @test2_helper()
-  // CHECK-NEXT: ret [[A]]* [[T0]]
+  // CHECK:      [[T0:%.*]] = call [[CC]]ptr @test2_helper()
+  // CHECK-NEXT: ret ptr [[T0]]
   return test2_helper();
 }
 
 id test3(void) {
   extern A *test3_helper(void);
-  // CHECK:      [[T0:%.*]] = call [[CC]][[A:%.*]]* @test3_helper()
-  // CHECK-NEXT: [[T1:%.*]] = bitcast [[A]]* [[T0]] to i8*
-  // CHECK-NEXT: ret i8* [[T1]]
+  // CHECK:      [[T0:%.*]] = call [[CC]]ptr @test3_helper()
+  // CHECK-NEXT: ret ptr [[T0]]
   return test3_helper();
 }

diff  --git a/clang/test/CodeGenObjC/arc-block-copy-escape.m b/clang/test/CodeGenObjC/arc-block-copy-escape.m
index 6ce623c70a8e3..886e97a8a10a7 100644
--- a/clang/test/CodeGenObjC/arc-block-copy-escape.m
+++ b/clang/test/CodeGenObjC/arc-block-copy-escape.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -fobjc-arc -fblocks -emit-llvm %s -o - | FileCheck -check-prefix=CHECK -check-prefix=CHECK-HEAP %s
-// RUN: %clang_cc1 -no-opaque-pointers -fobjc-arc -fblocks -fobjc-avoid-heapify-local-blocks -emit-llvm %s -o - | FileCheck -check-prefix=CHECK -check-prefix=CHECK-NOHEAP %s
+// RUN: %clang_cc1 -fobjc-arc -fblocks -emit-llvm %s -o - | FileCheck -check-prefix=CHECK -check-prefix=CHECK-HEAP %s
+// RUN: %clang_cc1 -fobjc-arc -fblocks -fobjc-avoid-heapify-local-blocks -emit-llvm %s -o - | FileCheck -check-prefix=CHECK -check-prefix=CHECK-NOHEAP %s
 
 
 typedef void (^block_t)(void);
@@ -11,7 +11,7 @@
 void test0(int i) {
   block_t block = ^{ use_int(i); };
   // CHECK-LABEL:      define {{.*}}void @test0(
-  // CHECK-HEAP:       call {{.*}}i8* @llvm.objc.retainBlock(i8* {{%.*}}) [[NUW:#[0-9]+]], !clang.arc.copy_on_escape
+  // CHECK-HEAP:       call {{.*}}ptr @llvm.objc.retainBlock(ptr {{%.*}}) [[NUW:#[0-9]+]], !clang.arc.copy_on_escape
   // CHECK-NOHEAP-NOT: @llvm.objc.retainBlock(
   // CHECK:            ret void
 }
@@ -19,8 +19,8 @@ void test0(int i) {
 void test1(int i) {
   id block = ^{ use_int(i); };
   // CHECK-LABEL:   define {{.*}}void @test1(
-  // CHECK-HEAP:    call {{.*}}i8* @llvm.objc.retainBlock(i8* {{%.*}}) [[NUW]]
-  // CHECK-NOHEAP:  call {{.*}}i8* @llvm.objc.retainBlock(i8*  {{%.*}}) [[NUW:#[0-9]+]]
+  // CHECK-HEAP:    call {{.*}}ptr @llvm.objc.retainBlock(ptr {{%.*}}) [[NUW]]
+  // CHECK-NOHEAP:  call {{.*}}ptr @llvm.objc.retainBlock(ptr  {{%.*}}) [[NUW:#[0-9]+]]
   // CHECK-NOT:     !clang.arc.copy_on_escape
   // CHECK:         ret void
 }

diff  --git a/clang/test/CodeGenObjC/arc-blocks-exceptions.m b/clang/test/CodeGenObjC/arc-blocks-exceptions.m
index 9318a74818403..821b818d4027d 100644
--- a/clang/test/CodeGenObjC/arc-blocks-exceptions.m
+++ b/clang/test/CodeGenObjC/arc-blocks-exceptions.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -fblocks -fobjc-arc -fobjc-runtime-has-weak -fexceptions -disable-llvm-passes -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -fblocks -fobjc-arc -fobjc-runtime-has-weak -fexceptions -disable-llvm-passes -o - %s | FileCheck %s
 
 void test1(_Bool c) {
   void test1_fn(void (^blk)(void));
@@ -6,30 +6,30 @@ void test1(_Bool c) {
   test1_fn(c ? ^{ (void)weakId; } : 0);
 
   // CHECK: [[CLEANUP_COND:%.*]] = alloca i1
-  // CHECK-NEXT: [[CLEANUP_SAVE:%.*]] = alloca i8**
+  // CHECK-NEXT: [[CLEANUP_SAVE:%.*]] = alloca ptr
 
-  // CHECK: store i1 true, i1* [[CLEANUP_COND]]
-  // CHECK-NEXT: store i8** {{.*}}, i8*** [[CLEANUP_SAVE]]
+  // CHECK: store i1 true, ptr [[CLEANUP_COND]]
+  // CHECK-NEXT: store ptr {{.*}}, ptr [[CLEANUP_SAVE]]
 
   // CHECK: invoke void @test1_fn(
   // CHECK-NEXT: to label %[[INVOKE_CONT:.*]] unwind label %[[LANDING_PAD_LAB:.*]]
 
   // CHECK: [[INVOKE_CONT]]:
-  // CHECK-NEXT: [[LOAD:%.*]] = load i1, i1* [[CLEANUP_COND]]
+  // CHECK-NEXT: [[LOAD:%.*]] = load i1, ptr [[CLEANUP_COND]]
   // CHECK-NEXT: br i1 [[LOAD]], label %[[END_OF_SCOPE_LAB:.*]], label
 
   // CHECK: [[END_OF_SCOPE_LAB]]:
-  // CHECK-NEXT: [[LOAD:%.*]] = load i8**, i8*** [[CLEANUP_SAVE]]
-  // CHECK-NEXT: call void @llvm.objc.destroyWeak(i8** [[LOAD]])
+  // CHECK-NEXT: [[LOAD:%.*]] = load ptr, ptr [[CLEANUP_SAVE]]
+  // CHECK-NEXT: call void @llvm.objc.destroyWeak(ptr [[LOAD]])
   // CHECK-NEXT: br label
 
   // CHECK: [[LANDING_PAD_LAB]]:
   //   /* some EH stuff */
-  // CHECK:      [[LOAD:%.*]] = load i1, i1* [[CLEANUP_COND]]
+  // CHECK:      [[LOAD:%.*]] = load i1, ptr [[CLEANUP_COND]]
   // CHECK-NEXT: br i1 [[LOAD]], label %[[EH_CLEANUP_LAB:.*]], label
 
   // CHECK: [[EH_CLEANUP_LAB]]:
-  // CHECK-NEXT: [[LOAD:%.*]] = load i8**, i8*** [[CLEANUP_SAVE]]
-  // CHECK-NEXT: call void @llvm.objc.destroyWeak(i8** [[LOAD]])
+  // CHECK-NEXT: [[LOAD:%.*]] = load ptr, ptr [[CLEANUP_SAVE]]
+  // CHECK-NEXT: call void @llvm.objc.destroyWeak(ptr [[LOAD]])
   // CHECK-NEXT: br label
 }

diff  --git a/clang/test/CodeGenObjC/arc-bridged-cast.m b/clang/test/CodeGenObjC/arc-bridged-cast.m
index 32dcfde8ce0b1..a769ce4292879 100644
--- a/clang/test/CodeGenObjC/arc-bridged-cast.m
+++ b/clang/test/CodeGenObjC/arc-bridged-cast.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -fblocks -fobjc-arc -O2 -disable-llvm-passes -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -fblocks -fobjc-arc -O2 -disable-llvm-passes -o - %s | FileCheck %s
 
 typedef const void *CFTypeRef;
 typedef const struct __CFString *CFStringRef;
@@ -18,12 +18,12 @@ @interface NSString
 void bridge_transfer_from_cf(int *i) {
   // CHECK: store i32 7
   *i = 7;
-  // CHECK: call i8* @CFCreateSomething()
+  // CHECK: call ptr @CFCreateSomething()
   id obj1 = (__bridge_transfer id)CFCreateSomething();
   // CHECK-NOT: retain
   // CHECK: store i32 11
   *i = 11;
-  // CHECK: call i8* @CFCreateSomething()
+  // CHECK: call ptr @CFCreateSomething()
   // CHECK-NOT: retain
   // CHECK: store i32 13
   (void)(__bridge_transfer id)CFCreateSomething(), *i = 13;
@@ -31,7 +31,6 @@ void bridge_transfer_from_cf(int *i) {
   // CHECK: store i32 17
   *i = 17;
   // CHECK: call void @llvm.objc.release
-  // CHECK-NEXT: bitcast
   // CHECK-NEXT: call void @llvm.lifetime.end
   // CHECK-NEXT: ret void
 }
@@ -40,18 +39,17 @@ void bridge_transfer_from_cf(int *i) {
 void bridge_from_cf(int *i) {
   // CHECK: store i32 7
   *i = 7;
-  // CHECK: call i8* @CFCreateSomething() [ "clang.arc.attachedcall"(i8* (i8*)* @llvm.objc.retainAutoreleasedReturnValue) ]
+  // CHECK: call ptr @CFCreateSomething() [ "clang.arc.attachedcall"(ptr @llvm.objc.retainAutoreleasedReturnValue) ]
   id obj1 = (__bridge id)CFCreateSomething();
   // CHECK: store i32 11
   *i = 11;
-  // CHECK: call i8* @CFCreateSomething()
+  // CHECK: call ptr @CFCreateSomething()
   // CHECK-NOT: release
   // CHECK: store i32 13
   (void)(__bridge id)CFCreateSomething(), *i = 13;
   // CHECK: store i32 17
   *i = 17;
   // CHECK: call void @llvm.objc.release
-  // CHECK-NEXT: bitcast
   // CHECK-NEXT: call void @llvm.lifetime.end
   // CHECK-NEXT: ret void
 }
@@ -59,16 +57,15 @@ void bridge_from_cf(int *i) {
 // CHECK-LABEL: define{{.*}} void @bridge_retained_of_cf
 void bridge_retained_of_cf(int *i) {
   *i = 7;
-  // CHECK: call i8* @CreateSomething() [ "clang.arc.attachedcall"(i8* (i8*)* @llvm.objc.retainAutoreleasedReturnValue) ]
+  // CHECK: call ptr @CreateSomething() [ "clang.arc.attachedcall"(ptr @llvm.objc.retainAutoreleasedReturnValue) ]
   CFTypeRef cf1 = (__bridge_retained CFTypeRef)CreateSomething();
   // CHECK: store i32 11
   *i = 11;
-  // CHECK: call i8* @CreateSomething() [ "clang.arc.attachedcall"(i8* (i8*)* @llvm.objc.retainAutoreleasedReturnValue) ]
+  // CHECK: call ptr @CreateSomething() [ "clang.arc.attachedcall"(ptr @llvm.objc.retainAutoreleasedReturnValue) ]
   (__bridge_retained CFTypeRef)CreateSomething(), *i = 13;
   // CHECK: store i32 13
   // CHECK: store i32 17
   *i = 17;
-  // CHECK-NEXT: bitcast
   // CHECK-NEXT: call void @llvm.lifetime.end
   // CHECK-NEXT: ret void
 }
@@ -78,23 +75,22 @@ void bridge_of_cf(int *i) {
   // CHECK: store i32 7
   *i = 7;
   // CHECK: call void @llvm.lifetime.start
-  // CHECK-NEXT: call i8* @CreateSomething()
+  // CHECK-NEXT: call ptr @CreateSomething()
   CFTypeRef cf1 = (__bridge CFTypeRef)CreateSomething();
   // CHECK-NOT: retain
   // CHECK: store i32 11
   *i = 11;
-  // CHECK: call i8* @CreateSomething
+  // CHECK: call ptr @CreateSomething
   (__bridge CFTypeRef)CreateSomething(), *i = 13;
   // CHECK: store i32 13
   // CHECK-NOT: release
   // CHECK: store i32 17
   *i = 17;
-  // CHECK-NEXT: bitcast
   // CHECK-NEXT: call void @llvm.lifetime.end
   // CHECK-NEXT: ret void
 }
 
-// CHECK-LABEL: define{{.*}} %struct.__CFString* @bridge_of_paren_expr()
+// CHECK-LABEL: define{{.*}} ptr @bridge_of_paren_expr()
 CFStringRef bridge_of_paren_expr(void) {
   // CHECK-NOT: "@llvm.objc"
   CFStringRef r = (__bridge CFStringRef)(CreateNSString());

diff  --git a/clang/test/CodeGenObjC/arc-i386.m b/clang/test/CodeGenObjC/arc-i386.m
index cd6b6e1118201..09edc7f1f357b 100644
--- a/clang/test/CodeGenObjC/arc-i386.m
+++ b/clang/test/CodeGenObjC/arc-i386.m
@@ -1,23 +1,23 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple i386-apple-darwin10 -emit-llvm -fblocks -fobjc-arc -o - %s | FileCheck %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple i386-apple-iossimulator6.0 -emit-llvm -fblocks -fobjc-arc -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple i386-apple-darwin10 -emit-llvm -fblocks -fobjc-arc -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple i386-apple-iossimulator6.0 -emit-llvm -fblocks -fobjc-arc -o - %s | FileCheck %s
 
 // <rdar://24531556>: implement objc_retainAutoreleasedReturnValue on i386
 
-// CHECK-LABEL: define{{.*}} i8* @test0()
+// CHECK-LABEL: define{{.*}} ptr @test0()
 id test0(void) {
   extern id test0_helper(void);
-  // CHECK:      [[T0:%.*]] = call i8* @test0_helper()
-  // CHECK-NEXT: ret i8* [[T0]]
+  // CHECK:      [[T0:%.*]] = call ptr @test0_helper()
+  // CHECK-NEXT: ret ptr [[T0]]
   return test0_helper();
 }
 
 // CHECK-LABEL: define{{.*}} void @test1()
 void test1(void) {
   extern id test1_helper(void);
-  // CHECK:      [[T0:%.*]] = call i8* @test1_helper()
+  // CHECK:      [[T0:%.*]] = call ptr @test1_helper()
   // CHECK-NEXT: call void asm sideeffect "mov
-  // CHECK-NEXT: [[T1:%.*]] = call i8* @llvm.objc.retainAutoreleasedReturnValue(i8* [[T0]])
-  // CHECK-NEXT: store i8* [[T1]],
+  // CHECK-NEXT: [[T1:%.*]] = call ptr @llvm.objc.retainAutoreleasedReturnValue(ptr [[T0]])
+  // CHECK-NEXT: store ptr [[T1]],
   // CHECK-NEXT: call void @llvm.objc.storeStrong(
   // CHECK-NEXT: ret void
   id x = test1_helper();
@@ -28,16 +28,15 @@ void test1(void) {
 @class A;
 A *test2(void) {
   extern A *test2_helper(void);
-  // CHECK:      [[T0:%.*]] = call [[A:%.*]]* @test2_helper()
-  // CHECK-NEXT: ret [[A]]* [[T0]]
+  // CHECK:      [[T0:%.*]] = call ptr @test2_helper()
+  // CHECK-NEXT: ret ptr [[T0]]
   return test2_helper();
 }
 
-// CHECK-LABEL: define{{.*}} i8* @test3()
+// CHECK-LABEL: define{{.*}} ptr @test3()
 id test3(void) {
   extern A *test3_helper(void);
-  // CHECK:      [[T0:%.*]] = call [[A]]* @test3_helper()
-  // CHECK-NEXT: [[T1:%.*]] = bitcast [[A]]* [[T0]] to i8*
-  // CHECK-NEXT: ret i8* [[T1]]
+  // CHECK:      [[T0:%.*]] = call ptr @test3_helper()
+  // CHECK-NEXT: ret ptr [[T0]]
   return test3_helper();
 }

diff  --git a/clang/test/CodeGenObjC/arc-linetable.m b/clang/test/CodeGenObjC/arc-linetable.m
index 14722cb89f984..60e0e0f70ac6d 100644
--- a/clang/test/CodeGenObjC/arc-linetable.m
+++ b/clang/test/CodeGenObjC/arc-linetable.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -fblocks -fobjc-arc -debug-info-kind=standalone -dwarf-version=4 -gno-column-info -disable-llvm-passes -triple x86_64-apple-darwin10 %s -o - | FileCheck %s
+// RUN: %clang_cc1 -emit-llvm -fblocks -fobjc-arc -debug-info-kind=standalone -dwarf-version=4 -gno-column-info -disable-llvm-passes -triple x86_64-apple-darwin10 %s -o - | FileCheck %s
 
 // Legend: EXP = Return expression, RET = ret instruction
 
@@ -34,7 +34,7 @@
 
 // CHECK: define {{.*}}testCleanupVoid
 // CHECK: icmp ne {{.*}}!dbg ![[SKIP1:[0-9]+]]
-// CHECK: store i32 0, i32* {{.*}}, !dbg ![[RET8:[0-9]+]]
+// CHECK: store i32 0, ptr {{.*}}, !dbg ![[RET8:[0-9]+]]
 // CHECK: @llvm.objc.storeStrong{{.*}}, !dbg ![[RET8]]
 // CHECK: ret {{.*}} !dbg ![[RET8]]
 

diff  --git a/clang/test/CodeGenObjC/arc-no-arc-exceptions.m b/clang/test/CodeGenObjC/arc-no-arc-exceptions.m
index 9483299c2b2ea..223cc43f835d0 100644
--- a/clang/test/CodeGenObjC/arc-no-arc-exceptions.m
+++ b/clang/test/CodeGenObjC/arc-no-arc-exceptions.m
@@ -1,6 +1,6 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -fobjc-arc -fblocks -fexceptions -fobjc-exceptions -O2 -disable-llvm-passes -o - %s | FileCheck %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -fobjc-arc -fblocks -fexceptions -fobjc-exceptions -disable-llvm-passes -o - %s | FileCheck -check-prefix=NO-METADATA %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -fobjc-arc -fblocks -fexceptions -fobjc-exceptions -O2 -disable-llvm-passes -o - %s -fobjc-arc-exceptions | FileCheck -check-prefix=NO-METADATA %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -fobjc-arc -fblocks -fexceptions -fobjc-exceptions -O2 -disable-llvm-passes -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -fobjc-arc -fblocks -fexceptions -fobjc-exceptions -disable-llvm-passes -o - %s | FileCheck -check-prefix=NO-METADATA %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -fobjc-arc -fblocks -fexceptions -fobjc-exceptions -O2 -disable-llvm-passes -o - %s -fobjc-arc-exceptions | FileCheck -check-prefix=NO-METADATA %s
 
 // The front-end should emit clang.arc.no_objc_arc_exceptions in -fobjc-arc-exceptions
 // mode when optimization is enabled, and not otherwise.
@@ -34,7 +34,7 @@ void test1(id x) {
 void NSLog(id, ...);
 
 // CHECK-LABEL: define{{.*}} void @test2(
-// CHECK: invoke void (i8*, ...) @NSLog(i8* noundef bitcast (%struct.__NSConstantString_tag* @_unnamed_cfstring_ to i8*), i32* noundef %{{.*}})
+// CHECK: invoke void (ptr, ...) @NSLog(ptr noundef @_unnamed_cfstring_, ptr noundef %{{.*}})
 // CHECK:   to label %{{.*}} unwind label %{{.*}}, !clang.arc.no_objc_arc_exceptions !
 // NO-METADATA-LABEL: define{{.*}} void @test2(
 // NO-METADATA-NOT: !clang.arc.no_objc_arc_exceptions
@@ -48,7 +48,7 @@ void test2(void) {
 }
 
 // CHECK-LABEL: define{{.*}} void @test3(
-// CHECK: invoke void %{{.*}}(i8* noundef %{{.*}})
+// CHECK: invoke void %{{.*}}(ptr noundef %{{.*}})
 // CHECK:   to label %{{.*}} unwind label %{{.*}}, !clang.arc.no_objc_arc_exceptions !
 // NO-METADATA-LABEL: define{{.*}} void @test3(
 // NO-METADATA-NOT: !clang.arc.no_objc_arc_exceptions
@@ -63,7 +63,7 @@ void test3(void) {
 }
 
 // CHECK-LABEL: define{{.*}} void @test4(
-// CHECK: invoke void %{{.*}}(i8* noundef %{{.*}})
+// CHECK: invoke void %{{.*}}(ptr noundef %{{.*}})
 // CHECK:   to label %{{.*}} unwind label %{{.*}}, !clang.arc.no_objc_arc_exceptions !
 // NO-METADATA-LABEL: define{{.*}} void @test4(
 // NO-METADATA-NOT: !clang.arc.no_objc_arc_exceptions

diff  --git a/clang/test/CodeGenObjC/arc-rv-attr.m b/clang/test/CodeGenObjC/arc-rv-attr.m
index 77aef00ad8450..89b9c49564f35 100644
--- a/clang/test/CodeGenObjC/arc-rv-attr.m
+++ b/clang/test/CodeGenObjC/arc-rv-attr.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple arm64-apple-ios9 -fobjc-runtime=ios-9.0 -fobjc-arc -O -disable-llvm-passes -emit-llvm -o - %s | FileCheck %s -check-prefix=CHECK
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-macosx10 -fobjc-runtime=ios-9.0 -fobjc-arc -O -disable-llvm-passes -emit-llvm -o - %s | FileCheck %s -check-prefix=CHECK
+// RUN: %clang_cc1 -triple arm64-apple-ios9 -fobjc-runtime=ios-9.0 -fobjc-arc -O -disable-llvm-passes -emit-llvm -o - %s | FileCheck %s -check-prefix=CHECK
+// RUN: %clang_cc1 -triple x86_64-apple-macosx10 -fobjc-runtime=ios-9.0 -fobjc-arc -O -disable-llvm-passes -emit-llvm -o - %s | FileCheck %s -check-prefix=CHECK
 
 @class A;
 
@@ -10,12 +10,10 @@ void test_assign(void) {
   x = makeA();
 }
 // CHECK-LABEL: define{{.*}} void @test_assign()
-// CHECK:         [[X:%.*]] = alloca i8*
-// CHECK:         [[T0:%.*]] = call [[A:.*]]* @makeA() [ "clang.arc.attachedcall"(i8* (i8*)* @llvm.objc.unsafeClaimAutoreleasedReturnValue) ]
+// CHECK:         [[X:%.*]] = alloca ptr
+// CHECK:         [[T0:%.*]] = call ptr @makeA() [ "clang.arc.attachedcall"(ptr @llvm.objc.unsafeClaimAutoreleasedReturnValue) ]
 // CHECK-NEXT:    call void (...) @llvm.objc.clang.arc.noop.use({{.*}} [[T0]])
-// CHECK-NEXT:    [[T1:%.*]] = bitcast [[A]]* [[T0]] to i8*
-// CHECK-NEXT:    store i8* [[T1]], i8** [[X]]
-// CHECK-NEXT:    bitcast
+// CHECK-NEXT:    store ptr [[T0]], ptr [[X]]
 // CHECK-NEXT:    lifetime.end
 // CHECK-NEXT:    ret void
 
@@ -24,16 +22,13 @@ void test_assign_assign(void) {
   x = y = makeA();
 }
 // CHECK-LABEL: define{{.*}} void @test_assign_assign()
-// CHECK:         [[X:%.*]] = alloca i8*
-// CHECK:         [[Y:%.*]] = alloca i8*
-// CHECK:         [[T0:%.*]] = call [[A]]* @makeA() [ "clang.arc.attachedcall"(i8* (i8*)* @llvm.objc.unsafeClaimAutoreleasedReturnValue) ]
+// CHECK:         [[X:%.*]] = alloca ptr
+// CHECK:         [[Y:%.*]] = alloca ptr
+// CHECK:         [[T0:%.*]] = call ptr @makeA() [ "clang.arc.attachedcall"(ptr @llvm.objc.unsafeClaimAutoreleasedReturnValue) ]
 // CHECK-NEXT:    call void (...) @llvm.objc.clang.arc.noop.use({{.*}} [[T0]])
-// CHECK-NEXT:    [[T1:%.*]] = bitcast [[A]]* [[T0]] to i8*
-// CHECK-NEXT:    store i8* [[T1]], i8** [[Y]]
-// CHECK-NEXT:    store i8* [[T1]], i8** [[X]]
-// CHECK-NEXT:    bitcast
+// CHECK-NEXT:    store ptr [[T0]], ptr [[Y]]
+// CHECK-NEXT:    store ptr [[T0]], ptr [[X]]
 // CHECK-NEXT:    lifetime.end
-// CHECK-NEXT:    bitcast
 // CHECK-NEXT:    lifetime.end
 // CHECK-NEXT:    ret void
 
@@ -43,20 +38,17 @@ void test_strong_assign_assign(void) {
   x = y = makeA();
 }
 // CHECK-LABEL: define{{.*}} void @test_strong_assign_assign()
-// CHECK:         [[X:%.*]] = alloca i8*
-// CHECK:         [[Y:%.*]] = alloca i8*
-// CHECK:         [[T0:%.*]] = call [[A]]* @makeA() [ "clang.arc.attachedcall"(i8* (i8*)* @llvm.objc.retainAutoreleasedReturnValue) ]
+// CHECK:         [[X:%.*]] = alloca ptr
+// CHECK:         [[Y:%.*]] = alloca ptr
+// CHECK:         [[T0:%.*]] = call ptr @makeA() [ "clang.arc.attachedcall"(ptr @llvm.objc.retainAutoreleasedReturnValue) ]
 // CHECK-NEXT:    call void (...) @llvm.objc.clang.arc.noop.use({{.*}} [[T0]])
-// CHECK-NEXT:    [[T1:%.*]] = bitcast [[A]]* [[T0]] to i8*
-// CHECK-NEXT:    store i8* [[T1]], i8** [[Y]]
-// CHECK-NEXT:    [[OLD:%.*]] = load i8*, i8** [[X]]
-// CHECK-NEXT:    store i8* [[T1]], i8** [[X]]
-// CHECK-NEXT:    call void @llvm.objc.release(i8* [[OLD]]
-// CHECK-NEXT:    bitcast
+// CHECK-NEXT:    store ptr [[T0]], ptr [[Y]]
+// CHECK-NEXT:    [[OLD:%.*]] = load ptr, ptr [[X]]
+// CHECK-NEXT:    store ptr [[T0]], ptr [[X]]
+// CHECK-NEXT:    call void @llvm.objc.release(ptr [[OLD]]
 // CHECK-NEXT:    lifetime.end
-// CHECK-NEXT:    [[T0:%.*]] = load i8*, i8** [[X]]
-// CHECK-NEXT:    call void @llvm.objc.release(i8* [[T0]])
-// CHECK-NEXT:    bitcast
+// CHECK-NEXT:    [[T0:%.*]] = load ptr, ptr [[X]]
+// CHECK-NEXT:    call void @llvm.objc.release(ptr [[T0]])
 // CHECK-NEXT:    lifetime.end
 // CHECK-NEXT:    ret void
 
@@ -66,20 +58,17 @@ void test_assign_strong_assign(void) {
   x = y = makeA();
 }
 // CHECK-LABEL: define{{.*}} void @test_assign_strong_assign()
-// CHECK:         [[X:%.*]] = alloca i8*
-// CHECK:         [[Y:%.*]] = alloca i8*
-// CHECK:         [[T0:%.*]] = call [[A]]* @makeA() [ "clang.arc.attachedcall"(i8* (i8*)* @llvm.objc.retainAutoreleasedReturnValue) ]
+// CHECK:         [[X:%.*]] = alloca ptr
+// CHECK:         [[Y:%.*]] = alloca ptr
+// CHECK:         [[T0:%.*]] = call ptr @makeA() [ "clang.arc.attachedcall"(ptr @llvm.objc.retainAutoreleasedReturnValue) ]
 // CHECK-NEXT:    call void (...) @llvm.objc.clang.arc.noop.use({{.*}} [[T0]])
-// CHECK-NEXT:    [[T1:%.*]] = bitcast [[A]]* [[T0]] to i8*
-// CHECK-NEXT:    [[OLD:%.*]] = load i8*, i8** [[Y]]
-// CHECK-NEXT:    store i8* [[T1]], i8** [[Y]]
-// CHECK-NEXT:    call void @llvm.objc.release(i8* [[OLD]]
-// CHECK-NEXT:    store i8* [[T1]], i8** [[X]]
-// CHECK-NEXT:    [[T0:%.*]] = load i8*, i8** [[Y]]
-// CHECK-NEXT:    call void @llvm.objc.release(i8* [[T0]])
-// CHECK-NEXT:    bitcast
+// CHECK-NEXT:    [[OLD:%.*]] = load ptr, ptr [[Y]]
+// CHECK-NEXT:    store ptr [[T0]], ptr [[Y]]
+// CHECK-NEXT:    call void @llvm.objc.release(ptr [[OLD]]
+// CHECK-NEXT:    store ptr [[T0]], ptr [[X]]
+// CHECK-NEXT:    [[T0:%.*]] = load ptr, ptr [[Y]]
+// CHECK-NEXT:    call void @llvm.objc.release(ptr [[T0]])
 // CHECK-NEXT:    lifetime.end
-// CHECK-NEXT:    bitcast
 // CHECK-NEXT:    lifetime.end
 // CHECK-NEXT:    ret void
 
@@ -87,12 +76,10 @@ void test_init(void) {
   __unsafe_unretained id x = makeA();
 }
 // CHECK-LABEL: define{{.*}} void @test_init()
-// CHECK:         [[X:%.*]] = alloca i8*
-// CHECK:         [[T0:%.*]] = call [[A]]* @makeA() [ "clang.arc.attachedcall"(i8* (i8*)* @llvm.objc.unsafeClaimAutoreleasedReturnValue) ]
+// CHECK:         [[X:%.*]] = alloca ptr
+// CHECK:         [[T0:%.*]] = call ptr @makeA() [ "clang.arc.attachedcall"(ptr @llvm.objc.unsafeClaimAutoreleasedReturnValue) ]
 // CHECK-NEXT:    call void (...) @llvm.objc.clang.arc.noop.use({{.*}} [[T0]])
-// CHECK-NEXT:    [[T1:%.*]] = bitcast [[A]]* [[T0]] to i8*
-// CHECK-NEXT:    store i8* [[T1]], i8** [[X]]
-// CHECK-NEXT:    bitcast
+// CHECK-NEXT:    store ptr [[T0]], ptr [[X]]
 // CHECK-NEXT:    lifetime.end
 // CHECK-NEXT:    ret void
 
@@ -101,16 +88,13 @@ void test_init_assignment(void) {
   __unsafe_unretained id y = x = makeA();
 }
 // CHECK-LABEL: define{{.*}} void @test_init_assignment()
-// CHECK:         [[X:%.*]] = alloca i8*
-// CHECK:         [[Y:%.*]] = alloca i8*
-// CHECK:         [[T0:%.*]] = call [[A]]* @makeA() [ "clang.arc.attachedcall"(i8* (i8*)* @llvm.objc.unsafeClaimAutoreleasedReturnValue) ]
+// CHECK:         [[X:%.*]] = alloca ptr
+// CHECK:         [[Y:%.*]] = alloca ptr
+// CHECK:         [[T0:%.*]] = call ptr @makeA() [ "clang.arc.attachedcall"(ptr @llvm.objc.unsafeClaimAutoreleasedReturnValue) ]
 // CHECK-NEXT:    call void (...) @llvm.objc.clang.arc.noop.use({{.*}} [[T0]])
-// CHECK-NEXT:    [[T1:%.*]] = bitcast [[A]]* [[T0]] to i8*
-// CHECK-NEXT:    store i8* [[T1]], i8** [[X]]
-// CHECK-NEXT:    store i8* [[T1]], i8** [[Y]]
-// CHECK-NEXT:    bitcast
+// CHECK-NEXT:    store ptr [[T0]], ptr [[X]]
+// CHECK-NEXT:    store ptr [[T0]], ptr [[Y]]
 // CHECK-NEXT:    lifetime.end
-// CHECK-NEXT:    bitcast
 // CHECK-NEXT:    lifetime.end
 // CHECK-NEXT:    ret void
 
@@ -119,18 +103,15 @@ void test_strong_init_assignment(void) {
   __strong id y = x = makeA();
 }
 // CHECK-LABEL: define{{.*}} void @test_strong_init_assignment()
-// CHECK:         [[X:%.*]] = alloca i8*
-// CHECK:         [[Y:%.*]] = alloca i8*
-// CHECK:         [[T0:%.*]] = call [[A]]* @makeA() [ "clang.arc.attachedcall"(i8* (i8*)* @llvm.objc.retainAutoreleasedReturnValue) ]
+// CHECK:         [[X:%.*]] = alloca ptr
+// CHECK:         [[Y:%.*]] = alloca ptr
+// CHECK:         [[T0:%.*]] = call ptr @makeA() [ "clang.arc.attachedcall"(ptr @llvm.objc.retainAutoreleasedReturnValue) ]
 // CHECK-NEXT:    call void (...) @llvm.objc.clang.arc.noop.use({{.*}} [[T0]])
-// CHECK-NEXT:    [[T1:%.*]] = bitcast [[A]]* [[T0]] to i8*
-// CHECK-NEXT:    store i8* [[T1]], i8** [[X]]
-// CHECK-NEXT:    store i8* [[T1]], i8** [[Y]]
-// CHECK-NEXT:    [[T0:%.*]] = load i8*, i8** [[Y]]
-// CHECK-NEXT:    call void @llvm.objc.release(i8* [[T0]])
-// CHECK-NEXT:    bitcast
+// CHECK-NEXT:    store ptr [[T0]], ptr [[X]]
+// CHECK-NEXT:    store ptr [[T0]], ptr [[Y]]
+// CHECK-NEXT:    [[T0:%.*]] = load ptr, ptr [[Y]]
+// CHECK-NEXT:    call void @llvm.objc.release(ptr [[T0]])
 // CHECK-NEXT:    lifetime.end
-// CHECK-NEXT:    bitcast
 // CHECK-NEXT:    lifetime.end
 // CHECK-NEXT:    ret void
 
@@ -139,20 +120,17 @@ void test_init_strong_assignment(void) {
   __unsafe_unretained id y = x = makeA();
 }
 // CHECK-LABEL: define{{.*}} void @test_init_strong_assignment()
-// CHECK:         [[X:%.*]] = alloca i8*
-// CHECK:         [[Y:%.*]] = alloca i8*
-// CHECK:         [[T0:%.*]] = call [[A]]* @makeA() [ "clang.arc.attachedcall"(i8* (i8*)* @llvm.objc.retainAutoreleasedReturnValue) ]
+// CHECK:         [[X:%.*]] = alloca ptr
+// CHECK:         [[Y:%.*]] = alloca ptr
+// CHECK:         [[T0:%.*]] = call ptr @makeA() [ "clang.arc.attachedcall"(ptr @llvm.objc.retainAutoreleasedReturnValue) ]
 // CHECK-NEXT:    call void (...) @llvm.objc.clang.arc.noop.use({{.*}} [[T0]])
-// CHECK-NEXT:    [[T1:%.*]] = bitcast [[A]]* [[T0]] to i8*
-// CHECK-NEXT:    [[OLD:%.*]] = load i8*, i8** [[X]]
-// CHECK-NEXT:    store i8* [[T1]], i8** [[X]]
-// CHECK-NEXT:    call void @llvm.objc.release(i8* [[OLD]])
-// CHECK-NEXT:    store i8* [[T1]], i8** [[Y]]
-// CHECK-NEXT:    bitcast
+// CHECK-NEXT:    [[OLD:%.*]] = load ptr, ptr [[X]]
+// CHECK-NEXT:    store ptr [[T0]], ptr [[X]]
+// CHECK-NEXT:    call void @llvm.objc.release(ptr [[OLD]])
+// CHECK-NEXT:    store ptr [[T0]], ptr [[Y]]
 // CHECK-NEXT:    lifetime.end
-// CHECK-NEXT:    [[T0:%.*]] = load i8*, i8** [[X]]
-// CHECK-NEXT:    call void @llvm.objc.release(i8* [[T0]])
-// CHECK-NEXT:    bitcast
+// CHECK-NEXT:    [[T0:%.*]] = load ptr, ptr [[X]]
+// CHECK-NEXT:    call void @llvm.objc.release(ptr [[T0]])
 // CHECK-NEXT:    lifetime.end
 // CHECK-NEXT:    ret void
 
@@ -160,7 +138,7 @@ void test_ignored(void) {
   makeA();
 }
 // CHECK-LABEL: define{{.*}} void @test_ignored()
-// CHECK:         [[T0:%.*]] = call [[A]]* @makeA() [ "clang.arc.attachedcall"(i8* (i8*)* @llvm.objc.unsafeClaimAutoreleasedReturnValue) ]
+// CHECK:         [[T0:%.*]] = call ptr @makeA() [ "clang.arc.attachedcall"(ptr @llvm.objc.unsafeClaimAutoreleasedReturnValue) ]
 // CHECK-NEXT:    call void (...) @llvm.objc.clang.arc.noop.use({{.*}} [[T0]])
 // CHECK-NEXT:    ret void
 
@@ -168,7 +146,7 @@ void test_cast_to_void(void) {
   (void) makeA();
 }
 // CHECK-LABEL: define{{.*}} void @test_cast_to_void()
-// CHECK:         [[T0:%.*]] = call [[A]]* @makeA() [ "clang.arc.attachedcall"(i8* (i8*)* @llvm.objc.unsafeClaimAutoreleasedReturnValue) ]
+// CHECK:         [[T0:%.*]] = call ptr @makeA() [ "clang.arc.attachedcall"(ptr @llvm.objc.unsafeClaimAutoreleasedReturnValue) ]
 // CHECK-NEXT:    call void (...) @llvm.objc.clang.arc.noop.use({{.*}} [[T0]])
 // CHECK-NEXT:    ret void
 

diff  --git a/clang/test/CodeGenObjC/arc-ternary-op.m b/clang/test/CodeGenObjC/arc-ternary-op.m
index 59fc5b27792c7..c62c913893b45 100644
--- a/clang/test/CodeGenObjC/arc-ternary-op.m
+++ b/clang/test/CodeGenObjC/arc-ternary-op.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -fblocks -fobjc-arc -fobjc-runtime-has-weak -O2 -disable-llvm-passes -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -fblocks -fobjc-arc -fobjc-runtime-has-weak -O2 -disable-llvm-passes -o - %s | FileCheck %s
 
 id g0, g1;
 
@@ -7,34 +7,32 @@ void test0(_Bool cond) {
 
   // CHECK-LABEL:      define{{.*}} void @test0(
   // CHECK:      [[COND:%.*]] = alloca i8,
-  // CHECK-NEXT: [[X:%.*]] = alloca i8*,
-  // CHECK-NEXT: [[RELVAL:%.*]] = alloca i8*
+  // CHECK-NEXT: [[X:%.*]] = alloca ptr,
+  // CHECK-NEXT: [[RELVAL:%.*]] = alloca ptr
   // CHECK-NEXT: [[RELCOND:%.*]] = alloca i1
   // CHECK-NEXT: zext
   // CHECK-NEXT: store
-  // CHECK-NEXT: [[XPTR1:%.*]] = bitcast i8** [[X]] to i8*
-  // CHECK-NEXT: call void @llvm.lifetime.start.p0i8(i64 8, i8* [[XPTR1]])
-  // CHECK-NEXT: [[T0:%.*]] = load i8, i8* [[COND]]
+  // CHECK-NEXT: call void @llvm.lifetime.start.p0(i64 8, ptr [[X]])
+  // CHECK-NEXT: [[T0:%.*]] = load i8, ptr [[COND]]
   // CHECK-NEXT: [[T1:%.*]] = trunc i8 [[T0]] to i1
-  // CHECK-NEXT: store i1 false, i1* [[RELCOND]]
+  // CHECK-NEXT: store i1 false, ptr [[RELCOND]]
   // CHECK-NEXT: br i1 [[T1]],
   // CHECK:      br label
-  // CHECK:      [[CALL:%.*]] = call i8* @test0_helper()
-  // CHECK-NEXT: store i8* [[CALL]], i8** [[RELVAL]]
-  // CHECK-NEXT: store i1 true, i1* [[RELCOND]]
+  // CHECK:      [[CALL:%.*]] = call ptr @test0_helper()
+  // CHECK-NEXT: store ptr [[CALL]], ptr [[RELVAL]]
+  // CHECK-NEXT: store i1 true, ptr [[RELCOND]]
   // CHECK-NEXT: br label
-  // CHECK:      [[T0:%.*]] = phi i8* [ null, {{%.*}} ], [ [[CALL]], {{%.*}} ]
-  // CHECK-NEXT: [[T1:%.*]] = call i8* @llvm.objc.retain(i8* [[T0]]) [[NUW:#[0-9]+]]
-  // CHECK-NEXT: store i8* [[T1]], i8** [[X]],
-  // CHECK-NEXT: [[REL:%.*]] = load i1, i1* [[RELCOND]]
+  // CHECK:      [[T0:%.*]] = phi ptr [ null, {{%.*}} ], [ [[CALL]], {{%.*}} ]
+  // CHECK-NEXT: [[T1:%.*]] = call ptr @llvm.objc.retain(ptr [[T0]]) [[NUW:#[0-9]+]]
+  // CHECK-NEXT: store ptr [[T1]], ptr [[X]],
+  // CHECK-NEXT: [[REL:%.*]] = load i1, ptr [[RELCOND]]
   // CHECK-NEXT: br i1 [[REL]],
-  // CHECK:      [[T0:%.*]] = load i8*, i8** [[RELVAL]]
-  // CHECK-NEXT: call void @llvm.objc.release(i8* [[T0]]) [[NUW]]
+  // CHECK:      [[T0:%.*]] = load ptr, ptr [[RELVAL]]
+  // CHECK-NEXT: call void @llvm.objc.release(ptr [[T0]]) [[NUW]]
   // CHECK-NEXT: br label
-  // CHECK:      [[T0:%.*]] = load i8*, i8** [[X]]
-  // CHECK-NEXT: call void @llvm.objc.release(i8* [[T0]]) [[NUW]]
-  // CHECK-NEXT: [[XPTR2:%.*]] = bitcast i8** [[X]] to i8*
-  // CHECK-NEXT: call void @llvm.lifetime.end.p0i8(i64 8, i8* [[XPTR2]])
+  // CHECK:      [[T0:%.*]] = load ptr, ptr [[X]]
+  // CHECK-NEXT: call void @llvm.objc.release(ptr [[T0]]) [[NUW]]
+  // CHECK-NEXT: call void @llvm.lifetime.end.p0(i64 8, ptr [[X]])
   // CHECK-NEXT: ret void
   id x = (cond ? 0 : test0_helper());
 }
@@ -48,65 +46,61 @@ void test1(int cond) {
 
   // CHECK-LABEL:    define{{.*}} void @test1(
   // CHECK:      [[COND:%.*]] = alloca i32
-  // CHECK-NEXT: [[STRONG:%.*]] = alloca i8*
-  // CHECK-NEXT: [[WEAK:%.*]] = alloca i8*
-  // CHECK-NEXT: [[TEMP1:%.*]] = alloca i8*
-  // CHECK-NEXT: [[TEMP2:%.*]] = alloca i8*
-  // CHECK-NEXT: [[CONDCLEANUPSAVE:%.*]] = alloca i8*
+  // CHECK-NEXT: [[STRONG:%.*]] = alloca ptr
+  // CHECK-NEXT: [[WEAK:%.*]] = alloca ptr
+  // CHECK-NEXT: [[TEMP1:%.*]] = alloca ptr
+  // CHECK-NEXT: [[TEMP2:%.*]] = alloca ptr
+  // CHECK-NEXT: [[CONDCLEANUPSAVE:%.*]] = alloca ptr
   // CHECK-NEXT: [[CONDCLEANUP:%.*]] = alloca i1
   // CHECK-NEXT: store i32
-  // CHECK-NEXT: [[STRONGPTR1:%.*]] = bitcast i8** [[STRONG]] to i8*
-  // CHECK-NEXT: call void @llvm.lifetime.start.p0i8(i64 8, i8* [[STRONGPTR1]])
-  // CHECK-NEXT: store i8* null, i8** [[STRONG]]
-  // CHECK-NEXT: [[WEAKPTR1:%.*]] = bitcast i8** [[WEAK]] to i8*
-  // CHECK-NEXT: call void @llvm.lifetime.start.p0i8(i64 8, i8* [[WEAKPTR1]])
-  // CHECK-NEXT: call i8* @llvm.objc.initWeak(i8** [[WEAK]], i8* null)
-
-  // CHECK-NEXT: [[T0:%.*]] = load i32, i32* [[COND]]
+  // CHECK-NEXT: call void @llvm.lifetime.start.p0(i64 8, ptr [[STRONG]])
+  // CHECK-NEXT: store ptr null, ptr [[STRONG]]
+  // CHECK-NEXT: call void @llvm.lifetime.start.p0(i64 8, ptr [[WEAK]])
+  // CHECK-NEXT: call ptr @llvm.objc.initWeak(ptr [[WEAK]], ptr null)
+
+  // CHECK-NEXT: [[T0:%.*]] = load i32, ptr [[COND]]
   // CHECK-NEXT: [[T1:%.*]] = icmp ne i32 [[T0]], 0
-  // CHECK:      [[ARG:%.*]] = phi i8**
-  // CHECK-NEXT: [[T0:%.*]] = icmp eq i8** [[ARG]], null
-  // CHECK-NEXT: [[T1:%.*]] = select i1 [[T0]], i8** null, i8** [[TEMP1]]
+  // CHECK:      [[ARG:%.*]] = phi ptr
+  // CHECK-NEXT: [[T0:%.*]] = icmp eq ptr [[ARG]], null
+  // CHECK-NEXT: [[T1:%.*]] = select i1 [[T0]], ptr null, ptr [[TEMP1]]
   // CHECK-NEXT: br i1 [[T0]],
-  // CHECK:      [[T0:%.*]] = load i8*, i8** [[ARG]]
-  // CHECK-NEXT: store i8* [[T0]], i8** [[TEMP1]]
+  // CHECK:      [[T0:%.*]] = load ptr, ptr [[ARG]]
+  // CHECK-NEXT: store ptr [[T0]], ptr [[TEMP1]]
   // CHECK-NEXT: br label
-  // CHECK:      [[W:%.*]] = phi i8* [ [[T0]], {{%.*}} ], [ undef, {{%.*}} ]
-  // CHECK-NEXT: call void @test1_sink(i8** noundef [[T1]])
-  // CHECK-NEXT: [[T0:%.*]] = icmp eq i8** [[ARG]], null
+  // CHECK:      [[W:%.*]] = phi ptr [ [[T0]], {{%.*}} ], [ undef, {{%.*}} ]
+  // CHECK-NEXT: call void @test1_sink(ptr noundef [[T1]])
+  // CHECK-NEXT: [[T0:%.*]] = icmp eq ptr [[ARG]], null
   // CHECK-NEXT: br i1 [[T0]],
-  // CHECK:      [[T0:%.*]] = load i8*, i8** [[TEMP1]]
-  // CHECK-NEXT: [[T1:%.*]] = call i8* @llvm.objc.retain(i8* [[T0]])
-  // CHECK-NEXT: call void (...) @llvm.objc.clang.arc.use(i8* [[W]]) [[NUW]]
-  // CHECK-NEXT: [[T2:%.*]] = load i8*, i8** [[ARG]]
-  // CHECK-NEXT: store i8* [[T1]], i8** [[ARG]]
-  // CHECK-NEXT: call void @llvm.objc.release(i8* [[T2]])
+  // CHECK:      [[T0:%.*]] = load ptr, ptr [[TEMP1]]
+  // CHECK-NEXT: [[T1:%.*]] = call ptr @llvm.objc.retain(ptr [[T0]])
+  // CHECK-NEXT: call void (...) @llvm.objc.clang.arc.use(ptr [[W]]) [[NUW]]
+  // CHECK-NEXT: [[T2:%.*]] = load ptr, ptr [[ARG]]
+  // CHECK-NEXT: store ptr [[T1]], ptr [[ARG]]
+  // CHECK-NEXT: call void @llvm.objc.release(ptr [[T2]])
   // CHECK-NEXT: br label
 
-  // CHECK:      [[T0:%.*]] = load i32, i32* [[COND]]
+  // CHECK:      [[T0:%.*]] = load i32, ptr [[COND]]
   // CHECK-NEXT: [[T1:%.*]] = icmp ne i32 [[T0]], 0
-  // CHECK:      [[ARG:%.*]] = phi i8**
-  // CHECK-NEXT: [[T0:%.*]] = icmp eq i8** [[ARG]], null
-  // CHECK-NEXT: [[T1:%.*]] = select i1 [[T0]], i8** null, i8** [[TEMP2]]
-  // CHECK-NEXT: store i1 false, i1* [[CONDCLEANUP]]
+  // CHECK:      [[ARG:%.*]] = phi ptr
+  // CHECK-NEXT: [[T0:%.*]] = icmp eq ptr [[ARG]], null
+  // CHECK-NEXT: [[T1:%.*]] = select i1 [[T0]], ptr null, ptr [[TEMP2]]
+  // CHECK-NEXT: store i1 false, ptr [[CONDCLEANUP]]
   // CHECK-NEXT: br i1 [[T0]],
-  // CHECK:      [[T0:%.*]] = call i8* @llvm.objc.loadWeakRetained(i8** [[ARG]])
-  // CHECK-NEXT: store i8* [[T0]], i8** [[CONDCLEANUPSAVE]]
-  // CHECK-NEXT: store i1 true, i1* [[CONDCLEANUP]]
-  // CHECK-NEXT: store i8* [[T0]], i8** [[TEMP2]]
+  // CHECK:      [[T0:%.*]] = call ptr @llvm.objc.loadWeakRetained(ptr [[ARG]])
+  // CHECK-NEXT: store ptr [[T0]], ptr [[CONDCLEANUPSAVE]]
+  // CHECK-NEXT: store i1 true, ptr [[CONDCLEANUP]]
+  // CHECK-NEXT: store ptr [[T0]], ptr [[TEMP2]]
   // CHECK-NEXT: br label
-  // CHECK:      call void @test1_sink(i8** noundef [[T1]])
-  // CHECK-NEXT: [[T0:%.*]] = icmp eq i8** [[ARG]], null
+  // CHECK:      call void @test1_sink(ptr noundef [[T1]])
+  // CHECK-NEXT: [[T0:%.*]] = icmp eq ptr [[ARG]], null
   // CHECK-NEXT: br i1 [[T0]],
-  // CHECK:      [[T0:%.*]] = load i8*, i8** [[TEMP2]]
-  // CHECK-NEXT: call i8* @llvm.objc.storeWeak(i8** [[ARG]], i8* [[T0]])
+  // CHECK:      [[T0:%.*]] = load ptr, ptr [[TEMP2]]
+  // CHECK-NEXT: call ptr @llvm.objc.storeWeak(ptr [[ARG]], ptr [[T0]])
   // CHECK-NEXT: br label
 
-  // CHECK:      call void @llvm.objc.destroyWeak(i8** [[WEAK]])
-  // CHECK:      [[WEAKPTR2:%.*]] = bitcast i8** [[WEAK]] to i8*
-  // CHECK:      call void @llvm.lifetime.end.p0i8(i64 8, i8* [[WEAKPTR2]])
-  // CHECK:      [[STRONGPTR2:%.*]] = bitcast i8** [[STRONG]] to i8*
-  // CHECK:      call void @llvm.lifetime.end.p0i8(i64 8, i8* [[STRONGPTR2]])
+  // CHECK:      call void @llvm.objc.destroyWeak(ptr [[WEAK]])
+  // CHECK:      call void @llvm.lifetime.end.p0(i64 8, ptr [[WEAK]])
+  // CHECK:      call void @llvm.lifetime.end.p0(i64 8, ptr [[STRONG]])
   // CHECK:      ret void
 }
 
@@ -122,31 +116,31 @@ void test2(int cond) {
 
   // CHECK-LABEL:    define{{.*}} void @test2(
   // CHECK:      [[COND:%.*]] = alloca i32,
-  // CHECK:      alloca i8*
-  // CHECK:      [[CLEANUP_SAVE:%.*]] = alloca i8*
+  // CHECK:      alloca ptr
+  // CHECK:      [[CLEANUP_SAVE:%.*]] = alloca ptr
   // CHECK:      [[RUN_CLEANUP:%.*]] = alloca i1
   //   Evaluate condition; cleanup disabled by default.
-  // CHECK:      [[T0:%.*]] = load i32, i32* [[COND]],
+  // CHECK:      [[T0:%.*]] = load i32, ptr [[COND]],
   // CHECK-NEXT: icmp ne i32 [[T0]], 0
-  // CHECK-NEXT: store i1 false, i1* [[RUN_CLEANUP]]
+  // CHECK-NEXT: store i1 false, ptr [[RUN_CLEANUP]]
   // CHECK-NEXT: br i1
   //   Within true branch, cleanup enabled.
-  // CHECK:      [[T1:%.*]] = call i8* @test2_producer() [ "clang.arc.attachedcall"(i8* (i8*)* @llvm.objc.retainAutoreleasedReturnValue) ]
-  // CHECK-NEXT: call void (...) @llvm.objc.clang.arc.noop.use(i8* [[T1]])
-  // CHECK-NEXT: store i8* [[T1]], i8** [[CLEANUP_SAVE]]
-  // CHECK-NEXT: store i1 true, i1* [[RUN_CLEANUP]]
+  // CHECK:      [[T1:%.*]] = call ptr @test2_producer() [ "clang.arc.attachedcall"(ptr @llvm.objc.retainAutoreleasedReturnValue) ]
+  // CHECK-NEXT: call void (...) @llvm.objc.clang.arc.noop.use(ptr [[T1]])
+  // CHECK-NEXT: store ptr [[T1]], ptr [[CLEANUP_SAVE]]
+  // CHECK-NEXT: store i1 true, ptr [[RUN_CLEANUP]]
   // CHECK-NEXT: br label
   //   Join point for conditional operator; retain immediately.
-  // CHECK:      [[T0:%.*]] = phi i8* [ [[T1]], {{%.*}} ], [ null, {{%.*}} ]
-  // CHECK-NEXT: [[RESULT:%.*]] = call i8* @llvm.objc.retain(i8* [[T0]])
+  // CHECK:      [[T0:%.*]] = phi ptr [ [[T1]], {{%.*}} ], [ null, {{%.*}} ]
+  // CHECK-NEXT: [[RESULT:%.*]] = call ptr @llvm.objc.retain(ptr [[T0]])
   //   Leaving full-expression; run conditional cleanup.
-  // CHECK-NEXT: [[T0:%.*]] = load i1, i1* [[RUN_CLEANUP]]
+  // CHECK-NEXT: [[T0:%.*]] = load i1, ptr [[RUN_CLEANUP]]
   // CHECK-NEXT: br i1 [[T0]]
-  // CHECK:      [[T0:%.*]] = load i8*, i8** [[CLEANUP_SAVE]]
-  // CHECK-NEXT: call void @llvm.objc.release(i8* [[T0]])
+  // CHECK:      [[T0:%.*]] = load ptr, ptr [[CLEANUP_SAVE]]
+  // CHECK-NEXT: call void @llvm.objc.release(ptr [[T0]])
   // CHECK-NEXT: br label
   //   And way down at the end of the loop:
-  // CHECK:      call void @llvm.objc.release(i8* [[RESULT]])
+  // CHECK:      call void @llvm.objc.release(ptr [[RESULT]])
 }
 
 void test3(int cond) {
@@ -154,53 +148,53 @@ void test3(int cond) {
   test2(cond);
 
   // CHECK: define{{.*}} void @test3(
-  // CHECK: %[[P:.*]] = alloca i8**, align 8
-  // CHECK: %[[_COMPOUNDLITERAL:.*]] = alloca [2 x i8*], align 8
+  // CHECK: %[[P:.*]] = alloca ptr, align 8
+  // CHECK: %[[_COMPOUNDLITERAL:.*]] = alloca [2 x ptr], align 8
   // CHECK: %[[CLEANUP_COND:.*]] = alloca i1, align 1
-  // CHECK: %[[_COMPOUNDLITERAL1:.*]] = alloca [2 x i8*], align 8
+  // CHECK: %[[_COMPOUNDLITERAL1:.*]] = alloca [2 x ptr], align 8
   // CHECK: %[[CLEANUP_COND4:.*]] = alloca i1, align 1
 
-  // CHECK: %[[ARRAYINIT_BEGIN:.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* %[[_COMPOUNDLITERAL]], i64 0, i64 0
-  // CHECK: %[[V2:.*]] = load i8*, i8** @g0, align 8
-  // CHECK: %[[V3:.*]] = call i8* @llvm.objc.retain(i8* %[[V2]])
-  // CHECK: store i8* %[[V3]], i8** %[[ARRAYINIT_BEGIN]], align 8
-  // CHECK: %[[ARRAYINIT_ELEMENT:.*]] = getelementptr inbounds i8*, i8** %[[ARRAYINIT_BEGIN]], i64 1
-  // CHECK: %[[V4:.*]] = load i8*, i8** @g1, align 8
-  // CHECK: %[[V5:.*]] = call i8* @llvm.objc.retain(i8* %[[V4]])
-  // CHECK: store i8* %[[V5]], i8** %[[ARRAYINIT_ELEMENT]], align 8
-  // CHECK: store i1 true, i1* %[[CLEANUP_COND]], align 1
-  // CHECK: %[[ARRAYDECAY:.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* %[[_COMPOUNDLITERAL]], i64 0, i64 0
-
-  // CHECK: %[[ARRAYINIT_BEGIN2:.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* %[[_COMPOUNDLITERAL1]], i64 0, i64 0
-  // CHECK: %[[V6:.*]] = load i8*, i8** @g1, align 8
-  // CHECK: %[[V7:.*]] = call i8* @llvm.objc.retain(i8* %[[V6]])
-  // CHECK: store i8* %[[V7]], i8** %[[ARRAYINIT_BEGIN2]], align 8
-  // CHECK: %[[ARRAYINIT_ELEMENT3:.*]] = getelementptr inbounds i8*, i8** %[[ARRAYINIT_BEGIN2]], i64 1
-  // CHECK: %[[V8:.*]] = load i8*, i8** @g0, align 8
-  // CHECK: %[[V9:.*]] = call i8* @llvm.objc.retain(i8* %[[V8]])
-  // CHECK: store i8* %[[V9]], i8** %[[ARRAYINIT_ELEMENT3]], align 8
-  // CHECK: store i1 true, i1* %[[CLEANUP_COND4]], align 1
-  // CHECK: %[[ARRAYDECAY5:.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* %[[_COMPOUNDLITERAL1]], i64 0, i64 0
-
-  // CHECK: %[[COND6:.*]] = phi i8** [ %[[ARRAYDECAY]], %{{.*}} ], [ %[[ARRAYDECAY5]], %{{.*}} ]
-  // CHECK: store i8** %[[COND6]], i8*** %[[P]], align 8
+  // CHECK: %[[ARRAYINIT_BEGIN:.*]] = getelementptr inbounds [2 x ptr], ptr %[[_COMPOUNDLITERAL]], i64 0, i64 0
+  // CHECK: %[[V2:.*]] = load ptr, ptr @g0, align 8
+  // CHECK: %[[V3:.*]] = call ptr @llvm.objc.retain(ptr %[[V2]])
+  // CHECK: store ptr %[[V3]], ptr %[[ARRAYINIT_BEGIN]], align 8
+  // CHECK: %[[ARRAYINIT_ELEMENT:.*]] = getelementptr inbounds ptr, ptr %[[ARRAYINIT_BEGIN]], i64 1
+  // CHECK: %[[V4:.*]] = load ptr, ptr @g1, align 8
+  // CHECK: %[[V5:.*]] = call ptr @llvm.objc.retain(ptr %[[V4]])
+  // CHECK: store ptr %[[V5]], ptr %[[ARRAYINIT_ELEMENT]], align 8
+  // CHECK: store i1 true, ptr %[[CLEANUP_COND]], align 1
+  // CHECK: %[[ARRAYDECAY:.*]] = getelementptr inbounds [2 x ptr], ptr %[[_COMPOUNDLITERAL]], i64 0, i64 0
+
+  // CHECK: %[[ARRAYINIT_BEGIN2:.*]] = getelementptr inbounds [2 x ptr], ptr %[[_COMPOUNDLITERAL1]], i64 0, i64 0
+  // CHECK: %[[V6:.*]] = load ptr, ptr @g1, align 8
+  // CHECK: %[[V7:.*]] = call ptr @llvm.objc.retain(ptr %[[V6]])
+  // CHECK: store ptr %[[V7]], ptr %[[ARRAYINIT_BEGIN2]], align 8
+  // CHECK: %[[ARRAYINIT_ELEMENT3:.*]] = getelementptr inbounds ptr, ptr %[[ARRAYINIT_BEGIN2]], i64 1
+  // CHECK: %[[V8:.*]] = load ptr, ptr @g0, align 8
+  // CHECK: %[[V9:.*]] = call ptr @llvm.objc.retain(ptr %[[V8]])
+  // CHECK: store ptr %[[V9]], ptr %[[ARRAYINIT_ELEMENT3]], align 8
+  // CHECK: store i1 true, ptr %[[CLEANUP_COND4]], align 1
+  // CHECK: %[[ARRAYDECAY5:.*]] = getelementptr inbounds [2 x ptr], ptr %[[_COMPOUNDLITERAL1]], i64 0, i64 0
+
+  // CHECK: %[[COND6:.*]] = phi ptr [ %[[ARRAYDECAY]], %{{.*}} ], [ %[[ARRAYDECAY5]], %{{.*}} ]
+  // CHECK: store ptr %[[COND6]], ptr %[[P]], align 8
   // CHECK: call void @test2(
 
-  // CHECK: %[[ARRAY_BEGIN:.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* %[[_COMPOUNDLITERAL1]], i32 0, i32 0
-  // CHECK: %[[V11:.*]] = getelementptr inbounds i8*, i8** %[[ARRAY_BEGIN]], i64 2
+  // CHECK: %[[ARRAY_BEGIN:.*]] = getelementptr inbounds [2 x ptr], ptr %[[_COMPOUNDLITERAL1]], i32 0, i32 0
+  // CHECK: %[[V11:.*]] = getelementptr inbounds ptr, ptr %[[ARRAY_BEGIN]], i64 2
 
-  // CHECK: %[[ARRAYDESTROY_ELEMENTPAST:.*]] = phi i8** [ %[[V11]], %{{.*}} ], [ %[[ARRAYDESTROY_ELEMENT:.*]], %{{.*}} ]
-  // CHECK: %[[ARRAYDESTROY_ELEMENT]] = getelementptr inbounds i8*, i8** %[[ARRAYDESTROY_ELEMENTPAST]], i64 -1
-  // CHECK: %[[V12:.*]] = load i8*, i8** %[[ARRAYDESTROY_ELEMENT]], align 8
-  // CHECK: call void @llvm.objc.release(i8* %[[V12]])
+  // CHECK: %[[ARRAYDESTROY_ELEMENTPAST:.*]] = phi ptr [ %[[V11]], %{{.*}} ], [ %[[ARRAYDESTROY_ELEMENT:.*]], %{{.*}} ]
+  // CHECK: %[[ARRAYDESTROY_ELEMENT]] = getelementptr inbounds ptr, ptr %[[ARRAYDESTROY_ELEMENTPAST]], i64 -1
+  // CHECK: %[[V12:.*]] = load ptr, ptr %[[ARRAYDESTROY_ELEMENT]], align 8
+  // CHECK: call void @llvm.objc.release(ptr %[[V12]])
 
-  // CHECK: %[[ARRAY_BEGIN10:.*]] = getelementptr inbounds [2 x i8*], [2 x i8*]* %[[_COMPOUNDLITERAL]], i32 0, i32 0
-  // CHECK: %[[V13:.*]] = getelementptr inbounds i8*, i8** %[[ARRAY_BEGIN10]], i64 2
+  // CHECK: %[[ARRAY_BEGIN10:.*]] = getelementptr inbounds [2 x ptr], ptr %[[_COMPOUNDLITERAL]], i32 0, i32 0
+  // CHECK: %[[V13:.*]] = getelementptr inbounds ptr, ptr %[[ARRAY_BEGIN10]], i64 2
 
-  // CHECK: %[[ARRAYDESTROY_ELEMENTPAST12:.*]] = phi i8** [ %[[V13]], %{{.*}} ], [ %[[ARRAYDESTROY_ELEMENT13:.*]], %{{.*}} ]
-  // CHECK: %[[ARRAYDESTROY_ELEMENT13]] = getelementptr inbounds i8*, i8** %[[ARRAYDESTROY_ELEMENTPAST12]], i64 -1
-  // CHECK: %[[V14:.*]] = load i8*, i8** %[[ARRAYDESTROY_ELEMENT13]], align 8
-  // CHECK: call void @llvm.objc.release(i8* %[[V14]])
+  // CHECK: %[[ARRAYDESTROY_ELEMENTPAST12:.*]] = phi ptr [ %[[V13]], %{{.*}} ], [ %[[ARRAYDESTROY_ELEMENT13:.*]], %{{.*}} ]
+  // CHECK: %[[ARRAYDESTROY_ELEMENT13]] = getelementptr inbounds ptr, ptr %[[ARRAYDESTROY_ELEMENTPAST12]], i64 -1
+  // CHECK: %[[V14:.*]] = load ptr, ptr %[[ARRAYDESTROY_ELEMENT13]], align 8
+  // CHECK: call void @llvm.objc.release(ptr %[[V14]])
 }
 
 // CHECK: attributes [[NUW]] = { nounwind }

diff  --git a/clang/test/CodeGenObjC/arc-unopt.m b/clang/test/CodeGenObjC/arc-unopt.m
index 2822ca2166e6e..8f247c64b9597 100644
--- a/clang/test/CodeGenObjC/arc-unopt.m
+++ b/clang/test/CodeGenObjC/arc-unopt.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -fobjc-runtime-has-weak -fblocks -fobjc-arc -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -fobjc-runtime-has-weak -fblocks -fobjc-arc -o - %s | FileCheck %s
 
 // A test to ensure that we generate fused calls at -O0.
 
@@ -7,34 +7,32 @@
   extern Test0 *test0_helper;
   return test0_helper;
 
-  // CHECK:      [[LD:%.*]] = load [[TEST0:%.*]]*, [[TEST0:%.*]]** @test0_helper
-  // CHECK-NEXT: [[T0:%.*]] = bitcast [[TEST0]]* [[LD]] to i8*
-  // CHECK-NEXT: [[T1:%.*]] = tail call i8* @llvm.objc.retainAutoreleaseReturnValue(i8* [[T0]])
-  // CHECK-NEXT: [[T2:%.*]] = bitcast i8* [[T1]] to [[TEST0]]*
-  // CHECK-NEXT: ret [[TEST0]]* [[T2]]
+  // CHECK:      [[LD:%.*]] = load ptr, ptr @test0_helper
+  // CHECK-NEXT: [[T1:%.*]] = tail call ptr @llvm.objc.retainAutoreleaseReturnValue(ptr [[LD]])
+  // CHECK-NEXT: ret ptr [[T1]]
 }
 
 id test1(void) {
   extern id test1_helper;
   return test1_helper;
 
-  // CHECK:      [[LD:%.*]] = load i8*, i8** @test1_helper
-  // CHECK-NEXT: [[T0:%.*]] = tail call i8* @llvm.objc.retainAutoreleaseReturnValue(i8* [[LD]])
-  // CHECK-NEXT: ret i8* [[T0]]
+  // CHECK:      [[LD:%.*]] = load ptr, ptr @test1_helper
+  // CHECK-NEXT: [[T0:%.*]] = tail call ptr @llvm.objc.retainAutoreleaseReturnValue(ptr [[LD]])
+  // CHECK-NEXT: ret ptr [[T0]]
 }
 
 void test2(void) {
-  // CHECK:      [[X:%.*]] = alloca i8*
-  // CHECK-NEXT: store i8* null, i8** [[X]]
-  // CHECK-NEXT: call void @llvm.objc.destroyWeak(i8** [[X]])
+  // CHECK:      [[X:%.*]] = alloca ptr
+  // CHECK-NEXT: store ptr null, ptr [[X]]
+  // CHECK-NEXT: call void @llvm.objc.destroyWeak(ptr [[X]])
   // CHECK-NEXT: ret void
   __weak id x;
 }
 
 id test3(void) {
   extern id test3_helper(void);
-  // CHECK:      [[T0:%.*]] = call i8* @test3_helper()
-  // CHECK-NEXT: ret i8* [[T0]]
+  // CHECK:      [[T0:%.*]] = call ptr @test3_helper()
+  // CHECK-NEXT: ret ptr [[T0]]
   return test3_helper();
 }
 
@@ -42,9 +40,8 @@ @interface Test4 { id x; } @end
 @interface Test4_sub : Test4 { id y; } @end
 Test4 *test4(void) {
   extern Test4_sub *test4_helper(void);
-  // CHECK:      [[T0:%.*]] = call [[TEST4S:%.*]]* @test4_helper()
-  // CHECK-NEXT: [[T1:%.*]] = bitcast [[TEST4S]]* [[T0]] to [[TEST4:%.*]]*
-  // CHECK-NEXT: ret [[TEST4]]* [[T1]]
+  // CHECK:      [[T0:%.*]] = call ptr @test4_helper()
+  // CHECK-NEXT: ret ptr [[T0]]
   return test4_helper();
 }
 
@@ -56,14 +53,12 @@ void test5(void) {
     y = 0;
 
 // CHECK-LABEL:    define{{.*}} void @test5()
-// CHECK:      [[X:%.*]] = alloca [[TEST5:%.*]]*,
-// CHECK-NEXT: [[Y:%.*]] = alloca [[TEST5:%.*]]*,
-// CHECK-NEXT: store [[TEST5]]* null, [[TEST5]]** [[X]],
-// CHECK-NEXT: store [[TEST5]]* null, [[TEST5]]** [[Y]],
-// CHECK-NEXT: [[T0:%.*]] = load [[TEST5]]*, [[TEST5]]** [[Y]],
-// CHECK-NEXT: [[T1:%.*]] = bitcast [[TEST5]]** [[X]] to i8**
-// CHECK-NEXT: [[T2:%.*]] = bitcast [[TEST5]]* [[T0]] to i8*
-// CHECK-NEXT: call void @llvm.objc.storeStrong(i8** [[T1]], i8* [[T2]])
-// CHECK-NEXT: [[T3:%.*]] = icmp ne [[TEST5]]* [[T0]], null
+// CHECK:      [[X:%.*]] = alloca ptr,
+// CHECK-NEXT: [[Y:%.*]] = alloca ptr,
+// CHECK-NEXT: store ptr null, ptr [[X]],
+// CHECK-NEXT: store ptr null, ptr [[Y]],
+// CHECK-NEXT: [[T0:%.*]] = load ptr, ptr [[Y]],
+// CHECK-NEXT: call void @llvm.objc.storeStrong(ptr [[X]], ptr [[T0]])
+// CHECK-NEXT: [[T3:%.*]] = icmp ne ptr [[T0]], null
 // CHECK-NEXT: br i1 [[T3]],
 }

diff  --git a/clang/test/CodeGenObjC/arc-unoptimized-byref-var.m b/clang/test/CodeGenObjC/arc-unoptimized-byref-var.m
index 66fa21ea02fe4..870580c984f7c 100644
--- a/clang/test/CodeGenObjC/arc-unoptimized-byref-var.m
+++ b/clang/test/CodeGenObjC/arc-unoptimized-byref-var.m
@@ -1,16 +1,16 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -fblocks -fobjc-arc -fobjc-runtime-has-weak -o - %s | FileCheck -check-prefix=CHECK-UNOPT %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -fblocks -fobjc-arc -fobjc-runtime-has-weak -o - %s | FileCheck -check-prefix=CHECK-UNOPT %s
 // rdar://12530881
 
 void test19(void) {
    __block id x;
    ^{ (void)x; };
 // CHECK-UNOPT-LABEL: define internal void @__Block_byref_object_copy
-// CHECK-UNOPT: [[X:%.*]] = getelementptr inbounds [[BYREF_T:%.*]], [[BYREF_T:%.*]]* [[VAR:%.*]], i32 0, i32 6
-// CHECK-UNOPT: [[X2:%.*]] = getelementptr inbounds [[BYREF_T:%.*]], [[BYREF_T:%.*]]* [[VAR1:%.*]], i32 0, i32 6
-// CHECK-UNOPT-NEXT: [[SIX:%.*]] = load i8*, i8** [[X2]], align 8
-// CHECK-UNOPT-NEXT: store i8* null, i8** [[X]], align 8
-// CHECK-UNOPT-NEXT: call void @llvm.objc.storeStrong(i8** [[X]], i8* [[SIX]]) [[NUW:#[0-9]+]]
-// CHECK-UNOPT-NEXT: call void @llvm.objc.storeStrong(i8** [[X2]], i8* null) [[NUW]]
+// CHECK-UNOPT: [[X:%.*]] = getelementptr inbounds [[BYREF_T:%.*]], ptr [[VAR:%.*]], i32 0, i32 6
+// CHECK-UNOPT: [[X2:%.*]] = getelementptr inbounds [[BYREF_T:%.*]], ptr [[VAR1:%.*]], i32 0, i32 6
+// CHECK-UNOPT-NEXT: [[SIX:%.*]] = load ptr, ptr [[X2]], align 8
+// CHECK-UNOPT-NEXT: store ptr null, ptr [[X]], align 8
+// CHECK-UNOPT-NEXT: call void @llvm.objc.storeStrong(ptr [[X]], ptr [[SIX]]) [[NUW:#[0-9]+]]
+// CHECK-UNOPT-NEXT: call void @llvm.objc.storeStrong(ptr [[X2]], ptr null) [[NUW]]
 // CHECK-UNOPT-NEXT: ret void
 }
 

diff  --git a/clang/test/CodeGenObjC/arc-weak-property.m b/clang/test/CodeGenObjC/arc-weak-property.m
index 765be75805e60..b62fdec3a0603 100644
--- a/clang/test/CodeGenObjC/arc-weak-property.m
+++ b/clang/test/CodeGenObjC/arc-weak-property.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -fobjc-runtime-has-weak -fblocks -fobjc-arc -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -fobjc-runtime-has-weak -fblocks -fobjc-arc -o - %s | FileCheck %s
 // rdar://8899430
 
 @interface WeakPropertyTest {
@@ -11,45 +11,39 @@ @implementation WeakPropertyTest
 @synthesize PROP;
 @end
 
-// CHECK:     define internal i8* @"\01-[WeakPropertyTest PROP]"
-// CHECK:       [[SELF:%.*]] = alloca [[WPT:%.*]]*,
-// CHECK-NEXT:  [[CMD:%.*]] = alloca i8*,
-// CHECK-NEXT:  store [[WPT]]* {{%.*}}, [[WPT]]** [[SELF]]
-// CHECK-NEXT:  store i8* {{%.*}}, i8** [[CMD]]
-// CHECK-NEXT:  [[T0:%.*]] = load [[WPT]]*, [[WPT]]** [[SELF]]
-// CHECK-NEXT:  [[T1:%.*]] = load i64, i64* @"OBJC_IVAR_$_WeakPropertyTest.PROP"
-// CHECK-NEXT:  [[T2:%.*]] = bitcast [[WPT]]* [[T0]] to i8*
-// CHECK-NEXT:  [[T3:%.*]] = getelementptr inbounds i8, i8* [[T2]], i64 [[T1]]
-// CHECK-NEXT:  [[T4:%.*]] = bitcast i8* [[T3]] to i8**
-// CHECK-NEXT:  [[T5:%.*]] = call i8* @llvm.objc.loadWeakRetained(i8** [[T4]])
-// CHECK-NEXT:  [[T6:%.*]] = tail call i8* @llvm.objc.autoreleaseReturnValue(i8* [[T5]])
-// CHECK-NEXT:  ret i8* [[T6]]
+// CHECK:     define internal ptr @"\01-[WeakPropertyTest PROP]"
+// CHECK:       [[SELF:%.*]] = alloca ptr,
+// CHECK-NEXT:  [[CMD:%.*]] = alloca ptr,
+// CHECK-NEXT:  store ptr {{%.*}}, ptr [[SELF]]
+// CHECK-NEXT:  store ptr {{%.*}}, ptr [[CMD]]
+// CHECK-NEXT:  [[T0:%.*]] = load ptr, ptr [[SELF]]
+// CHECK-NEXT:  [[T1:%.*]] = load i64, ptr @"OBJC_IVAR_$_WeakPropertyTest.PROP"
+// CHECK-NEXT:  [[T3:%.*]] = getelementptr inbounds i8, ptr [[T0]], i64 [[T1]]
+// CHECK-NEXT:  [[T5:%.*]] = call ptr @llvm.objc.loadWeakRetained(ptr [[T3]])
+// CHECK-NEXT:  [[T6:%.*]] = tail call ptr @llvm.objc.autoreleaseReturnValue(ptr [[T5]])
+// CHECK-NEXT:  ret ptr [[T6]]
 
 // CHECK:     define internal void @"\01-[WeakPropertyTest setPROP:]"
-// CHECK:       [[SELF:%.*]] = alloca [[WPT:%.*]]*,
-// CHECK-NEXT:  [[CMD:%.*]] = alloca i8*,
-// CHECK-NEXT:  [[PROP:%.*]] = alloca i8*,
-// CHECK-NEXT:  store [[WPT]]* {{%.*}}, [[WPT]]** [[SELF]]
-// CHECK-NEXT:  store i8* {{%.*}}, i8** [[CMD]]
-// CHECK-NEXT:  store i8* {{%.*}}, i8** [[PROP]]
-// CHECK-NEXT:  [[V:%.*]] = load i8*, i8** [[PROP]]
-// CHECK-NEXT:  [[T0:%.*]] = load [[WPT]]*, [[WPT]]** [[SELF]]
-// CHECK-NEXT:  [[T1:%.*]] = load i64, i64* @"OBJC_IVAR_$_WeakPropertyTest.PROP"
-// CHECK-NEXT:  [[T2:%.*]] = bitcast [[WPT]]* [[T0]] to i8*
-// CHECK-NEXT:  [[T3:%.*]] = getelementptr inbounds i8, i8* [[T2]], i64 [[T1]]
-// CHECK-NEXT:  [[T4:%.*]] = bitcast i8* [[T3]] to i8**
-// CHECK-NEXT:  call i8* @llvm.objc.storeWeak(i8** [[T4]], i8* [[V]])
+// CHECK:       [[SELF:%.*]] = alloca ptr,
+// CHECK-NEXT:  [[CMD:%.*]] = alloca ptr,
+// CHECK-NEXT:  [[PROP:%.*]] = alloca ptr,
+// CHECK-NEXT:  store ptr {{%.*}}, ptr [[SELF]]
+// CHECK-NEXT:  store ptr {{%.*}}, ptr [[CMD]]
+// CHECK-NEXT:  store ptr {{%.*}}, ptr [[PROP]]
+// CHECK-NEXT:  [[V:%.*]] = load ptr, ptr [[PROP]]
+// CHECK-NEXT:  [[T0:%.*]] = load ptr, ptr [[SELF]]
+// CHECK-NEXT:  [[T1:%.*]] = load i64, ptr @"OBJC_IVAR_$_WeakPropertyTest.PROP"
+// CHECK-NEXT:  [[T3:%.*]] = getelementptr inbounds i8, ptr [[T0]], i64 [[T1]]
+// CHECK-NEXT:  call ptr @llvm.objc.storeWeak(ptr [[T3]], ptr [[V]])
 // CHECK-NEXT:  ret void
 
 // CHECK:     define internal void @"\01-[WeakPropertyTest .cxx_destruct]"
-// CHECK:       [[SELF:%.*]] = alloca [[WPT:%.*]]*,
-// CHECK-NEXT:  [[CMD:%.*]] = alloca i8*,
-// CHECK-NEXT:  store [[WPT]]* {{%.*}}, [[WPT]]** [[SELF]]
-// CHECK-NEXT:  store i8* {{%.*}}, i8** [[CMD]]
-// CHECK-NEXT:  [[T0:%.*]] = load [[WPT]]*, [[WPT]]** [[SELF]]
-// CHECK-NEXT:  [[T1:%.*]] = load i64, i64* @"OBJC_IVAR_$_WeakPropertyTest.PROP"
-// CHECK-NEXT:  [[T2:%.*]] = bitcast [[WPT]]* [[T0]] to i8*
-// CHECK-NEXT:  [[T3:%.*]] = getelementptr inbounds i8, i8* [[T2]], i64 [[T1]]
-// CHECK-NEXT:  [[T4:%.*]] = bitcast i8* [[T3]] to i8**
-// CHECK-NEXT:  call void @llvm.objc.destroyWeak(i8** [[T4]])
+// CHECK:       [[SELF:%.*]] = alloca ptr,
+// CHECK-NEXT:  [[CMD:%.*]] = alloca ptr,
+// CHECK-NEXT:  store ptr {{%.*}}, ptr [[SELF]]
+// CHECK-NEXT:  store ptr {{%.*}}, ptr [[CMD]]
+// CHECK-NEXT:  [[T0:%.*]] = load ptr, ptr [[SELF]]
+// CHECK-NEXT:  [[T1:%.*]] = load i64, ptr @"OBJC_IVAR_$_WeakPropertyTest.PROP"
+// CHECK-NEXT:  [[T3:%.*]] = getelementptr inbounds i8, ptr [[T0]], i64 [[T1]]
+// CHECK-NEXT:  call void @llvm.objc.destroyWeak(ptr [[T3]])
 // CHECK-NEXT:  ret void

diff  --git a/clang/test/CodeGenObjC/arc-weak.m b/clang/test/CodeGenObjC/arc-weak.m
index d0a22acce9bd8..9cc622cd20eda 100644
--- a/clang/test/CodeGenObjC/arc-weak.m
+++ b/clang/test/CodeGenObjC/arc-weak.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -fblocks -fobjc-arc -fobjc-runtime-has-weak -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -fblocks -fobjc-arc -fobjc-runtime-has-weak -o - %s | FileCheck %s
 
 __attribute((objc_root_class)) @interface A @end
 @interface B : A @end
@@ -9,12 +9,8 @@ void test0(__weak B **src) {
   __weak A *dest = *src;
 }
 // CHECK-LABEL: define{{.*}} void @test0
-// CHECK:       [[SRC:%.*]] = alloca [[B:%.*]]**, align 8
-// CHECK:       [[DEST:%.*]] = alloca [[A:%.*]]*, align 8
-// CHECK:       [[T0:%.*]] = load [[B]]**, [[B]]*** [[SRC]], align 8
-// CHECK-NEXT:  [[T1:%.*]] = bitcast [[B]]** [[T0]] to [[A]]**
-// CHECK-NEXT:  [[T2:%.*]] = bitcast [[A]]** [[DEST]] to i8**
-// CHECK-NEXT:  [[T3:%.*]] = bitcast [[A]]** [[T1]] to i8**
-// CHECK-NEXT:  call void @llvm.objc.copyWeak(i8** [[T2]], i8** [[T3]])
-// CHECK-NEXT:  [[T0:%.*]] = bitcast [[A]]** [[DEST]] to i8**
-// CHECK:       call void @llvm.objc.destroyWeak(i8** [[T0]])
+// CHECK:       [[SRC:%.*]] = alloca ptr, align 8
+// CHECK:       [[DEST:%.*]] = alloca ptr, align 8
+// CHECK:       [[T0:%.*]] = load ptr, ptr [[SRC]], align 8
+// CHECK-NEXT:  call void @llvm.objc.copyWeak(ptr [[DEST]], ptr [[T0]])
+// CHECK:       call void @llvm.objc.destroyWeak(ptr [[DEST]])

diff  --git a/clang/test/CodeGenObjC/arc-with-atthrow.m b/clang/test/CodeGenObjC/arc-with-atthrow.m
index 40a10d63295be..7fddad22ed178 100644
--- a/clang/test/CodeGenObjC/arc-with-atthrow.m
+++ b/clang/test/CodeGenObjC/arc-with-atthrow.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -fobjc-arc -fobjc-exceptions -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -fobjc-arc -fobjc-exceptions -o - %s | FileCheck %s
 // pr10411
 // rdar://10042689
 
@@ -10,10 +10,10 @@ void test(void) {
 // TODO: We should probably emit this specific pattern without the reclaim.
 
 // CHECK-LABEL:    define{{.*}} void @test()
-// CHECK:      [[T0:%.*]] = call i8* @make()
-// CHECK-NEXT: [[T1:%.*]] = notail call i8* @llvm.objc.retainAutoreleasedReturnValue(i8* [[T0]])
-// CHECK-NEXT: [[T2:%.*]] = call i8* @llvm.objc.autorelease(i8* [[T1]])
-// CHECK-NEXT: call void @objc_exception_throw(i8* [[T2]]) [[NR:#[0-9]+]]
+// CHECK:      [[T0:%.*]] = call ptr @make()
+// CHECK-NEXT: [[T1:%.*]] = notail call ptr @llvm.objc.retainAutoreleasedReturnValue(ptr [[T0]])
+// CHECK-NEXT: [[T2:%.*]] = call ptr @llvm.objc.autorelease(ptr [[T1]])
+// CHECK-NEXT: call void @objc_exception_throw(ptr [[T2]]) [[NR:#[0-9]+]]
 // CHECK-NEXT: unreachable
 
 // CHECK: attributes [[NR]] = { noreturn }

diff  --git a/clang/test/CodeGenObjC/arc.ll b/clang/test/CodeGenObjC/arc.ll
index cfc88c3c7eb71..90187de0a6e7e 100644
--- a/clang/test/CodeGenObjC/arc.ll
+++ b/clang/test/CodeGenObjC/arc.ll
@@ -1,25 +1,25 @@
 ; RUN: %clang_cc1 -triple x86_64-apple-darwin10 -Os -emit-llvm -fobjc-arc -o - %s | FileCheck %s
 
-declare i8* @llvm.objc.retain(i8*)
-declare void @llvm.objc.release(i8*)
+declare ptr @llvm.objc.retain(ptr)
+declare void @llvm.objc.release(ptr)
 
 ; CHECK-LABEL: define void @test(
 ; CHECK-NOT: @objc_
 ; CHECK: }
-define void @test(i8* %x, i1* %p) nounwind {
+define void @test(ptr %x, ptr %p) nounwind {
 entry:
   br label %loop
 
 loop:
-  call i8* @llvm.objc.retain(i8* %x)
-  %q = load i1, i1* %p
+  call ptr @llvm.objc.retain(ptr %x)
+  %q = load i1, ptr %p
   br i1 %q, label %loop.more, label %exit
 
 loop.more:
-  call void @llvm.objc.release(i8* %x)
+  call void @llvm.objc.release(ptr %x)
   br label %loop
 
 exit:
-  call void @llvm.objc.release(i8* %x)
+  call void @llvm.objc.release(ptr %x)
   ret void
 }

diff  --git a/clang/test/CodeGenObjC/arm-atomic-scalar-setter-getter.m b/clang/test/CodeGenObjC/arm-atomic-scalar-setter-getter.m
index a6a8dc5ecb830..e179bdacb2ef7 100644
--- a/clang/test/CodeGenObjC/arm-atomic-scalar-setter-getter.m
+++ b/clang/test/CodeGenObjC/arm-atomic-scalar-setter-getter.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple armv7-apple-darwin10 -emit-llvm -o - %s | FileCheck -check-prefix=CHECK-ARM %s
+// RUN: %clang_cc1 -triple armv7-apple-darwin10 -emit-llvm -o - %s | FileCheck -check-prefix=CHECK-ARM %s
 // rdar://7761305
 
 @interface I
@@ -8,6 +8,6 @@ @interface I
 @implementation I
 @synthesize LONG_PROP;
 @end
-// CHECK-ARM: call void @objc_copyStruct(i8* noundef %{{.*}}, i8* noundef %{{.*}}, i32 noundef 8, i1 noundef zeroext true, i1 noundef zeroext false)
-// CHECK-ARM: call void @objc_copyStruct(i8* noundef %{{.*}}, i8* noundef %{{.*}}, i32 noundef 8, i1 noundef zeroext true, i1 noundef zeroext false)
+// CHECK-ARM: call void @objc_copyStruct(ptr noundef %{{.*}}, ptr noundef %{{.*}}, i32 noundef 8, i1 noundef zeroext true, i1 noundef zeroext false)
+// CHECK-ARM: call void @objc_copyStruct(ptr noundef %{{.*}}, ptr noundef %{{.*}}, i32 noundef 8, i1 noundef zeroext true, i1 noundef zeroext false)
 

diff  --git a/clang/test/CodeGenObjC/arm64-int32-ivar.m b/clang/test/CodeGenObjC/arm64-int32-ivar.m
index e84a838ffe048..76d0543b20b11 100644
--- a/clang/test/CodeGenObjC/arm64-int32-ivar.m
+++ b/clang/test/CodeGenObjC/arm64-int32-ivar.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple arm64-apple-ios -emit-llvm  -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple arm64-apple-ios -emit-llvm  -o - %s | FileCheck %s
 // rdar://12617764
 
 // CHECK: @"OBJC_IVAR_$_I.IVAR2" = global i32 8
@@ -11,7 +11,7 @@ @interface I
 @end
 
 @implementation I
-// CHECK: [[IVAR:%.*]] = load i32, i32* @"OBJC_IVAR_$_I.IVAR2"
+// CHECK: [[IVAR:%.*]] = load i32, ptr @"OBJC_IVAR_$_I.IVAR2"
 // CHECK: [[CONV:%.*]] = sext i32 [[IVAR]] to i64
 - (id) METH { return IVAR2; }
 @end

diff  --git a/clang/test/CodeGenObjC/atomic-aggregate-property.m b/clang/test/CodeGenObjC/atomic-aggregate-property.m
index a37659c8e8151..b574cc3ce4a66 100644
--- a/clang/test/CodeGenObjC/atomic-aggregate-property.m
+++ b/clang/test/CodeGenObjC/atomic-aggregate-property.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -no-enable-noundef-analysis -triple x86_64-apple-darwin10  -fobjc-gc -emit-llvm -o - %s | FileCheck -check-prefix CHECK-LP64 %s
-// RUN: %clang_cc1 -no-opaque-pointers -no-enable-noundef-analysis -x objective-c++ -triple x86_64-apple-darwin10  -fobjc-gc -emit-llvm -o - %s | FileCheck -check-prefix CHECK-LP64 %s
+// RUN: %clang_cc1 -no-enable-noundef-analysis -triple x86_64-apple-darwin10  -fobjc-gc -emit-llvm -o - %s | FileCheck -check-prefix CHECK-LP64 %s
+// RUN: %clang_cc1 -no-enable-noundef-analysis -x objective-c++ -triple x86_64-apple-darwin10  -fobjc-gc -emit-llvm -o - %s | FileCheck -check-prefix CHECK-LP64 %s
 // rdar: // 7849824
 // <rdar://problem/12547611>
 
@@ -29,22 +29,22 @@ @implementation A
 @synthesize a;
 @end
 // CHECK-LP64: define internal double @"\01-[A x]"(
-// CHECK-LP64: load atomic i64, i64* {{%.*}} unordered, align 8
+// CHECK-LP64: load atomic i64, ptr {{%.*}} unordered, align 8
 
 // CHECK-LP64: define internal void @"\01-[A setX:]"(
-// CHECK-LP64: store atomic i64 {{%.*}}, i64* {{%.*}} unordered, align 8
+// CHECK-LP64: store atomic i64 {{%.*}}, ptr {{%.*}} unordered, align 8
 
 // CHECK-LP64: define internal void @"\01-[A y]"(
-// CHECK-LP64: call void @objc_copyStruct(i8* {{%.*}}, i8* {{%.*}}, i64 32, i1 zeroext true, i1 zeroext false)
+// CHECK-LP64: call void @objc_copyStruct(ptr {{%.*}}, ptr {{%.*}}, i64 32, i1 zeroext true, i1 zeroext false)
 
 // CHECK-LP64: define internal void @"\01-[A setY:]"(
-// CHECK-LP64: call void @objc_copyStruct(i8* {{%.*}}, i8* {{%.*}}, i64 32, i1 zeroext true, i1 zeroext false)
+// CHECK-LP64: call void @objc_copyStruct(ptr {{%.*}}, ptr {{%.*}}, i64 32, i1 zeroext true, i1 zeroext false)
 
 // CHECK-LP64: define internal void @"\01-[A z]"(
-// CHECK-LP64: call i8* @objc_memmove_collectable(
+// CHECK-LP64: call ptr @objc_memmove_collectable(
 
 // CHECK-LP64: define internal void @"\01-[A setZ:]"(
-// CHECK-LP64: call i8* @objc_memmove_collectable(
+// CHECK-LP64: call ptr @objc_memmove_collectable(
 
 // CHECK-LP64: define internal void @"\01-[A a]"(
 // (do nothing)

diff  --git a/clang/test/CodeGenObjC/attr-exception.m b/clang/test/CodeGenObjC/attr-exception.m
index 8769630af2f98..5a038796d24f4 100644
--- a/clang/test/CodeGenObjC/attr-exception.m
+++ b/clang/test/CodeGenObjC/attr-exception.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -fexceptions -fobjc-exceptions -o - %s | FileCheck %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -fexceptions -fobjc-exceptions -fvisibility=hidden -o - %s | FileCheck -check-prefix=CHECK-HIDDEN %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -fexceptions -fobjc-exceptions -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -fexceptions -fobjc-exceptions -fvisibility=hidden -o - %s | FileCheck -check-prefix=CHECK-HIDDEN %s
 
 __attribute__((objc_root_class)) 
 @interface Root {
@@ -13,8 +13,8 @@ @interface A : Root
 
 @implementation A
 @end
-// CHECK: @"OBJC_EHTYPE_$_A" ={{.*}} global {{%.*}} { i8** getelementptr inbounds (i8*, i8** @objc_ehtype_vtable, i32 2)
-// CHECK-HIDDEN: @"OBJC_EHTYPE_$_A" = hidden global {{%.*}} { i8** getelementptr inbounds (i8*, i8** @objc_ehtype_vtable, i32 2)
+// CHECK: @"OBJC_EHTYPE_$_A" ={{.*}} global {{%.*}} { ptr getelementptr inbounds (ptr, ptr @objc_ehtype_vtable, i32 2)
+// CHECK-HIDDEN: @"OBJC_EHTYPE_$_A" = hidden global {{%.*}} { ptr getelementptr inbounds (ptr, ptr @objc_ehtype_vtable, i32 2)
 
 __attribute__((objc_exception))
 __attribute__((visibility("default")))
@@ -23,5 +23,5 @@ @interface B : Root
 
 @implementation B
 @end
-// CHECK: @"OBJC_EHTYPE_$_B" ={{.*}} global {{%.*}} { i8** getelementptr inbounds (i8*, i8** @objc_ehtype_vtable, i32 2)
-// CHECK-HIDDEN: @"OBJC_EHTYPE_$_B" ={{.*}} global {{%.*}} { i8** getelementptr inbounds (i8*, i8** @objc_ehtype_vtable, i32 2)
+// CHECK: @"OBJC_EHTYPE_$_B" ={{.*}} global {{%.*}} { ptr getelementptr inbounds (ptr, ptr @objc_ehtype_vtable, i32 2)
+// CHECK-HIDDEN: @"OBJC_EHTYPE_$_B" ={{.*}} global {{%.*}} { ptr getelementptr inbounds (ptr, ptr @objc_ehtype_vtable, i32 2)

diff  --git a/clang/test/CodeGenObjC/attr-noreturn.m b/clang/test/CodeGenObjC/attr-noreturn.m
index 7645fd9a7cb93..70a117616fda9 100644
--- a/clang/test/CodeGenObjC/attr-noreturn.m
+++ b/clang/test/CodeGenObjC/attr-noreturn.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -o - %s | FileCheck %s -check-prefix=CHECK -check-prefix=CHECK-MRC
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -fobjc-arc -o - %s | FileCheck %s -check-prefix=CHECK -check-prefix=CHECK-ARC
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -o - %s | FileCheck %s -check-prefix=CHECK -check-prefix=CHECK-MRC
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -fobjc-arc -o - %s | FileCheck %s -check-prefix=CHECK -check-prefix=CHECK-ARC
 
 __attribute__((objc_root_class))
 @interface Root
@@ -22,14 +22,14 @@ void testInstanceMethod(Derived *x) {
   [x fail];
 }
 // CHECK-LABEL: @testInstanceMethod
-// CHECK: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*)*)(i8* {{.*}}, i8* {{.*}}){{$}}
+// CHECK: call void @objc_msgSend(ptr {{.*}}, ptr {{.*}}){{$}}
 
 // A direct call of a class method will normally never have a null receiver.
 void testClassMethod(void) {
   [Derived abort];
 }
 // CHECK-LABEL: @testClassMethod
-// CHECK: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*)*)(i8* {{.*}}, i8* {{.*}}) [[NORETURN:#[0-9]+]]
+// CHECK: call void @objc_msgSend(ptr {{.*}}, ptr {{.*}}) [[NORETURN:#[0-9]+]]
 
 __attribute__((weak_import))
 @interface WeakMiddle : Base
@@ -44,7 +44,7 @@ void testWeakImport(void) {
   [WeakDerived abort];
 }
 // CHECK-LABEL: @testWeakImport
-// CHECK: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*)*)(i8* {{.*}}, i8* {{.*}}){{$}}
+// CHECK: call void @objc_msgSend(ptr {{.*}}, ptr {{.*}}){{$}}
 
 @interface Derived (MyMethods)
 @end
@@ -59,8 +59,8 @@ - (void) testSelfInstanceMethod {
   [self fail];
 }
 // CHECK-LABEL: [Derived(MyMethods) testSelfInstanceMethod]
-// CHECK-MRC: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*)*)(i8* {{.*}}, i8* {{.*}}){{$}}
-// CHECK-ARC: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*)*)(i8* {{.*}}, i8* {{.*}}) [[NORETURN]]
+// CHECK-MRC: call void @objc_msgSend(ptr {{.*}}, ptr {{.*}}){{$}}
+// CHECK-ARC: call void @objc_msgSend(ptr {{.*}}, ptr {{.*}}) [[NORETURN]]
 
 // The ARC rule doesn't apply in -init methods.
 - (id) initWhileTestingSelfInstanceMethod {
@@ -69,15 +69,15 @@ - (id) initWhileTestingSelfInstanceMethod {
   return self;
 }
 // CHECK-LABEL: [Derived(MyMethods) initWhileTestingSelfInstanceMethod]
-// CHECK: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*)*)(i8* {{.*}}, i8* {{.*}}){{$}}
+// CHECK: call void @objc_msgSend(ptr {{.*}}, ptr {{.*}}){{$}}
 
 // Same thing applies to class methods.
 + (void) testSelfClassMethod {
   [self abort];
 }
 // CHECK-LABEL: [Derived(MyMethods) testSelfClassMethod]
-// CHECK-MRC: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*)*)(i8* {{.*}}, i8* {{.*}}){{$}}
-// CHECK-ARC: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*)*)(i8* {{.*}}, i8* {{.*}}) [[NORETURN]]
+// CHECK-MRC: call void @objc_msgSend(ptr {{.*}}, ptr {{.*}}){{$}}
+// CHECK-ARC: call void @objc_msgSend(ptr {{.*}}, ptr {{.*}}) [[NORETURN]]
 
 // Super invocations may never be used with a null pointer; this is a
 // constraint on user code when it isn't enforced by the ARC const-self
@@ -86,13 +86,13 @@ - (void) testSuperInstanceMethod {
   [super fail];
 }
 // CHECK-LABEL: [Derived(MyMethods) testSuperInstanceMethod]
-// CHECK: call void bitcast (i8* ([[SUPER_T:%.*]]*, i8*, ...)* @objc_msgSendSuper2 to void ([[SUPER_T]]*, i8*)*)([[SUPER_T]]* {{.*}}, i8* {{.*}}) [[NORETURN]]
+// CHECK: call void @objc_msgSendSuper2(ptr {{.*}}, ptr {{.*}}) [[NORETURN]]
 
 + (void) testSuperClassMethod {
   [super abort];
 }
 // CHECK-LABEL: [Derived(MyMethods) testSuperClassMethod]
-// CHECK: call void bitcast (i8* ([[SUPER_T]]*, i8*, ...)* @objc_msgSendSuper2 to void ([[SUPER_T]]*, i8*)*)([[SUPER_T]]* {{.*}}, i8* {{.*}}) [[NORETURN]]
+// CHECK: call void @objc_msgSendSuper2(ptr {{.*}}, ptr {{.*}}) [[NORETURN]]
 @end
 
 // CHECK: attributes [[NORETURN]] = { noreturn }

diff  --git a/clang/test/CodeGenObjC/attr-objc-arc-inert.m b/clang/test/CodeGenObjC/attr-objc-arc-inert.m
index e830250b62421..bbfe6b3c17d94 100644
--- a/clang/test/CodeGenObjC/attr-objc-arc-inert.m
+++ b/clang/test/CodeGenObjC/attr-objc-arc-inert.m
@@ -1,7 +1,7 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -fblocks -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -fblocks -emit-llvm -o - %s | FileCheck %s
 
-// CHECK: @{{.*}} = private global %{{.*}} { i32* getelementptr inbounds ([0 x i32], [0 x i32]* @__CFConstantStringClassReference, i32 0, i32 0), i32 1992, i8* getelementptr inbounds ([4 x i8], [4 x i8]* @{{.*}}, i32 0, i32 0), i64 3 }, section "__DATA,__cfstring", align 8 #[[ATTRNUM0:[0-9]+]]
-// CHECK: @{{.*}} = internal constant { i8**, i32, i32, i8*, %{{.*}}* } { i8** @_NSConcreteGlobalBlock, i32 1342177280, i32 0, i8* bitcast (i32 (i8*)* @{{.*}} to i8*), %{{.*}}* bitcast ({ i64, i64, i8*, i8* }* @{{.*}} to %{{.*}}*) }, align 8 #[[ATTRNUM0]]
+// CHECK: @{{.*}} = private global %{{.*}} { ptr @__CFConstantStringClassReference, i32 1992, ptr @{{.*}}, i64 3 }, section "__DATA,__cfstring", align 8 #[[ATTRNUM0:[0-9]+]]
+// CHECK: @{{.*}} = internal constant { ptr, i32, i32, ptr, ptr } { ptr @_NSConcreteGlobalBlock, i32 1342177280, i32 0, ptr @{{.*}}, ptr @{{.*}} }, align 8 #[[ATTRNUM0]]
 
 @class NSString;
 

diff  --git a/clang/test/CodeGenObjC/attr-objc-runtime-visible.m b/clang/test/CodeGenObjC/attr-objc-runtime-visible.m
index 4899ad9c5074a..21415cc8ac772 100644
--- a/clang/test/CodeGenObjC/attr-objc-runtime-visible.m
+++ b/clang/test/CodeGenObjC/attr-objc-runtime-visible.m
@@ -1,6 +1,6 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin -fobjc-runtime=macosx-10.9.0 -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin -fobjc-runtime=macosx-10.9.0 -emit-llvm %s -o - | FileCheck %s
 
-// RUN: %clang_cc1 -no-opaque-pointers -triple i386-apple-darwin -fobjc-runtime=macosx-fragile-10.9.0 -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -triple i386-apple-darwin -fobjc-runtime=macosx-fragile-10.9.0 -emit-llvm %s -o - | FileCheck %s
 
 @interface Root
 +(Class)class;
@@ -12,8 +12,8 @@ @interface A : Root
 @end
 
 // CHECK: [[CLASSNAME:@.*]] = private unnamed_addr constant [22 x i8] c"MyRuntimeVisibleClass
-// CHECK: define{{.*}} i8* @getClass() #0 {
+// CHECK: define{{.*}} ptr @getClass() #0 {
 Class getClass(void) {
-  // CHECK: call i8* @objc_lookUpClass(i8* getelementptr inbounds ([22 x i8], [22 x i8]* [[CLASSNAME]], i32 0, i32 0)) #2
+  // CHECK: call ptr @objc_lookUpClass(ptr [[CLASSNAME]]) #2
   return [A class];
 }

diff  --git a/clang/test/CodeGenObjC/autorelease.m b/clang/test/CodeGenObjC/autorelease.m
index a724d221c0771..d7aa831596304 100644
--- a/clang/test/CodeGenObjC/autorelease.m
+++ b/clang/test/CodeGenObjC/autorelease.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -fobjc-arc -fobjc-runtime=macosx-10.7 -fexceptions -fobjc-exceptions -o - %s | FileCheck %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin11 -emit-llvm -fobjc-runtime=macosx-10.7 -fexceptions -fobjc-exceptions -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -fobjc-arc -fobjc-runtime=macosx-10.7 -fexceptions -fobjc-exceptions -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin11 -emit-llvm -fobjc-runtime=macosx-10.7 -fexceptions -fobjc-exceptions -o - %s | FileCheck %s
 // rdar://8881826
 // rdar://9412038
 
@@ -25,8 +25,8 @@ + (id) MyAlloc {
 }
 @end
 
-// CHECK: call i8* @llvm.objc.autoreleasePoolPush
-// CHECK: [[T:%.*]] = load i8*, i8** [[A:%.*]]
+// CHECK: call ptr @llvm.objc.autoreleasePoolPush
+// CHECK: [[T:%.*]] = load ptr, ptr [[A:%.*]]
 // CHECK: call void @llvm.objc.autoreleasePoolPop
 
 // rdar://13660038
@@ -39,15 +39,15 @@ int tryTo(int (*f)(void)) {
     return 0;
   }
 }
-// CHECK-LABEL:    define{{.*}} i32 @tryTo(i32 ()*
+// CHECK-LABEL:    define{{.*}} i32 @tryTo(ptr
 // CHECK:      [[RET:%.*]] = alloca i32,
-// CHECK:      [[T0:%.*]] = call i8* @llvm.objc.autoreleasePoolPush()
-// CHECK-NEXT: [[T1:%.*]] = load i32 ()*, i32 ()** {{%.*}},
+// CHECK:      [[T0:%.*]] = call ptr @llvm.objc.autoreleasePoolPush()
+// CHECK-NEXT: [[T1:%.*]] = load ptr, ptr {{%.*}},
 // CHECK-NEXT: [[T2:%.*]] = invoke i32 [[T1]]()
-// CHECK:      store i32 [[T2]], i32* [[RET]]
-// CHECK:      invoke void @objc_autoreleasePoolPop(i8* [[T0]])
-// CHECK:      landingpad { i8*, i32 }
-// CHECK-NEXT:   catch i8* null
-// CHECK:      call i8* @objc_begin_catch
-// CHECK-NEXT: store i32 0, i32* [[RET]]
+// CHECK:      store i32 [[T2]], ptr [[RET]]
+// CHECK:      invoke void @objc_autoreleasePoolPop(ptr [[T0]])
+// CHECK:      landingpad { ptr, i32 }
+// CHECK-NEXT:   catch ptr null
+// CHECK:      call ptr @objc_begin_catch
+// CHECK-NEXT: store i32 0, ptr [[RET]]
 // CHECK:      call void @objc_end_catch()

diff  --git a/clang/test/CodeGenObjC/availability-cf-link-guard.m b/clang/test/CodeGenObjC/availability-cf-link-guard.m
index 770eef87aed01..cb21c2e79ade3 100644
--- a/clang/test/CodeGenObjC/availability-cf-link-guard.m
+++ b/clang/test/CodeGenObjC/availability-cf-link-guard.m
@@ -1,21 +1,21 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-macosx10.11 -emit-llvm -o - %s | FileCheck --check-prefixes=CHECK,CHECK_LINK_OPT %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-macosx10.11 -emit-llvm -o - -D USE_BUILTIN %s | FileCheck --check-prefixes=CHECK,CHECK_LINK_OPT %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-macosx10.11 -emit-llvm -o - -D DEF_CF %s | FileCheck --check-prefixes=CHECK_CF,CHECK_LINK_OPT %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-macosx10.12 -emit-llvm -o - %s | FileCheck --check-prefix=CHECK_NO_GUARD %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-linux -emit-llvm -o - %s | FileCheck --check-prefix=CHECK_NO_GUARD %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-macos10.15 -DCHECK_OS="macos 10.15.1" -emit-llvm -o - %s | FileCheck --check-prefix=CHECK_NO_GUARD %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple arm64-apple-ios13.0 -DCHECK_OS="ios 14" -emit-llvm -o - %s | FileCheck --check-prefix=CHECK_NO_GUARD %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple arm64-apple-tvos13.0 -DCHECK_OS="tvos 14" -emit-llvm -o - %s | FileCheck --check-prefix=CHECK_NO_GUARD %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple arm64-apple-watchos6.0 -DCHECK_OS="watchos 7" -emit-llvm -o - %s | FileCheck --check-prefix=CHECK_NO_GUARD %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple arm64-apple-ios12.0 -DCHECK_OS="ios 13" -emit-llvm -o - %s | FileCheck --check-prefixes=CHECK,CHECK_LINK_OPT %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple arm64-apple-tvos12.0 -DCHECK_OS="tvos 13" -emit-llvm -o - %s | FileCheck --check-prefixes=CHECK,CHECK_LINK_OPT %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple arm64-apple-watchos5.0 -DCHECK_OS="watchos 6" -emit-llvm -o - %s | FileCheck --check-prefixes=CHECK,CHECK_LINK_OPT %s
+// RUN: %clang_cc1 -triple x86_64-apple-macosx10.11 -emit-llvm -o - %s | FileCheck --check-prefixes=CHECK,CHECK_LINK_OPT %s
+// RUN: %clang_cc1 -triple x86_64-apple-macosx10.11 -emit-llvm -o - -D USE_BUILTIN %s | FileCheck --check-prefixes=CHECK,CHECK_LINK_OPT %s
+// RUN: %clang_cc1 -triple x86_64-apple-macosx10.11 -emit-llvm -o - -D DEF_CF %s | FileCheck --check-prefixes=CHECK_CF,CHECK_LINK_OPT %s
+// RUN: %clang_cc1 -triple x86_64-apple-macosx10.12 -emit-llvm -o - %s | FileCheck --check-prefix=CHECK_NO_GUARD %s
+// RUN: %clang_cc1 -triple x86_64-unknown-linux -emit-llvm -o - %s | FileCheck --check-prefix=CHECK_NO_GUARD %s
+// RUN: %clang_cc1 -triple x86_64-apple-macos10.15 -DCHECK_OS="macos 10.15.1" -emit-llvm -o - %s | FileCheck --check-prefix=CHECK_NO_GUARD %s
+// RUN: %clang_cc1 -triple arm64-apple-ios13.0 -DCHECK_OS="ios 14" -emit-llvm -o - %s | FileCheck --check-prefix=CHECK_NO_GUARD %s
+// RUN: %clang_cc1 -triple arm64-apple-tvos13.0 -DCHECK_OS="tvos 14" -emit-llvm -o - %s | FileCheck --check-prefix=CHECK_NO_GUARD %s
+// RUN: %clang_cc1 -triple arm64-apple-watchos6.0 -DCHECK_OS="watchos 7" -emit-llvm -o - %s | FileCheck --check-prefix=CHECK_NO_GUARD %s
+// RUN: %clang_cc1 -triple arm64-apple-ios12.0 -DCHECK_OS="ios 13" -emit-llvm -o - %s | FileCheck --check-prefixes=CHECK,CHECK_LINK_OPT %s
+// RUN: %clang_cc1 -triple arm64-apple-tvos12.0 -DCHECK_OS="tvos 13" -emit-llvm -o - %s | FileCheck --check-prefixes=CHECK,CHECK_LINK_OPT %s
+// RUN: %clang_cc1 -triple arm64-apple-watchos5.0 -DCHECK_OS="watchos 6" -emit-llvm -o - %s | FileCheck --check-prefixes=CHECK,CHECK_LINK_OPT %s
 
 #ifdef DEF_CF
 struct CFBundle;
 typedef struct CFBundle *CFBundleRef;
 unsigned CFBundleGetVersionNumber(CFBundleRef bundle);
-// CHECK_CF: declare i32 @CFBundleGetVersionNumber(%struct.CFBundle* noundef)
+// CHECK_CF: declare i32 @CFBundleGetVersionNumber(ptr noundef)
 // CHECK_CF: @__clang_at_available_requires_core_foundation_framework
 // CHECK_CF-NEXT: call {{.*}}@CFBundleGetVersionNumber
 #endif
@@ -39,10 +39,10 @@ void use_at_available(void) {
 
 // CHECK: @llvm.compiler.used{{.*}}@__clang_at_available_requires_core_foundation_framework
 
-// CHECK: declare i32 @CFBundleGetVersionNumber(i8*)
+// CHECK: declare i32 @CFBundleGetVersionNumber(ptr)
 
 // CHECK-LABEL: linkonce hidden void @__clang_at_available_requires_core_foundation_framework
-// CHECK: call i32 @CFBundleGetVersionNumber(i8* null)
+// CHECK: call i32 @CFBundleGetVersionNumber(ptr null)
 // CHECK-NEXT: unreachable
 
 // CHECK_NO_GUARD-NOT: __clang_at_available_requires_core_foundation_framework

diff  --git a/clang/test/CodeGenObjC/bitfield-access.m b/clang/test/CodeGenObjC/bitfield-access.m
index 273eba2b7fbb2..0ba73776b2125 100644
--- a/clang/test/CodeGenObjC/bitfield-access.m
+++ b/clang/test/CodeGenObjC/bitfield-access.m
@@ -1,7 +1,7 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple i386-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -o %t1 %s
+// RUN: %clang_cc1 -triple i386-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -o %t1 %s
 // RUN: FileCheck -check-prefix=CHECK-I386 < %t1 %s
 
-// RUN: %clang_cc1 -no-opaque-pointers -triple armv6-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 -target-abi apcs-gnu -emit-llvm -o %t2 %s
+// RUN: %clang_cc1 -triple armv6-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 -target-abi apcs-gnu -emit-llvm -o %t2 %s
 // RUN: FileCheck -check-prefix=CHECK-ARM < %t2 %s
 
 @interface I0 { 
@@ -15,7 +15,7 @@ @interface I0 {
 // end of the structure.
 //
 // CHECK-I386-LABEL: define{{.*}} i32 @f0(
-// CHECK-I386:   [[t0_0:%.*]] = load i8, i8* {{.*}}, align 1
+// CHECK-I386:   [[t0_0:%.*]] = load i8, ptr {{.*}}, align 1
 // CHECK-I386:   lshr i8 [[t0_0]], 7
 // CHECK-I386: }
 int f0(I0 *a) {
@@ -26,8 +26,7 @@ int f0(I0 *a) {
 //
 // CHECK-ARM-LABEL: define{{.*}} i32 @f1(
 // CHECK-ARM:    [[t1_ptr:%.*]] = getelementptr
-// CHECK-ARM:    [[t1_base:%.*]] = bitcast i8* [[t1_ptr]] to i40*
-// CHECK-ARM:    [[t1_0:%.*]] = load i40, i40* [[t1_base]], align 1
+// CHECK-ARM:    [[t1_0:%.*]] = load i40, ptr [[t1_ptr]], align 1
 // CHECK-ARM:    [[t1_1:%.*]] = lshr i40 [[t1_0]], 1
 // CHECK-ARM:    [[t1_2:%.*]] = and i40 [[t1_1]],
 // CHECK-ARM:                   trunc i40 [[t1_2]] to i32

diff  --git a/clang/test/CodeGenObjC/block-byref-variable-layout.m b/clang/test/CodeGenObjC/block-byref-variable-layout.m
index eac168d88bcdd..63c338bc4be66 100644
--- a/clang/test/CodeGenObjC/block-byref-variable-layout.m
+++ b/clang/test/CodeGenObjC/block-byref-variable-layout.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -fblocks -fobjc-arc -fobjc-runtime-has-weak -triple x86_64-apple-darwin -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -fblocks -fobjc-arc -fobjc-runtime-has-weak -triple x86_64-apple-darwin -emit-llvm %s -o - | FileCheck %s
 
 // rdar://12759433
 @class NSString;
@@ -8,7 +8,7 @@ void Test12759433(void) {
  void (^block)(void) = ^{ uuByref = 0; };
  block();
 }
-// CHECK: %struct.__block_byref_uuByref = type { i8*, %struct.__block_byref_uuByref*, i32, i32, [[ZERO:%.*]]* }
+// CHECK: %struct.__block_byref_uuByref = type { ptr, ptr, i32, i32, ptr }
 int main(void) {
   __block __weak id wid;
   __block long XXX;
@@ -21,20 +21,20 @@ int main(void) {
   } import;
   void (^c)(void) = ^{
 // Inline flag for BYREF variable layout (1107296256): BLOCK_BYREF_HAS_COPY_DISPOSE BLOCK_BYREF_LAYOUT_WEAK
-// CHECK: store i32 1107296256, i32* [[T0:%.*]]
+// CHECK: store i32 1107296256, ptr [[T0:%.*]]
   wid = 0;
 
 // Inline flag for BYREF variable layout (536870912): BLOCK_BYREF_LAYOUT_NON_OBJECT
-// CHECK: store i32 536870912, i32* [[T1:%.*]]
+// CHECK: store i32 536870912, ptr [[T1:%.*]]
     XXX = 12345;
 
 // Inline flag for BYREF variable layout (838860800): BLOCK_BYREF_HAS_COPY_DISPOSE BLOCK_BYREF_LAYOUT_STRONG
-// CHECK: store i32 838860800, i32* [[T2:%.*]]
+// CHECK: store i32 838860800, ptr [[T2:%.*]]
     ID = 0;
 
 // Inline flag for BYREF variable layout (268435456): BLOCK_BYREF_LAYOUT_EXTENDED
 // BYREF variable layout: BL_NON_OBJECT_WORD:3, BL_UNRETAINED:1, BL_OPERATOR:0
-// CHECK: store i32 268435456, i32* [[T3:%.*]]
+// CHECK: store i32 268435456, ptr [[T3:%.*]]
     import.iD = 3.14;
 
   };

diff  --git a/clang/test/CodeGenObjC/blocks-1.m b/clang/test/CodeGenObjC/blocks-1.m
index 141bd5be4373e..6a25e5c854e27 100644
--- a/clang/test/CodeGenObjC/blocks-1.m
+++ b/clang/test/CodeGenObjC/blocks-1.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers %s -emit-llvm -o - -fobjc-gc -fblocks -triple i386-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 | FileCheck %s --check-prefix=CHECK --check-prefix=OBJC
-// RUN: %clang_cc1 -no-opaque-pointers -x objective-c++ %s -emit-llvm -o - -fobjc-gc -fblocks -triple i386-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 | FileCheck %s --check-prefix=CHECK --check-prefix=OBJCXX
+// RUN: %clang_cc1 %s -emit-llvm -o - -fobjc-gc -fblocks -triple i386-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 | FileCheck %s --check-prefix=CHECK --check-prefix=OBJC
+// RUN: %clang_cc1 -x objective-c++ %s -emit-llvm -o - -fobjc-gc -fblocks -triple i386-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 | FileCheck %s --check-prefix=CHECK --check-prefix=OBJCXX
 
 // OBJC-LABEL: define{{.*}} void @test1(
 // OBJCXX-LABEL: define{{.*}} void @_Z5test1P12NSDictionary(
@@ -12,13 +12,13 @@
 
 // OBJC-LABEL: define{{.*}} void @foo(
 // OBJCXX-LABEL: define{{.*}} void @_Z3foov(
-// CHECK: call i8* @objc_read_weak(
-// CHECK: call i8* @objc_assign_weak(
+// CHECK: call ptr @objc_read_weak(
+// CHECK: call ptr @objc_assign_weak(
 // CHECK: call void @_Block_object_dispose(
 
 // OBJC-LABEL: define{{.*}} void @test2(
 // OBJCXX-LABEL: define{{.*}} void @_Z5test2v(
-// CHECK: call i8* @objc_assign_weak(
+// CHECK: call ptr @objc_assign_weak(
 // CHECK: call void @_Block_object_dispose(
 
 // CHECK-LABEL: define linkonce_odr hidden void @__copy_helper_block_

diff  --git a/clang/test/CodeGenObjC/builtin-memfns.m b/clang/test/CodeGenObjC/builtin-memfns.m
index 7a5e5a63090a3..ed78d2fd556db 100644
--- a/clang/test/CodeGenObjC/builtin-memfns.m
+++ b/clang/test/CodeGenObjC/builtin-memfns.m
@@ -1,10 +1,10 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-macosx10.8.0 -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-macosx10.8.0 -emit-llvm -o - %s | FileCheck %s
 
 void *memcpy(void *restrict s1, const void *restrict s2, unsigned long n);
 
 // PR13697
 void test1(int *a, id b) {
   // CHECK: @test1
-  // CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* {{.*}}, i8* {{.*}}, i64 8, i1 false)
+  // CHECK: call void @llvm.memcpy.p0.p0.i64(ptr {{.*}}, ptr {{.*}}, i64 8, i1 false)
   memcpy(a, b, 8);
 }

diff  --git a/clang/test/CodeGenObjC/builtins.m b/clang/test/CodeGenObjC/builtins.m
index 131e94b23fea5..4ad9d479968e5 100644
--- a/clang/test/CodeGenObjC/builtins.m
+++ b/clang/test/CodeGenObjC/builtins.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -o - %s | FileCheck %s
 
 id objc_msgSend(id, SEL, ...);
 
@@ -6,4 +6,4 @@ void test0(id receiver, SEL sel, const char *str) {
   short s = ((short (*)(id, SEL, const char*)) objc_msgSend)(receiver, sel, str);
 }
 // CHECK-LABEL: define{{.*}} void @test0(
-// CHECK:   call signext i16 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i16 (i8*, i8*, i8*)*)(
+// CHECK:   call signext i16 @objc_msgSend(

diff  --git a/clang/test/CodeGenObjC/category-super-class-meth.m b/clang/test/CodeGenObjC/category-super-class-meth.m
index 50401c384d73f..dabd0c202239d 100644
--- a/clang/test/CodeGenObjC/category-super-class-meth.m
+++ b/clang/test/CodeGenObjC/category-super-class-meth.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers %s -emit-llvm -triple x86_64-apple-darwin -o - | FileCheck %s
+// RUN: %clang_cc1 %s -emit-llvm -triple x86_64-apple-darwin -o - | FileCheck %s
 // rdar://12459358
 @interface NSObject 
 -(id)copy;
@@ -21,9 +21,8 @@ -(id)copy { return [super copy]; }  // ok: instance method in category
 +(id)copy { return [super copy]; }  // BAD: class method in category
 @end
 
-// CHECK: define internal i8* @"\01+[Sub2(Category) copy]
-// CHECK: [[ONE:%.*]] = load %struct._class_t*, %struct._class_t** @"OBJC_CLASSLIST_SUP_REFS_$_.3"
-// CHECK: [[TWO:%.*]] = bitcast %struct._class_t* [[ONE]] to i8*
-// CHECK: [[THREE:%.*]] = getelementptr inbounds %struct._objc_super, %struct._objc_super* [[OBJC_SUPER:%.*]], i32 0, i32 1
-// CHECK: store i8* [[TWO]], i8** [[THREE]]
-// CHECK: [[FOUR:%.*]] = load i8*, i8** @OBJC_SELECTOR_REFERENCES_
+// CHECK: define internal ptr @"\01+[Sub2(Category) copy]
+// CHECK: [[ONE:%.*]] = load ptr, ptr @"OBJC_CLASSLIST_SUP_REFS_$_.3"
+// CHECK: [[THREE:%.*]] = getelementptr inbounds %struct._objc_super, ptr [[OBJC_SUPER:%.*]], i32 0, i32 1
+// CHECK: store ptr [[ONE]], ptr [[THREE]]
+// CHECK: [[FOUR:%.*]] = load ptr, ptr @OBJC_SELECTOR_REFERENCES_

diff  --git a/clang/test/CodeGenObjC/compound-literal-property-access.m b/clang/test/CodeGenObjC/compound-literal-property-access.m
index 1a1a93b2e8316..df3a5eaec08ae 100644
--- a/clang/test/CodeGenObjC/compound-literal-property-access.m
+++ b/clang/test/CodeGenObjC/compound-literal-property-access.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers %s -emit-llvm -triple x86_64-apple-darwin -o - | FileCheck %s
+// RUN: %clang_cc1 %s -emit-llvm -triple x86_64-apple-darwin -o - | FileCheck %s
 // rdar://20407999
 
 typedef __attribute__((__ext_vector_type__(2))) float vector_float2;
@@ -17,5 +17,5 @@ -(void)getForce {
 @end
 
 // CHECK: [[CL:%.*]] = alloca <2 x float>, align 8
-// CHECK: store <2 x float> [[VECINIT:%.*]], <2 x float>* [[CL]]
-// CHECK: [[FOURTEEN:%.*]] = load <2 x float>, <2 x float>* [[CL]]
+// CHECK: store <2 x float> [[VECINIT:%.*]], ptr [[CL]]
+// CHECK: [[FOURTEEN:%.*]] = load <2 x float>, ptr [[CL]]

diff  --git a/clang/test/CodeGenObjC/constant-non-fragile-ivar-offset.m b/clang/test/CodeGenObjC/constant-non-fragile-ivar-offset.m
index 61c7a199be6d3..788b3220af306 100644
--- a/clang/test/CodeGenObjC/constant-non-fragile-ivar-offset.m
+++ b/clang/test/CodeGenObjC/constant-non-fragile-ivar-offset.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-macosx10.14.0 -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-macosx10.14.0 -emit-llvm %s -o - | FileCheck %s
 
 // CHECK: @"OBJC_IVAR_$_StaticLayout.static_layout_ivar" = hidden constant i64 20
 // CHECK: @"OBJC_IVAR_$_NotStaticLayout.not_static_layout_ivar" = hidden global i64 12
@@ -16,7 +16,7 @@ @implementation StaticLayout {
 }
 -(void)meth {
   static_layout_ivar = 0;
-  // CHECK-NOT: load i64, i64* @"OBJC_IVAR_$_StaticLayout
+  // CHECK-NOT: load i64, ptr @"OBJC_IVAR_$_StaticLayout
 }
 @end
 
@@ -33,6 +33,6 @@ @implementation NotStaticLayout {
 }
 -(void)meth {
   not_static_layout_ivar = 0;
-  // CHECK: load i64, i64* @"OBJC_IVAR_$_NotStaticLayout.not_static_layout_ivar
+  // CHECK: load i64, ptr @"OBJC_IVAR_$_NotStaticLayout.not_static_layout_ivar
 }
 @end

diff  --git a/clang/test/CodeGenObjC/constant-strings.m b/clang/test/CodeGenObjC/constant-strings.m
index 172df36b51ed7..fb5865ea754f6 100644
--- a/clang/test/CodeGenObjC/constant-strings.m
+++ b/clang/test/CodeGenObjC/constant-strings.m
@@ -1,28 +1,28 @@
 // REQUIRES: x86-registered-target
 
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-macho -emit-llvm -o %t %s
+// RUN: %clang_cc1 -triple x86_64-macho -emit-llvm -o %t %s
 // RUN: FileCheck --check-prefix=CHECK-NEXT < %t %s
 
 // Check that we set alignment 1 on the string.
 //
 // CHECK-NEXT: @.str = {{.*}}constant [13 x i8] c"Hello World!\00", section "__TEXT,__cstring,cstring_literals", align 1
 
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-macho -fobjc-runtime=gcc -emit-llvm -o %t %s
+// RUN: %clang_cc1 -triple x86_64-macho -fobjc-runtime=gcc -emit-llvm -o %t %s
 // RUN: FileCheck --check-prefix=CHECK-GNU < %t %s
 // CHECK-GNU: NXConstantString
 
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-macho -fobjc-runtime=gcc -fconstant-string-class NSConstantString -emit-llvm -o %t %s
+// RUN: %clang_cc1 -triple x86_64-macho -fobjc-runtime=gcc -fconstant-string-class NSConstantString -emit-llvm -o %t %s
 // RUN: FileCheck --check-prefix=CHECK-GNU-WITH-CLASS < %t %s
 // CHECK-GNU-WITH-CLASS: NSConstantString
 //
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-freebsd -fobjc-runtime=gnustep-2.0 -emit-llvm -o %t %s
+// RUN: %clang_cc1 -triple x86_64-unknown-freebsd -fobjc-runtime=gnustep-2.0 -emit-llvm -o %t %s
 // RUN: FileCheck --check-prefix=CHECK-GNUSTEP2 < %t %s
 
-// CHECK-GNUSTEP2: @._OBJC_CLASS_NSConstantString = external global i8*
+// CHECK-GNUSTEP2: @._OBJC_CLASS_NSConstantString = external global ptr
 // CHECK-GNUSTEP2: @0 = private unnamed_addr constant [13 x i8] c"Hello World!\00", align 1
-// CHECK-GNUSTEP2: @.objc_string = private global { i8**, i32, i32, i32, i32, i8* } { i8** @._OBJC_CLASS_NSConstantString, i32 0, i32 12, i32 12, i32 0, i8* getelementptr inbounds ([13 x i8], [13 x i8]* @0, i64 0, i64 0) }, section "__objc_constant_string", align 8
-// CHECK-GNUSTEP2: @b ={{.*}} global i8* inttoptr (i64 -3340545023602065388 to i8*), align 8
-// CHECK-GNUSTEP2: @.objc_str_Hello_World = linkonce_odr hidden global { i8**, i32, i32, i32, i32, i8* } { i8** @._OBJC_CLASS_NSConstantString, i32 0, i32 11, i32 11, i32 0, i8* getelementptr inbounds ([12 x i8], [12 x i8]* @1, i64 0, i64 0) }, section "__objc_constant_string", comdat, align 8
+// CHECK-GNUSTEP2: @.objc_string = private global { ptr, i32, i32, i32, i32, ptr } { ptr @._OBJC_CLASS_NSConstantString, i32 0, i32 12, i32 12, i32 0, ptr @0 }, section "__objc_constant_string", align 8
+// CHECK-GNUSTEP2: @b ={{.*}} global ptr inttoptr (i64 -3340545023602065388 to ptr), align 8
+// CHECK-GNUSTEP2: @.objc_str_Hello_World = linkonce_odr hidden global { ptr, i32, i32, i32, i32, ptr } { ptr @._OBJC_CLASS_NSConstantString, i32 0, i32 11, i32 11, i32 0, ptr @1 }, section "__objc_constant_string", comdat, align 8
 // CHECK-GNUSTEP2: @c =
 // CHECK-SAME-GNUSTEP2: @.objc_str_Hello_World
 //

diff  --git a/clang/test/CodeGenObjC/debug-info-default-synth-ivar.m b/clang/test/CodeGenObjC/debug-info-default-synth-ivar.m
index a69967c3dab95..fb2641265bde3 100644
--- a/clang/test/CodeGenObjC/debug-info-default-synth-ivar.m
+++ b/clang/test/CodeGenObjC/debug-info-default-synth-ivar.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -debug-info-kind=limited %s -o %t
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -debug-info-kind=limited %s -o %t
 // RUN: grep DW_TAG_member %t | count 5
 // rdar://8493239
 

diff  --git a/clang/test/CodeGenObjC/debug-info-getter-name.m b/clang/test/CodeGenObjC/debug-info-getter-name.m
index dbdf4dc22fca9..3e0f52be86462 100644
--- a/clang/test/CodeGenObjC/debug-info-getter-name.m
+++ b/clang/test/CodeGenObjC/debug-info-getter-name.m
@@ -1,5 +1,5 @@
 // REQUIRES: x86-registered-target
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -triple x86_64-apple-darwin10 -fexceptions -fobjc-exceptions -debug-info-kind=limited %s -o - | FileCheck %s
+// RUN: %clang_cc1 -emit-llvm -triple x86_64-apple-darwin10 -fexceptions -fobjc-exceptions -debug-info-kind=limited %s -o - | FileCheck %s
 
 // CHECK:  !DISubprogram(name: "-[InstanceVariablesEverywhereButTheInterface someString]"
 

diff  --git a/clang/test/CodeGenObjC/debug-info-nested-blocks.m b/clang/test/CodeGenObjC/debug-info-nested-blocks.m
index b66a6dd301aaf..3f35ef7003e21 100644
--- a/clang/test/CodeGenObjC/debug-info-nested-blocks.m
+++ b/clang/test/CodeGenObjC/debug-info-nested-blocks.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -dwarf-version=2 -debug-info-kind=limited -fblocks -o - -x objective-c %s| FileCheck %s
+// RUN: %clang_cc1 -emit-llvm -dwarf-version=2 -debug-info-kind=limited -fblocks -o - -x objective-c %s| FileCheck %s
 // This code triggered a bug where a dbg.declare intrinsic ended up with the
 // wrong parent and subsequently failed the Verifier.
 void baz(id b);
@@ -21,6 +21,6 @@ void bar(void) {
 // Verify that debug info for BlockPointerDbgLoc is emitted for the
 // innermost block.
 //
-// CHECK: define {{.*}}void @__bar_block_invoke_3(i8* noundef %.block_descriptor)
-// CHECK: %[[BLOCKADDR:.*]] = alloca <{{.*}}>*, align
+// CHECK: define {{.*}}void @__bar_block_invoke_3(ptr noundef %.block_descriptor)
+// CHECK: %[[BLOCKADDR:.*]] = alloca ptr, align
 // CHECK: call void @llvm.dbg.declare(metadata {{.*}}%[[BLOCKADDR]]

diff  --git a/clang/test/CodeGenObjC/debug-info-objc-property-dwarf5.m b/clang/test/CodeGenObjC/debug-info-objc-property-dwarf5.m
index 9ccd6db9c0a53..2b3a86fe39971 100644
--- a/clang/test/CodeGenObjC/debug-info-objc-property-dwarf5.m
+++ b/clang/test/CodeGenObjC/debug-info-objc-property-dwarf5.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -debug-info-kind=standalone -dwarf-version=5 %s -o - | FileCheck %s
+// RUN: %clang_cc1 -emit-llvm -debug-info-kind=standalone -dwarf-version=5 %s -o - | FileCheck %s
 
 @protocol NSObject
 @end

diff  --git a/clang/test/CodeGenObjC/direct-method-ret-mismatch.m b/clang/test/CodeGenObjC/direct-method-ret-mismatch.m
index 51a31e0b1a306..889a6d68da0d7 100644
--- a/clang/test/CodeGenObjC/direct-method-ret-mismatch.m
+++ b/clang/test/CodeGenObjC/direct-method-ret-mismatch.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -fobjc-arc -triple x86_64-apple-darwin10 %s -o - | FileCheck %s
+// RUN: %clang_cc1 -emit-llvm -fobjc-arc -triple x86_64-apple-darwin10 %s -o - | FileCheck %s
 
 __attribute__((objc_root_class))
 @interface Root
@@ -6,13 +6,13 @@ - (Root *)method __attribute__((objc_direct));
 @end
 
 @implementation Root
-// CHECK-LABEL: define internal i8* @"\01-[Root something]"(
+// CHECK-LABEL: define internal ptr @"\01-[Root something]"(
 - (id)something {
   // CHECK: %{{[^ ]*}} = call {{.*}} @"\01-[Root method]"
   return [self method];
 }
 
-// CHECK-LABEL: define hidden i8* @"\01-[Root method]"(
+// CHECK-LABEL: define hidden ptr @"\01-[Root method]"(
 - (id)method {
   return self;
 }

diff  --git a/clang/test/CodeGenObjC/disable-direct-method.m b/clang/test/CodeGenObjC/disable-direct-method.m
index 6ea003e59c7e5..3fca36f18271f 100644
--- a/clang/test/CodeGenObjC/disable-direct-method.m
+++ b/clang/test/CodeGenObjC/disable-direct-method.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -triple x86_64-apple-darwin10 -fobjc-disable-direct-methods-for-testing %s -o - | FileCheck %s
+// RUN: %clang_cc1 -emit-llvm -triple x86_64-apple-darwin10 -fobjc-disable-direct-methods-for-testing %s -o - | FileCheck %s
 
 @interface Y
 @property (direct) int direct_property;
@@ -17,5 +17,5 @@ -(void)m __attribute__((objc_direct));
 void f(X *x) {
   [x m];
 
-  // CHECK: call void bitcast ({{.*}} @objc_msgSend to {{.*}})
+  // CHECK: call void @objc_msgSend
 }

diff  --git a/clang/test/CodeGenObjC/encode-test-2.m b/clang/test/CodeGenObjC/encode-test-2.m
index 93fe82bf4a7ce..c82bf4c1142b6 100644
--- a/clang/test/CodeGenObjC/encode-test-2.m
+++ b/clang/test/CodeGenObjC/encode-test-2.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple=i686-apple-darwin9 -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple=i686-apple-darwin9 -emit-llvm -o - %s | FileCheck %s
 
 // CHECK: private unnamed_addr constant [7 x i8] c"@\22<X>\22\00",
 // CHECK: private unnamed_addr constant [10 x i8] c"@\22<X><Y>\22\00",
@@ -7,7 +7,7 @@
 // CHECK: private unnamed_addr constant [13 x i8] c"{Intf=@@@@#}\00",
 
 // CHECK: @[[PROP_NAME_ATTR:.*]] = private unnamed_addr constant [5 x i8] c"T@,D\00",
-// CHECK: @"_OBJC_$_PROP_LIST_C0" = internal global { i32, i32, [1 x %{{.*}}] } { i32 8, i32 1, [1 x %{{.*}}] [%{{.*}} { {{.*}}, i8* getelementptr inbounds ([5 x i8], [5 x i8]* @[[PROP_NAME_ATTR]], i32 0, i32 0) }] },
+// CHECK: @"_OBJC_$_PROP_LIST_C0" = internal global { i32, i32, [1 x %{{.*}}] } { i32 8, i32 1, [1 x %{{.*}}] [%{{.*}} { {{.*}}, ptr @[[PROP_NAME_ATTR]] }] },
 
 @protocol X, Y, Z;
 @class Foo;

diff  --git a/clang/test/CodeGenObjC/encode-test-5.m b/clang/test/CodeGenObjC/encode-test-5.m
index 0de3e759ee9fa..997220b2c206c 100644
--- a/clang/test/CodeGenObjC/encode-test-5.m
+++ b/clang/test/CodeGenObjC/encode-test-5.m
@@ -1,21 +1,21 @@
-// RUN: %clang_cc1 -no-opaque-pointers -x objective-c -triple=x86_64-apple-darwin9 -emit-llvm -o - < %s | FileCheck %s
+// RUN: %clang_cc1 -x objective-c -triple=x86_64-apple-darwin9 -emit-llvm -o - < %s | FileCheck %s
 
 // CHECK-DAG: @[[complex_int:.*]] = private unnamed_addr constant [3 x i8] c"ji\00", align 1
-// CHECK-DAG: @a ={{.*}} global i8* getelementptr inbounds ([3 x i8], [3 x i8]* @[[complex_int]], i32 0, i32 0), align 8
+// CHECK-DAG: @a ={{.*}} global ptr @[[complex_int]], align 8
 char *a = @encode(_Complex int);
 
 // CHECK-DAG: @[[complex_float:.*]] = private unnamed_addr constant [3 x i8] c"jf\00", align 1
-// CHECK-DAG: @b ={{.*}} global i8* getelementptr inbounds ([3 x i8], [3 x i8]* @[[complex_float]], i32 0, i32 0), align 8
+// CHECK-DAG: @b ={{.*}} global ptr @[[complex_float]], align 8
 char *b = @encode(_Complex float);
 
 // CHECK-DAG: @[[complex_double:.*]] = private unnamed_addr constant [3 x i8] c"jd\00", align 1
-// CHECK-DAG: @c ={{.*}} global i8* getelementptr inbounds ([3 x i8], [3 x i8]* @[[complex_double]], i32 0, i32 0), align 8
+// CHECK-DAG: @c ={{.*}} global ptr @[[complex_double]], align 8
 char *c = @encode(_Complex double);
 
 // CHECK-DAG: @[[int_128:.*]] = private unnamed_addr constant [2 x i8] c"t\00", align 1
-// CHECK-DAG: @e ={{.*}} global i8* getelementptr inbounds ([2 x i8], [2 x i8]* @[[int_128]], i32 0, i32 0), align 8
+// CHECK-DAG: @e ={{.*}} global ptr @[[int_128]], align 8
 char *e = @encode(__int128_t);
 
 // CHECK-DAG: @[[uint_128:.*]] = private unnamed_addr constant [2 x i8] c"T\00", align 1
-// CHECK-DAG: @f ={{.*}} global i8* getelementptr inbounds ([2 x i8], [2 x i8]* @[[uint_128]], i32 0, i32 0), align 8
+// CHECK-DAG: @f ={{.*}} global ptr @[[uint_128]], align 8
 char *f = @encode(__uint128_t);

diff  --git a/clang/test/CodeGenObjC/encode-test-6.m b/clang/test/CodeGenObjC/encode-test-6.m
index fd4f2b9931f95..7c07289210e4d 100644
--- a/clang/test/CodeGenObjC/encode-test-6.m
+++ b/clang/test/CodeGenObjC/encode-test-6.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin -emit-llvm -o %t %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin -emit-llvm -o %t %s
 // RUN: FileCheck < %t %s
 // rdar://11777609
 
@@ -62,5 +62,5 @@ @implementation SCNCamera
     return e;
 }
 // CHECK: @e ={{.*}} global [2 x i8] c"i\00", align 1
-// CHECK: define{{.*}} i8* @Test()
-// CHECK: ret i8* getelementptr inbounds ([2 x i8], [2 x i8]* @e, i64 0, i64 0)
+// CHECK: define{{.*}} ptr @Test()
+// CHECK: ret ptr @e

diff  --git a/clang/test/CodeGenObjC/encode-test.m b/clang/test/CodeGenObjC/encode-test.m
index 528f240202d2d..3e03ea81993cc 100644
--- a/clang/test/CodeGenObjC/encode-test.m
+++ b/clang/test/CodeGenObjC/encode-test.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-apple-darwin9 -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -o %t %s
+// RUN: %clang_cc1 -triple i686-apple-darwin9 -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -o %t %s
 // RUN: FileCheck < %t %s
 //
 // CHECK: @OBJC_METH_VAR_TYPE_{{.*}} = private unnamed_addr constant [16 x i8] c"v12 at 0:4[3[4@]]8\00"
@@ -187,7 +187,7 @@ @implementation Derived1X @end
 // CHECK-LABEL: @test_strlen(
 // CHECK: %[[i:.*]] = alloca i32
 // CHECK: %[[call:.*]] = call i32 @strlen
-// CHECK: store i32 %[[call]], i32* %[[i]]
+// CHECK: store i32 %[[call]], ptr %[[i]]
 void test_strlen(void) {
   const char array[] = @encode(int);
   int i = strlen(array);

diff  --git a/clang/test/CodeGenObjC/extern-void-class-decl.m b/clang/test/CodeGenObjC/extern-void-class-decl.m
index 43eee02fd4e92..a01b5b74fc9db 100644
--- a/clang/test/CodeGenObjC/extern-void-class-decl.m
+++ b/clang/test/CodeGenObjC/extern-void-class-decl.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-macosx10.14.0 %s -emit-llvm -o - | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-macosx10.14.0 %s -emit-llvm -o - | FileCheck %s
 
 // rdar://45077269
 
@@ -10,5 +10,5 @@ @implementation f @end
 // Check that we override the initializer for c, and that OBJC_CLASS_$_f gets
 // the right definition.
 
-// CHECK: @c ={{.*}} global i8* bitcast (%struct._class_t* @"OBJC_CLASS_$_f" to i8*)
+// CHECK: @c ={{.*}} global ptr @"OBJC_CLASS_$_f"
 // CHECK: @"OBJC_CLASS_$_f" ={{.*}} global %struct._class_t

diff  --git a/clang/test/CodeGenObjC/forward-protocol-metadata-symbols.m b/clang/test/CodeGenObjC/forward-protocol-metadata-symbols.m
index 646913050bdde..88c07c1a41034 100644
--- a/clang/test/CodeGenObjC/forward-protocol-metadata-symbols.m
+++ b/clang/test/CodeGenObjC/forward-protocol-metadata-symbols.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-10.14 -emit-llvm -x objective-c %s -o - | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-10.14 -emit-llvm -x objective-c %s -o - | FileCheck %s
 // rdar://16203115
 
 @interface NSObject @end
@@ -23,12 +23,12 @@ int main(void) {
 // CHECK: @"_OBJC_CLASS_PROTOCOLS_$_A" = internal global
 // CHECK: @"_OBJC_PROTOCOL_REFERENCE_$_P0" = weak hidden global
 
-// CHECK: llvm.used = appending global [3 x i8*]
+// CHECK: llvm.used = appending global [3 x ptr]
 // CHECK-SAME: "_OBJC_PROTOCOL_$_P0"
 // CHECK-SAME: "_OBJC_LABEL_PROTOCOL_$_P0"
 // CHECK-SAME: "_OBJC_PROTOCOL_REFERENCE_$_P0"
 
-// CHECK: llvm.compiler.used = appending global [7 x i8*]
+// CHECK: llvm.compiler.used = appending global [7 x ptr]
 // CHECK-SAME: OBJC_CLASS_NAME_
 // CHECK-SAME: OBJC_METH_VAR_NAME_
 // CHECK-SAME: OBJC_METH_VAR_TYPE_

diff  --git a/clang/test/CodeGenObjC/gc-weak-attribute.m b/clang/test/CodeGenObjC/gc-weak-attribute.m
index 551b42941c5f6..500fd112eef82 100644
--- a/clang/test/CodeGenObjC/gc-weak-attribute.m
+++ b/clang/test/CodeGenObjC/gc-weak-attribute.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -fobjc-gc -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -fobjc-gc -emit-llvm -o - %s | FileCheck %s
 // rdar://10073896
 
 @interface I
@@ -17,12 +17,12 @@ @implementation I
 @synthesize wRandom;
 @synthesize wAnother;
 @end
-// CHECK:  call i8* @objc_read_weak
-// CHECK:  call i8* @objc_assign_weak
-// CHECK:  call i8* @objc_read_weak
-// CHECK:  call i8* @objc_assign_weak
-// CHECK:  call i8* @objc_read_weak
-// CHECK:  call i8* @objc_assign_weak
-// CHECK:  call i8* @objc_read_weak
-// CHECK:  call i8* @objc_assign_weak
+// CHECK:  call ptr @objc_read_weak
+// CHECK:  call ptr @objc_assign_weak
+// CHECK:  call ptr @objc_read_weak
+// CHECK:  call ptr @objc_assign_weak
+// CHECK:  call ptr @objc_read_weak
+// CHECK:  call ptr @objc_assign_weak
+// CHECK:  call ptr @objc_read_weak
+// CHECK:  call ptr @objc_assign_weak
 

diff  --git a/clang/test/CodeGenObjC/gc.m b/clang/test/CodeGenObjC/gc.m
index a57ecd89c5998..a55e9bea5578e 100644
--- a/clang/test/CodeGenObjC/gc.m
+++ b/clang/test/CodeGenObjC/gc.m
@@ -1,15 +1,15 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -fobjc-gc -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -fobjc-gc -emit-llvm -o - %s | FileCheck %s
 
 void test0(void) {
   extern id test0_helper(void);
   __attribute__((objc_precise_lifetime)) id x = test0_helper();
   test0_helper();
   // CHECK-LABEL: define{{.*}} void @test0()
-  // CHECK:      [[T0:%.*]] = call i8* @test0_helper()
-  // CHECK-NEXT: store i8* [[T0]], i8** [[X:%.*]], align 8
-  // CHECK-NEXT: call i8* @test0_helper()
-  // CHECK-NEXT: [[T0:%.*]] = load i8*, i8** [[X]], align 8
-  // CHECK-NEXT: call void asm sideeffect "", "r"(i8* [[T0]]) [[NUW:#[0-9]+]]
+  // CHECK:      [[T0:%.*]] = call ptr @test0_helper()
+  // CHECK-NEXT: store ptr [[T0]], ptr [[X:%.*]], align 8
+  // CHECK-NEXT: call ptr @test0_helper()
+  // CHECK-NEXT: [[T0:%.*]] = load ptr, ptr [[X]], align 8
+  // CHECK-NEXT: call void asm sideeffect "", "r"(ptr [[T0]]) [[NUW:#[0-9]+]]
   // CHECK-NEXT: ret void
 }
 

diff  --git a/clang/test/CodeGenObjC/getter-property-mismatch.m b/clang/test/CodeGenObjC/getter-property-mismatch.m
index aed5f911a5b9e..64a1d8cbc84d6 100644
--- a/clang/test/CodeGenObjC/getter-property-mismatch.m
+++ b/clang/test/CodeGenObjC/getter-property-mismatch.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers %s -emit-llvm -triple x86_64-apple-darwin -o - | FileCheck %s
+// RUN: %clang_cc1 %s -emit-llvm -triple x86_64-apple-darwin -o - | FileCheck %s
 // rdar://11323676
 
 @interface NSDictionary @end
@@ -13,6 +13,5 @@ @implementation CalDAVAddManagedAttachmentsTaskGroup
 @synthesize filenamesToServerLocation=_filenamesToServerLocation;
 @end
 
-// CHECK:  [[CALL:%.*]] = tail call i8* @objc_getProperty
-// CHECK:  [[ONE:%.*]] = bitcast i8* [[CALL:%.*]] to [[T1:%.*]]*
-// CHECK:  ret [[T1]]* [[ONE]]
+// CHECK:  [[CALL:%.*]] = tail call ptr @objc_getProperty
+// CHECK:  ret ptr [[CALL:%.*]]

diff  --git a/clang/test/CodeGenObjC/getter-property-type-mismatch.m b/clang/test/CodeGenObjC/getter-property-type-mismatch.m
index 60466c996dd9f..d09a00c7ea9aa 100644
--- a/clang/test/CodeGenObjC/getter-property-type-mismatch.m
+++ b/clang/test/CodeGenObjC/getter-property-type-mismatch.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10  -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10  -emit-llvm -o - %s | FileCheck %s
 // rdar://11515196
 
 @interface NSArray @end
@@ -29,7 +29,6 @@ - (void)applicationDidFinishLaunching
 
 @end
 
-// CHECK: define internal [[RET:%.*]]* @"\01-[BPXLAppDelegate arrayOfThings
-// CHECK: [[THREE:%.*]] = bitcast [[OPQ:%.*]]* [[TWO:%.*]] to [[RET]]*
-// CHECK: ret [[RET]]* [[THREE]]
+// CHECK: define internal ptr @"\01-[BPXLAppDelegate arrayOfThings
+// CHECK: ret ptr [[TWO:%.*]]
 

diff  --git a/clang/test/CodeGenObjC/gnu-deterministic-selectors.m b/clang/test/CodeGenObjC/gnu-deterministic-selectors.m
index 2ff1d00214f5b..03747663b5f1a 100644
--- a/clang/test/CodeGenObjC/gnu-deterministic-selectors.m
+++ b/clang/test/CodeGenObjC/gnu-deterministic-selectors.m
@@ -1,10 +1,10 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-freebsd -fobjc-runtime=gnustep-1.5 %s -emit-llvm -o - | FileCheck %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-freebsd -fobjc-runtime=gcc %s -emit-llvm -o - | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-unknown-freebsd -fobjc-runtime=gnustep-1.5 %s -emit-llvm -o - | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-unknown-freebsd -fobjc-runtime=gcc %s -emit-llvm -o - | FileCheck %s
 
 // Check that these selectors are emitted in alphabetical order.
 // The order doesn't actually matter, only that it doesn't vary across runs.
 // Clang sorts them when targeting a GCC-like ABI to guarantee determinism.
-// CHECK: @.objc_selector_list = internal global [6 x { i8*, i8* }] [{ i8*, i8* } { i8* getelementptr inbounds ([2 x i8], [2 x i8]* @.objc_sel_namea, i64 0, i64 0), i8* null }, { i8*, i8* } { i8* getelementptr inbounds ([2 x i8], [2 x i8]* @.objc_sel_nameg, i64 0, i64 0), i8* null }, { i8*, i8* } { i8* getelementptr inbounds ([2 x i8], [2 x i8]* @.objc_sel_namej, i64 0, i64 0), i8* null }, { i8*, i8* } { i8* getelementptr inbounds ([2 x i8], [2 x i8]* @.objc_sel_namel, i64 0, i64 0), i8* null }, { i8*, i8* } { i8* getelementptr inbounds ([2 x i8], [2 x i8]* @.objc_sel_namez, i64 0, i64 0), i8* null }, { i8*, i8* } zeroinitializer], align 8
+// CHECK: @.objc_selector_list = internal global [6 x { ptr, ptr }] [{ ptr, ptr } { ptr @.objc_sel_namea, ptr null }, { ptr, ptr } { ptr @.objc_sel_nameg, ptr null }, { ptr, ptr } { ptr @.objc_sel_namej, ptr null }, { ptr, ptr } { ptr @.objc_sel_namel, ptr null }, { ptr, ptr } { ptr @.objc_sel_namez, ptr null }, { ptr, ptr } zeroinitializer], align 8
 
 
 void f(void) {

diff  --git a/clang/test/CodeGenObjC/gnu-empty-protocol-v3.m b/clang/test/CodeGenObjC/gnu-empty-protocol-v3.m
index 360e7312dfbc5..d7ec0e65880d3 100644
--- a/clang/test/CodeGenObjC/gnu-empty-protocol-v3.m
+++ b/clang/test/CodeGenObjC/gnu-empty-protocol-v3.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-pc-linux-gnu -fobjc-runtime=gnustep-1.9 -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple i686-pc-linux-gnu -fobjc-runtime=gnustep-1.9 -emit-llvm -o - %s | FileCheck %s
 
 @protocol X;
 
@@ -10,17 +10,17 @@ @implementation Z
 @end
 
 
-// CHECK:      @.objc_protocol_list = internal global { i8*, i32, [0 x i8*] } zeroinitializer, align 4
-// CHECK:      @.objc_method_list = internal global { i32, [0 x { i8*, i8* }] } zeroinitializer, align 4
+// CHECK:      @.objc_protocol_list = internal global { ptr, i32, [0 x ptr] } zeroinitializer, align 4
+// CHECK:      @.objc_method_list = internal global { i32, [0 x { ptr, ptr }] } zeroinitializer, align 4
 // CHECK:      @.objc_protocol_name = private unnamed_addr constant [2 x i8] c"X\00", align 1
-// CHECK:      @._OBJC_PROTOCOL_X = internal global { i8*, i8*, { i8*, i32, [0 x i8*] }*, i8*, i8*, i8*, i8*, i8*, i8* } { 
-// CHECK-SAME:     i8* inttoptr (i32 3 to i8*),
-// CHECK-SAME:     i8* getelementptr inbounds ([2 x i8], [2 x i8]* @.objc_protocol_name, i32 0, i32 0),
-// CHECK-SAME:     { i8*, i32, [0 x i8*] }* @.objc_protocol_list
-// CHECK-SAME:     { i32, [0 x { i8*, i8* }] }* @.objc_method_list
-// CHECK-SAME:     { i32, [0 x { i8*, i8* }] }* @.objc_method_list
-// CHECK-SAME:     { i32, [0 x { i8*, i8* }] }* @.objc_method_list
-// CHECK-SAME:     { i32, [0 x { i8*, i8* }] }* @.objc_method_list
-// CHECK-SAME:     i8* null
-// CHECK-SAME:     i8* null
+// CHECK:      @._OBJC_PROTOCOL_X = internal global { ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr } { 
+// CHECK-SAME:     ptr inttoptr (i32 3 to ptr),
+// CHECK-SAME:     ptr @.objc_protocol_name,
+// CHECK-SAME:     ptr @.objc_protocol_list
+// CHECK-SAME:     ptr @.objc_method_list
+// CHECK-SAME:     ptr @.objc_method_list
+// CHECK-SAME:     ptr @.objc_method_list
+// CHECK-SAME:     ptr @.objc_method_list
+// CHECK-SAME:     ptr null
+// CHECK-SAME:     ptr null
 // CHECK-SAME: }, align 4

diff  --git a/clang/test/CodeGenObjC/gnu-exceptions.m b/clang/test/CodeGenObjC/gnu-exceptions.m
index fcf8132d8ee98..cde4c3786a9a0 100644
--- a/clang/test/CodeGenObjC/gnu-exceptions.m
+++ b/clang/test/CodeGenObjC/gnu-exceptions.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-pc-linux-gnu -emit-llvm -fexceptions -fobjc-exceptions -fobjc-runtime=gcc -o - %s | FileCheck %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-freebsd -emit-llvm -fexceptions -fobjc-exceptions -fobjc-runtime=gnustep-1.7 -o - %s | FileCheck -check-prefix=NEW-ABI %s
+// RUN: %clang_cc1 -triple x86_64-pc-linux-gnu -emit-llvm -fexceptions -fobjc-exceptions -fobjc-runtime=gcc -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-unknown-freebsd -emit-llvm -fexceptions -fobjc-exceptions -fobjc-runtime=gnustep-1.7 -o - %s | FileCheck -check-prefix=NEW-ABI %s
 
 void opaque(void);
 void log(int i);
@@ -7,7 +7,7 @@
 @class C;
 
 // CHECK: define{{.*}} void @test0() [[TF:#[0-9]+]]
-// CHECK-SAME: personality i8* bitcast (i32 (...)* @__gnu_objc_personality_v0 to i8*)
+// CHECK-SAME: personality ptr @__gnu_objc_personality_v0
 void test0(void) {
   @try {
     // CHECK: invoke void @opaque()
@@ -16,8 +16,8 @@ void test0(void) {
     // CHECK: call void @log(i32 noundef 1)
 
   } @catch (C *c) {
-    // CHECK:      landingpad { i8*, i32 }
-    // CHECK-NEXT:   catch i8* getelementptr inbounds ([2 x i8], [2 x i8]* @0, i64 0, i64 0)
+    // CHECK:      landingpad { ptr, i32 }
+    // CHECK-NEXT:   catch ptr @0
     // CHECK:      br i1
 
     // CHECK: call void @log(i32 noundef 0)

diff  --git a/clang/test/CodeGenObjC/gnu-init.m b/clang/test/CodeGenObjC/gnu-init.m
index 0f7fb022b3bac..d0aa6fdc4ac9f 100644
--- a/clang/test/CodeGenObjC/gnu-init.m
+++ b/clang/test/CodeGenObjC/gnu-init.m
@@ -1,7 +1,7 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-freebsd -S -emit-llvm -fno-use-init-array -fobjc-runtime=gnustep-2.0 -o - %s | FileCheck %s -check-prefix=CHECK-NEW
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-pc-windows-msvc -S -emit-llvm -fno-use-init-array -fobjc-runtime=gnustep-2.0 -o - %s | FileCheck %s -check-prefix=CHECK-WIN
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-freebsd -S -emit-llvm -fno-use-init-array -fobjc-runtime=gnustep-1.8 -o - %s | FileCheck %s -check-prefix=CHECK-OLD
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-freebsd -S -emit-llvm -fobjc-runtime=gnustep-2.0 -o - %s | FileCheck %s -check-prefix=CHECK-INIT_ARRAY
+// RUN: %clang_cc1 -triple x86_64-unknown-freebsd -S -emit-llvm -fno-use-init-array -fobjc-runtime=gnustep-2.0 -o - %s | FileCheck %s -check-prefix=CHECK-NEW
+// RUN: %clang_cc1 -triple x86_64-pc-windows-msvc -S -emit-llvm -fno-use-init-array -fobjc-runtime=gnustep-2.0 -o - %s | FileCheck %s -check-prefix=CHECK-WIN
+// RUN: %clang_cc1 -triple x86_64-unknown-freebsd -S -emit-llvm -fno-use-init-array -fobjc-runtime=gnustep-1.8 -o - %s | FileCheck %s -check-prefix=CHECK-OLD
+// RUN: %clang_cc1 -triple x86_64-unknown-freebsd -S -emit-llvm -fobjc-runtime=gnustep-2.0 -o - %s | FileCheck %s -check-prefix=CHECK-INIT_ARRAY
 
 // Almost minimal Objective-C file, check that it emits calls to the correct
 // runtime entry points.
@@ -18,43 +18,43 @@ @implementation X @end
 // CHECK-NEW-SAME: @._OBJC_CLASS_X, section "__objc_classes"
 
 // Check that we emit the section start and end symbols as hidden globals.
-// CHECK-NEW: @__start___objc_selectors = external hidden global i8*
-// CHECK-NEW: @__stop___objc_selectors = external hidden global i8*
-// CHECK-NEW: @__start___objc_classes = external hidden global i8*
-// CHECK-NEW: @__stop___objc_classes = external hidden global i8*
-// CHECK-NEW: @__start___objc_class_refs = external hidden global i8*
-// CHECK-NEW: @__stop___objc_class_refs = external hidden global i8*
-// CHECK-NEW: @__start___objc_cats = external hidden global i8*
-// CHECK-NEW: @__stop___objc_cats = external hidden global i8*
-// CHECK-NEW: @__start___objc_protocols = external hidden global i8*
-// CHECK-NEW: @__stop___objc_protocols = external hidden global i8*
-// CHECK-NEW: @__start___objc_protocol_refs = external hidden global i8*
-// CHECK-NEW: @__stop___objc_protocol_refs = external hidden global i8*
-// CHECK-NEW: @__start___objc_class_aliases = external hidden global i8*
-// CHECK-NEW: @__stop___objc_class_aliases = external hidden global i8*
-// CHECK-NEW: @__start___objc_constant_string = external hidden global i8*
-// CHECK-NEW: @__stop___objc_constant_string = external hidden global i8*
+// CHECK-NEW: @__start___objc_selectors = external hidden global ptr
+// CHECK-NEW: @__stop___objc_selectors = external hidden global ptr
+// CHECK-NEW: @__start___objc_classes = external hidden global ptr
+// CHECK-NEW: @__stop___objc_classes = external hidden global ptr
+// CHECK-NEW: @__start___objc_class_refs = external hidden global ptr
+// CHECK-NEW: @__stop___objc_class_refs = external hidden global ptr
+// CHECK-NEW: @__start___objc_cats = external hidden global ptr
+// CHECK-NEW: @__stop___objc_cats = external hidden global ptr
+// CHECK-NEW: @__start___objc_protocols = external hidden global ptr
+// CHECK-NEW: @__stop___objc_protocols = external hidden global ptr
+// CHECK-NEW: @__start___objc_protocol_refs = external hidden global ptr
+// CHECK-NEW: @__stop___objc_protocol_refs = external hidden global ptr
+// CHECK-NEW: @__start___objc_class_aliases = external hidden global ptr
+// CHECK-NEW: @__stop___objc_class_aliases = external hidden global ptr
+// CHECK-NEW: @__start___objc_constant_string = external hidden global ptr
+// CHECK-NEW: @__stop___objc_constant_string = external hidden global ptr
 
 // Check that we emit the init structure correctly, including in a comdat.
-// CHECK-NEW: @.objc_init = linkonce_odr hidden global { i64, i8**, i8**, i8**, i8**, i8**, i8**, i8**, i8**, i8**, i8**, i8**, i8**, i8**, i8**, i8**, i8** } { i64 0, i8** @__start___objc_selectors, i8** @__stop___objc_selectors, i8** @__start___objc_classes, i8** @__stop___objc_classes, i8** @__start___objc_class_refs, i8** @__stop___objc_class_refs, i8** @__start___objc_cats, i8** @__stop___objc_cats, i8** @__start___objc_protocols, i8** @__stop___objc_protocols, i8** @__start___objc_protocol_refs, i8** @__stop___objc_protocol_refs, i8** @__start___objc_class_aliases, i8** @__stop___objc_class_aliases, i8** @__start___objc_constant_string, i8** @__stop___objc_constant_string }, comdat, align 8
+// CHECK-NEW: @.objc_init = linkonce_odr hidden global { i64, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr } { i64 0, ptr @__start___objc_selectors, ptr @__stop___objc_selectors, ptr @__start___objc_classes, ptr @__stop___objc_classes, ptr @__start___objc_class_refs, ptr @__stop___objc_class_refs, ptr @__start___objc_cats, ptr @__stop___objc_cats, ptr @__start___objc_protocols, ptr @__stop___objc_protocols, ptr @__start___objc_protocol_refs, ptr @__stop___objc_protocol_refs, ptr @__start___objc_class_aliases, ptr @__stop___objc_class_aliases, ptr @__start___objc_constant_string, ptr @__stop___objc_constant_string }, comdat, align 8
 
 // Check that the load function is manually inserted into .ctors.
-// CHECK-NEW: @.objc_ctor = linkonce hidden global void ()* @.objcv2_load_function, section ".ctors", comdat
-// CHECK-INIT_ARRAY: @.objc_ctor = linkonce hidden global void ()* @.objcv2_load_function, section ".init_array", comdat
+// CHECK-NEW: @.objc_ctor = linkonce hidden global ptr @.objcv2_load_function, section ".ctors", comdat
+// CHECK-INIT_ARRAY: @.objc_ctor = linkonce hidden global ptr @.objcv2_load_function, section ".init_array", comdat
 
 
 // Make sure that we provide null versions of everything so the __start /
 // __stop symbols work.
-// CHECK-NEW: @.objc_null_selector = linkonce_odr hidden global { i8*, i8* } zeroinitializer, section "__objc_selectors", comdat, align 8
-// CHECK-NEW: @.objc_null_category = linkonce_odr hidden global { i8*, i8*, i8*, i8*, i8*, i8*, i8* } zeroinitializer, section "__objc_cats", comdat, align 8
-// CHECK-NEW: @.objc_null_protocol = linkonce_odr hidden global { i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8* } zeroinitializer, section "__objc_protocols", comdat, align 8
-// CHECK-NEW: @.objc_null_protocol_ref = linkonce_odr hidden global { i8* } zeroinitializer, section "__objc_protocol_refs", comdat, align 8
-// CHECK-NEW: @.objc_null_class_alias = linkonce_odr hidden global { i8*, i8* } zeroinitializer, section "__objc_class_aliases", comdat, align 8
-// CHECK-NEW: @.objc_null_constant_string = linkonce_odr hidden global { i8*, i32, i32, i32, i32, i8* } zeroinitializer, section "__objc_constant_string", comdat, align 8
+// CHECK-NEW: @.objc_null_selector = linkonce_odr hidden global { ptr, ptr } zeroinitializer, section "__objc_selectors", comdat, align 8
+// CHECK-NEW: @.objc_null_category = linkonce_odr hidden global { ptr, ptr, ptr, ptr, ptr, ptr, ptr } zeroinitializer, section "__objc_cats", comdat, align 8
+// CHECK-NEW: @.objc_null_protocol = linkonce_odr hidden global { ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr } zeroinitializer, section "__objc_protocols", comdat, align 8
+// CHECK-NEW: @.objc_null_protocol_ref = linkonce_odr hidden global { ptr } zeroinitializer, section "__objc_protocol_refs", comdat, align 8
+// CHECK-NEW: @.objc_null_class_alias = linkonce_odr hidden global { ptr, ptr } zeroinitializer, section "__objc_class_aliases", comdat, align 8
+// CHECK-NEW: @.objc_null_constant_string = linkonce_odr hidden global { ptr, i32, i32, i32, i32, ptr } zeroinitializer, section "__objc_constant_string", comdat, align 8
 // Make sure that the null symbols are not going to be removed, even by linking.
-// CHECK-NEW: @llvm.used = appending global [8 x i8*] [i8* bitcast ({ { i8*, i8*, i8*, i64, i64, i64, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i64, i8* }*, i8*, i8*, i64, i64, i64, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i64, i8* }** @._OBJC_INIT_CLASS_X to i8*), i8* bitcast (void ()** @.objc_ctor to i8*), i8* bitcast ({ i8*, i8* }* @.objc_null_selector to i8*), i8* bitcast ({ i8*, i8*, i8*, i8*, i8*, i8*, i8* }* @.objc_null_category to i8*), i8* bitcast ({ i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8* }* @.objc_null_protocol to i8*), i8* bitcast ({ i8* }* @.objc_null_protocol_ref to i8*), i8* bitcast ({ i8*, i8* }* @.objc_null_class_alias to i8*), i8* bitcast ({ i8*, i32, i32, i32, i32, i8* }* @.objc_null_constant_string to i8*)], section "llvm.metadata"
+// CHECK-NEW: @llvm.used = appending global [8 x ptr] [ptr @._OBJC_INIT_CLASS_X, ptr @.objc_ctor, ptr @.objc_null_selector, ptr @.objc_null_category, ptr @.objc_null_protocol, ptr @.objc_null_protocol_ref, ptr @.objc_null_class_alias, ptr @.objc_null_constant_string], section "llvm.metadata"
 // Make sure that the load function and the reference to it are marked as used.
-// CHECK-NEW: @llvm.compiler.used = appending global [1 x i8*] [i8* bitcast (void ()* @.objcv2_load_function to i8*)], section "llvm.metadata"
+// CHECK-NEW: @llvm.compiler.used = appending global [1 x ptr] [ptr @.objcv2_load_function], section "llvm.metadata"
 
 // Check that we emit the load function in a comdat and that it does the right thing.
 // CHECK-NEW: define linkonce_odr hidden void @.objcv2_load_function() comdat {
@@ -63,12 +63,12 @@ @implementation X @end
 // CHECK-NEW-SAME: @.objc_init
 // CHECK-NEW-NEXT: ret void
 
-// CHECK-OLD: @4 = internal global { i64, i64, i8*, { i64, { i8*, i8* }*, i16, i16, [4 x i8*] }* } { i64 9, i64 32, i8* getelementptr inbounds ([{{[0-9]+}} x i8], [{{[0-9]+}} x i8]* @.objc_source_file_name, i64 0, i64 0), { i64, { i8*, i8* }*, i16, i16, [4 x i8*] }* @3 }, align 8
-// CHECK-OLD: @llvm.global_ctors = appending global [1 x { i32, void ()*, i8* }] [{ i32, void ()*, i8* } { i32 65535, void ()* @.objc_load_function, i8* null }]
+// CHECK-OLD: @4 = internal global { i64, i64, ptr, ptr } { i64 9, i64 32, ptr @.objc_source_file_name, ptr @3 }, align 8
+// CHECK-OLD: @llvm.global_ctors = appending global [1 x { i32, ptr, ptr }] [{ i32, ptr, ptr } { i32 65535, ptr @.objc_load_function, ptr null }]
 
 // CHECK-OLD: define internal void @.objc_load_function() {
 // CHECK-OLD-NEXT: entry:
-// CHECK-OLD-NEXT: call void ({ i64, i64, i8*, { i64, { i8*, i8* }*, i16, i16, [4 x i8*] }* }*, ...) @__objc_exec_class({ i64, i64, i8*, { i64, { i8*, i8* }*, i16, i16, [4 x i8*] }* }* @4)
+// CHECK-OLD-NEXT: call void (ptr, ...) @__objc_exec_class(ptr @4)
 
 
 
@@ -88,14 +88,14 @@ @implementation X @end
 // CHECK-WIN-DAG: @"__stop.objcrt$CAL" = linkonce_odr hidden global %.objc_section_sentinel zeroinitializer, section ".objcrt$CAL$z", comdat, align 8
 // CHECK-WIN-DAG: @"__start_.objcrt$STR" = linkonce_odr hidden global %.objc_section_sentinel zeroinitializer, section ".objcrt$STR$a", comdat, align 8
 // CHECK-WIN-DAG: @"__stop.objcrt$STR" = linkonce_odr hidden global %.objc_section_sentinel zeroinitializer, section ".objcrt$STR$z", comdat, align 8
-// CHECK-WIN-DAG: @.objc_init = linkonce_odr hidden global { i64, %.objc_section_sentinel*, %.objc_section_sentinel*, %.objc_section_sentinel*, %.objc_section_sentinel*, %.objc_section_sentinel*, %.objc_section_sentinel*, %.objc_section_sentinel*, %.objc_section_sentinel*, %.objc_section_sentinel*, %.objc_section_sentinel*, %.objc_section_sentinel*, %.objc_section_sentinel*, %.objc_section_sentinel*, %.objc_section_sentinel*, %.objc_section_sentinel*, %.objc_section_sentinel* } { i64 0, %.objc_section_sentinel* @"__start_.objcrt$SEL", %.objc_section_sentinel* @"__stop.objcrt$SEL", %.objc_section_sentinel* @"__start_.objcrt$CLS", %.objc_section_sentinel* @"__stop.objcrt$CLS", %.objc_section_sentinel* @"__start_.objcrt$CLR", %.objc_section_sentinel* @"__stop.objcrt$CLR", %.objc_section_sentinel* @"__start_.objcrt$CAT", %.objc_section_sentinel* @"__stop.objcrt$CAT", %.objc_section_sentinel* @"__start_.objcrt$PCL", %.objc_section_sentinel* @"__stop.objcrt$PCL", %.objc_section_sentinel* @"__start_.objcrt$PCR", %.objc_section_sentinel* @"__stop.objcrt$PCR", %.objc_section_sentinel* @"__start_.objcrt$CAL", %.objc_section_sentinel* @"__stop.objcrt$CAL", %.objc_section_sentinel* @"__start_.objcrt$STR", %.objc_section_sentinel* @"__stop.objcrt$STR" }, comdat, align 8
+// CHECK-WIN-DAG: @.objc_init = linkonce_odr hidden global { i64, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr } { i64 0, ptr @"__start_.objcrt$SEL", ptr @"__stop.objcrt$SEL", ptr @"__start_.objcrt$CLS", ptr @"__stop.objcrt$CLS", ptr @"__start_.objcrt$CLR", ptr @"__stop.objcrt$CLR", ptr @"__start_.objcrt$CAT", ptr @"__stop.objcrt$CAT", ptr @"__start_.objcrt$PCL", ptr @"__stop.objcrt$PCL", ptr @"__start_.objcrt$PCR", ptr @"__stop.objcrt$PCR", ptr @"__start_.objcrt$CAL", ptr @"__stop.objcrt$CAL", ptr @"__start_.objcrt$STR", ptr @"__stop.objcrt$STR" }, comdat, align 8
 
 // Make sure our init variable is in the correct section for late library init.
-// CHECK-WIN: @.objc_ctor = linkonce hidden global void ()* @.objcv2_load_function, section ".CRT$XCLz", comdat
+// CHECK-WIN: @.objc_ctor = linkonce hidden global ptr @.objcv2_load_function, section ".CRT$XCLz", comdat
 
 // We shouldn't have emitted any null placeholders on Windows.
-// CHECK-WIN: @llvm.used = appending global [2 x i8*] [i8* bitcast ({ { i8*, i8*, i8*, i32, i32, i32, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i32, i8* }*, i8*, i8*, i32, i32, i32, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i32, i8* }** @"$_OBJC_INIT_CLASS_X" to i8*), i8* bitcast (void ()** @.objc_ctor to i8*)], section "llvm.metadata"
-// CHECK-WIN: @llvm.compiler.used = appending global [1 x i8*] [i8* bitcast (void ()* @.objcv2_load_function to i8*)], section "llvm.metadata"
+// CHECK-WIN: @llvm.used = appending global [2 x ptr] [ptr @"$_OBJC_INIT_CLASS_X", ptr @.objc_ctor], section "llvm.metadata"
+// CHECK-WIN: @llvm.compiler.used = appending global [1 x ptr] [ptr @.objcv2_load_function], section "llvm.metadata"
 
 // Check our load function is in a comdat.
 // CHECK-WIN: define linkonce_odr hidden void @.objcv2_load_function() comdat {

diff  --git a/clang/test/CodeGenObjC/gnu-method-only-once.m b/clang/test/CodeGenObjC/gnu-method-only-once.m
index 186adca2f9533..639209e02116e 100644
--- a/clang/test/CodeGenObjC/gnu-method-only-once.m
+++ b/clang/test/CodeGenObjC/gnu-method-only-once.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-freebsd -S -emit-llvm -fobjc-runtime=gnustep-2.0 -o - %s | FileCheck %s -check-prefix=CHECK-NEW
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-freebsd -S -emit-llvm -fobjc-runtime=gnustep-1.8 -o - %s | FileCheck %s -check-prefix=CHECK-OLD
+// RUN: %clang_cc1 -triple x86_64-unknown-freebsd -S -emit-llvm -fobjc-runtime=gnustep-2.0 -o - %s | FileCheck %s -check-prefix=CHECK-NEW
+// RUN: %clang_cc1 -triple x86_64-unknown-freebsd -S -emit-llvm -fobjc-runtime=gnustep-1.8 -o - %s | FileCheck %s -check-prefix=CHECK-OLD
 
 // Clang 9 or 10 changed the handling of method lists so that methods provided
 // from synthesised properties showed up in the method list, where previously
@@ -19,5 +19,5 @@ @implementation X
 @end
 
 // Check that the method list has precisely 2 methods.
-// CHECK-NEW: @.objc_method_list = internal global { i8*, i32, i64, [2 x
-// CHECK-OLD: @.objc_method_list = internal global { i8*, i32, [2 x
+// CHECK-NEW: @.objc_method_list = internal global { ptr, i32, i64, [2 x
+// CHECK-OLD: @.objc_method_list = internal global { ptr, i32, [2 x

diff  --git a/clang/test/CodeGenObjC/gnu-nil-receiver.m b/clang/test/CodeGenObjC/gnu-nil-receiver.m
index 27f3126aff8b2..7a1ee8afca6d6 100644
--- a/clang/test/CodeGenObjC/gnu-nil-receiver.m
+++ b/clang/test/CodeGenObjC/gnu-nil-receiver.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-freebsd -fobjc-arc -S -emit-llvm -fobjc-runtime=gnustep-2.0 -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-unknown-freebsd -fobjc-arc -S -emit-llvm -fobjc-runtime=gnustep-2.0 -o - %s | FileCheck %s
 
 typedef struct {
   int x[12];
@@ -34,13 +34,13 @@ - (void) test_super_send {
 @end
 
 // CHECK-LABEL: define{{.*}} void @test_loop_zeroing(
-// CHECK:         [[P:%.*]] = alloca i8*,
+// CHECK:         [[P:%.*]] = alloca ptr,
 // CHECK:         [[BIG:%.*]] = alloca %struct.Big,
 // CHECK:         br label %for.cond
 //
 // CHECK:       for.cond:
-// CHECK-NEXT:    [[RECEIVER:%.*]] = load i8*, i8** [[P]],
-// CHECK-NEXT:    [[ISNIL:%.*]] = icmp eq i8* [[RECEIVER]], null
+// CHECK-NEXT:    [[RECEIVER:%.*]] = load ptr, ptr [[P]],
+// CHECK-NEXT:    [[ISNIL:%.*]] = icmp eq ptr [[RECEIVER]], null
 // CHECK-NEXT:    br i1 [[ISNIL]], label %nilReceiverCleanup, label %msgSend
 //
 // CHECK:       msgSend:
@@ -49,8 +49,7 @@ - (void) test_super_send {
 // CHECK:         br label %continue
 //
 // CHECK:       nilReceiverCleanup:
-// CHECK-NEXT:    [[T0:%.*]] = bitcast %struct.Big* [[BIG]] to i8*
-// CHECK-NEXT:    call void @llvm.memset.p0i8.i64(i8* align 4 [[T0]], i8 0, i64 48, i1 false)
+// CHECK-NEXT:    call void @llvm.memset.p0.i64(ptr align 4 [[BIG]], i8 0, i64 48, i1 false)
 // CHECK-NEXT:    br label %continue
 //
 // CHECK:       continue:
@@ -62,16 +61,16 @@ void test_loop_zeroing(id<P> p) {
 }
 
 // CHECK-LABEL: define{{.*}} void @test_zeroing_and_consume(
-// CHECK:         [[P:%.*]] = alloca i8*,
-// CHECK:         [[Q:%.*]] = alloca i8*,
+// CHECK:         [[P:%.*]] = alloca ptr,
+// CHECK:         [[Q:%.*]] = alloca ptr,
 // CHECK:         [[BIG:%.*]] = alloca %struct.Big,
 // CHECK:         br label %for.cond
 //
 // CHECK:       for.cond:
-// CHECK-NEXT:    [[RECEIVER:%.*]] = load i8*, i8** [[P]],
-// CHECK-NEXT:    [[Q_LOAD:%.*]] = load i8*, i8** [[Q]],
-// CHECK-NEXT:    [[Q_RETAIN:%.*]] = call i8* @llvm.objc.retain(i8* [[Q_LOAD]])
-// CHECK-NEXT:    [[ISNIL:%.*]] = icmp eq i8* [[RECEIVER]], null
+// CHECK-NEXT:    [[RECEIVER:%.*]] = load ptr, ptr [[P]],
+// CHECK-NEXT:    [[Q_LOAD:%.*]] = load ptr, ptr [[Q]],
+// CHECK-NEXT:    [[Q_RETAIN:%.*]] = call ptr @llvm.objc.retain(ptr [[Q_LOAD]])
+// CHECK-NEXT:    [[ISNIL:%.*]] = icmp eq ptr [[RECEIVER]], null
 // CHECK-NEXT:    br i1 [[ISNIL]], label %nilReceiverCleanup, label %msgSend
 //
 // CHECK:       msgSend:
@@ -80,9 +79,8 @@ void test_loop_zeroing(id<P> p) {
 // CHECK:         br label %continue
 //
 // CHECK:       nilReceiverCleanup:
-// CHECK-NEXT:    call void @llvm.objc.release(i8* [[Q_RETAIN]])
-// CHECK-NEXT:    [[T0:%.*]] = bitcast %struct.Big* [[BIG]] to i8*
-// CHECK-NEXT:    call void @llvm.memset.p0i8.i64(i8* align 4 [[T0]], i8 0, i64 48, i1 false)
+// CHECK-NEXT:    call void @llvm.objc.release(ptr [[Q_RETAIN]])
+// CHECK-NEXT:    call void @llvm.memset.p0.i64(ptr align 4 [[BIG]], i8 0, i64 48, i1 false)
 // CHECK-NEXT:    br label %continue
 //
 // CHECK:       continue:
@@ -94,9 +92,9 @@ void test_zeroing_and_consume(id<P> p, id q) {
 }
 
 // CHECK-LABEL: define{{.*}} void @test_complex(
-// CHECK:         [[P:%.*]] = alloca i8*,
-// CHECK:         [[RECEIVER:%.*]] = load i8*, i8** [[P]],
-// CHECK-NEXT:    [[ISNIL:%.*]] = icmp eq i8* [[RECEIVER]], null
+// CHECK:         [[P:%.*]] = alloca ptr,
+// CHECK:         [[RECEIVER:%.*]] = load ptr, ptr [[P]],
+// CHECK-NEXT:    [[ISNIL:%.*]] = icmp eq ptr [[RECEIVER]], null
 // CHECK-NEXT:    br i1 [[ISNIL]], label %continue, label %msgSend
 // CHECK:       msgSend:
 // CHECK:         @objc_msg_lookup_sender

diff  --git a/clang/test/CodeGenObjC/gnustep2-category.m b/clang/test/CodeGenObjC/gnustep2-category.m
index 05ffba4e786b3..6114a5979dbc6 100644
--- a/clang/test/CodeGenObjC/gnustep2-category.m
+++ b/clang/test/CodeGenObjC/gnustep2-category.m
@@ -1,8 +1,8 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-freebsd -S -emit-llvm -fobjc-runtime=gnustep-2.0 -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-unknown-freebsd -S -emit-llvm -fobjc-runtime=gnustep-2.0 -o - %s | FileCheck %s
 
 
 // Check that we have a method list that refers to the correct thing method:
-// CHECK: internal global { i8*, i32, i64, [1 x { i8* (i8*, i8*, ...)*, i8*, i8* }] } { i8* null, i32 1, i64 24, 
+// CHECK: internal global { ptr, i32, i64, [1 x { ptr, ptr, ptr }] } { ptr null, i32 1, i64 24, 
 // CHECK-SAME: @_i_X_Cat_x 
 // CHECK-SAME: @".objc_selector_x_i16\010:8"
 
@@ -10,9 +10,9 @@
 // CHECK: c"Ti,R\00"
 
 // Check that we emit a single-element property list of the correct form.
-// CHECK: internal global { i32, i32, i8*, [1 x { i8*, i8*, i8*, i8*, i8* }] }
+// CHECK: internal global { i32, i32, ptr, [1 x { ptr, ptr, ptr, ptr, ptr }] }
 
-// CHECK: @.objc_category_XCat = internal global { i8*, i8*, i8*, i8*, i8*, i8*, i8* }
+// CHECK: @.objc_category_XCat = internal global { ptr, ptr, ptr, ptr, ptr, ptr, ptr }
 // CHECK-SAME: section "__objc_cats", align 8
 
 @interface X @end

diff  --git a/clang/test/CodeGenObjC/gnustep2-class.m b/clang/test/CodeGenObjC/gnustep2-class.m
index 490b422bce839..28f9ad12c6b44 100644
--- a/clang/test/CodeGenObjC/gnustep2-class.m
+++ b/clang/test/CodeGenObjC/gnustep2-class.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-freebsd -S -emit-llvm -fobjc-runtime=gnustep-2.0 -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-unknown-freebsd -S -emit-llvm -fobjc-runtime=gnustep-2.0 -o - %s | FileCheck %s
 
 @interface Super @end
 
@@ -23,17 +23,17 @@ - (id)meth { return ivar2; }
 // CHECK: @"__objc_ivar_offset_X.y.\01" = hidden global i32 16
 //
 // Check that we get a sensible metaclass method list.
-// CHECK: internal global { i8*, i32, i64, [1 x { i8* (i8*, i8*, ...)*, i8*, i8* }] }
+// CHECK: internal global { ptr, i32, i64, [1 x { ptr, ptr, ptr }] }
 // CHECK-SAME: @_c_X__clsMeth
 
 // Check that we get a metaclass and that it is not an exposed symbol:
 // CHECK: @._OBJC_METACLASS_X = internal global
 
 // Check that we get a reference to the superclass symbol:
-// CHECK: @._OBJC_CLASS_Super = external global i8*
+// CHECK: @._OBJC_CLASS_Super = external global ptr
 
 // Check that we get an ivar list with all three ivars, in the correct order
-// CHECK: private global { i32, i64, [3 x { i8*, i8*, i32*, i32, i32 }] }
+// CHECK: private global { i32, i64, [3 x { ptr, ptr, ptr, i32, i32 }] }
 // CHECK-SAME: @__objc_ivar_offset_X.ivar1.i
 // CHECK-SAME: @"__objc_ivar_offset_X.ivar2.\01"
 // CHECK-SAME: @"__objc_ivar_offset_X.y.\01"
@@ -41,10 +41,10 @@ - (id)meth { return ivar2; }
 // Check that we get some plausible property metadata.
 // CHECK: private unnamed_addr constant [5 x i8] c"Ti,R\00", align 1
 // CHECK: private unnamed_addr constant [6 x i8] c"T@,Vy\00", align 1
-// CHECK: = internal global { i32, i32, i8*, [2 x { i8*, i8*, i8*, i8*, i8* }] } { i32 2, i32 40, i8* null,
+// CHECK: = internal global { i32, i32, ptr, [2 x { ptr, ptr, ptr, ptr, ptr }] } { i32 2, i32 40, ptr null,
 
 // Check that we get a class structure.
-// CHECK: @._OBJC_CLASS_X = global { { i8*, i8*, i8*, i64, i64, i64, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i64, i8* }*, i8*, i8*, i64, i64, i64, { i32, i64, [3 x { i8*, i8*, i32*, i32, i32 }] }*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i8*, i64, { i32, i32, i8*, [2 x { i8*, i8*, i8*, i8*, i8* }] }* }
+// CHECK: @._OBJC_CLASS_X = global { ptr, ptr, ptr, i64, i64, i64, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, i64, ptr }
 // CHECK-SAME: @._OBJC_METACLASS_X
 // CHECK-SAME: @._OBJC_CLASS_Super
 

diff  --git a/clang/test/CodeGenObjC/gnustep2-ivar-offset.m b/clang/test/CodeGenObjC/gnustep2-ivar-offset.m
index a9d96a4ff17e9..1f838cdd7563f 100644
--- a/clang/test/CodeGenObjC/gnustep2-ivar-offset.m
+++ b/clang/test/CodeGenObjC/gnustep2-ivar-offset.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-freebsd -S -emit-llvm -fobjc-runtime=gnustep-2.0 -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-unknown-freebsd -S -emit-llvm -fobjc-runtime=gnustep-2.0 -o - %s | FileCheck %s
 
 
 @class NSString;
@@ -25,6 +25,6 @@ @interface ANObject {
 @implementation ANObject @end
 
 // Check that the ivar metadata contains 3 entries of the correct form and correctly sets the size.
-// CHECK: @.objc_ivar_list = private global { i32, i64, [4 x { i8*, i8*, i32*, i32, i32 }] } { i32 4, i64 32,
+// CHECK: @.objc_ivar_list = private global { i32, i64, [4 x { ptr, ptr, ptr, i32, i32 }] } { i32 4, i64 32,
 // Check that we emit 1 as the size of _Bool, not 0.
 // CHECK-SAME:  @__objc_ivar_offset_ANObject.boolIvar.B, i32 1, i32 4

diff  --git a/clang/test/CodeGenObjC/gnustep2-proto.m b/clang/test/CodeGenObjC/gnustep2-proto.m
index 1e56e157e1509..d5b61a43655e8 100644
--- a/clang/test/CodeGenObjC/gnustep2-proto.m
+++ b/clang/test/CodeGenObjC/gnustep2-proto.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-freebsd -S -emit-llvm -fobjc-runtime=gnustep-2.0 -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-unknown-freebsd -S -emit-llvm -fobjc-runtime=gnustep-2.0 -o - %s | FileCheck %s
 
 @protocol X
 @optional
@@ -11,10 +11,10 @@ + (void*)y;
 @end
 
 // Check that we get some plausible-looking method lists.
-// CHECK: internal global { i32, i32, [2 x { i8*, i8* }] } { i32 2, i32 16, 
+// CHECK: internal global { i32, i32, [2 x { ptr, ptr }] } { i32 2, i32 16, 
 // CHECK-SAME: @".objc_selector_reqProp_i16\010:8"
 // CHECK-SAME: @".objc_selector_setReqProp:_v20\010:8i16"
-// CHECK: internal global { i32, i32, [3 x { i8*, i8* }] } { i32 3, i32 16,
+// CHECK: internal global { i32, i32, [3 x { ptr, ptr }] } { i32 3, i32 16,
 // CHECK-SAME: @".objc_selector_x_\0116\010:8"
 // CHECK-SAME: @".objc_selector_optProp_i16\010:8"
 // CHECK-SAME: @".objc_selector_setOptProp:_v20\010:8i16"
@@ -30,7 +30,7 @@ + (void*)y;
 
 
 // Check that we load from the indirection variable on protocol references.
-// CHECK: define{{.*}} i8* @x()
+// CHECK: define{{.*}} ptr @x()
 // CHECK:   = load 
 // CHECK-SAME: @._OBJC_REF_PROTOCOL_X, align 8
 void *x(void)

diff  --git a/clang/test/CodeGenObjC/id-isa-codegen.m b/clang/test/CodeGenObjC/id-isa-codegen.m
index 1d591888bc8cd..46c0b7181ee52 100644
--- a/clang/test/CodeGenObjC/id-isa-codegen.m
+++ b/clang/test/CodeGenObjC/id-isa-codegen.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5  -emit-llvm -o - %s | FileCheck -check-prefix CHECK-LP64 %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple i386-apple-darwin9 -fobjc-runtime=macosx-fragile-10.5  -emit-llvm -o - %s | FileCheck -check-prefix CHECK-LP32 %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5  -emit-llvm -o - %s | FileCheck -check-prefix CHECK-LP64 %s
+// RUN: %clang_cc1 -triple i386-apple-darwin9 -fobjc-runtime=macosx-fragile-10.5  -emit-llvm -o - %s | FileCheck -check-prefix CHECK-LP32 %s
 
 typedef struct objc_class *Class;
 
@@ -64,10 +64,8 @@ + (void)geneticallyAlterCat:(Cat *)cat {
     ((id)cat)->isa = dynamicSubclass;
 }
 @end
-// CHECK-LP64: %{{.*}} = load i8*, i8** %
-// CHECK-NEXT: %{{.*}} = bitcast i8* %{{.*}} to i8**
-// CHECK-NEXT: store i8* %{{.*}}, i8** %{{.*}}
+// CHECK-LP64: %{{.*}} = load ptr, ptr %
+// CHECK-NEXT: store ptr %{{.*}}, ptr %{{.*}}
 
-// CHECK-LP32: %{{.*}} = load i8*, i8** %
-// CHECK-NEXT: %{{.*}} = bitcast i8* %{{.*}} to i8**
-// CHECK-NEXT: store i8* %{{.*}}, i8** %{{.*}}
+// CHECK-LP32: %{{.*}} = load ptr, ptr %
+// CHECK-NEXT: store ptr %{{.*}}, ptr %{{.*}}

diff  --git a/clang/test/CodeGenObjC/implicit-objc_msgSend.m b/clang/test/CodeGenObjC/implicit-objc_msgSend.m
index 2c56108903b44..079a7f8b1d831 100644
--- a/clang/test/CodeGenObjC/implicit-objc_msgSend.m
+++ b/clang/test/CodeGenObjC/implicit-objc_msgSend.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -Wno-implicit-function-declaration -triple x86_64-apple-darwin9 -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -o %t %s
-// RUN: grep -F 'declare i8* @objc_msgSend(i8* noundef, i8* noundef, ...)' %t
+// RUN: %clang_cc1 -Wno-implicit-function-declaration -triple x86_64-apple-darwin9 -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -o %t %s
+// RUN: grep -F 'declare ptr @objc_msgSend(ptr noundef, ptr noundef, ...)' %t
 
 typedef struct objc_selector *SEL;
 id f0(id x, SEL s) {

diff  --git a/clang/test/CodeGenObjC/instance-method-metadata.m b/clang/test/CodeGenObjC/instance-method-metadata.m
index 8ebe893a5a068..e08de8fdacec2 100644
--- a/clang/test/CodeGenObjC/instance-method-metadata.m
+++ b/clang/test/CodeGenObjC/instance-method-metadata.m
@@ -1,5 +1,5 @@
 // REQUIRES: x86-registered-target
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -S %s -o - | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -S %s -o - | FileCheck %s
 
 // rdar://9072317
 

diff  --git a/clang/test/CodeGenObjC/ivar-base-as-invariant-load.m b/clang/test/CodeGenObjC/ivar-base-as-invariant-load.m
index 25ecea4ec5a6e..f48436e72a98b 100644
--- a/clang/test/CodeGenObjC/ivar-base-as-invariant-load.m
+++ b/clang/test/CodeGenObjC/ivar-base-as-invariant-load.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm  -triple x86_64-apple-darwin -x objective-c %s -o - | FileCheck %s
+// RUN: %clang_cc1 -emit-llvm  -triple x86_64-apple-darwin -x objective-c %s -o - | FileCheck %s
 // rdar://10840980
 
 @interface A {
@@ -23,7 +23,7 @@ - (id)init {
 
 @end
 
-// CHECK: [[T1:%.*]] = load i64, i64* @"OBJC_IVAR_$_A._flags", align 8, !invariant.load ![[MD_NUM:[0-9]+]]
-// CHECK: [[T2:%.*]] = load i64, i64* @"OBJC_IVAR_$_A._flags", align 8, !invariant.load ![[MD_NUM]]
-// CHECK: [[T3:%.*]] = load i64, i64* @"OBJC_IVAR_$_A._flags", align 8, !invariant.load ![[MD_NUM]]
+// CHECK: [[T1:%.*]] = load i64, ptr @"OBJC_IVAR_$_A._flags", align 8, !invariant.load ![[MD_NUM:[0-9]+]]
+// CHECK: [[T2:%.*]] = load i64, ptr @"OBJC_IVAR_$_A._flags", align 8, !invariant.load ![[MD_NUM]]
+// CHECK: [[T3:%.*]] = load i64, ptr @"OBJC_IVAR_$_A._flags", align 8, !invariant.load ![[MD_NUM]]
 //

diff  --git a/clang/test/CodeGenObjC/ivar-invariant.m b/clang/test/CodeGenObjC/ivar-invariant.m
index 6cc454311f027..d4cdcb97b78fc 100644
--- a/clang/test/CodeGenObjC/ivar-invariant.m
+++ b/clang/test/CodeGenObjC/ivar-invariant.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin -fblocks -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin -fblocks -emit-llvm -o - %s | FileCheck %s
 
 @interface NSObject
 + (id) new;
@@ -28,8 +28,8 @@ - (id) init
 }
 @end
 
-// CHECK: define internal i8* @"\01-[Derived init]"
-// CHECK: [[IVAR:%.*]] = load i64, i64* @"OBJC_IVAR_$_Derived.member", align 8, !invariant.load
+// CHECK: define internal ptr @"\01-[Derived init]"
+// CHECK: [[IVAR:%.*]] = load i64, ptr @"OBJC_IVAR_$_Derived.member", align 8, !invariant.load
 
 void * variant_load_1(int i) {
     void *ptr;
@@ -40,8 +40,8 @@ - (id) init
     return ptr;
 }
 
-// CHECK-LABEL: define{{.*}} i8* @variant_load_1(i32 noundef %i)
-// CHECK: [[IVAR:%.*]] = load i64, i64* @"OBJC_IVAR_$_Derived.member", align 8{{$}}
+// CHECK-LABEL: define{{.*}} ptr @variant_load_1(i32 noundef %i)
+// CHECK: [[IVAR:%.*]] = load i64, ptr @"OBJC_IVAR_$_Derived.member", align 8{{$}}
 
 @interface Container : Derived @end
 @implementation Container
@@ -51,8 +51,8 @@ - (void *) invariant_load_1
 }
 @end
 
-// CHECK-LABEL: define internal i8* @"\01-[Container invariant_load_1]"
-// CHECK: [[IVAR:%.*]] = load i64, i64* @"OBJC_IVAR_$_Derived.member", align 8, !invariant.load
+// CHECK-LABEL: define internal ptr @"\01-[Container invariant_load_1]"
+// CHECK: [[IVAR:%.*]] = load i64, ptr @"OBJC_IVAR_$_Derived.member", align 8, !invariant.load
 
 @interface ForBlock
 { 
@@ -61,8 +61,8 @@ @interface ForBlock
 }
 @end
 
-// CHECK-LABEL: define internal i8* @block_block_invoke
-// CHECK: load i64, i64* @"OBJC_IVAR_$_ForBlock.foo"
+// CHECK-LABEL: define internal ptr @block_block_invoke
+// CHECK: load i64, ptr @"OBJC_IVAR_$_ForBlock.foo"
 id (^block)(ForBlock*) = ^(ForBlock* a) {
   return a->foo;
 };

diff  --git a/clang/test/CodeGenObjC/local-static-block.m b/clang/test/CodeGenObjC/local-static-block.m
index 6c8ed88532861..d4962beeb5826 100644
--- a/clang/test/CodeGenObjC/local-static-block.m
+++ b/clang/test/CodeGenObjC/local-static-block.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -fblocks -triple x86_64-apple-darwin -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -fblocks -triple x86_64-apple-darwin -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -o - %s | FileCheck %s
 // rdar: // 8390455
 
 // CHECK: @ArrayRecurs = internal global
@@ -59,7 +59,7 @@ static void (^const block1)(int) = ^(int a){
 
 // CHECK-LABEL: define{{.*}} void @FUNC2(
 // CHECK: define internal void @_block_invoke{{.*}}(
-// CHECK: call void %{{.*}}(i8* noundef bitcast ({ i8**, i32, i32, i8*, %struct.__block_descriptor* }* @__block_literal_global{{.*}} to i8*), i32 noundef %{{.*}})
+// CHECK: call void %{{.*}}(ptr noundef @__block_literal_global{{.*}}, i32 noundef %{{.*}})
 
 void FUNC1(void)
 {

diff  --git a/clang/test/CodeGenObjC/matrix-type-builtins.m b/clang/test/CodeGenObjC/matrix-type-builtins.m
index acda3ac8f5c19..4b19820d2c85e 100644
--- a/clang/test/CodeGenObjC/matrix-type-builtins.m
+++ b/clang/test/CodeGenObjC/matrix-type-builtins.m
@@ -1,5 +1,5 @@
 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -fenable-matrix -emit-llvm -disable-llvm-optzns -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -fenable-matrix -emit-llvm -disable-llvm-optzns -o - %s | FileCheck %s
 
 // Check that we correctly deal with placeholder expressions.
 
@@ -40,22 +40,22 @@ @interface IntValue
 @end
 
 void test_column_major_load(PtrValue *Ptr, IntValue *Stride) {
-    // CHECK-LABEL: define{{.*}} void @test_column_major_load(%2* noundef %Ptr, %3* noundef %Stride) #4 {
-  // CHECK:         [[STRIDE:%.*]] = call i32 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i32 (i8*, i8*)*)
+    // CHECK-LABEL: define{{.*}} void @test_column_major_load(ptr noundef %Ptr, ptr noundef %Stride) #4 {
+  // CHECK:         [[STRIDE:%.*]] = call i32 @objc_msgSend
   // CHECK-NEXT:    [[STRIDE_EXT:%.*]] = sext i32 [[STRIDE]] to i64
-  // CHECK:         [[PTR:%.*]] = call i32* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i32* (i8*, i8*)*)
-  // CHECK-NEXT:    call <12 x i32> @llvm.matrix.column.major.load.v12i32.i64(i32* align 4 [[PTR]], i64 [[STRIDE_EXT]], i1 false, i32 3, i32 4)
+  // CHECK:         [[PTR:%.*]] = call ptr @objc_msgSend
+  // CHECK-NEXT:    call <12 x i32> @llvm.matrix.column.major.load.v12i32.i64(ptr align 4 [[PTR]], i64 [[STRIDE_EXT]], i1 false, i32 3, i32 4)
 
   u3x4 m = __builtin_matrix_column_major_load(Ptr.value, 3, 4, Stride.value);
 }
 
 void test_column_major_store(UnsignedMatrixValue *M, PtrValue *Ptr, IntValue *Stride) {
-    // CHECK-LABEL: define{{.*}} void @test_column_major_store(%1* noundef %M, %2* noundef %Ptr, %3* noundef %Stride) #3 {
-  // CHECK:         [[M:%.*]] = call <12 x i32> bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to <12 x i32> (i8*, i8*)*)
-  // CHECK:         [[PTR:%.*]] = call i32* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i32* (i8*, i8*)*)
-  // CHECK:         [[IDX:%.*]] = call i32 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i32 (i8*, i8*)*)
+    // CHECK-LABEL: define{{.*}} void @test_column_major_store(ptr noundef %M, ptr noundef %Ptr, ptr noundef %Stride) #3 {
+  // CHECK:         [[M:%.*]] = call <12 x i32> @objc_msgSend
+  // CHECK:         [[PTR:%.*]] = call ptr @objc_msgSend
+  // CHECK:         [[IDX:%.*]] = call i32 @objc_msgSend
   // CHECK-NEXT:    [[IDX_EXT:%.*]] = sext i32 [[IDX]] to i64
-  // CHECK-NEXT:    call void @llvm.matrix.column.major.store.v12i32.i64(<12 x i32> [[M]], i32* align 4 [[PTR]], i64 [[IDX_EXT]], i1 false, i32 3, i32 4)
+  // CHECK-NEXT:    call void @llvm.matrix.column.major.store.v12i32.i64(<12 x i32> [[M]], ptr align 4 [[PTR]], i64 [[IDX_EXT]], i1 false, i32 3, i32 4)
 
   __builtin_matrix_column_major_store(M.value, Ptr.value, Stride.value);
 }

diff  --git a/clang/test/CodeGenObjC/matrix-type-operators.m b/clang/test/CodeGenObjC/matrix-type-operators.m
index 36f294f6e8b93..179bb180daf90 100644
--- a/clang/test/CodeGenObjC/matrix-type-operators.m
+++ b/clang/test/CodeGenObjC/matrix-type-operators.m
@@ -1,5 +1,5 @@
 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -fenable-matrix -emit-llvm -disable-llvm-optzns -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -fenable-matrix -emit-llvm -disable-llvm-optzns -o - %s | FileCheck %s
 
 __attribute__((objc_root_class))
 @interface IntValue
@@ -12,19 +12,17 @@ @interface IntValue
 
 // CHECK-LABEL: @test_index_placeholders(
 // CHECK-NEXT:  entry:
-// CHECK:         [[IV:%.*]] = load %0*, %0** [[IV_ADDR:%.*]], align 8
-// CHECK-NEXT:    [[IV_PTR:%.*]] = bitcast %0* [[IV]] to i8*
-// CHECK-NEXT:    [[SEL:%.*]] = load i8*, i8** @OBJC_SELECTOR_REFERENCES_, align 8, !invariant.load !7
-// CHECK-NEXT:    [[CALL:%.*]] = call i32 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i32 (i8*, i8*)*)(i8* noundef [[IV_PTR]], i8* noundef [[SEL]])
+// CHECK:         [[IV:%.*]] = load ptr, ptr [[IV_ADDR:%.*]], align 8
+// CHECK-NEXT:    [[SEL:%.*]] = load ptr, ptr @OBJC_SELECTOR_REFERENCES_, align 8, !invariant.load !7
+// CHECK-NEXT:    [[CALL:%.*]] = call i32 @objc_msgSend(ptr noundef [[IV]], ptr noundef [[SEL]])
 // CHECK-NEXT:    [[CONV:%.*]] = sext i32 [[CALL]] to i64
-// CHECK-NEXT:    [[IV2:%.*]] = load %0*, %0** [[IV_ADDR]], align 8
-// CHECK-NEXT:    [[IV2_PTR:%.*]] = bitcast %0* [[IV2]] to i8*
-// CHECK-NEXT:    [[SEL2:%.*]] = load i8*, i8** @OBJC_SELECTOR_REFERENCES_, align 8, !invariant.load !7
-// CHECK-NEXT:    [[CALL1:%.*]] = call i32 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i32 (i8*, i8*)*)(i8* noundef [[IV2_PTR]], i8* noundef [[SEL2]])
+// CHECK-NEXT:    [[IV2:%.*]] = load ptr, ptr [[IV_ADDR]], align 8
+// CHECK-NEXT:    [[SEL2:%.*]] = load ptr, ptr @OBJC_SELECTOR_REFERENCES_, align 8, !invariant.load !7
+// CHECK-NEXT:    [[CALL1:%.*]] = call i32 @objc_msgSend(ptr noundef [[IV2]], ptr noundef [[SEL2]])
 // CHECK-NEXT:    [[CONV2:%.*]] = sext i32 [[CALL1]] to i64
 // CHECK-NEXT:    [[IDX1:%.*]] = mul i64 [[CONV2]], 4
 // CHECK-NEXT:    [[IDX2:%.*]] = add i64 [[IDX1]], [[CONV]]
-// CHECK-NEXT:    [[MAT:%.*]] = load <16 x double>, <16 x double>* {{.*}} align 8
+// CHECK-NEXT:    [[MAT:%.*]] = load <16 x double>, ptr {{.*}} align 8
 // CHECK-NEXT:    [[MATEXT:%.*]] = extractelement <16 x double> [[MAT]], i64 [[IDX2]]
 // CHECK-NEXT:    ret double [[MATEXT]]
 //
@@ -39,22 +37,19 @@ @interface MatrixValue
 @end
 
 // CHECK-LABEL: @test_base_and_index_placeholders(
-// CHECK:         [[IV:%.*]] = load %0*, %0** [[IV_ADDR:%.*]], align 8
-// CHECK-NEXT:    [[IV_PTR:%.*]] = bitcast %0* [[IV]] to i8*
-// CHECK-NEXT:    [[SEL:%.*]] = load i8*, i8** @OBJC_SELECTOR_REFERENCES_, align 8, !invariant.load !7
-// CHECK-NEXT:    [[CALL:%.*]] = call i32 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i32 (i8*, i8*)*)(i8* noundef [[IV_PTR]], i8* noundef [[SEL]])
+// CHECK:         [[IV:%.*]] = load ptr, ptr [[IV_ADDR:%.*]], align 8
+// CHECK-NEXT:    [[SEL:%.*]] = load ptr, ptr @OBJC_SELECTOR_REFERENCES_, align 8, !invariant.load !7
+// CHECK-NEXT:    [[CALL:%.*]] = call i32 @objc_msgSend(ptr noundef [[IV]], ptr noundef [[SEL]])
 // CHECK-NEXT:    [[CONV:%.*]] = sext i32 [[CALL]] to i64
-// CHECK-NEXT:    [[IV2:%.*]] = load %0*, %0** [[IV_ADDR]], align 8
-// CHECK-NEXT:    [[IV2_PTR:%.*]] = bitcast %0* [[IV2]] to i8*
-// CHECK-NEXT:    [[SEL2:%.*]] = load i8*, i8** @OBJC_SELECTOR_REFERENCES_, align 8, !invariant.load !7
-// CHECK-NEXT:    [[CALL1:%.*]] = call i32 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i32 (i8*, i8*)*)(i8* noundef [[IV2_PTR]], i8* noundef [[SEL2]])
+// CHECK-NEXT:    [[IV2:%.*]] = load ptr, ptr [[IV_ADDR]], align 8
+// CHECK-NEXT:    [[SEL2:%.*]] = load ptr, ptr @OBJC_SELECTOR_REFERENCES_, align 8, !invariant.load !7
+// CHECK-NEXT:    [[CALL1:%.*]] = call i32 @objc_msgSend(ptr noundef [[IV2]], ptr noundef [[SEL2]])
 // CHECK-NEXT:    [[CONV2:%.*]] = sext i32 [[CALL1]] to i64
 // CHECK-NEXT:    [[IDX1:%.*]] = mul i64 [[CONV2]], 4
 // CHECK-NEXT:    [[IDX2:%.*]] = add i64 [[IDX1]], [[CONV]]
-// CHECK-NEXT:    [[M:%.*]] = load %1*, %1** %m.addr, align 8
-// CHECK-NEXT:    [[M_PTR:%.*]] = bitcast %1* [[M]] to i8*
-// CHECK-NEXT:    [[SEL3:%.*]] = load i8*, i8** @OBJC_SELECTOR_REFERENCES_, align 8, !invariant.load !7
-// CHECK-NEXT:    [[MAT:%.*]] = call <16 x double> bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to <16 x double> (i8*, i8*)*)(i8* noundef [[M_PTR]], i8* noundef [[SEL3]])
+// CHECK-NEXT:    [[M:%.*]] = load ptr, ptr %m.addr, align 8
+// CHECK-NEXT:    [[SEL3:%.*]] = load ptr, ptr @OBJC_SELECTOR_REFERENCES_, align 8, !invariant.load !7
+// CHECK-NEXT:    [[MAT:%.*]] = call <16 x double> @objc_msgSend(ptr noundef [[M]], ptr noundef [[SEL3]])
 // CHECK-NEXT:    [[MATEXT:%.*]] = extractelement <16 x double> [[MAT]], i64 [[IDX2]]
 // CHECK-NEXT:    ret double [[MATEXT]]
 //

diff  --git a/clang/test/CodeGenObjC/misc-atomic-property.m b/clang/test/CodeGenObjC/misc-atomic-property.m
index d141bf682f493..4c8cc669705dd 100644
--- a/clang/test/CodeGenObjC/misc-atomic-property.m
+++ b/clang/test/CodeGenObjC/misc-atomic-property.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10  -emit-llvm -o - %s | FileCheck %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple i386-apple-darwin9 -fobjc-runtime=macosx-fragile-10.5  -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10  -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple i386-apple-darwin9 -fobjc-runtime=macosx-fragile-10.5  -emit-llvm -o - %s | FileCheck %s
 // rdar: //8808439
 
 typedef struct {

diff  --git a/clang/test/CodeGenObjC/mrc-weak.m b/clang/test/CodeGenObjC/mrc-weak.m
index 23079a52a9739..499a41a510ec5 100644
--- a/clang/test/CodeGenObjC/mrc-weak.m
+++ b/clang/test/CodeGenObjC/mrc-weak.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-10.10 -emit-llvm -fblocks -fobjc-weak -o - %s | FileCheck %s -check-prefix=CHECK -check-prefix=CHECK-MODERN
-// RUN: %clang_cc1 -no-opaque-pointers -triple i386-apple-darwin10 -fobjc-runtime=macosx-fragile-10.10 -emit-llvm -fblocks -fobjc-weak -o - %s | FileCheck %s -check-prefix=CHECK -check-prefix=CHECK-FRAGILE
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-10.10 -emit-llvm -fblocks -fobjc-weak -o - %s | FileCheck %s -check-prefix=CHECK -check-prefix=CHECK-MODERN
+// RUN: %clang_cc1 -triple i386-apple-darwin10 -fobjc-runtime=macosx-fragile-10.10 -emit-llvm -fblocks -fobjc-weak -o - %s | FileCheck %s -check-prefix=CHECK -check-prefix=CHECK-FRAGILE
 
 @interface Object
 - (instancetype) retain;
@@ -52,7 +52,7 @@ @implementation Foo
 
 void test1(__weak id x) {}
 // CHECK-LABEL: define{{.*}} void @test1
-// CHECK:      [[X:%.*]] = alloca i8*,
+// CHECK:      [[X:%.*]] = alloca ptr,
 // CHECK-NEXT: @llvm.objc.initWeak
 // CHECK-NEXT: @llvm.objc.destroyWeak
 // CHECK-NEXT: ret void
@@ -61,12 +61,12 @@ void test2(id y) {
   __weak id z = y;
 }
 // CHECK-LABEL: define{{.*}} void @test2
-// CHECK:      [[Y:%.*]] = alloca i8*,
-// CHECK-NEXT: [[Z:%.*]] = alloca i8*,
+// CHECK:      [[Y:%.*]] = alloca ptr,
+// CHECK-NEXT: [[Z:%.*]] = alloca ptr,
 // CHECK-NEXT: store
-// CHECK-NEXT: [[T0:%.*]] = load i8*, i8** [[Y]]
-// CHECK-NEXT: call i8* @llvm.objc.initWeak(i8** [[Z]], i8* [[T0]])
-// CHECK-NEXT: call void @llvm.objc.destroyWeak(i8** [[Z]])
+// CHECK-NEXT: [[T0:%.*]] = load ptr, ptr [[Y]]
+// CHECK-NEXT: call ptr @llvm.objc.initWeak(ptr [[Z]], ptr [[T0]])
+// CHECK-NEXT: call void @llvm.objc.destroyWeak(ptr [[Z]])
 // CHECK-NEXT: ret void
 
 void test3(id y) {
@@ -74,51 +74,49 @@ void test3(id y) {
   z = y;
 }
 // CHECK-LABEL: define{{.*}} void @test3
-// CHECK:      [[Y:%.*]] = alloca i8*,
-// CHECK-NEXT: [[Z:%.*]] = alloca i8*,
+// CHECK:      [[Y:%.*]] = alloca ptr,
+// CHECK-NEXT: [[Z:%.*]] = alloca ptr,
 // CHECK-NEXT: store
-// CHECK-NEXT: store i8* null, i8** [[Z]]
-// CHECK-NEXT: [[T0:%.*]] = load i8*, i8** [[Y]]
-// CHECK-NEXT: call i8* @llvm.objc.storeWeak(i8** [[Z]], i8* [[T0]])
-// CHECK-NEXT: call void @llvm.objc.destroyWeak(i8** [[Z]])
+// CHECK-NEXT: store ptr null, ptr [[Z]]
+// CHECK-NEXT: [[T0:%.*]] = load ptr, ptr [[Y]]
+// CHECK-NEXT: call ptr @llvm.objc.storeWeak(ptr [[Z]], ptr [[T0]])
+// CHECK-NEXT: call void @llvm.objc.destroyWeak(ptr [[Z]])
 // CHECK-NEXT: ret void
 
 void test4(__weak id *p) {
   id y = *p;
 }
 // CHECK-LABEL: define{{.*}} void @test4
-// CHECK:      [[P:%.*]] = alloca i8**,
-// CHECK-NEXT: [[Y:%.*]] = alloca i8*,
+// CHECK:      [[P:%.*]] = alloca ptr,
+// CHECK-NEXT: [[Y:%.*]] = alloca ptr,
 // CHECK-NEXT: store
-// CHECK-NEXT: [[T0:%.*]] = load i8**, i8*** [[P]]
-// CHECK-NEXT: [[T1:%.*]] = call i8* @llvm.objc.loadWeak(i8** [[T0]])
-// CHECK-NEXT: store i8* [[T1]], i8** [[Y]]
+// CHECK-NEXT: [[T0:%.*]] = load ptr, ptr [[P]]
+// CHECK-NEXT: [[T1:%.*]] = call ptr @llvm.objc.loadWeak(ptr [[T0]])
+// CHECK-NEXT: store ptr [[T1]], ptr [[Y]]
 // CHECK-NEXT: ret void
 
 void test5(__weak id *p) {
   id y = [*p retain];
 }
 // CHECK-LABEL: define{{.*}} void @test5
-// CHECK:      [[P:%.*]] = alloca i8**,
-// CHECK-NEXT: [[Y:%.*]] = alloca i8*,
+// CHECK:      [[P:%.*]] = alloca ptr,
+// CHECK-NEXT: [[Y:%.*]] = alloca ptr,
 // CHECK-NEXT: store
-// CHECK-NEXT: [[T0:%.*]] = load i8**, i8*** [[P]]
-// CHECK-NEXT: [[T1:%.*]] = call i8* @llvm.objc.loadWeakRetained(i8** [[T0]])
-// CHECK-NEXT: store i8* [[T1]], i8** [[Y]]
+// CHECK-NEXT: [[T0:%.*]] = load ptr, ptr [[P]]
+// CHECK-NEXT: [[T1:%.*]] = call ptr @llvm.objc.loadWeakRetained(ptr [[T0]])
+// CHECK-NEXT: store ptr [[T1]], ptr [[Y]]
 // CHECK-NEXT: ret void
 
 void test6(__weak Foo **p) {
   Foo *y = [*p retain];
 }
 // CHECK-LABEL: define{{.*}} void @test6
-// CHECK:      [[P:%.*]] = alloca [[FOO:%.*]]**,
-// CHECK-NEXT: [[Y:%.*]] = alloca [[FOO]]*,
+// CHECK:      [[P:%.*]] = alloca ptr,
+// CHECK-NEXT: [[Y:%.*]] = alloca ptr,
 // CHECK-NEXT: store
-// CHECK-NEXT: [[T0:%.*]] = load [[FOO]]**, [[FOO]]*** [[P]]
-// CHECK-NEXT: [[T1:%.*]] = bitcast [[FOO]]** [[T0]] to i8**
-// CHECK-NEXT: [[T2:%.*]] = call i8* @llvm.objc.loadWeakRetained(i8** [[T1]])
-// CHECK-NEXT: [[T3:%.*]] = bitcast i8* [[T2]] to [[FOO]]*
-// CHECK-NEXT: store [[FOO]]* [[T3]], [[FOO]]** [[Y]]
+// CHECK-NEXT: [[T0:%.*]] = load ptr, ptr [[P]]
+// CHECK-NEXT: [[T2:%.*]] = call ptr @llvm.objc.loadWeakRetained(ptr [[T0]])
+// CHECK-NEXT: store ptr [[T2]], ptr [[Y]]
 // CHECK-NEXT: ret void
 
 extern id get_object(void);
@@ -129,12 +127,9 @@ void test7(void) {
   use_block(^{ [p run ]; });
 }
 // CHECK-LABEL: define{{.*}} void @test7
-// CHECK:       [[P:%.*]] = alloca [[FOO]]*,
-// CHECK:       [[T0:%.*]] = call i8* @get_object()
-// CHECK-NEXT:  [[T1:%.*]] = bitcast i8* [[T0]] to [[FOO]]*
-// CHECK-NEXT:  [[T2:%.*]] = bitcast [[FOO]]** [[P]] to i8**
-// CHECK-NEXT:  [[T3:%.*]] = bitcast [[FOO]]* [[T1]] to i8*
-// CHECK-NEXT:  call i8* @llvm.objc.initWeak(i8** [[T2]], i8* [[T3]])
+// CHECK:       [[P:%.*]] = alloca ptr,
+// CHECK:       [[T0:%.*]] = call ptr @get_object()
+// CHECK-NEXT:  call ptr @llvm.objc.initWeak(ptr [[P]], ptr [[T0]])
 // CHECK:       call void @llvm.objc.copyWeak
 // CHECK:       call void @use_block
 // CHECK:       call void @llvm.objc.destroyWeak
@@ -150,7 +145,7 @@ void test8(void) {
   use_block(^{ [p run ]; });
 }
 // CHECK-LABEL: define{{.*}} void @test8
-// CHECK:       call i8* @llvm.objc.initWeak
+// CHECK:       call ptr @llvm.objc.initWeak
 // CHECK-NOT:   call void @llvm.objc.copyWeak
 // CHECK:       call void @use_block
 // CHECK:       call void @llvm.objc.destroyWeak

diff  --git a/clang/test/CodeGenObjC/newproperty-nested-synthesis-1.m b/clang/test/CodeGenObjC/newproperty-nested-synthesis-1.m
index d97a2f3670d0e..aa0c8c923971e 100644
--- a/clang/test/CodeGenObjC/newproperty-nested-synthesis-1.m
+++ b/clang/test/CodeGenObjC/newproperty-nested-synthesis-1.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -o %t %s
+// RUN: %clang_cc1 -emit-llvm -o %t %s
 // REQUIRES: LP64
 
 @interface Object

diff  --git a/clang/test/CodeGenObjC/next-objc-dispatch.m b/clang/test/CodeGenObjC/next-objc-dispatch.m
index 835509a55e5ba..e3a304d909505 100644
--- a/clang/test/CodeGenObjC/next-objc-dispatch.m
+++ b/clang/test/CodeGenObjC/next-objc-dispatch.m
@@ -1,16 +1,16 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin9 -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -o - %s \
+// RUN: %clang_cc1 -triple x86_64-apple-darwin9 -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -o - %s \
 // RUN:   -fobjc-dispatch-method=legacy | \
 // RUN:   FileCheck -check-prefix CHECK-FRAGILE_LEGACY %s
 //
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin9 -emit-llvm -o - %s    \
+// RUN: %clang_cc1 -triple x86_64-apple-darwin9 -emit-llvm -o - %s    \
 // RUN:   -fobjc-dispatch-method=legacy | \
 // RUN:   FileCheck -check-prefix CHECK-NONFRAGILE_LEGACY %s
 //
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin9 -emit-llvm -o - %s    \
+// RUN: %clang_cc1 -triple x86_64-apple-darwin9 -emit-llvm -o - %s    \
 // RUN:   -fobjc-dispatch-method=non-legacy | \
 // RUN:   FileCheck -check-prefix CHECK-NONFRAGILE_NONLEGACY %s
 //
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin9 -emit-llvm -o - %s    \
+// RUN: %clang_cc1 -triple x86_64-apple-darwin9 -emit-llvm -o - %s    \
 // RUN:   -fobjc-dispatch-method=mixed | \
 // RUN:   FileCheck -check-prefix CHECK-NONFRAGILE_MIXED %s
 //
@@ -27,25 +27,25 @@
 // they use some 
diff erent API calls (objc_msgSendSuper vs objc_msgSendSuper2).
 
 // CHECK-FRAGILE_LEGACY: ModuleID
-// CHECK-FRAGILE_LEGACY-NOT: declare i8* @objc_msgSendSuper2_fixup(
-// CHECK-FRAGILE_LEGACY-NOT: declare i8* @objc_msgSend_fixup(
-// CHECK-FRAGILE_LEGACY: declare i8* @objc_msgSendSuper(
-// CHECK-FRAGILE_LEGACY: declare i8* @objc_msgSend(
+// CHECK-FRAGILE_LEGACY-NOT: declare ptr @objc_msgSendSuper2_fixup(
+// CHECK-FRAGILE_LEGACY-NOT: declare ptr @objc_msgSend_fixup(
+// CHECK-FRAGILE_LEGACY: declare ptr @objc_msgSendSuper(
+// CHECK-FRAGILE_LEGACY: declare ptr @objc_msgSend(
 
 // CHECK-NONFRAGILE_LEGACY: ModuleID
-// CHECK-NONFRAGILE_LEGACY-NOT: declare i8* @objc_msgSendSuper2_fixup(
-// CHECK-NONFRAGILE_LEGACY-NOT: declare i8* @objc_msgSend_fixup(
-// CHECK-NONFRAGILE_LEGACY: declare i8* @objc_msgSendSuper2(
-// CHECK-NONFRAGILE_LEGACY: declare i8* @objc_msgSend(
+// CHECK-NONFRAGILE_LEGACY-NOT: declare ptr @objc_msgSendSuper2_fixup(
+// CHECK-NONFRAGILE_LEGACY-NOT: declare ptr @objc_msgSend_fixup(
+// CHECK-NONFRAGILE_LEGACY: declare ptr @objc_msgSendSuper2(
+// CHECK-NONFRAGILE_LEGACY: declare ptr @objc_msgSend(
 
 // CHECK-NONFRAGILE_NONLEGACY: ModuleID
-// CHECK-NONFRAGILE_NONLEGACY: declare i8* @objc_msgSendSuper2_fixup(
-// CHECK-NONFRAGILE_NONLEGACY: declare i8* @objc_msgSend_fixup(
+// CHECK-NONFRAGILE_NONLEGACY: declare ptr @objc_msgSendSuper2_fixup(
+// CHECK-NONFRAGILE_NONLEGACY: declare ptr @objc_msgSend_fixup(
 
-// CHECK-NONFRAGILE_MIXED: declare i8* @objc_msgSendSuper2_fixup(
-// CHECK-NONFRAGILE_MIXED: declare i8* @objc_msgSendSuper2(
-// CHECK-NONFRAGILE_MIXED: declare i8* @objc_msgSend_fixup(
-// CHECK-NONFRAGILE_MIXED: declare i8* @objc_msgSend(
+// CHECK-NONFRAGILE_MIXED: declare ptr @objc_msgSendSuper2_fixup(
+// CHECK-NONFRAGILE_MIXED: declare ptr @objc_msgSendSuper2(
+// CHECK-NONFRAGILE_MIXED: declare ptr @objc_msgSend_fixup(
+// CHECK-NONFRAGILE_MIXED: declare ptr @objc_msgSend(
 
 @interface NSObject
 + (id)alloc;

diff  --git a/clang/test/CodeGenObjC/non-runtime-protocol.m b/clang/test/CodeGenObjC/non-runtime-protocol.m
index a6d0d5f1b0caa..889bf0835270c 100644
--- a/clang/test/CodeGenObjC/non-runtime-protocol.m
+++ b/clang/test/CodeGenObjC/non-runtime-protocol.m
@@ -1,29 +1,29 @@
-// RUN: not %clang_cc1 -no-opaque-pointers -emit-llvm -fobjc-arc -triple x86_64-apple-darwin10 %s -DPROTOEXPR -o - 2>&1 \
+// RUN: not %clang_cc1 -emit-llvm -fobjc-arc -triple x86_64-apple-darwin10 %s -DPROTOEXPR -o - 2>&1 \
 // RUN:     | FileCheck -check-prefix=PROTOEXPR %s
 
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -fobjc-arc -triple x86_64-apple-darwin10 %s -DREDUNDANCY -o - \
+// RUN: %clang_cc1 -emit-llvm -fobjc-arc -triple x86_64-apple-darwin10 %s -DREDUNDANCY -o - \
 // RUN:     | FileCheck -check-prefix=REDUNDANCY1 %s
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -fobjc-arc -triple x86_64-apple-darwin10 %s -DREDUNDANCY -o - \
+// RUN: %clang_cc1 -emit-llvm -fobjc-arc -triple x86_64-apple-darwin10 %s -DREDUNDANCY -o - \
 // RUN:     | FileCheck -check-prefix=REDUNDANCY2 %s
 
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -fobjc-arc -triple x86_64-apple-darwin10 %s -DBASE -o - \
+// RUN: %clang_cc1 -emit-llvm -fobjc-arc -triple x86_64-apple-darwin10 %s -DBASE -o - \
 // RUN:     | FileCheck -check-prefix=NONFRAGILE %s
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -fobjc-arc -triple x86_64-apple-darwin10 %s -DINHERITANCE -o - \
+// RUN: %clang_cc1 -emit-llvm -fobjc-arc -triple x86_64-apple-darwin10 %s -DINHERITANCE -o - \
 // RUN:     | FileCheck -check-prefix=INHERITANCE %s
 
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -triple x86_64-apple-darwin -fobjc-runtime=macosx-fragile-10.5 %s -DBASE -o - \
+// RUN: %clang_cc1 -emit-llvm -triple x86_64-apple-darwin -fobjc-runtime=macosx-fragile-10.5 %s -DBASE -o - \
 // RUN:     | FileCheck -check-prefix=FRAGILE %s
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -triple x86_64-apple-darwin -fobjc-runtime=macosx-fragile-10.5 %s -DINHERITANCE -o - \
+// RUN: %clang_cc1 -emit-llvm -triple x86_64-apple-darwin -fobjc-runtime=macosx-fragile-10.5 %s -DINHERITANCE -o - \
 // RUN:     | FileCheck -check-prefix=FRAGILEINHERITANCE %s
 
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -triple x86_64-linux-gnu -fobjc-runtime=gnustep %s -DBASE -o - \
+// RUN: %clang_cc1 -emit-llvm -triple x86_64-linux-gnu -fobjc-runtime=gnustep %s -DBASE -o - \
 // RUN:     | FileCheck -check-prefix=GNU %s
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -triple x86_64-linux-gnu -fobjc-runtime=gnustep %s -DINHERITANCE -o - \
+// RUN: %clang_cc1 -emit-llvm -triple x86_64-linux-gnu -fobjc-runtime=gnustep %s -DINHERITANCE -o - \
 // RUN:     | FileCheck -check-prefix=GNUINHERITANCE %s
 //
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -triple x86_64-linux-gnu -fobjc-runtime=gnustep-2 %s -DBASE -o - \
+// RUN: %clang_cc1 -emit-llvm -triple x86_64-linux-gnu -fobjc-runtime=gnustep-2 %s -DBASE -o - \
 // RUN:     | FileCheck -check-prefix=GNU2 %s
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -triple x86_64-linux-gnu -fobjc-runtime=gnustep-2 %s -DINHERITANCE -o - \
+// RUN: %clang_cc1 -emit-llvm -triple x86_64-linux-gnu -fobjc-runtime=gnustep-2 %s -DINHERITANCE -o - \
 // RUN:     | FileCheck -check-prefix=GNU2INHERITANCE %s
 
 __attribute__((objc_root_class))
@@ -71,8 +71,8 @@ @protocol NonRuntimeProtocol
 - (void)doThing;
 + (void)doClassThing;
 @end
-// NONFRAGILE: @"_OBJC_METACLASS_RO_$_NonRuntimeImplementer" {{.*}} %struct._objc_protocol_list* null
-// NONFRAGILE: @"_OBJC_CLASS_RO_$_NonRuntimeImplementer" {{.*}} %struct._objc_protocol_list* null
+// NONFRAGILE: @"_OBJC_METACLASS_RO_$_NonRuntimeImplementer" {{.*}} ptr null
+// NONFRAGILE: @"_OBJC_CLASS_RO_$_NonRuntimeImplementer" {{.*}} ptr null
 @interface NonRuntimeImplementer : Root <NonRuntimeProtocol>
 - (void)doThing;
 + (void)doClassThing;

diff  --git a/clang/test/CodeGenObjC/nonlazy-msgSend.m b/clang/test/CodeGenObjC/nonlazy-msgSend.m
index 06db9a31f36c9..eb1dd2aac0979 100644
--- a/clang/test/CodeGenObjC/nonlazy-msgSend.m
+++ b/clang/test/CodeGenObjC/nonlazy-msgSend.m
@@ -1,6 +1,6 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin9 -fobjc-runtime=macosx-fragile-10.5 -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin9 -fobjc-runtime=macosx-fragile-10.5 -emit-llvm %s -o - | FileCheck %s
 
-// CHECK: declare i8* @objc_msgSend(i8*, i8*, ...) [[NLB:#[0-9]+]]
+// CHECK: declare ptr @objc_msgSend(ptr, ptr, ...) [[NLB:#[0-9]+]]
 void f0(id x) {
   [x foo];
 }

diff  --git a/clang/test/CodeGenObjC/nontrivial-c-struct-within-struct-name.m b/clang/test/CodeGenObjC/nontrivial-c-struct-within-struct-name.m
index 4c1e210765d7e..c2a8ab88f2fda 100644
--- a/clang/test/CodeGenObjC/nontrivial-c-struct-within-struct-name.m
+++ b/clang/test/CodeGenObjC/nontrivial-c-struct-within-struct-name.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple arm64-apple-ios11 -fobjc-arc -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple arm64-apple-ios11 -fobjc-arc -emit-llvm -o - %s | FileCheck %s
 
 @class I;
 
@@ -22,23 +22,23 @@ void f(void) {
   Baz baz = {bar};
 }
 
-// CHECK: define linkonce_odr hidden void @__destructor_8_S_S_s0(i8** noundef %[[DST:.*]])
-// CHECK: %[[DST_ADDR:.*]] = alloca i8**, align 8
-// CHECK: store i8** %[[DST]], i8*** %[[DST_ADDR]], align 8
-// CHECK: %[[V0:.*]] = load i8**, i8*** %[[DST_ADDR]], align 8
-// CHECK: call void @__destructor_8_S_s0(i8** %[[V0]])
+// CHECK: define linkonce_odr hidden void @__destructor_8_S_S_s0(ptr noundef %[[DST:.*]])
+// CHECK: %[[DST_ADDR:.*]] = alloca ptr, align 8
+// CHECK: store ptr %[[DST]], ptr %[[DST_ADDR]], align 8
+// CHECK: %[[V0:.*]] = load ptr, ptr %[[DST_ADDR]], align 8
+// CHECK: call void @__destructor_8_S_s0(ptr %[[V0]])
 // CHECK: ret void
 
-// CHECK: define linkonce_odr hidden void @__destructor_8_S_s0(i8** noundef %[[DST:.*]])
-// CHECK: %[[DST_ADDR:.*]] = alloca i8**, align 8
-// CHECK: store i8** %[[DST]], i8*** %[[DST_ADDR]], align 8
-// CHECK: %[[V0:.*]] = load i8**, i8*** %[[DST_ADDR]], align 8
-// CHECK: call void @__destructor_8_s0(i8** %[[V0]])
+// CHECK: define linkonce_odr hidden void @__destructor_8_S_s0(ptr noundef %[[DST:.*]])
+// CHECK: %[[DST_ADDR:.*]] = alloca ptr, align 8
+// CHECK: store ptr %[[DST]], ptr %[[DST_ADDR]], align 8
+// CHECK: %[[V0:.*]] = load ptr, ptr %[[DST_ADDR]], align 8
+// CHECK: call void @__destructor_8_s0(ptr %[[V0]])
 // CHECK: ret void
 
-// CHECK: define linkonce_odr hidden void @__destructor_8_s0(i8** noundef %dst)
-// CHECK: %[[DST_ADDR:.*]] = alloca i8**, align 8
-// CHECK: store i8** %[[DST]], i8*** %[[DST_ADDR]], align 8
-// CHECK: %[[V0:.*]] = load i8**, i8*** %[[DST_ADDR]], align 8
-// CHECK: call void @llvm.objc.storeStrong(i8** %[[V0]], i8* null)
+// CHECK: define linkonce_odr hidden void @__destructor_8_s0(ptr noundef %dst)
+// CHECK: %[[DST_ADDR:.*]] = alloca ptr, align 8
+// CHECK: store ptr %[[DST]], ptr %[[DST_ADDR]], align 8
+// CHECK: %[[V0:.*]] = load ptr, ptr %[[DST_ADDR]], align 8
+// CHECK: call void @llvm.objc.storeStrong(ptr %[[V0]], ptr null)
 // CHECK: ret void

diff  --git a/clang/test/CodeGenObjC/nontrivial-struct-param-init.m b/clang/test/CodeGenObjC/nontrivial-struct-param-init.m
index dc4b27e9d0095..79c97d9818543 100644
--- a/clang/test/CodeGenObjC/nontrivial-struct-param-init.m
+++ b/clang/test/CodeGenObjC/nontrivial-struct-param-init.m
@@ -1,17 +1,16 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple i386-apple-watchos6.0-simulator -emit-llvm -fblocks -fobjc-arc -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple i386-apple-watchos6.0-simulator -emit-llvm -fblocks -fobjc-arc -o - %s | FileCheck %s
 
-// CHECK: %[[STRUCT_S:.*]] = type { i8* }
+// CHECK: %[[STRUCT_S:.*]] = type { ptr }
 
 typedef struct {
   id x;
 } S;
 
-// CHECK: define{{.*}} void @test0(i8* %[[A_0:.*]])
+// CHECK: define{{.*}} void @test0(ptr %[[A_0:.*]])
 // CHECK: %[[A:.*]] = alloca %[[STRUCT_S]], align 4
-// CHECK: %[[X:.*]] = getelementptr inbounds %[[STRUCT_S]], %[[STRUCT_S]]* %[[A]], i32 0, i32 0
-// CHECK: store i8* %[[A_0]], i8** %[[X]], align 4
-// CHECK: %[[V0:.*]] = bitcast %[[STRUCT_S]]* %[[A]] to i8**
-// CHECK: call void @__destructor_4_s0(i8** %[[V0]]) #2
+// CHECK: %[[X:.*]] = getelementptr inbounds %[[STRUCT_S]], ptr %[[A]], i32 0, i32 0
+// CHECK: store ptr %[[A_0]], ptr %[[X]], align 4
+// CHECK: call void @__destructor_4_s0(ptr %[[A]]) #2
 
 void test0(S a) {
 }

diff  --git a/clang/test/CodeGenObjC/null-objc-empty-vtable.m b/clang/test/CodeGenObjC/null-objc-empty-vtable.m
index 9348dbe91a476..de35778fc4be1 100644
--- a/clang/test/CodeGenObjC/null-objc-empty-vtable.m
+++ b/clang/test/CodeGenObjC/null-objc-empty-vtable.m
@@ -1,10 +1,10 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-macosx10.9 -emit-llvm -o - %s | FileCheck -check-prefix CHECK-OSX %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple thumbv7-apple-ios3.0.0  -emit-llvm -o - %s | FileCheck -check-prefix CHECK-IOS %s
+// RUN: %clang_cc1 -triple x86_64-apple-macosx10.9 -emit-llvm -o - %s | FileCheck -check-prefix CHECK-OSX %s
+// RUN: %clang_cc1 -triple thumbv7-apple-ios3.0.0  -emit-llvm -o - %s | FileCheck -check-prefix CHECK-IOS %s
 // rdar://14802916
 
 @interface I
 @end
 
 @implementation I @end
-// CHECK-OSX: %struct._class_t* null, %struct._objc_cache* @_objc_empty_cache, i8* (i8*, i8*)** null
-// CHECK-IOS: %struct._class_t* null, %struct._objc_cache* @_objc_empty_cache, i8* (i8*, i8*)** null
+// CHECK-OSX: ptr null, ptr @_objc_empty_cache, ptr null
+// CHECK-IOS: ptr null, ptr @_objc_empty_cache, ptr null

diff  --git a/clang/test/CodeGenObjC/objc-alloc-init.m b/clang/test/CodeGenObjC/objc-alloc-init.m
index 2fe8a8860ef50..0e8260ad858bd 100644
--- a/clang/test/CodeGenObjC/objc-alloc-init.m
+++ b/clang/test/CodeGenObjC/objc-alloc-init.m
@@ -1,7 +1,7 @@
-// RUN: %clang_cc1 -no-opaque-pointers %s -fobjc-exceptions -fexceptions -fobjc-runtime=macosx-10.14.4 -emit-llvm -O0 -o - | FileCheck %s --check-prefix=OPTIMIZED --check-prefix=EITHER
-// RUN: %clang_cc1 -no-opaque-pointers %s -fobjc-exceptions -fexceptions -fobjc-runtime=macosx-10.14.3 -emit-llvm -O0 -o - | FileCheck %s --check-prefix=NOT_OPTIMIZED --check-prefix=EITHER
-// RUN: %clang_cc1 -no-opaque-pointers %s -fobjc-exceptions -fexceptions -fobjc-runtime=ios-12.2 -emit-llvm -O0 -o - | FileCheck %s --check-prefix=OPTIMIZED --check-prefix=EITHER
-// RUN: %clang_cc1 -no-opaque-pointers %s -fobjc-exceptions -fexceptions -fobjc-runtime=ios-12.1 -emit-llvm -O0 -o - | FileCheck %s --check-prefix=NOT_OPTIMIZED --check-prefix=EITHER
+// RUN: %clang_cc1 %s -fobjc-exceptions -fexceptions -fobjc-runtime=macosx-10.14.4 -emit-llvm -O0 -o - | FileCheck %s --check-prefix=OPTIMIZED --check-prefix=EITHER
+// RUN: %clang_cc1 %s -fobjc-exceptions -fexceptions -fobjc-runtime=macosx-10.14.3 -emit-llvm -O0 -o - | FileCheck %s --check-prefix=NOT_OPTIMIZED --check-prefix=EITHER
+// RUN: %clang_cc1 %s -fobjc-exceptions -fexceptions -fobjc-runtime=ios-12.2 -emit-llvm -O0 -o - | FileCheck %s --check-prefix=OPTIMIZED --check-prefix=EITHER
+// RUN: %clang_cc1 %s -fobjc-exceptions -fexceptions -fobjc-runtime=ios-12.1 -emit-llvm -O0 -o - | FileCheck %s --check-prefix=NOT_OPTIMIZED --check-prefix=EITHER
 
 @interface X
 +(X *)alloc;
@@ -10,15 +10,15 @@ -(X *)init;
 
 void f(void) {
   [[X alloc] init];
-  // OPTIMIZED: call i8* @objc_alloc_init(
-  // NOT_OPTIMIZED: call i8* @objc_alloc(
+  // OPTIMIZED: call ptr @objc_alloc_init(
+  // NOT_OPTIMIZED: call ptr @objc_alloc(
 
   @try {
     [[X alloc] init];
   } @catch (X *x) {
   }
-  // OPTIMIZED: invoke i8* @objc_alloc_init(
-  // NOT_OPTIMIZED: invoke i8* @objc_alloc(
+  // OPTIMIZED: invoke ptr @objc_alloc_init(
+  // NOT_OPTIMIZED: invoke ptr @objc_alloc(
 }
 
 @interface Y : X
@@ -33,16 +33,16 @@ +(Class)class {
 }
 +(void)meth {
   [[self alloc] init];
-  // OPTIMIZED: call i8* @objc_alloc_init(
-  // NOT_OPTIMIZED: call i8* @objc_alloc(
+  // OPTIMIZED: call ptr @objc_alloc_init(
+  // NOT_OPTIMIZED: call ptr @objc_alloc(
 }
 + (void)meth2 {
   [[[self class] alloc] init];
-  // OPTIMIZED: call i8* @objc_alloc_init(
-  // NOT_OPTIMIZED: call i8* @objc_alloc(
+  // OPTIMIZED: call ptr @objc_alloc_init(
+  // NOT_OPTIMIZED: call ptr @objc_alloc(
 }
 -(void)instanceMeth {
-  // EITHER-NOT: call i8* @objc_alloc
+  // EITHER-NOT: call ptr @objc_alloc
   // EITHER: call {{.*}} @objc_msgSend
   // EITHER: call {{.*}} @objc_msgSend
   [[(id)self alloc] init];

diff  --git a/clang/test/CodeGenObjC/objc-asm-attribute-test.m b/clang/test/CodeGenObjC/objc-asm-attribute-test.m
index 875de5b5014eb..fbb4b4d915d6c 100644
--- a/clang/test/CodeGenObjC/objc-asm-attribute-test.m
+++ b/clang/test/CodeGenObjC/objc-asm-attribute-test.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -triple x86_64-apple-darwin %s -o - | FileCheck %s
+// RUN: %clang_cc1 -emit-llvm -triple x86_64-apple-darwin %s -o - | FileCheck %s
 // rdar://16462586
 
 __attribute__((objc_runtime_name("MySecretNamespace.Protocol")))
@@ -64,5 +64,5 @@ id Test16877359(void) {
 // CHECK: private unnamed_addr constant [50 x i8] c"T@\22<MySecretNamespace.Protocol3>\22,&,V_idProtoProp\00"
 
 // CHECK: @"OBJC_CLASS_$_foo" = external global %struct._class_t
-// CHECK: define internal i8* @"\01-[Message MyMethod]"
-// CHECK: [[IVAR:%.*]] = load i64, i64* @"OBJC_IVAR_$_MySecretNamespace.Message.MyIVAR"
+// CHECK: define internal ptr @"\01-[Message MyMethod]"
+// CHECK: [[IVAR:%.*]] = load i64, ptr @"OBJC_IVAR_$_MySecretNamespace.Message.MyIVAR"

diff  --git a/clang/test/CodeGenObjC/objc-dispatch-null-check.m b/clang/test/CodeGenObjC/objc-dispatch-null-check.m
index 08e9c2169145d..d7ee2441cc252 100644
--- a/clang/test/CodeGenObjC/objc-dispatch-null-check.m
+++ b/clang/test/CodeGenObjC/objc-dispatch-null-check.m
@@ -1,6 +1,6 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin9 -fobjc-dispatch-method=non-legacy -fobjc-arc -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin9 -fobjc-dispatch-method=non-legacy -fobjc-arc -emit-llvm -o - %s | FileCheck %s
 
-// CHECK: %[[STRUCT_STRONG:.*]] = type { i8* }
+// CHECK: %[[STRUCT_STRONG:.*]] = type { ptr }
 
 typedef struct {
   id x;
@@ -14,19 +14,17 @@ - (void)passStrong:(Strong)a;
 
 // CHECK-LABEL: define{{.*}} void @test0(
 // CHECK: %[[AGG_TMP:.*]] = alloca %[[STRUCT_STRONG]], align 8
-// CHECK: %[[CALL:.*]] = call i8* @getStrong()
-// CHECK-NEXT: %[[COERCE_DIVE:.*]] = getelementptr inbounds %[[STRUCT_STRONG]], %[[STRUCT_STRONG]]* %[[AGG_TMP]], i32 0, i32 0
-// CHECK-NEXT: store i8* %[[CALL]], i8** %[[COERCE_DIVE]], align 8
-
-// CHECK: %[[MSGSEND_FN:.*]] = load i8*, i8**
-// CHECK: %[[V5:.*]] = bitcast i8* %[[MSGSEND_FN]] to void (i8*, i8*, i8*)*
-// CHECK: %[[COERCE_DIVE1:.*]] = getelementptr inbounds %[[STRUCT_STRONG]], %[[STRUCT_STRONG]]* %[[AGG_TMP]], i32 0, i32 0
-// CHECK: %[[V6:.*]] = load i8*, i8** %[[COERCE_DIVE1]], align 8
-// CHECK: call void %[[V5]]({{.*}}, i8* %[[V6]])
+// CHECK: %[[CALL:.*]] = call ptr @getStrong()
+// CHECK-NEXT: %[[COERCE_DIVE:.*]] = getelementptr inbounds %[[STRUCT_STRONG]], ptr %[[AGG_TMP]], i32 0, i32 0
+// CHECK-NEXT: store ptr %[[CALL]], ptr %[[COERCE_DIVE]], align 8
+
+// CHECK: %[[MSGSEND_FN:.*]] = load ptr, ptr
+// CHECK: %[[COERCE_DIVE1:.*]] = getelementptr inbounds %[[STRUCT_STRONG]], ptr %[[AGG_TMP]], i32 0, i32 0
+// CHECK: %[[V6:.*]] = load ptr, ptr %[[COERCE_DIVE1]], align 8
+// CHECK: call void %[[MSGSEND_FN]]({{.*}}, ptr %[[V6]])
 // CHECK: br
 
-// CHECK: %[[V7:.*]] = bitcast %[[STRUCT_STRONG]]* %[[AGG_TMP]] to i8**
-// CHECK: call void @__destructor_8_s0(i8** %[[V7]])
+// CHECK: call void @__destructor_8_s0(ptr %[[AGG_TMP]])
 // CHECK: br
 
 void test0(I0 *a) {

diff  --git a/clang/test/CodeGenObjC/objc-gc-aggr-assign.m b/clang/test/CodeGenObjC/objc-gc-aggr-assign.m
index e82fa33f8b056..3b46588db0be7 100644
--- a/clang/test/CodeGenObjC/objc-gc-aggr-assign.m
+++ b/clang/test/CodeGenObjC/objc-gc-aggr-assign.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 -fobjc-gc -emit-llvm -o - %s | FileCheck -check-prefix CHECK-C %s
-// RUN: %clang_cc1 -no-opaque-pointers -x objective-c++ -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 -fobjc-gc -emit-llvm -o - %s | FileCheck -check-prefix CHECK-CP %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 -fobjc-gc -emit-llvm -o - %s | FileCheck -check-prefix CHECK-C %s
+// RUN: %clang_cc1 -x objective-c++ -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 -fobjc-gc -emit-llvm -o - %s | FileCheck -check-prefix CHECK-CP %s
 
 static int count;
 
@@ -52,11 +52,11 @@ void foo(Derived* src, Derived* dest) {
 }
 #endif
 
-// CHECK-C: call i8* @objc_memmove_collectable
-// CHECK-C: call i8* @objc_memmove_collectable
-// CHECK-C: call i8* @objc_memmove_collectable
+// CHECK-C: call ptr @objc_memmove_collectable
+// CHECK-C: call ptr @objc_memmove_collectable
+// CHECK-C: call ptr @objc_memmove_collectable
 
-// CHECK-CP: call i8* @objc_memmove_collectable
-// CHECK-CP: call i8* @objc_memmove_collectable
-// CHECK-CP: call i8* @objc_memmove_collectable
-// CHECK-CP: call i8* @objc_memmove_collectable
+// CHECK-CP: call ptr @objc_memmove_collectable
+// CHECK-CP: call ptr @objc_memmove_collectable
+// CHECK-CP: call ptr @objc_memmove_collectable
+// CHECK-CP: call ptr @objc_memmove_collectable

diff  --git a/clang/test/CodeGenObjC/objc-literal-debugger-test.m b/clang/test/CodeGenObjC/objc-literal-debugger-test.m
index 0522805391412..62aa8574091b4 100644
--- a/clang/test/CodeGenObjC/objc-literal-debugger-test.m
+++ b/clang/test/CodeGenObjC/objc-literal-debugger-test.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -fdebugger-objc-literal -emit-llvm -o - %s | FileCheck %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -fdebugger-objc-literal -emit-llvm -o - %s -DINCLUDE_INTERFACES=1 | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -fdebugger-objc-literal -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -fdebugger-objc-literal -emit-llvm -o - %s -DINCLUDE_INTERFACES=1 | FileCheck %s
 
 // We need two 
diff erent RUN lines here because the first time a class/method is absent,
 // it will be added for -fdebugger-objc-literal.
@@ -50,6 +50,6 @@ int main(void) {
 #endif
 }
 
-// CHECK: declare i8* @objc_msgSend(i8*, i8*, ...) [[NLB:#[0-9]+]]
+// CHECK: declare ptr @objc_msgSend(ptr, ptr, ...) [[NLB:#[0-9]+]]
 
 // CHECK: attributes [[NLB]] = { nonlazybind }

diff  --git a/clang/test/CodeGenObjC/objc-non-trivial-struct-nrvo.m b/clang/test/CodeGenObjC/objc-non-trivial-struct-nrvo.m
index da320b2b3bfa3..effba8c958441 100644
--- a/clang/test/CodeGenObjC/objc-non-trivial-struct-nrvo.m
+++ b/clang/test/CodeGenObjC/objc-non-trivial-struct-nrvo.m
@@ -1,9 +1,9 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -fobjc-arc -fobjc-runtime-has-weak -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -fobjc-arc -fobjc-runtime-has-weak -o - %s | FileCheck %s
 
 // CHECK: %[[STRUCT_TRIVIAL:.*]] = type { i32 }
 // CHECK: %[[STRUCT_TRIVIALBIG:.*]] = type { [64 x i32] }
-// CHECK: %[[STRUCT_STRONG:.*]] = type { i8* }
-// CHECK: %[[STRUCT_WEAK:.*]] = type { i8* }
+// CHECK: %[[STRUCT_STRONG:.*]] = type { ptr }
+// CHECK: %[[STRUCT_WEAK:.*]] = type { ptr }
 
 typedef struct {
   int x;
@@ -23,7 +23,7 @@
 
 // CHECK: define{{.*}} i32 @testTrivial()
 // CHECK: %[[RETVAL:.*]] = alloca %[[STRUCT_TRIVIAL]], align 4
-// CHECK-NEXT: call void @func0(%[[STRUCT_TRIVIAL]]* noundef %[[RETVAL]])
+// CHECK-NEXT: call void @func0(ptr noundef %[[RETVAL]])
 // CHECK-NOT: memcpy
 // CHECK: ret i32 %
 
@@ -37,8 +37,8 @@ Trivial testTrivial(void) {
 
 void func1(TrivialBig *);
 
-// CHECK: define{{.*}} void @testTrivialBig(%[[STRUCT_TRIVIALBIG]]* noalias sret(%[[STRUCT_TRIVIALBIG]]) align 4 %[[AGG_RESULT:.*]])
-// CHECK: call void @func1(%[[STRUCT_TRIVIALBIG]]* noundef %[[AGG_RESULT]])
+// CHECK: define{{.*}} void @testTrivialBig(ptr noalias sret(%[[STRUCT_TRIVIALBIG]]) align 4 %[[AGG_RESULT:.*]])
+// CHECK: call void @func1(ptr noundef %[[AGG_RESULT]])
 // CHECK-NEXT: ret void
 
 TrivialBig testTrivialBig(void) {
@@ -47,38 +47,34 @@ TrivialBig testTrivialBig(void) {
   return a;
 }
 
-// CHECK: define{{.*}} i8* @testStrong()
+// CHECK: define{{.*}} ptr @testStrong()
 // CHECK: %[[RETVAL:.*]] = alloca %[[STRUCT_STRONG]], align 8
 // CHECK: %[[NRVO:.*]] = alloca i1, align 1
-// CHECK: %[[V0:.*]] = bitcast %[[STRUCT_STRONG]]* %[[RETVAL]] to i8**
-// CHECK: call void @__default_constructor_8_s0(i8** %[[V0]])
-// CHECK: store i1 true, i1* %[[NRVO]], align 1
-// CHECK: %[[NRVO_VAL:.*]] = load i1, i1* %[[NRVO]], align 1
+// CHECK: call void @__default_constructor_8_s0(ptr %[[RETVAL]])
+// CHECK: store i1 true, ptr %[[NRVO]], align 1
+// CHECK: %[[NRVO_VAL:.*]] = load i1, ptr %[[NRVO]], align 1
 // CHECK: br i1 %[[NRVO_VAL]],
 
-// CHECK: %[[V1:.*]] = bitcast %[[STRUCT_STRONG]]* %[[RETVAL]] to i8**
-// CHECK: call void @__destructor_8_s0(i8** %[[V1]])
+// CHECK: call void @__destructor_8_s0(ptr %[[RETVAL]])
 // CHECK: br
 
-// CHECK: %[[COERCE_DIVE:.*]] = getelementptr inbounds %[[STRUCT_STRONG]], %[[STRUCT_STRONG]]* %[[RETVAL]], i32 0, i32 0
-// CHECK: %[[V2:.*]] = load i8*, i8** %[[COERCE_DIVE]], align 8
-// CHECK: ret i8* %[[V2]]
+// CHECK: %[[COERCE_DIVE:.*]] = getelementptr inbounds %[[STRUCT_STRONG]], ptr %[[RETVAL]], i32 0, i32 0
+// CHECK: %[[V2:.*]] = load ptr, ptr %[[COERCE_DIVE]], align 8
+// CHECK: ret ptr %[[V2]]
 
 Strong testStrong(void) {
   Strong a;
   return a;
 }
 
-// CHECK: define{{.*}} void @testWeak(%[[STRUCT_WEAK]]* noalias sret(%[[STRUCT_WEAK]]) align 8 %[[AGG_RESULT:.*]])
+// CHECK: define{{.*}} void @testWeak(ptr noalias sret(%[[STRUCT_WEAK]]) align 8 %[[AGG_RESULT:.*]])
 // CHECK: %[[NRVO:.*]] = alloca i1, align 1
-// CHECK: %[[V0:.*]] = bitcast %[[STRUCT_WEAK]]* %[[AGG_RESULT]] to i8**
-// CHECK: call void @__default_constructor_8_w0(i8** %[[V0]])
-// CHECK: store i1 true, i1* %[[NRVO]], align 1
-// CHECK: %[[NRVO_VAL:.*]] = load i1, i1* %[[NRVO]], align 1
+// CHECK: call void @__default_constructor_8_w0(ptr %[[AGG_RESULT]])
+// CHECK: store i1 true, ptr %[[NRVO]], align 1
+// CHECK: %[[NRVO_VAL:.*]] = load i1, ptr %[[NRVO]], align 1
 // CHECK: br i1 %[[NRVO_VAL]],
 
-// CHECK: %[[V1:.*]] = bitcast %[[STRUCT_WEAK]]* %[[AGG_RESULT]] to i8**
-// CHECK: call void @__destructor_8_w0(i8** %[[V1]])
+// CHECK: call void @__destructor_8_w0(ptr %[[AGG_RESULT]])
 // CHECK: br
 
 // CHECK-NOT: call
@@ -105,16 +101,14 @@ Weak testWeak2(int c) {
     return b;
 }
 
-// CHECK: define internal void @"\01-[C1 foo1]"(%[[STRUCT_WEAK]]* noalias sret(%[[STRUCT_WEAK]]) align 8 %[[AGG_RESULT:.*]], %{{.*}}* noundef %{{.*}}, i8* noundef %{{.*}})
+// CHECK: define internal void @"\01-[C1 foo1]"(ptr noalias sret(%[[STRUCT_WEAK]]) align 8 %[[AGG_RESULT:.*]], ptr noundef %{{.*}}, ptr noundef %{{.*}})
 // CHECK: %[[NRVO:.*]] = alloca i1, align 1
-// CHECK: %[[V0:.*]] = bitcast %[[STRUCT_WEAK]]* %[[AGG_RESULT]] to i8**
-// CHECK: call void @__default_constructor_8_w0(i8** %[[V0]])
-// CHECK: store i1 true, i1* %[[NRVO]], align 1
-// CHECK: %[[NRVO_VAL:.*]] = load i1, i1* %[[NRVO]], align 1
+// CHECK: call void @__default_constructor_8_w0(ptr %[[AGG_RESULT]])
+// CHECK: store i1 true, ptr %[[NRVO]], align 1
+// CHECK: %[[NRVO_VAL:.*]] = load i1, ptr %[[NRVO]], align 1
 // CHECK: br i1 %[[NRVO_VAL]],
 
-// CHECK: %[[V1:.*]] = bitcast %[[STRUCT_WEAK]]* %[[AGG_RESULT]] to i8**
-// CHECK: call void @__destructor_8_w0(i8** %[[V1]])
+// CHECK: call void @__destructor_8_w0(ptr %[[AGG_RESULT]])
 // CHECK: br
 
 // CHECK-NOT: call

diff  --git a/clang/test/CodeGenObjC/objc-read-weak-byref.m b/clang/test/CodeGenObjC/objc-read-weak-byref.m
index 5637a596f9939..87b05ef97ad17 100644
--- a/clang/test/CodeGenObjC/objc-read-weak-byref.m
+++ b/clang/test/CodeGenObjC/objc-read-weak-byref.m
@@ -1,6 +1,6 @@
-// RUN: %clang_cc1 -no-opaque-pointers -fblocks -fobjc-gc -triple x86_64-apple-darwin -fobjc-runtime=macosx-fragile-10.5 -emit-llvm %s -o - | \
+// RUN: %clang_cc1 -fblocks -fobjc-gc -triple x86_64-apple-darwin -fobjc-runtime=macosx-fragile-10.5 -emit-llvm %s -o - | \
 // RUN: FileCheck %s
-// RUN: %clang_cc1 -no-opaque-pointers -fblocks -fobjc-gc -triple i386-apple-darwin -fobjc-runtime=macosx-fragile-10.5 -emit-llvm %s -o - | \
+// RUN: %clang_cc1 -fblocks -fobjc-gc -triple i386-apple-darwin -fobjc-runtime=macosx-fragile-10.5 -emit-llvm %s -o - | \
 // RUN: FileCheck %s
 
 @interface NSObject 
@@ -18,5 +18,5 @@ int main(void) {
     return 0;
 }
 
-// CHECK: call i8* @objc_read_weak
-// CHECK: call i8* @objc_read_weak
+// CHECK: call ptr @objc_read_weak
+// CHECK: call ptr @objc_read_weak

diff  --git a/clang/test/CodeGenObjC/objc-runtime-name.m b/clang/test/CodeGenObjC/objc-runtime-name.m
index d590d30c06a32..c838b8c2c6f1a 100644
--- a/clang/test/CodeGenObjC/objc-runtime-name.m
+++ b/clang/test/CodeGenObjC/objc-runtime-name.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple i386-apple-macosx10.13.0 -fobjc-runtime=macosx-fragile-10.13.0 -fobjc-subscripting-legacy-runtime -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple i386-apple-macosx10.13.0 -fobjc-runtime=macosx-fragile-10.13.0 -fobjc-subscripting-legacy-runtime -emit-llvm -o - %s | FileCheck %s
 
 // Check that the runtime name is emitted and used instead of the class
 // identifier.
@@ -8,7 +8,7 @@
 // CHECK: module asm {{.*}}lazy_reference .objc_class_name_XYZ
 
 // CHECK: @[[OBJC_CLASS_NAME:.*]] = private unnamed_addr constant [4 x i8] c"XYZ{{.*}}, section "__TEXT,__cstring,cstring_literals",
-// CHECK: = private global {{.*}} bitcast ([4 x i8]* @[[OBJC_CLASS_NAME]] to {{.*}}), section "__OBJC,__cls_refs,literal_pointers,no_dead_strip"
+// CHECK: = private global {{.*}} @[[OBJC_CLASS_NAME]], section "__OBJC,__cls_refs,literal_pointers,no_dead_strip"
 
 __attribute__((objc_root_class,objc_runtime_name("XYZ")))
 @interface A

diff  --git a/clang/test/CodeGenObjC/objc2-protocol-metadata.m b/clang/test/CodeGenObjC/objc2-protocol-metadata.m
index 1809dc47380e3..38f723709e029 100644
--- a/clang/test/CodeGenObjC/objc2-protocol-metadata.m
+++ b/clang/test/CodeGenObjC/objc2-protocol-metadata.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-macosx10.10 -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-macosx10.10 -emit-llvm -o - %s | FileCheck %s
 // rdar://20286356
 
 @protocol P1
@@ -14,4 +14,4 @@ @implementation INTF
 + ClsP  { return 0; }
 @end
 
-// CHECK: %struct._protocol_t = type { i8*, i8*, %struct._objc_protocol_list*, %struct.__method_list_t*, %struct.__method_list_t*, %struct.__method_list_t*, %struct.__method_list_t*, %struct._prop_list_t*, i32, i32, i8**, i8*, %struct._prop_list_t* }
+// CHECK: %struct._protocol_t = type { ptr, ptr, ptr, ptr, ptr, ptr, ptr, ptr, i32, i32, ptr, ptr, ptr }

diff  --git a/clang/test/CodeGenObjC/objc2-strong-cast-block-import.m b/clang/test/CodeGenObjC/objc2-strong-cast-block-import.m
index 92ca283badac3..bdb3804787ed7 100644
--- a/clang/test/CodeGenObjC/objc2-strong-cast-block-import.m
+++ b/clang/test/CodeGenObjC/objc2-strong-cast-block-import.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -fobjc-gc-only -fblocks  -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -fobjc-gc-only -fblocks  -emit-llvm -o - %s | FileCheck %s
 // rdar://10150823
 
 @interface Test {
@@ -20,6 +20,6 @@ - (void)testWithObjectInBlock {
 }
 
 @end
-// CHECK: call i8* @objc_assign_strongCast
-// CHECK: call i8* @objc_assign_strongCast
+// CHECK: call ptr @objc_assign_strongCast
+// CHECK: call ptr @objc_assign_strongCast
 

diff  --git a/clang/test/CodeGenObjC/objc2-weak-block-call.m b/clang/test/CodeGenObjC/objc2-weak-block-call.m
index cffc325579eca..8519b13a5eacf 100644
--- a/clang/test/CodeGenObjC/objc2-weak-block-call.m
+++ b/clang/test/CodeGenObjC/objc2-weak-block-call.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -fblocks -fobjc-gc -triple x86_64-apple-darwin -fobjc-runtime=macosx-fragile-10.5 -emit-llvm %s -o - | FileCheck -check-prefix CHECK-LP64 %s
-// RUN: %clang_cc1 -no-opaque-pointers -fblocks -fobjc-gc -triple i386-apple-darwin -fobjc-runtime=macosx-fragile-10.5 -emit-llvm %s -o - | FileCheck -check-prefix CHECK-LP32 %s
+// RUN: %clang_cc1 -fblocks -fobjc-gc -triple x86_64-apple-darwin -fobjc-runtime=macosx-fragile-10.5 -emit-llvm %s -o - | FileCheck -check-prefix CHECK-LP64 %s
+// RUN: %clang_cc1 -fblocks -fobjc-gc -triple i386-apple-darwin -fobjc-runtime=macosx-fragile-10.5 -emit-llvm %s -o - | FileCheck -check-prefix CHECK-LP32 %s
 
 @interface MyView
 - (void)MyView_sharedInit;
@@ -19,6 +19,6 @@ - (void)MyView_sharedInit {
 }
 @end
 
-// CHECK-LP64: call i8* @objc_read_weak
-// CHECK-LP32: call i8* @objc_read_weak
+// CHECK-LP64: call ptr @objc_read_weak
+// CHECK-LP32: call ptr @objc_read_weak
 

diff  --git a/clang/test/CodeGenObjC/objc_copyStruct.m b/clang/test/CodeGenObjC/objc_copyStruct.m
index 97cadbd94e245..7bbad866e2b1f 100644
--- a/clang/test/CodeGenObjC/objc_copyStruct.m
+++ b/clang/test/CodeGenObjC/objc_copyStruct.m
@@ -1,5 +1,5 @@
-// RUN: %clang -Xclang -no-opaque-pointers -target x86_64-unknown-windows-msvc -fobjc-runtime=ios -Wno-objc-root-class -S -o - -emit-llvm %s | FileCheck %s
-// RUN: %clang -Xclang -no-opaque-pointers -target x86_64-apple-ios -fobjc-runtime=ios -Wno-objc-root-class -S -o - -emit-llvm %s | FileCheck %s
+// RUN: %clang -target x86_64-unknown-windows-msvc -fobjc-runtime=ios -Wno-objc-root-class -S -o - -emit-llvm %s | FileCheck %s
+// RUN: %clang -target x86_64-apple-ios -fobjc-runtime=ios -Wno-objc-root-class -S -o - -emit-llvm %s | FileCheck %s
 
 struct S {
   float f, g;
@@ -12,5 +12,5 @@ @interface I
 @implementation I
 @end
 
-// CHECK: declare {{.*}}void @objc_copyStruct(i8*, i8*, i64, i1, i1)
+// CHECK: declare {{.*}}void @objc_copyStruct(ptr, ptr, i64, i1, i1)
 

diff  --git a/clang/test/CodeGenObjC/objfw.m b/clang/test/CodeGenObjC/objfw.m
index da31d69a498c9..15284f43e53b3 100644
--- a/clang/test/CodeGenObjC/objfw.m
+++ b/clang/test/CodeGenObjC/objfw.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-pc-linux-gnu -fobjc-runtime=objfw -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-pc-linux-gnu -fobjc-runtime=objfw -emit-llvm -o - %s | FileCheck %s
 
 // Test the ObjFW runtime.
 
@@ -9,7 +9,6 @@ void test0(void) {
   [Test0 test];
 }
 // CHECK-LABEL:    define{{.*}} void @test0()
-// CHECK:      [[T0:%.*]] = call i8* (i8*, i8*, ...)* @objc_msg_lookup(i8* bitcast (i64* @_OBJC_CLASS_Test0 to i8*),
-// CHECK-NEXT: [[T1:%.*]] = bitcast i8* (i8*, i8*, ...)* [[T0]] to void (i8*, i8*)*
-// CHECK-NEXT: call void [[T1]](i8* noundef bitcast (i64* @_OBJC_CLASS_Test0 to i8*), 
+// CHECK:      [[T0:%.*]] = call ptr @objc_msg_lookup(ptr @_OBJC_CLASS_Test0,
+// CHECK-NEXT: call void [[T0]](ptr noundef @_OBJC_CLASS_Test0, 
 // CHECK-NEXT: ret void

diff  --git a/clang/test/CodeGenObjC/optimize-ivar-offset-load.m b/clang/test/CodeGenObjC/optimize-ivar-offset-load.m
index 95d534ec224d0..b943dbe678c84 100644
--- a/clang/test/CodeGenObjC/optimize-ivar-offset-load.m
+++ b/clang/test/CodeGenObjC/optimize-ivar-offset-load.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10  -O0 -emit-llvm %s -o -  | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10  -O0 -emit-llvm %s -o -  | FileCheck %s
 // rdar://16095748
 
 @interface MyNSObject 
@@ -26,11 +26,9 @@ - (void)application
     }
 }
 @end
-// CHECK: [[IVAR:%.*]]  = load i64, i64* @"OBJC_IVAR_$_SampleClass._value", align 8
-// CHECK: [[THREE:%.*]] = bitcast [[ONE:%.*]]* [[CALL:%.*]] to i8*
-// CHECK: [[ADDPTR:%.*]] = getelementptr inbounds i8, i8* [[THREE]], i64 [[IVAR]]
-// CHECK: [[FOUR:%.*]] = bitcast i8* [[ADDPTR]] to i32*
-// CHECK: [[FIVE:%.*]] = load i32, i32* [[FOUR]], align 4
+// CHECK: [[IVAR:%.*]]  = load i64, ptr @"OBJC_IVAR_$_SampleClass._value", align 8
+// CHECK: [[ADDPTR:%.*]] = getelementptr inbounds i8, ptr [[CALL:%.*]], i64 [[IVAR]]
+// CHECK: [[FIVE:%.*]] = load i32, ptr [[ADDPTR]], align 4
 // CHECK:   call void @foo(i32 noundef [[FIVE]])
 
 @implementation SampleClass
@@ -44,8 +42,8 @@ - (void) SampleClassApplication
     }
 }
 @end
-// CHECK: [[ZERO:%.*]] = load i8*, i8** @OBJC_SELECTOR_REFERENCES_, align 8, !invariant.load
-// CHECK: [[IVAR:%.*]] = load i64, i64* @"OBJC_IVAR_$_SampleClass._value", align 8, !invariant.load
+// CHECK: [[ZERO:%.*]] = load ptr, ptr @OBJC_SELECTOR_REFERENCES_, align 8, !invariant.load
+// CHECK: [[IVAR:%.*]] = load i64, ptr @"OBJC_IVAR_$_SampleClass._value", align 8, !invariant.load
 
 @interface Sample : SampleClass @end
 
@@ -59,6 +57,6 @@ - (void) SampleApplication
     }
 }
 @end
-// CHECK: [[ZERO:%.*]] = load i8*, i8** @OBJC_SELECTOR_REFERENCES_, align 8, !invariant.load 
-// CHECK: [[IVAR:%.*]] = load i64, i64* @"OBJC_IVAR_$_SampleClass._value", align 8, !invariant.load
+// CHECK: [[ZERO:%.*]] = load ptr, ptr @OBJC_SELECTOR_REFERENCES_, align 8, !invariant.load 
+// CHECK: [[IVAR:%.*]] = load i64, ptr @"OBJC_IVAR_$_SampleClass._value", align 8, !invariant.load
 

diff  --git a/clang/test/CodeGenObjC/pass-by-value-noalias.m b/clang/test/CodeGenObjC/pass-by-value-noalias.m
index df2faac8d003c..ed94d4c80b525 100644
--- a/clang/test/CodeGenObjC/pass-by-value-noalias.m
+++ b/clang/test/CodeGenObjC/pass-by-value-noalias.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -fpass-by-value-is-noalias -triple arm64-apple-iphoneos -emit-llvm -disable-llvm-optzns -fobjc-runtime-has-weak -fobjc-arc -fobjc-dispatch-method=mixed %s -o - 2>&1 | FileCheck --check-prefix=WITH_NOALIAS %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple arm64-apple-iphoneos -emit-llvm -disable-llvm-optzns -fobjc-runtime-has-weak -fobjc-arc -fobjc-dispatch-method=mixed %s -o - 2>&1 | FileCheck --check-prefix=NO_NOALIAS %s
+// RUN: %clang_cc1 -fpass-by-value-is-noalias -triple arm64-apple-iphoneos -emit-llvm -disable-llvm-optzns -fobjc-runtime-has-weak -fobjc-arc -fobjc-dispatch-method=mixed %s -o - 2>&1 | FileCheck --check-prefix=WITH_NOALIAS %s
+// RUN: %clang_cc1 -triple arm64-apple-iphoneos -emit-llvm -disable-llvm-optzns -fobjc-runtime-has-weak -fobjc-arc -fobjc-dispatch-method=mixed %s -o - 2>&1 | FileCheck --check-prefix=NO_NOALIAS %s
 
 @interface Bar
 @property char value;
@@ -17,6 +17,6 @@ @interface Bar
   Bar *__weak f;
 };
 
-// WITH_NOALIAS: define{{.*}} void @take(%struct.Foo* noundef %arg)
-// NO_NOALIAS: define{{.*}} void @take(%struct.Foo* noundef %arg)
+// WITH_NOALIAS: define{{.*}} void @take(ptr noundef %arg)
+// NO_NOALIAS: define{{.*}} void @take(ptr noundef %arg)
 void take(struct Foo arg) {}

diff  --git a/clang/test/CodeGenObjC/personality.m b/clang/test/CodeGenObjC/personality.m
index 3ebbf1b103d34..6ec67ace3f1ac 100644
--- a/clang/test/CodeGenObjC/personality.m
+++ b/clang/test/CodeGenObjC/personality.m
@@ -1,59 +1,59 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=gnustep-1.7 -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GNUSTEP-1_7
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=gnustep -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GNUSTEP
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC-SEH
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC-SJLJ
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW-SEH
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW-SJLJ
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=gnustep-1.7 -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GNUSTEP-1_7
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=gnustep -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GNUSTEP
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC-SEH
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC-SJLJ
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW-SEH
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW-SJLJ
 
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fobjc-runtime=gnustep-1.7 -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fobjc-runtime=gnustep -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
+// RUN: %clang_cc1 -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
+// RUN: %clang_cc1 -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
+// RUN: %clang_cc1 -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
+// RUN: %clang_cc1 -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
+// RUN: %clang_cc1 -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fobjc-runtime=gnustep-1.7 -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
+// RUN: %clang_cc1 -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fobjc-runtime=gnustep -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
+// RUN: %clang_cc1 -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
+// RUN: %clang_cc1 -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
 
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE-MINGW-DWARF
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE-MINGW-SEH
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE-MINGW-SJLJ
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=gnustep-1.7 -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GNUSTEP-1_7
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=gnustep -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GNUSTEP
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC-SEH
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC-SJLJ
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW-SEH
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW-SJLJ
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE-MINGW-DWARF
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE-MINGW-SEH
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE-MINGW-SJLJ
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=gnustep-1.7 -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GNUSTEP-1_7
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=gnustep -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GNUSTEP
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC-SEH
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC-SJLJ
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW-SEH
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW-SJLJ
 
 void g(void);
 
-// CHECK-MACOSX-FRAGILE: personality i8* bitcast (i32 (...)* @__gcc_personality_v0 to i8*)
-// CHECK-NS: personality i8* bitcast (i32 (...)* @__objc_personality_v0 to i8*)
-// CHECK-GNUSTEP-1_7: personality i8* bitcast (i32 (...)* @__gnustep_objc_personality_v0 to i8*)
-// CHECK-GNUSTEP: personality i8* bitcast (i32 (...)* @__gnu_objc_personality_v0 to i8*)
-// CHECK-GCC: personality i8* bitcast (i32 (...)* @__gnu_objc_personality_v0 to i8*)
-// CHECK-GCC-SEH: personality i8* bitcast (i32 (...)* @__gnu_objc_personality_seh0 to i8*)
-// CHECK-GCC-SJLJ: personality i8* bitcast (i32 (...)* @__gnu_objc_personality_sj0 to i8*)
-// CHECK-OBJFW: personality i8* bitcast (i32 (...)* @__gnu_objc_personality_v0 to i8*)
-// CHECK-OBJFW-SEH: personality i8* bitcast (i32 (...)* @__gnu_objc_personality_seh0 to i8*)
-// CHECK-OBJFW-SJLJ: personality i8* bitcast (i32 (...)* @__gnu_objc_personality_sj0 to i8*)
+// CHECK-MACOSX-FRAGILE: personality ptr @__gcc_personality_v0
+// CHECK-NS: personality ptr @__objc_personality_v0
+// CHECK-GNUSTEP-1_7: personality ptr @__gnustep_objc_personality_v0
+// CHECK-GNUSTEP: personality ptr @__gnu_objc_personality_v0
+// CHECK-GCC: personality ptr @__gnu_objc_personality_v0
+// CHECK-GCC-SEH: personality ptr @__gnu_objc_personality_seh0
+// CHECK-GCC-SJLJ: personality ptr @__gnu_objc_personality_sj0
+// CHECK-OBJFW: personality ptr @__gnu_objc_personality_v0
+// CHECK-OBJFW-SEH: personality ptr @__gnu_objc_personality_seh0
+// CHECK-OBJFW-SJLJ: personality ptr @__gnu_objc_personality_sj0
 
-// CHECK-WIN-MSVC: personality i8* bitcast (i32 (...)* @__CxxFrameHandler3  to i8*)
+// CHECK-WIN-MSVC: personality ptr @__CxxFrameHandler3
 
-// CHECK-MACOSX-FRAGILE-MINGW-DWARF: personality i8* bitcast (i32 (...)* @__gcc_personality_v0 to i8*)
-// CHECK-MACOSX-FRAGILE-MINGW-SEH: personality i8* bitcast (i32 (...)* @__gcc_personality_seh0 to i8*)
-// CHECK-MACOSX-FRAGILE-MINGW-SJLJ: personality i8* bitcast (i32 (...)* @__gcc_personality_sj0 to i8*)
+// CHECK-MACOSX-FRAGILE-MINGW-DWARF: personality ptr @__gcc_personality_v0
+// CHECK-MACOSX-FRAGILE-MINGW-SEH: personality ptr @__gcc_personality_seh0
+// CHECK-MACOSX-FRAGILE-MINGW-SJLJ: personality ptr @__gcc_personality_sj0
 
 void f(void) {
   @try {
@@ -63,8 +63,8 @@ void f(void) {
 }
 
 #if defined(__SEH_EXCEPTIONS__)
-// CHECK-WIN-SEH-X86: personality i8* bitcast (i32 (...)* @_except_handler3 to i8*)
-// CHECK-WIN-SEH-X64: personality i8* bitcast (i32 (...)* @__C_specific_handler to i8*)
+// CHECK-WIN-SEH-X86: personality ptr @_except_handler3
+// CHECK-WIN-SEH-X64: personality ptr @__C_specific_handler
 
 void h(void) {
   __try {

diff  --git a/clang/test/CodeGenObjC/ppc32-varargs-id.m b/clang/test/CodeGenObjC/ppc32-varargs-id.m
index 65a268d069e70..7b1348245c6d4 100644
--- a/clang/test/CodeGenObjC/ppc32-varargs-id.m
+++ b/clang/test/CodeGenObjC/ppc32-varargs-id.m
@@ -1,5 +1,5 @@
 // REQUIRES: powerpc-registered-target
-// RUN: %clang_cc1 -no-opaque-pointers -triple powerpc-unknown-openbsd -fblocks -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple powerpc-unknown-openbsd -fblocks -emit-llvm -o - %s | FileCheck %s
 
 #include <stdarg.h>
 
@@ -8,13 +8,12 @@ id testObject(va_list ap) {
 }
 // CHECK: @testObject
 // CHECK: using_regs:
-// CHECK-NEXT: getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{[0-9]+}}, i32 0, i32 4
-// CHECK-NEXT: load i8*, i8** %{{[0-9]+}}, align 4
+// CHECK-NEXT: getelementptr inbounds %struct.__va_list_tag, ptr %{{[0-9]+}}, i32 0, i32 4
+// CHECK-NEXT: load ptr, ptr %{{[0-9]+}}, align 4
 // CHECK-NEXT: mul i8 %numUsedRegs, 4
-// CHECK-NEXT: getelementptr inbounds i8, i8* %{{[0-9]+}}, i8 %{{[0-9]+}}
-// CHECK-NEXT: bitcast i8* %{{[0-9]+}} to i8**
+// CHECK-NEXT: getelementptr inbounds i8, ptr %{{[0-9]+}}, i8 %{{[0-9]+}}
 // CHECK-NEXT: add i8 %numUsedRegs, 1
-// CHECK-NEXT: store i8 %{{[0-9]+}}, i8* %gpr, align 4
+// CHECK-NEXT: store i8 %{{[0-9]+}}, ptr %gpr, align 4
 // CHECK-NEXT: br label %cont
 
 typedef void (^block)(void);
@@ -23,11 +22,10 @@ block testBlock(va_list ap) {
 }
 // CHECK: @testBlock
 // CHECK: using_regs:
-// CHECK-NEXT: getelementptr inbounds %struct.__va_list_tag, %struct.__va_list_tag* %{{[0-9]+}}, i32 0, i32 4
-// CHECK-NEXT: load i8*, i8** %{{[0-9]+}}, align 4
+// CHECK-NEXT: getelementptr inbounds %struct.__va_list_tag, ptr %{{[0-9]+}}, i32 0, i32 4
+// CHECK-NEXT: load ptr, ptr %{{[0-9]+}}, align 4
 // CHECK-NEXT: mul i8 %numUsedRegs, 4
-// CHECK-NEXT: getelementptr inbounds i8, i8* %{{[0-9]+}}, i8 %{{[0-9]+}}
-// CHECK-NEXT: bitcast i8* %{{[0-9]+}} to void ()**
+// CHECK-NEXT: getelementptr inbounds i8, ptr %{{[0-9]+}}, i8 %{{[0-9]+}}
 // CHECK-NEXT: add i8 %numUsedRegs, 1
-// CHECK-NEXT: store i8 %{{[0-9]+}}, i8* %gpr, align 4
+// CHECK-NEXT: store i8 %{{[0-9]+}}, ptr %gpr, align 4
 // CHECK-NEXT: br label %cont

diff  --git a/clang/test/CodeGenObjC/property-array-type.m b/clang/test/CodeGenObjC/property-array-type.m
index cc3529360ccce..81939617e0813 100644
--- a/clang/test/CodeGenObjC/property-array-type.m
+++ b/clang/test/CodeGenObjC/property-array-type.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10  -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10  -emit-llvm -o - %s | FileCheck %s
 // rdar://15610943
 
 struct _GLKMatrix4
@@ -24,7 +24,7 @@ - (void)viewDidLoad {
 }
 @end
 
-// CHECK: [[M:%.*]] = getelementptr inbounds %struct._GLKMatrix4, %struct._GLKMatrix4* [[TMP:%.*]], i32 0, i32 0
-// CHECK: [[ARRAYDECAY:%.*]] = getelementptr inbounds [16 x float], [16 x float]* [[M]], i64 0, i64 0
-// CHECK: [[SIX:%.*]] = load i8*, i8** @OBJC_SELECTOR_REFERENCES
-// CHECK:  call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*, float*)*)(i8* noundef [[SEVEN:%.*]], i8* noundef [[SIX]], float* noundef [[ARRAYDECAY]])
+// CHECK: [[M:%.*]] = getelementptr inbounds %struct._GLKMatrix4, ptr [[TMP:%.*]], i32 0, i32 0
+// CHECK: [[ARRAYDECAY:%.*]] = getelementptr inbounds [16 x float], ptr [[M]], i64 0, i64 0
+// CHECK: [[SIX:%.*]] = load ptr, ptr @OBJC_SELECTOR_REFERENCES
+// CHECK:  call void @objc_msgSend(ptr noundef [[SEVEN:%.*]], ptr noundef [[SIX]], ptr noundef [[ARRAYDECAY]])

diff  --git a/clang/test/CodeGenObjC/property-atomic-bool.m b/clang/test/CodeGenObjC/property-atomic-bool.m
index 5664da86f74db..4eb3f1b7cb71c 100644
--- a/clang/test/CodeGenObjC/property-atomic-bool.m
+++ b/clang/test/CodeGenObjC/property-atomic-bool.m
@@ -1,21 +1,21 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-macosx10 -emit-llvm -x objective-c %s -o - | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-macosx10 -emit-llvm -x objective-c %s -o - | FileCheck %s
 
 // CHECK: define internal zeroext i1 @"\01-[A0 p]"(
-// CHECK:   %[[ATOMIC_LOAD:.*]] = load atomic i8, i8* %{{.*}} seq_cst, align 1
+// CHECK:   %[[ATOMIC_LOAD:.*]] = load atomic i8, ptr %{{.*}} seq_cst, align 1
 // CHECK:   %[[TOBOOL:.*]] = trunc i8 %[[ATOMIC_LOAD]] to i1
 // CHECK:   ret i1 %[[TOBOOL]]
 
 // CHECK: define internal void @"\01-[A0 setP:]"({{.*}} i1 noundef zeroext {{.*}})
-// CHECK:   store atomic i8 %{{.*}}, i8* %{{.*}} seq_cst, align 1
+// CHECK:   store atomic i8 %{{.*}}, ptr %{{.*}} seq_cst, align 1
 // CHECK:   ret void
 
 // CHECK: define internal zeroext i1 @"\01-[A1 p]"(
-// CHECK:   %[[ATOMIC_LOAD:.*]] = load atomic i8, i8* %{{.*}} unordered, align 1
+// CHECK:   %[[ATOMIC_LOAD:.*]] = load atomic i8, ptr %{{.*}} unordered, align 1
 // CHECK:   %[[TOBOOL:.*]] = trunc i8 %load to i1
 // CHECK:   ret i1 %[[TOBOOL]]
 
 // CHECK: define internal void @"\01-[A1 setP:]"({{.*}} i1 noundef zeroext %p)
-// CHECK:   store atomic i8 %{{.*}}, i8* %{{.*}} unordered, align 1
+// CHECK:   store atomic i8 %{{.*}}, ptr %{{.*}} unordered, align 1
 // CHECK:   ret void
 
 @interface A0

diff  --git a/clang/test/CodeGenObjC/protocol-property-synth.m b/clang/test/CodeGenObjC/protocol-property-synth.m
index 7f3a3959d5392..c998d631677a7 100644
--- a/clang/test/CodeGenObjC/protocol-property-synth.m
+++ b/clang/test/CodeGenObjC/protocol-property-synth.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -o %t %s
+// RUN: %clang_cc1 -emit-llvm -o %t %s
 
 @interface BaseClass {
     id _delegate;

diff  --git a/clang/test/CodeGenObjC/related-result-type.m b/clang/test/CodeGenObjC/related-result-type.m
index 7b552a608ff30..677ac78cb7665 100644
--- a/clang/test/CodeGenObjC/related-result-type.m
+++ b/clang/test/CodeGenObjC/related-result-type.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -emit-llvm -o - %s | FileCheck %s
 
 @interface NSObject
 + (id)alloc;
@@ -14,7 +14,6 @@ void test1(void) {
   // CHECK: {{call.*@objc_msgSend}}
   // CHECK: {{call.*@objc_msgSend}}
   // CHECK: {{call.*@objc_msgSend}}
-  // CHECK: bitcast i8*
   NSString *str1 = [[[NSString alloc] init] retain];
 }
 
@@ -23,7 +22,6 @@ void test2(void) {
   // CHECK: {{call.*@objc_msgSend}}
   // CHECK: {{call.*@objc_msgSend}}
   // CHECK: {{call.*@objc_msgSend}}
-  // CHECK: bitcast i8*
   NSString *str1 = NSString.alloc.init.retain;
 }
 
@@ -32,10 +30,9 @@ - (id)init;
 @end
 
 @implementation Test2
-// CHECK: define internal {{.*}}i8* @"\01-[Test2 init]"
+// CHECK: define internal {{.*}}ptr @"\01-[Test2 init]"
 - (id)init {
   // CHECK: {{call.*@objc_msgSendSuper}}
-  // CHECK-NEXT: bitcast i8*
   return [super init];
 }
 @end
@@ -45,10 +42,9 @@ - (id)init;
 @end
 
 @implementation Test3
-// CHECK: define internal {{.*}}i8* @"\01-[Test3 init]"
+// CHECK: define internal {{.*}}ptr @"\01-[Test3 init]"
 - (id)init {
   // CHECK: {{call.*@objc_msgSendSuper}}
-  // CHECK-NEXT: bitcast i8*
   return [super init];
 }
 @end

diff  --git a/clang/test/CodeGenObjC/return-objc-object.mm b/clang/test/CodeGenObjC/return-objc-object.mm
index 1cead8c3b492f..24bb5a6cdf195 100644
--- a/clang/test/CodeGenObjC/return-objc-object.mm
+++ b/clang/test/CodeGenObjC/return-objc-object.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64 -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64 -emit-llvm -o - %s | FileCheck %s
 
 @protocol P1 @end
 @interface NSOperationQueue 
@@ -15,5 +15,5 @@ void call_once() {
   f();
   f1();
 }
-// CHECK: call noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) %0* @_Z1fv()
-// CHECK: call noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) %0* @_Z2f1v()  
+// CHECK: call noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) ptr @_Z1fv()
+// CHECK: call noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) ptr @_Z2f1v()  

diff  --git a/clang/test/CodeGenObjC/runtime-abi-match.m b/clang/test/CodeGenObjC/runtime-abi-match.m
index e0f1ad2ce985d..c8408914bd9ca 100644
--- a/clang/test/CodeGenObjC/runtime-abi-match.m
+++ b/clang/test/CodeGenObjC/runtime-abi-match.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple thumbv7--windows-itanium -fobjc-runtime=ios -O1 -fexceptions -fobjc-exceptions -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -triple thumbv7--windows-itanium -fobjc-runtime=ios -O1 -fexceptions -fobjc-exceptions -emit-llvm %s -o - | FileCheck %s
 // REQUIRES: arm-registered-target
 
 void (*f)(id);
@@ -18,8 +18,8 @@ void i(void) {
   }
 }
 
-// CHECK: call arm_aapcs_vfpcc i8* @objc_begin_catch
+// CHECK: call arm_aapcs_vfpcc ptr @objc_begin_catch
 // CHECK: call arm_aapcs_vfpcc void @objc_end_catch
-// CHECK-NOT: call i8* @objc_begin_catch
+// CHECK-NOT: call ptr @objc_begin_catch
 // CHECK-NOT: call void @objc_end_catch
 

diff  --git a/clang/test/CodeGenObjC/selector-ref-invariance.m b/clang/test/CodeGenObjC/selector-ref-invariance.m
index 902371e93efd7..cd37b5d8e77d4 100644
--- a/clang/test/CodeGenObjC/selector-ref-invariance.m
+++ b/clang/test/CodeGenObjC/selector-ref-invariance.m
@@ -1,9 +1,9 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin11 -emit-llvm -fblocks -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin11 -emit-llvm -fblocks -o - %s | FileCheck %s
 
 // rdar://6027699
 
 void test(id x) {
-// CHECK: load i8*, i8** @OBJC_SELECTOR_REFERENCES_, align 8, !invariant.load
+// CHECK: load ptr, ptr @OBJC_SELECTOR_REFERENCES_, align 8, !invariant.load
 // CHECK: @objc_msgSend
   [x foo];
 }

diff  --git a/clang/test/CodeGenObjC/stret-1.m b/clang/test/CodeGenObjC/stret-1.m
index f3511e00b14d2..ab3f664fb5fec 100644
--- a/clang/test/CodeGenObjC/stret-1.m
+++ b/clang/test/CodeGenObjC/stret-1.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -fblocks -triple arm64-apple-darwin %s -emit-llvm -o - | FileCheck %s
+// RUN: %clang_cc1 -fblocks -triple arm64-apple-darwin %s -emit-llvm -o - | FileCheck %s
 // rdar://12416433
 
 struct stret { int x[100]; };
@@ -14,19 +14,18 @@ int main(int argc, const char **argv)
 {
     struct stret s;
     s = [(id)(argc&~255) method];
-    // CHECK: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (%struct.stret*, i8*, i8*)*)(%struct.stret* sret(%struct.stret) align 4 [[T0:%[^,]+]]
-    // CHECK: [[T0P:%.*]] = bitcast %struct.stret* [[T0]] to i8*
-    // CHECK: call void @llvm.memset.p0i8.i64(i8* align 4 [[T0P]], i8 0, i64 400, i1 false)
+    // CHECK: call void @objc_msgSend(ptr sret(%struct.stret) align 4 [[T0:%[^,]+]]
+    // CHECK: call void @llvm.memset.p0.i64(ptr align 4 [[T0]], i8 0, i64 400, i1 false)
 
     s = [Test method];
-    // CHECK: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (%struct.stret*, i8*, i8*)*)(%struct.stret* sret(%struct.stret) align 4 [[T1:%[^,]+]]
-    // CHECK-NOT: call void @llvm.memset.p0i8.i64(
+    // CHECK: call void @objc_msgSend(ptr sret(%struct.stret) align 4 [[T1:%[^,]+]]
+    // CHECK-NOT: call void @llvm.memset.p0.i64(
 
     [(id)(argc&~255) method];
-    // CHECK: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (%struct.stret*, i8*, i8*)*)(%struct.stret* sret(%struct.stret) align 4 [[T1:%[^,]+]]
-    // CHECK-NOT: call void @llvm.memset.p0i8.i64(
+    // CHECK: call void @objc_msgSend(ptr sret(%struct.stret) align 4 [[T1:%[^,]+]]
+    // CHECK-NOT: call void @llvm.memset.p0.i64(
 
     [Test method];
-    // CHECK: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (%struct.stret*, i8*, i8*)*)(%struct.stret* sret(%struct.stret) align 4 [[T1:%[^,]+]]
-    // CHECK-NOT: call void @llvm.memset.p0i8.i64(
+    // CHECK: call void @objc_msgSend(ptr sret(%struct.stret) align 4 [[T1:%[^,]+]]
+    // CHECK-NOT: call void @llvm.memset.p0.i64(
 }

diff  --git a/clang/test/CodeGenObjC/super-dotsyntax-struct-property.m b/clang/test/CodeGenObjC/super-dotsyntax-struct-property.m
index 6d0329b14a7a0..d1984f8f092fe 100644
--- a/clang/test/CodeGenObjC/super-dotsyntax-struct-property.m
+++ b/clang/test/CodeGenObjC/super-dotsyntax-struct-property.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10  -emit-llvm %s -o -  | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10  -emit-llvm %s -o -  | FileCheck %s
 // rdar: // 8203426
 
 
@@ -44,4 +44,4 @@ - (void)setFrame:(CGRect)frame
 
 @end
 // CHECK-NOT: declare void @objc_msgSendSuper2_stret
-// CHECK: declare i8* @objc_msgSendSuper2
+// CHECK: declare ptr @objc_msgSendSuper2

diff  --git a/clang/test/CodeGenObjC/super-message-fragileabi.m b/clang/test/CodeGenObjC/super-message-fragileabi.m
index 450ce1f7bd1fb..c1b6fecf3c793 100644
--- a/clang/test/CodeGenObjC/super-message-fragileabi.m
+++ b/clang/test/CodeGenObjC/super-message-fragileabi.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple i386-apple-darwin9 -fobjc-runtime=macosx-fragile-10.5 -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -triple i386-apple-darwin9 -fobjc-runtime=macosx-fragile-10.5 -emit-llvm %s -o - | FileCheck %s
 
 @class  Some;
 
@@ -24,7 +24,7 @@ - (id)initSome:(Some *)arg {
  {
 	;
  }
-// CHECK: load %struct._objc_class*, %struct._objc_class** getelementptr inbounds (%struct._objc_class, %struct._objc_class* @OBJC_CLASS_BetterTable, i32 0, i32 1)
+// CHECK: load ptr, ptr getelementptr inbounds (%struct._objc_class, ptr @OBJC_CLASS_BetterTable, i32 0, i32 1)
 
  return self;
 }

diff  --git a/clang/test/CodeGenObjC/synchronized.m b/clang/test/CodeGenObjC/synchronized.m
index 9337fd77fb86f..25feb958c73ab 100644
--- a/clang/test/CodeGenObjC/synchronized.m
+++ b/clang/test/CodeGenObjC/synchronized.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -triple i686-apple-darwin9 -fobjc-runtime=macosx-fragile-10.5 -o - %s -O2 | FileCheck %s
+// RUN: %clang_cc1 -emit-llvm -triple i686-apple-darwin9 -fobjc-runtime=macosx-fragile-10.5 -o - %s -O2 | FileCheck %s
 
 @interface MyClass
 {
@@ -22,19 +22,19 @@ - (void)method
 
 // CHECK-LABEL: define{{.*}} void @foo(
 void foo(id a) {
-  // CHECK: [[A:%.*]] = alloca i8*
-  // CHECK: [[SYNC:%.*]] = alloca i8*
+  // CHECK: [[A:%.*]] = alloca ptr
+  // CHECK: [[SYNC:%.*]] = alloca ptr
 
-  // CHECK:      store i8* [[AVAL:%.*]], i8** [[A]]
-  // CHECK-NEXT: call i32 @objc_sync_enter(i8* [[AVAL]])
-  // CHECK-NEXT: store i8* [[AVAL]], i8** [[SYNC]]
+  // CHECK:      store ptr [[AVAL:%.*]], ptr [[A]]
+  // CHECK-NEXT: call i32 @objc_sync_enter(ptr [[AVAL]])
+  // CHECK-NEXT: store ptr [[AVAL]], ptr [[SYNC]]
   // CHECK-NEXT: call void @objc_exception_try_enter
   // CHECK:      call i32 @_setjmp
   @synchronized(a) {
     // This is unreachable, but the optimizers can't know that.
-    // CHECK: call void asm sideeffect "", "=*m,=*m,=*m"(i8** nonnull elementtype(i8*) [[A]], i8** nonnull elementtype(i8*) [[SYNC]]
+    // CHECK: call void asm sideeffect "", "=*m,=*m,=*m"(ptr nonnull elementtype(ptr) [[A]], ptr nonnull elementtype(ptr) [[SYNC]]
     // CHECK: call i32 @objc_sync_exit
-    // CHECK: call i8* @objc_exception_extract
+    // CHECK: call ptr @objc_exception_extract
     // CHECK: call void @objc_exception_throw
     // CHECK: unreachable
 
@@ -52,7 +52,7 @@ int f0(id a) {
   // that x isn't stored to within the synchronized block.
 
   // CHECK: [[X:%.*]] = alloca i32
-  // CHECK: store i32 1, i32* [[X]]
+  // CHECK: store i32 1, ptr [[X]]
   int x = 0;
   @synchronized((x++, a)) {    
   }

diff  --git a/clang/test/CodeGenObjC/synthesize_ivar.m b/clang/test/CodeGenObjC/synthesize_ivar.m
index 4b9840a0a59af..92f6096b7e4db 100644
--- a/clang/test/CodeGenObjC/synthesize_ivar.m
+++ b/clang/test/CodeGenObjC/synthesize_ivar.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -o %t %s
+// RUN: %clang_cc1 -emit-llvm -o %t %s
 
 // PR13820
 // REQUIRES: LP64

diff  --git a/clang/test/CodeGenObjC/tentative-cfconstantstring.m b/clang/test/CodeGenObjC/tentative-cfconstantstring.m
index 1cff9d5d6e3ed..b4c1bd77c0ba2 100644
--- a/clang/test/CodeGenObjC/tentative-cfconstantstring.m
+++ b/clang/test/CodeGenObjC/tentative-cfconstantstring.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -o - %s | FileCheck %s
 // rdar://13598026
 
 @interface NSObject @end
@@ -32,11 +32,10 @@ -(void)someMethod {
 @end
 
 // CHECK: @__CFConstantStringClassReference ={{.*}} global [24 x i32] zeroinitializer, align 16
-// CHECK: @_unnamed_cfstring_{{.*}} = private global %struct.__NSConstantString_tag { i32* getelementptr inbounds ([24 x i32], [24 x i32]* @__CFConstantStringClassReference, i32 0, i32 0)
+// CHECK: @_unnamed_cfstring_{{.*}} = private global %struct.__NSConstantString_tag { ptr @__CFConstantStringClassReference
 
 // CHECK-LABEL: define internal void @_inlineFunction()
-// CHECK:  [[ZERO:%.*]] = load %struct._class_t*, %struct._class_t** @"OBJC_CLASSLIST_REFERENCES_
-// CHECK-NEXT:   [[OBJ:%.*]] = bitcast %struct._class_t* [[ZERO]] to i8*
-// CHECK-NEXT:   [[SEL:%.*]] = load i8*, i8** @OBJC_SELECTOR_REFERENCES_
-// CHECK-NEXT:   call void (i8*, i8*, [[T:%.*]]*, ...) bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*, [[T:%.*]]*, ...)*)(i8* noundef [[OBJ]], i8* noundef [[SEL]], [[T:%.*]]* noundef bitcast (%struct.__NSConstantString_tag* @_unnamed_cfstring_{{.*}} to [[T:%.*]]*))
+// CHECK:  [[ZERO:%.*]] = load ptr, ptr @"OBJC_CLASSLIST_REFERENCES_
+// CHECK-NEXT:   [[SEL:%.*]] = load ptr, ptr @OBJC_SELECTOR_REFERENCES_
+// CHECK-NEXT:   call void (ptr, ptr, ptr, ...) @objc_msgSend(ptr noundef [[ZERO]], ptr noundef [[SEL]], ptr noundef @_unnamed_cfstring_{{.*}})
 // CHECK-NEXT:   ret void

diff  --git a/clang/test/CodeGenObjC/terminate.m b/clang/test/CodeGenObjC/terminate.m
index 0c7562ea58d33..aa24e355fb0c6 100644
--- a/clang/test/CodeGenObjC/terminate.m
+++ b/clang/test/CodeGenObjC/terminate.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin -fobjc-runtime=macosx-fragile-10.8 -emit-llvm -fexceptions -fobjc-exceptions -o - %s | FileCheck %s -check-prefix=CHECK-WITH
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin -fobjc-runtime=macosx-fragile-10.7 -emit-llvm -fexceptions -fobjc-exceptions -o - %s | FileCheck %s -check-prefix=CHECK-WITHOUT
+// RUN: %clang_cc1 -triple x86_64-apple-darwin -fobjc-runtime=macosx-fragile-10.8 -emit-llvm -fexceptions -fobjc-exceptions -o - %s | FileCheck %s -check-prefix=CHECK-WITH
+// RUN: %clang_cc1 -triple x86_64-apple-darwin -fobjc-runtime=macosx-fragile-10.7 -emit-llvm -fexceptions -fobjc-exceptions -o - %s | FileCheck %s -check-prefix=CHECK-WITHOUT
 
 void destroy(void**);
 
@@ -10,22 +10,22 @@ void test0(void) {
   test0_helper();
 
   // CHECK-WITH-LABEL:       define{{.*}} void @test0()
-  // CHECK-WITH-SAME:    personality i8* bitcast (i32 (...)* @__gcc_personality_v0 to i8*)
-  // CHECK-WITH:         [[PTR:%.*]] = alloca i8*,
-  // CHECK-WITH:         call void @destroy(i8** noundef [[PTR]])
+  // CHECK-WITH-SAME:    personality ptr @__gcc_personality_v0
+  // CHECK-WITH:         [[PTR:%.*]] = alloca ptr,
+  // CHECK-WITH:         call void @destroy(ptr noundef [[PTR]])
   // CHECK-WITH-NEXT:    ret void
-  // CHECK-WITH:         invoke void @destroy(i8** noundef [[PTR]])
-  // CHECK-WITH:         landingpad { i8*, i32 }
-  // CHECK-WITH-NEXT:      catch i8* null
+  // CHECK-WITH:         invoke void @destroy(ptr noundef [[PTR]])
+  // CHECK-WITH:         landingpad { ptr, i32 }
+  // CHECK-WITH-NEXT:      catch ptr null
   // CHECK-WITH-NEXT:    call void @objc_terminate()
 
   // CHECK-WITHOUT-LABEL:    define{{.*}} void @test0()
-  // CHECK-WITHOUT-SAME: personality i8* bitcast (i32 (...)* @__gcc_personality_v0 to i8*)
-  // CHECK-WITHOUT:      [[PTR:%.*]] = alloca i8*,
-  // CHECK-WITHOUT:      call void @destroy(i8** noundef [[PTR]])
+  // CHECK-WITHOUT-SAME: personality ptr @__gcc_personality_v0
+  // CHECK-WITHOUT:      [[PTR:%.*]] = alloca ptr,
+  // CHECK-WITHOUT:      call void @destroy(ptr noundef [[PTR]])
   // CHECK-WITHOUT-NEXT: ret void
-  // CHECK-WITHOUT:      invoke void @destroy(i8** noundef [[PTR]])
-  // CHECK-WITHOUT:      landingpad { i8*, i32 }
-  // CHECK-WITHOUT-NEXT:   catch i8* null
+  // CHECK-WITHOUT:      invoke void @destroy(ptr noundef [[PTR]])
+  // CHECK-WITHOUT:      landingpad { ptr, i32 }
+  // CHECK-WITHOUT-NEXT:   catch ptr null
   // CHECK-WITHOUT-NEXT: call void @abort()
 }

diff  --git a/clang/test/CodeGenObjC/ubsan-bool.m b/clang/test/CodeGenObjC/ubsan-bool.m
index 4fa133e4a0aa9..fa5883c01cf4f 100644
--- a/clang/test/CodeGenObjC/ubsan-bool.m
+++ b/clang/test/CodeGenObjC/ubsan-bool.m
@@ -1,6 +1,6 @@
-// RUN: %clang_cc1 -no-opaque-pointers -no-enable-noundef-analysis -x objective-c -emit-llvm -triple x86_64-apple-macosx10.10.0 -fsanitize=bool %s -o - -w | FileCheck %s -check-prefixes=SHARED,OBJC
-// RUN: %clang_cc1 -no-opaque-pointers -no-enable-noundef-analysis -x objective-c++ -emit-llvm -triple x86_64-apple-macosx10.10.0 -fsanitize=bool %s -o - -w | FileCheck %s -check-prefixes=SHARED,OBJC
-// RUN: %clang_cc1 -no-opaque-pointers -no-enable-noundef-analysis -x c -emit-llvm -triple x86_64-apple-macosx10.10.0 -fsanitize=bool %s -o - | FileCheck %s -check-prefixes=SHARED,C
+// RUN: %clang_cc1 -no-enable-noundef-analysis -x objective-c -emit-llvm -triple x86_64-apple-macosx10.10.0 -fsanitize=bool %s -o - -w | FileCheck %s -check-prefixes=SHARED,OBJC
+// RUN: %clang_cc1 -no-enable-noundef-analysis -x objective-c++ -emit-llvm -triple x86_64-apple-macosx10.10.0 -fsanitize=bool %s -o - -w | FileCheck %s -check-prefixes=SHARED,OBJC
+// RUN: %clang_cc1 -no-enable-noundef-analysis -x c -emit-llvm -triple x86_64-apple-macosx10.10.0 -fsanitize=bool %s -o - | FileCheck %s -check-prefixes=SHARED,C
 
 typedef signed char BOOL;
 
@@ -19,7 +19,7 @@ BOOL f1(void) {
 
 // SHARED-LABEL: f2
 BOOL f2(struct S1 *s) {
-  // OBJC: [[LOAD:%.*]] = load i8, i8* {{.*}}
+  // OBJC: [[LOAD:%.*]] = load i8, ptr {{.*}}
   // OBJC: [[SHL:%.*]] = shl i8 [[LOAD]], 7
   // OBJC: [[ASHR:%.*]] = ashr i8 [[SHL]], 7
   // OBJC: icmp ule i8 [[ASHR]], 1, !nosanitize
@@ -43,9 +43,9 @@ @implementation I1
 
 // Check the synthesized getter.
 // OBJC-LABEL: define internal signext i8 @"\01-[I1 b1]"
-// OBJC: [[IVAR:%.*]] = load i64, i64* @"OBJC_IVAR_$_I1.b1"
-// OBJC: [[ADDR:%.*]] = getelementptr inbounds i8, i8* {{.*}}, i64 [[IVAR]]
-// OBJC: [[LOAD:%.*]] = load i8, i8* {{.*}}
+// OBJC: [[IVAR:%.*]] = load i64, ptr @"OBJC_IVAR_$_I1.b1"
+// OBJC: [[ADDR:%.*]] = getelementptr inbounds i8, ptr {{.*}}, i64 [[IVAR]]
+// OBJC: [[LOAD:%.*]] = load i8, ptr {{.*}}
 // OBJC: [[SHL:%.*]] = shl i8 [[LOAD]], 7
 // OBJC: [[ASHR:%.*]] = ashr i8 [[SHL]], 7
 // OBJC: icmp ule i8 [[ASHR]], 1, !nosanitize
@@ -54,7 +54,7 @@ @implementation I1
 // Also check direct accesses to the ivar.
 // OBJC-LABEL: f3
 BOOL f3(I1 *i) {
-  // OBJC: [[LOAD:%.*]] = load i8, i8* {{.*}}
+  // OBJC: [[LOAD:%.*]] = load i8, ptr {{.*}}
   // OBJC: [[SHL:%.*]] = shl i8 [[LOAD]], 7
   // OBJC: [[ASHR:%.*]] = ashr i8 [[SHL]], 7
   // OBJC: icmp ule i8 [[ASHR]], 1, !nosanitize

diff  --git a/clang/test/CodeGenObjC/ubsan-nonnull-and-nullability.m b/clang/test/CodeGenObjC/ubsan-nonnull-and-nullability.m
index ce8d13ad5a628..b5ef3fa6e6f76 100644
--- a/clang/test/CodeGenObjC/ubsan-nonnull-and-nullability.m
+++ b/clang/test/CodeGenObjC/ubsan-nonnull-and-nullability.m
@@ -1,24 +1,24 @@
 // REQUIRES: asserts
-// RUN: %clang_cc1 -no-opaque-pointers -x objective-c -emit-llvm -triple x86_64-apple-macosx10.10.0 -fsanitize=nullability-return,returns-nonnull-attribute,nullability-arg,nonnull-attribute %s -o - -w | FileCheck %s
+// RUN: %clang_cc1 -x objective-c -emit-llvm -triple x86_64-apple-macosx10.10.0 -fsanitize=nullability-return,returns-nonnull-attribute,nullability-arg,nonnull-attribute %s -o - -w | FileCheck %s
 
 // If both the annotation and the attribute are present, prefer the attribute,
 // since it actually affects IRGen.
 
-// CHECK-LABEL: define{{.*}} nonnull i32* @f1
+// CHECK-LABEL: define{{.*}} nonnull ptr @f1
 __attribute__((returns_nonnull)) int *_Nonnull f1(int *_Nonnull p) {
   // CHECK: entry:
-  // CHECK-NEXT: [[SLOC_PTR:%.*]] = alloca i8*
-  // CHECK-NEXT: [[ADDR:%.*]] = alloca i32*
-  // CHECK-NEXT: store i8* null, i8** [[SLOC_PTR]]
-  // CHECK-NEXT: store i32* [[P:%.*]], i32** [[ADDR]]
+  // CHECK-NEXT: [[SLOC_PTR:%.*]] = alloca ptr
+  // CHECK-NEXT: [[ADDR:%.*]] = alloca ptr
+  // CHECK-NEXT: store ptr null, ptr [[SLOC_PTR]]
+  // CHECK-NEXT: store ptr [[P:%.*]], ptr [[ADDR]]
   // CHECK-NEXT: store {{.*}} [[SLOC_PTR]]
-  // CHECK-NEXT: [[ARG:%.*]] = load i32*, i32** [[ADDR]]
+  // CHECK-NEXT: [[ARG:%.*]] = load ptr, ptr [[ADDR]]
   // CHECK-NEXT: [[SLOC:%.*]] = load {{.*}} [[SLOC_PTR]]
-  // CHECK-NEXT: [[SLOC_NONNULL:%.*]] = icmp ne i8* [[SLOC]], null
+  // CHECK-NEXT: [[SLOC_NONNULL:%.*]] = icmp ne ptr [[SLOC]], null
   // CHECK-NEXT: br i1 [[SLOC_NONNULL]], label %nullcheck
   // 
   // CHECK: nullcheck:
-  // CHECK-NEXT: [[ICMP:%.*]] = icmp ne i32* [[ARG]], null, !nosanitize
+  // CHECK-NEXT: [[ICMP:%.*]] = icmp ne ptr [[ARG]], null, !nosanitize
   // CHECK-NEXT: br i1 [[ICMP]], label %[[CONT:.+]], label %[[HANDLE:[^,]+]]
   // CHECK: [[HANDLE]]:
   // CHECK:      call void @__ubsan_handle_nonnull_return
@@ -26,7 +26,7 @@
   // CHECK: [[CONT]]:
   // CHECK-NEXT:   br label %no.nullcheck
   // CHECK: no.nullcheck:
-  // CHECK-NEXT: ret i32* [[ARG]]
+  // CHECK-NEXT: ret ptr [[ARG]]
   return p;
 }
 
@@ -41,7 +41,7 @@ void call_f2(void) {
 }
 
 // If the return value isn't meant to be checked, make sure we don't check it.
-// CHECK-LABEL: define{{.*}} i32* @f3
+// CHECK-LABEL: define{{.*}} ptr @f3
 int *f3(int *p) {
   // CHECK-NOT: return.sloc
   // CHECK-NOT: call{{.*}}ubsan
@@ -51,11 +51,11 @@ void call_f2(void) {
 // Check for a valid "return" source location, even when there is no return
 // statement, to avoid accidentally calling the runtime.
 
-// CHECK-LABEL: define{{.*}} nonnull i32* @f4
+// CHECK-LABEL: define{{.*}} nonnull ptr @f4
 __attribute__((returns_nonnull)) int *f4(void) {
-  // CHECK: store i8* null, i8** [[SLOC_PTR:%.*]]
+  // CHECK: store ptr null, ptr [[SLOC_PTR:%.*]]
   // CHECK: [[SLOC:%.*]] = load {{.*}} [[SLOC_PTR]]
-  // CHECK: [[SLOC_NONNULL:%.*]] = icmp ne i8* [[SLOC]], null
+  // CHECK: [[SLOC_NONNULL:%.*]] = icmp ne ptr [[SLOC]], null
   // CHECK: br i1 [[SLOC_NONNULL]], label %nullcheck
   // CHECK: nullcheck:
 }

diff  --git a/clang/test/CodeGenObjC/ubsan-nullability-return-unreachable.m b/clang/test/CodeGenObjC/ubsan-nullability-return-unreachable.m
index a00739c25441c..ccfa2254f6564 100644
--- a/clang/test/CodeGenObjC/ubsan-nullability-return-unreachable.m
+++ b/clang/test/CodeGenObjC/ubsan-nullability-return-unreachable.m
@@ -1,6 +1,6 @@
-// RUN: %clang_cc1 -no-opaque-pointers -fsanitize=nullability-return -emit-llvm %s -o - -triple x86_64-apple-macosx10.10.0 -Wno-objc-root-class | FileCheck %s
+// RUN: %clang_cc1 -fsanitize=nullability-return -emit-llvm %s -o - -triple x86_64-apple-macosx10.10.0 -Wno-objc-root-class | FileCheck %s
 
-// CHECK-LABEL: define internal i8* @"\01-[I init]"
+// CHECK-LABEL: define internal ptr @"\01-[I init]"
 // CHECK: unreachable
 // CHECK-NEXT: }
 

diff  --git a/clang/test/CodeGenObjC/variadic-sends.m b/clang/test/CodeGenObjC/variadic-sends.m
index 4139239dfa2db..0382ac7ad7b84 100644
--- a/clang/test/CodeGenObjC/variadic-sends.m
+++ b/clang/test/CodeGenObjC/variadic-sends.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple i386-unknown-unknown -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -o - %s | FileCheck -check-prefix=CHECK-X86-32 %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-unknown -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -o - %s | FileCheck -check-prefix=CHECK-X86-64 %s
+// RUN: %clang_cc1 -triple i386-unknown-unknown -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -o - %s | FileCheck -check-prefix=CHECK-X86-32 %s
+// RUN: %clang_cc1 -triple x86_64-unknown-unknown -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -o - %s | FileCheck -check-prefix=CHECK-X86-64 %s
 
 @interface A
 -(void) im0;
@@ -8,33 +8,33 @@ -(void) im2: (int) x, ...;
 @end
 
 void f0(A *a) {
-  // CHECK-X86-32: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*)*)
-  // CHECK-X86-64: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*)*)
+  // CHECK-X86-32: call void @objc_msgSend
+  // CHECK-X86-64: call void @objc_msgSend
   [a im0];
 }
 
 void f1(A *a) {
-  // CHECK-X86-32: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*, i32)*)
-  // CHECK-X86-64: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*, i32)*)
+  // CHECK-X86-32: call void @objc_msgSend
+  // CHECK-X86-64: call void @objc_msgSend
   [a im1: 1];
 }
 
 void f2(A *a) {
-  // CHECK-X86-32: call void (i8*, i8*, i32, ...) bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*, i32, ...)*)
-  // CHECK-X86-64: call void (i8*, i8*, i32, ...) bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*, i32, ...)*)
+  // CHECK-X86-32: call void (ptr, ptr, i32, ...) @objc_msgSend
+  // CHECK-X86-64: call void (ptr, ptr, i32, ...) @objc_msgSend
   [a im2: 1, 2];
 }
 
 @interface B : A @end
 @implementation B : A
 -(void) foo {
-  // CHECK-X86-32: call void bitcast (i8* (%struct._objc_super*, i8*, ...)* @objc_msgSendSuper to void (%struct._objc_super*, i8*, i32)*)
-  // CHECK-X86-64: call void bitcast (i8* (%struct._objc_super*, i8*, ...)* @objc_msgSendSuper to void (%struct._objc_super*, i8*, i32)*)
+  // CHECK-X86-32: call void @objc_msgSendSuper
+  // CHECK-X86-64: call void @objc_msgSendSuper
   [super im1: 1];
 }
 -(void) bar {
-  // CHECK-X86-32: call void (%struct._objc_super*, i8*, i32, ...) bitcast (i8* (%struct._objc_super*, i8*, ...)* @objc_msgSendSuper to void (%struct._objc_super*, i8*, i32, ...)*)
-  // CHECK-X86-64: call void (%struct._objc_super*, i8*, i32, ...) bitcast (i8* (%struct._objc_super*, i8*, ...)* @objc_msgSendSuper to void (%struct._objc_super*, i8*, i32, ...)*)
+  // CHECK-X86-32: call void (ptr, ptr, i32, ...) @objc_msgSendSuper
+  // CHECK-X86-64: call void (ptr, ptr, i32, ...) @objc_msgSendSuper
   [super im2: 1, 2];
 }
 

diff  --git a/clang/test/CodeGenObjC/weak-in-c-struct.m b/clang/test/CodeGenObjC/weak-in-c-struct.m
index 38ff778eb374e..3ec08c2b9f18b 100644
--- a/clang/test/CodeGenObjC/weak-in-c-struct.m
+++ b/clang/test/CodeGenObjC/weak-in-c-struct.m
@@ -1,11 +1,11 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple arm64-apple-ios11 -fobjc-arc -fblocks -fobjc-runtime=ios-11.0 -emit-llvm -o - %s | FileCheck -check-prefix=ARM64 -check-prefix=COMMON %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple thumbv7-apple-ios10 -fobjc-arc -fblocks -fobjc-runtime=ios-10.0 -emit-llvm -o - %s | FileCheck -check-prefix=COMMON %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-macosx10.13 -fobjc-arc -fblocks -fobjc-runtime=macosx-10.13.0 -emit-llvm -o - %s | FileCheck -check-prefix=COMMON %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple i386-apple-macosx10.13.0 -fobjc-arc -fblocks -fobjc-runtime=macosx-fragile-10.13.0 -emit-llvm -o - %s | FileCheck -check-prefix=COMMON %s
+// RUN: %clang_cc1 -triple arm64-apple-ios11 -fobjc-arc -fblocks -fobjc-runtime=ios-11.0 -emit-llvm -o - %s | FileCheck -check-prefix=ARM64 -check-prefix=COMMON %s
+// RUN: %clang_cc1 -triple thumbv7-apple-ios10 -fobjc-arc -fblocks -fobjc-runtime=ios-10.0 -emit-llvm -o - %s | FileCheck -check-prefix=COMMON %s
+// RUN: %clang_cc1 -triple x86_64-apple-macosx10.13 -fobjc-arc -fblocks -fobjc-runtime=macosx-10.13.0 -emit-llvm -o - %s | FileCheck -check-prefix=COMMON %s
+// RUN: %clang_cc1 -triple i386-apple-macosx10.13.0 -fobjc-arc -fblocks -fobjc-runtime=macosx-fragile-10.13.0 -emit-llvm -o - %s | FileCheck -check-prefix=COMMON %s
 
 typedef void (^BlockTy)(void);
 
-// COMMON: %[[STRUCT_WEAK:.*]] = type { i32, i8* }
+// COMMON: %[[STRUCT_WEAK:.*]] = type { i32, ptr }
 
 typedef struct {
   int f0;
@@ -17,30 +17,23 @@
 
 // ARM64: define{{.*}} void @test_constructor_destructor_Weak()
 // ARM64: %[[T:.*]] = alloca %[[STRUCT_WEAK]], align 8
-// ARM64: %[[V0:.*]] = bitcast %[[STRUCT_WEAK]]* %[[T]] to i8**
-// ARM64: call void @__default_constructor_8_w8(i8** %[[V0]])
-// ARM64: %[[V1:.*]] = bitcast %[[STRUCT_WEAK]]* %[[T]] to i8**
-// ARM64: call void @__destructor_8_w8(i8** %[[V1]])
+// ARM64: call void @__default_constructor_8_w8(ptr %[[T]])
+// ARM64: call void @__destructor_8_w8(ptr %[[T]])
 // ARM64: ret void
 
-// ARM64: define linkonce_odr hidden void @__default_constructor_8_w8(i8** noundef %[[DST:.*]])
-// ARM64: %[[DST_ADDR:.*]] = alloca i8**, align 8
-// ARM64: store i8** %[[DST]], i8*** %[[DST_ADDR]], align 8
-// ARM64: %[[V0:.*]] = load i8**, i8*** %[[DST_ADDR]], align 8
-// ARM64: %[[V1]] = bitcast i8** %[[V0]] to i8*
-// ARM64: %[[V2:.*]] = getelementptr inbounds i8, i8* %[[V1]], i64 8
-// ARM64: %[[V3:.*]] = bitcast i8* %[[V2]] to i8**
-// ARM64: %[[V4:.*]] = bitcast i8** %[[V3]] to i8*
-// ARM64: call void @llvm.memset.p0i8.i64(i8* align 8 %[[V4]], i8 0, i64 8, i1 false)
-
-// ARM64: define linkonce_odr hidden void @__destructor_8_w8(i8** noundef %[[DST:.*]])
-// ARM64: %[[DST_ADDR:.*]] = alloca i8**, align 8
-// ARM64: store i8** %[[DST]], i8*** %[[DST_ADDR]], align 8
-// ARM64: %[[V0:.*]] = load i8**, i8*** %[[DST_ADDR]], align 8
-// ARM64: %[[V1:.*]] = bitcast i8** %[[V0]] to i8*
-// ARM64: %[[V2:.*]] = getelementptr inbounds i8, i8* %[[V1]], i64 8
-// ARM64: %[[V3:.*]] = bitcast i8* %[[V2]] to i8**
-// ARM64: call void @llvm.objc.destroyWeak(i8** %[[V3]])
+// ARM64: define linkonce_odr hidden void @__default_constructor_8_w8(ptr noundef %[[DST:.*]])
+// ARM64: %[[DST_ADDR:.*]] = alloca ptr, align 8
+// ARM64: store ptr %[[DST]], ptr %[[DST_ADDR]], align 8
+// ARM64: %[[V0:.*]] = load ptr, ptr %[[DST_ADDR]], align 8
+// ARM64: %[[V2:.*]] = getelementptr inbounds i8, ptr %[[V0]], i64 8
+// ARM64: call void @llvm.memset.p0.i64(ptr align 8 %[[V2]], i8 0, i64 8, i1 false)
+
+// ARM64: define linkonce_odr hidden void @__destructor_8_w8(ptr noundef %[[DST:.*]])
+// ARM64: %[[DST_ADDR:.*]] = alloca ptr, align 8
+// ARM64: store ptr %[[DST]], ptr %[[DST_ADDR]], align 8
+// ARM64: %[[V0:.*]] = load ptr, ptr %[[DST_ADDR]], align 8
+// ARM64: %[[V2:.*]] = getelementptr inbounds i8, ptr %[[V0]], i64 8
+// ARM64: call void @llvm.objc.destroyWeak(ptr %[[V2]])
 
 @interface C
 - (void)m:(Weak)a;
@@ -50,146 +43,117 @@ void test_constructor_destructor_Weak(void) {
   Weak t;
 }
 
-// ARM64: define{{.*}} void @test_copy_constructor_Weak(%[[STRUCT_WEAK]]* noundef %{{.*}})
-// ARM64: call void @__copy_constructor_8_8_t0w4_w8(i8** %{{.*}}, i8** %{{.*}})
-// ARM64: call void @__destructor_8_w8(i8** %{{.*}})
-
-// ARM64: define linkonce_odr hidden void @__copy_constructor_8_8_t0w4_w8(i8** noundef %[[DST:.*]], i8** noundef %[[SRC:.*]])
-// ARM64: %[[DST_ADDR:.*]] = alloca i8**, align 8
-// ARM64: %[[SRC_ADDR:.*]] = alloca i8**, align 8
-// ARM64: store i8** %[[DST]], i8*** %[[DST_ADDR]], align 8
-// ARM64: store i8** %[[SRC]], i8*** %[[SRC_ADDR]], align 8
-// ARM64: %[[V0:.*]] = load i8**, i8*** %[[DST_ADDR]], align 8
-// ARM64: %[[V1:.*]] = load i8**, i8*** %[[SRC_ADDR]], align 8
-// ARM64: %[[V2:.*]] = bitcast i8** %[[V0]] to i32*
-// ARM64: %[[V3:.*]] = bitcast i8** %[[V1]] to i32*
-// ARM64: %[[V4:.*]] = load i32, i32* %[[V3]], align 8
-// ARM64: store i32 %[[V4]], i32* %[[V2]], align 8
-// ARM64: %[[V5:.*]] = bitcast i8** %[[V0]] to i8*
-// ARM64: %[[V6:.*]] = getelementptr inbounds i8, i8* %[[V5]], i64 8
-// ARM64: %[[V7:.*]] = bitcast i8* %[[V6]] to i8**
-// ARM64: %[[V8:.*]] = bitcast i8** %[[V1]] to i8*
-// ARM64: %[[V9:.*]] = getelementptr inbounds i8, i8* %[[V8]], i64 8
-// ARM64: %[[V10:.*]] = bitcast i8* %[[V9]] to i8**
-// ARM64: call void @llvm.objc.copyWeak(i8** %[[V7]], i8** %[[V10]])
+// ARM64: define{{.*}} void @test_copy_constructor_Weak(ptr noundef %{{.*}})
+// ARM64: call void @__copy_constructor_8_8_t0w4_w8(ptr %{{.*}}, ptr %{{.*}})
+// ARM64: call void @__destructor_8_w8(ptr %{{.*}})
+
+// ARM64: define linkonce_odr hidden void @__copy_constructor_8_8_t0w4_w8(ptr noundef %[[DST:.*]], ptr noundef %[[SRC:.*]])
+// ARM64: %[[DST_ADDR:.*]] = alloca ptr, align 8
+// ARM64: %[[SRC_ADDR:.*]] = alloca ptr, align 8
+// ARM64: store ptr %[[DST]], ptr %[[DST_ADDR]], align 8
+// ARM64: store ptr %[[SRC]], ptr %[[SRC_ADDR]], align 8
+// ARM64: %[[V0:.*]] = load ptr, ptr %[[DST_ADDR]], align 8
+// ARM64: %[[V1:.*]] = load ptr, ptr %[[SRC_ADDR]], align 8
+// ARM64: %[[V4:.*]] = load i32, ptr %[[V1]], align 8
+// ARM64: store i32 %[[V4]], ptr %[[V0]], align 8
+// ARM64: %[[V6:.*]] = getelementptr inbounds i8, ptr %[[V0]], i64 8
+// ARM64: %[[V9:.*]] = getelementptr inbounds i8, ptr %[[V1]], i64 8
+// ARM64: call void @llvm.objc.copyWeak(ptr %[[V6]], ptr %[[V9]])
 
 void test_copy_constructor_Weak(Weak *s) {
   Weak t = *s;
 }
 
-// ARM64: define{{.*}} void @test_copy_assignment_Weak(%[[STRUCT_WEAK]]* noundef %{{.*}}, %[[STRUCT_WEAK]]* noundef %{{.*}})
-// ARM64: call void @__copy_assignment_8_8_t0w4_w8(i8** %{{.*}}, i8** %{{.*}})
-
-// ARM64: define linkonce_odr hidden void @__copy_assignment_8_8_t0w4_w8(i8** noundef %[[DST:.*]], i8** noundef %[[SRC:.*]])
-// ARM64: %[[DST_ADDR:.*]] = alloca i8**, align 8
-// ARM64: %[[SRC_ADDR:.*]] = alloca i8**, align 8
-// ARM64: store i8** %[[DST]], i8*** %[[DST_ADDR]], align 8
-// ARM64: store i8** %[[SRC]], i8*** %[[SRC_ADDR]], align 8
-// ARM64: %[[V0:.*]] = load i8**, i8*** %[[DST_ADDR]], align 8
-// ARM64: %[[V1:.*]] = load i8**, i8*** %[[SRC_ADDR]], align 8
-// ARM64: %[[V2:.*]] = bitcast i8** %[[V0]] to i32*
-// ARM64: %[[V3:.*]] = bitcast i8** %[[V1]] to i32*
-// ARM64: %[[V4:.*]] = load i32, i32* %[[V3]], align 8
-// ARM64: store i32 %[[V4]], i32* %[[V2]], align 8
-// ARM64: %[[V5:.*]] = bitcast i8** %[[V0]] to i8*
-// ARM64: %[[V6:.*]] = getelementptr inbounds i8, i8* %[[V5]], i64 8
-// ARM64: %[[V7:.*]] = bitcast i8* %[[V6]] to i8**
-// ARM64: %[[V8:.*]] = bitcast i8** %[[V1]] to i8*
-// ARM64: %[[V9:.*]] = getelementptr inbounds i8, i8* %[[V8]], i64 8
-// ARM64: %[[V10:.*]] = bitcast i8* %[[V9]] to i8**
-// ARM64: %[[V11:.*]] = call i8* @llvm.objc.loadWeakRetained(i8** %[[V10]])
-// ARM64: %[[V12:.*]] = call i8* @llvm.objc.storeWeak(i8** %[[V7]], i8* %[[V11]])
-// ARM64: call void @llvm.objc.release(i8* %[[V11]])
+// ARM64: define{{.*}} void @test_copy_assignment_Weak(ptr noundef %{{.*}}, ptr noundef %{{.*}})
+// ARM64: call void @__copy_assignment_8_8_t0w4_w8(ptr %{{.*}}, ptr %{{.*}})
+
+// ARM64: define linkonce_odr hidden void @__copy_assignment_8_8_t0w4_w8(ptr noundef %[[DST:.*]], ptr noundef %[[SRC:.*]])
+// ARM64: %[[DST_ADDR:.*]] = alloca ptr, align 8
+// ARM64: %[[SRC_ADDR:.*]] = alloca ptr, align 8
+// ARM64: store ptr %[[DST]], ptr %[[DST_ADDR]], align 8
+// ARM64: store ptr %[[SRC]], ptr %[[SRC_ADDR]], align 8
+// ARM64: %[[V0:.*]] = load ptr, ptr %[[DST_ADDR]], align 8
+// ARM64: %[[V1:.*]] = load ptr, ptr %[[SRC_ADDR]], align 8
+// ARM64: %[[V4:.*]] = load i32, ptr %[[V1]], align 8
+// ARM64: store i32 %[[V4]], ptr %[[V0]], align 8
+// ARM64: %[[V6:.*]] = getelementptr inbounds i8, ptr %[[V0]], i64 8
+// ARM64: %[[V9:.*]] = getelementptr inbounds i8, ptr %[[V1]], i64 8
+// ARM64: %[[V11:.*]] = call ptr @llvm.objc.loadWeakRetained(ptr %[[V9]])
+// ARM64: %[[V12:.*]] = call ptr @llvm.objc.storeWeak(ptr %[[V6]], ptr %[[V11]])
+// ARM64: call void @llvm.objc.release(ptr %[[V11]])
 
 void test_copy_assignment_Weak(Weak *d, Weak *s) {
   *d = *s;
 }
 
-// ARM64: define internal void @__Block_byref_object_copy_(i8* noundef %0, i8* noundef %1)
-// ARM64: call void @__move_constructor_8_8_t0w4_w8(i8** %{{.*}}, i8** %{{.*}})
-
-// ARM64: define linkonce_odr hidden void @__move_constructor_8_8_t0w4_w8(i8** noundef %[[DST:.*]], i8** noundef %[[SRC:.*]])
-// ARM64: %[[DST_ADDR:.*]] = alloca i8**, align 8
-// ARM64: %[[SRC_ADDR:.*]] = alloca i8**, align 8
-// ARM64: store i8** %[[DST]], i8*** %[[DST_ADDR]], align 8
-// ARM64: store i8** %[[SRC]], i8*** %[[SRC_ADDR]], align 8
-// ARM64: %[[V0:.*]] = load i8**, i8*** %[[DST_ADDR]], align 8
-// ARM64: %[[V1:.*]] = load i8**, i8*** %[[SRC_ADDR]], align 8
-// ARM64: %[[V2:.*]] = bitcast i8** %[[V0]] to i32*
-// ARM64: %[[V3:.*]] = bitcast i8** %[[V1]] to i32*
-// ARM64: %[[V4:.*]] = load i32, i32* %[[V3]], align 8
-// ARM64: store i32 %[[V4]], i32* %[[V2]], align 8
-// ARM64: %[[V5:.*]] = bitcast i8** %[[V0]] to i8*
-// ARM64: %[[V6:.*]] = getelementptr inbounds i8, i8* %[[V5]], i64 8
-// ARM64: %[[V7:.*]] = bitcast i8* %[[V6]] to i8**
-// ARM64: %[[V8:.*]] = bitcast i8** %[[V1]] to i8*
-// ARM64: %[[V9:.*]] = getelementptr inbounds i8, i8* %[[V8]], i64 8
-// ARM64: %[[V10:.*]] = bitcast i8* %[[V9]] to i8**
-// ARM64: call void @llvm.objc.moveWeak(i8** %[[V7]], i8** %[[V10]])
+// ARM64: define internal void @__Block_byref_object_copy_(ptr noundef %0, ptr noundef %1)
+// ARM64: call void @__move_constructor_8_8_t0w4_w8(ptr %{{.*}}, ptr %{{.*}})
+
+// ARM64: define linkonce_odr hidden void @__move_constructor_8_8_t0w4_w8(ptr noundef %[[DST:.*]], ptr noundef %[[SRC:.*]])
+// ARM64: %[[DST_ADDR:.*]] = alloca ptr, align 8
+// ARM64: %[[SRC_ADDR:.*]] = alloca ptr, align 8
+// ARM64: store ptr %[[DST]], ptr %[[DST_ADDR]], align 8
+// ARM64: store ptr %[[SRC]], ptr %[[SRC_ADDR]], align 8
+// ARM64: %[[V0:.*]] = load ptr, ptr %[[DST_ADDR]], align 8
+// ARM64: %[[V1:.*]] = load ptr, ptr %[[SRC_ADDR]], align 8
+// ARM64: %[[V4:.*]] = load i32, ptr %[[V1]], align 8
+// ARM64: store i32 %[[V4]], ptr %[[V0]], align 8
+// ARM64: %[[V6:.*]] = getelementptr inbounds i8, ptr %[[V0]], i64 8
+// ARM64: %[[V9:.*]] = getelementptr inbounds i8, ptr %[[V1]], i64 8
+// ARM64: call void @llvm.objc.moveWeak(ptr %[[V6]], ptr %[[V9]])
 
 void test_move_constructor_Weak(void) {
   __block Weak t;
   BlockTy b = ^{ (void)t; };
 }
 
-// ARM64: define{{.*}} void @test_move_assignment_Weak(%[[STRUCT_WEAK]]* noundef %{{.*}})
-// ARM64: call void @__move_assignment_8_8_t0w4_w8(i8** %{{.*}}, i8** %{{.*}})
-
-// ARM64: define linkonce_odr hidden void @__move_assignment_8_8_t0w4_w8(i8** noundef %[[DST:.*]], i8** noundef %[[SRC:.*]])
-// ARM64: %[[DST_ADDR:.*]] = alloca i8**, align 8
-// ARM64: %[[SRC_ADDR:.*]] = alloca i8**, align 8
-// ARM64: store i8** %[[DST]], i8*** %[[DST_ADDR]], align 8
-// ARM64: store i8** %[[SRC]], i8*** %[[SRC_ADDR]], align 8
-// ARM64: %[[V0:.*]] = load i8**, i8*** %[[DST_ADDR]], align 8
-// ARM64: %[[V1:.*]] = load i8**, i8*** %[[SRC_ADDR]], align 8
-// ARM64: %[[V2:.*]] = bitcast i8** %[[V0]] to i32*
-// ARM64: %[[V3:.*]] = bitcast i8** %[[V1]] to i32*
-// ARM64: %[[V4:.*]] = load i32, i32* %[[V3]], align 8
-// ARM64: store i32 %[[V4]], i32* %[[V2]], align 8
-// ARM64: %[[V5:.*]] = bitcast i8** %[[V0]] to i8*
-// ARM64: %[[V6:.*]] = getelementptr inbounds i8, i8* %[[V5]], i64 8
-// ARM64: %[[V7:.*]] = bitcast i8* %[[V6]] to i8**
-// ARM64: %[[V8:.*]] = bitcast i8** %[[V1]] to i8*
-// ARM64: %[[V9:.*]] = getelementptr inbounds i8, i8* %[[V8]], i64 8
-// ARM64: %[[V10:.*]] = bitcast i8* %[[V9]] to i8**
-// ARM64: %[[V11:.*]] = call i8* @llvm.objc.loadWeakRetained(i8** %[[V10]])
-// ARM64: %[[V12:.*]] = call i8* @llvm.objc.storeWeak(i8** %[[V7]], i8* %[[V11]])
-// ARM64: call void @llvm.objc.destroyWeak(i8** %[[V10]])
-// ARM64: call void @llvm.objc.release(i8* %[[V11]])
+// ARM64: define{{.*}} void @test_move_assignment_Weak(ptr noundef %{{.*}})
+// ARM64: call void @__move_assignment_8_8_t0w4_w8(ptr %{{.*}}, ptr %{{.*}})
+
+// ARM64: define linkonce_odr hidden void @__move_assignment_8_8_t0w4_w8(ptr noundef %[[DST:.*]], ptr noundef %[[SRC:.*]])
+// ARM64: %[[DST_ADDR:.*]] = alloca ptr, align 8
+// ARM64: %[[SRC_ADDR:.*]] = alloca ptr, align 8
+// ARM64: store ptr %[[DST]], ptr %[[DST_ADDR]], align 8
+// ARM64: store ptr %[[SRC]], ptr %[[SRC_ADDR]], align 8
+// ARM64: %[[V0:.*]] = load ptr, ptr %[[DST_ADDR]], align 8
+// ARM64: %[[V1:.*]] = load ptr, ptr %[[SRC_ADDR]], align 8
+// ARM64: %[[V4:.*]] = load i32, ptr %[[V1]], align 8
+// ARM64: store i32 %[[V4]], ptr %[[V0]], align 8
+// ARM64: %[[V6:.*]] = getelementptr inbounds i8, ptr %[[V0]], i64 8
+// ARM64: %[[V9:.*]] = getelementptr inbounds i8, ptr %[[V1]], i64 8
+// ARM64: %[[V11:.*]] = call ptr @llvm.objc.loadWeakRetained(ptr %[[V9]])
+// ARM64: %[[V12:.*]] = call ptr @llvm.objc.storeWeak(ptr %[[V6]], ptr %[[V11]])
+// ARM64: call void @llvm.objc.destroyWeak(ptr %[[V9]])
+// ARM64: call void @llvm.objc.release(ptr %[[V11]])
 
 void test_move_assignment_Weak(Weak *p) {
   *p = getWeak();
 }
 
-// COMMON: define{{.*}} void @test_parameter_Weak(%[[STRUCT_WEAK]]* noundef %[[A:.*]])
-// COMMON: %[[V0:.*]] = bitcast %[[STRUCT_WEAK]]* %[[A]] to i8**
-// COMMON: call void @__destructor_{{.*}}(i8** %[[V0]])
+// COMMON: define{{.*}} void @test_parameter_Weak(ptr noundef %[[A:.*]])
+// COMMON: call void @__destructor_{{.*}}(ptr %[[A]])
 
 void test_parameter_Weak(Weak a) {
 }
 
-// COMMON: define{{.*}} void @test_argument_Weak(%[[STRUCT_WEAK]]* noundef %[[A:.*]])
-// COMMON: %[[A_ADDR:.*]] = alloca %[[STRUCT_WEAK]]*
+// COMMON: define{{.*}} void @test_argument_Weak(ptr noundef %[[A:.*]])
+// COMMON: %[[A_ADDR:.*]] = alloca ptr
 // COMMON: %[[AGG_TMP:.*]] = alloca %[[STRUCT_WEAK]]
-// COMMON: store %[[STRUCT_WEAK]]* %[[A]], %[[STRUCT_WEAK]]** %[[A_ADDR]]
-// COMMON: %[[V0:.*]] = load %[[STRUCT_WEAK]]*, %[[STRUCT_WEAK]]** %[[A_ADDR]]
-// COMMON: %[[V1:.*]] = bitcast %[[STRUCT_WEAK]]* %[[AGG_TMP]] to i8**
-// COMMON: %[[V2:.*]] = bitcast %[[STRUCT_WEAK]]* %[[V0]] to i8**
-// COMMON: call void @__copy_constructor_{{.*}}(i8** %[[V1]], i8** %[[V2]])
-// COMMON: call void @calleeWeak(%[[STRUCT_WEAK]]* noundef %[[AGG_TMP]])
+// COMMON: store ptr %[[A]], ptr %[[A_ADDR]]
+// COMMON: %[[V0:.*]] = load ptr, ptr %[[A_ADDR]]
+// COMMON: call void @__copy_constructor_{{.*}}(ptr %[[AGG_TMP]], ptr %[[V0]])
+// COMMON: call void @calleeWeak(ptr noundef %[[AGG_TMP]])
 // COMMON-NEXT: ret
 
 void test_argument_Weak(Weak *a) {
   calleeWeak(*a);
 }
 
-// COMMON: define{{.*}} void @test_return_Weak(%[[STRUCT_WEAK]]* noalias sret(%[[STRUCT_WEAK]]) align {{.*}} %[[AGG_RESULT:.*]], %[[STRUCT_WEAK]]* noundef %[[A:.*]])
-// COMMON: %[[A_ADDR:.*]] = alloca %[[STRUCT_WEAK]]*
-// COMMON: store %[[STRUCT_WEAK]]* %[[A]], %[[STRUCT_WEAK]]** %[[A_ADDR]]
-// COMMON: %[[V0:.*]] = load %[[STRUCT_WEAK]]*, %[[STRUCT_WEAK]]** %[[A_ADDR]]
-// COMMON: %[[V1:.*]] = bitcast %[[STRUCT_WEAK]]* %[[AGG_RESULT]] to i8**
-// COMMON: %[[V2:.*]] = bitcast %[[STRUCT_WEAK]]* %[[V0]] to i8**
-// COMMON: call void @__copy_constructor_{{.*}}(i8** %[[V1]], i8** %[[V2]])
+// COMMON: define{{.*}} void @test_return_Weak(ptr noalias sret(%[[STRUCT_WEAK]]) align {{.*}} %[[AGG_RESULT:.*]], ptr noundef %[[A:.*]])
+// COMMON: %[[A_ADDR:.*]] = alloca ptr
+// COMMON: store ptr %[[A]], ptr %[[A_ADDR]]
+// COMMON: %[[V0:.*]] = load ptr, ptr %[[A_ADDR]]
+// COMMON: call void @__copy_constructor_{{.*}}(ptr %[[AGG_RESULT]], ptr %[[V0]])
 // COMMON: ret void
 
 Weak test_return_Weak(Weak *a) {
@@ -200,11 +164,10 @@ Weak test_return_Weak(Weak *a) {
 // COMMON: %[[AGG_TMP:.*]] = alloca %[[STRUCT_WEAK]]
 // COMMON: br i1
 
-// COMMON: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*, %[[STRUCT_WEAK]]*)*)({{.*}}, %[[STRUCT_WEAK]]* noundef %[[AGG_TMP]])
+// COMMON: call void @objc_msgSend({{.*}}, ptr noundef %[[AGG_TMP]])
 // COMMON: br
 
-// COMMON: %[[V6:.*]] = bitcast %[[STRUCT_WEAK]]* %[[AGG_TMP]] to i8**
-// COMMON: call void @__destructor_{{.*}}(i8** %[[V6]])
+// COMMON: call void @__destructor_{{.*}}(ptr %[[AGG_TMP]])
 // COMMON: br
 
 void test_null_receiver(C *c) {

diff  --git a/clang/test/CodeGenObjC/x86_64-struct-return-gc.m b/clang/test/CodeGenObjC/x86_64-struct-return-gc.m
index 31379c4ded514..6bc2929c7596f 100644
--- a/clang/test/CodeGenObjC/x86_64-struct-return-gc.m
+++ b/clang/test/CodeGenObjC/x86_64-struct-return-gc.m
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 -fobjc-gc -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 -fobjc-gc -emit-llvm -o - %s | FileCheck %s
 struct Coerce {
   id a;
 };
@@ -9,8 +9,8 @@
 void Coerce_test(void) {
   struct Coerce c;
   
-  // CHECK: call i8* @coerce_func
-  // CHECK: call i8* @objc_memmove_collectable(
+  // CHECK: call ptr @coerce_func
+  // CHECK: call ptr @objc_memmove_collectable(
   c = coerce_func();
 }
 
@@ -25,7 +25,7 @@ void Coerce_test(void) {
 void Indirect_test(void) {
   struct Indirect i;
   
-  // CHECK: call void @indirect_func(%struct.Indirect* sret
-  // CHECK: call i8* @objc_memmove_collectable(
+  // CHECK: call void @indirect_func(ptr sret
+  // CHECK: call ptr @objc_memmove_collectable(
   i = indirect_func();
 }

diff  --git a/clang/test/CodeGenObjCXX/arc-cxx11-member-init.mm b/clang/test/CodeGenObjCXX/arc-cxx11-member-init.mm
index 28127e3cba973..f28bf670e5585 100644
--- a/clang/test/CodeGenObjCXX/arc-cxx11-member-init.mm
+++ b/clang/test/CodeGenObjCXX/arc-cxx11-member-init.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers  -triple x86_64-apple-darwin10 -fobjc-arc -std=c++11 -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1  -triple x86_64-apple-darwin10 -fobjc-arc -std=c++11 -emit-llvm -o - %s | FileCheck %s
 // rdar://16299964
   
 @interface NSObject
@@ -22,13 +22,11 @@ - (void)applicationDidFinishLaunching
 }
 @end
 
-// CHECK: [[mClipData:%.*]] = getelementptr inbounds %class.XClipboardDataSet, %class.XClipboardDataSet*
-// CHECK: [[CLS:%.*]] = load %struct._class_t*, %struct._class_t** @"OBJC_CLASSLIST_REFERENCES_$_"
-// CHECK: [[RECV:%.*]] = bitcast %struct._class_t* [[CLS]] to i8*
-// CHECK: [[SEL:%.*]] = load i8*, i8** @OBJC_SELECTOR_REFERENCES_
-// CHECK: [[CALL:%.*]] = call noundef i8* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i8* (i8*, i8*)*)(i8* noundef [[RECV]], i8* noundef [[SEL]])
-// CHECK: [[THREE:%.*]] = bitcast i8* [[CALL]] to [[T:%.*]]*
-// CHECK: store [[T]]* [[THREE]], [[T]]** [[mClipData]], align 8
+// CHECK: [[mClipData:%.*]] = getelementptr inbounds %class.XClipboardDataSet, ptr
+// CHECK: [[CLS:%.*]] = load ptr, ptr @"OBJC_CLASSLIST_REFERENCES_$_"
+// CHECK: [[SEL:%.*]] = load ptr, ptr @OBJC_SELECTOR_REFERENCES_
+// CHECK: [[CALL:%.*]] = call noundef ptr @objc_msgSend(ptr noundef [[CLS]], ptr noundef [[SEL]])
+// CHECK: store ptr [[CALL]], ptr [[mClipData]], align 8
 
 // rdar://18950072
 struct Butt { };
@@ -42,4 +40,4 @@ @interface Foo {
 @end
 @implementation Foo
 @end
-// CHECK-NOT: define internal noundef i8* @"\01-[Foo .cxx_construct
+// CHECK-NOT: define internal noundef ptr @"\01-[Foo .cxx_construct

diff  --git a/clang/test/CodeGenObjCXX/arc-forwarded-lambda-call.mm b/clang/test/CodeGenObjCXX/arc-forwarded-lambda-call.mm
index 5f85b9617294d..523370c88a9f1 100644
--- a/clang/test/CodeGenObjCXX/arc-forwarded-lambda-call.mm
+++ b/clang/test/CodeGenObjCXX/arc-forwarded-lambda-call.mm
@@ -1,13 +1,13 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-macosx10.12.0 -emit-llvm -disable-llvm-passes -O3 -fblocks -fobjc-arc -fobjc-runtime-has-weak -std=c++11 -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-macosx10.12.0 -emit-llvm -disable-llvm-passes -O3 -fblocks -fobjc-arc -fobjc-runtime-has-weak -std=c++11 -o - %s | FileCheck %s
 
 void test0(id x) {
   extern void test0_helper(id (^)(void));
   test0_helper([=]() { return x; });
-  // CHECK-LABEL: define internal noundef i8* @___Z5test0P11objc_object_block_invoke
-  // CHECK: [[T0:%.*]] = call noundef i8* @"_ZZ5test0P11objc_objectENK3$_0clEv"{{.*}} [ "clang.arc.attachedcall"(i8* (i8*)* @llvm.objc.retainAutoreleasedReturnValue) ]
-  // CHECK-NEXT: call void (...) @llvm.objc.clang.arc.noop.use(i8* [[T0]])
-  // CHECK-NEXT: [[T2:%.*]] = tail call i8* @llvm.objc.autoreleaseReturnValue(i8* [[T0]])
-  // CHECK-NEXT: ret i8* [[T2]]
+  // CHECK-LABEL: define internal noundef ptr @___Z5test0P11objc_object_block_invoke
+  // CHECK: [[T0:%.*]] = call noundef ptr @"_ZZ5test0P11objc_objectENK3$_0clEv"{{.*}} [ "clang.arc.attachedcall"(ptr @llvm.objc.retainAutoreleasedReturnValue) ]
+  // CHECK-NEXT: call void (...) @llvm.objc.clang.arc.noop.use(ptr [[T0]])
+  // CHECK-NEXT: [[T2:%.*]] = tail call ptr @llvm.objc.autoreleaseReturnValue(ptr [[T0]])
+  // CHECK-NEXT: ret ptr [[T2]]
 }
 
 // Check that the delegating block invoke function doesn't destruct the Weak
@@ -26,11 +26,11 @@ void test0(id x) {
 void test1() {
   extern void test1_helper(id (*)(void));
   test1_helper([](){ return test1_rv; });
-  // CHECK-LABEL: define internal noundef i8* @"_ZZ5test1vEN3$_08__invokeEv"
-  // CHECK: [[T0:%.*]] = call noundef i8* @"_ZZ5test1vENK3$_0clEv"{{.*}} [ "clang.arc.attachedcall"(i8* (i8*)* @llvm.objc.retainAutoreleasedReturnValue) ]
-  // CHECK-NEXT: call void (...) @llvm.objc.clang.arc.noop.use(i8* [[T0]])
-  // CHECK-NEXT: [[T2:%.*]] = tail call i8* @llvm.objc.autoreleaseReturnValue(i8* [[T0]])
-  // CHECK-NEXT: ret i8* [[T2]]
+  // CHECK-LABEL: define internal noundef ptr @"_ZZ5test1vEN3$_08__invokeEv"
+  // CHECK: [[T0:%.*]] = call noundef ptr @"_ZZ5test1vENK3$_0clEv"{{.*}} [ "clang.arc.attachedcall"(ptr @llvm.objc.retainAutoreleasedReturnValue) ]
+  // CHECK-NEXT: call void (...) @llvm.objc.clang.arc.noop.use(ptr [[T0]])
+  // CHECK-NEXT: [[T2:%.*]] = tail call ptr @llvm.objc.autoreleaseReturnValue(ptr [[T0]])
+  // CHECK-NEXT: ret ptr [[T2]]
 }
 
 struct Weak {

diff  --git a/clang/test/CodeGenObjCXX/arc-indirect.mm b/clang/test/CodeGenObjCXX/arc-indirect.mm
index 9aed3d60287e1..c592d898df629 100644
--- a/clang/test/CodeGenObjCXX/arc-indirect.mm
+++ b/clang/test/CodeGenObjCXX/arc-indirect.mm
@@ -1,6 +1,6 @@
-// RUN: %clang_cc1 -no-opaque-pointers -std=c++11 -triple i686-unknown-windows-msvc -fobjc-runtime=gnustep -fobjc-arc -Wno-objc-root-class -emit-llvm -o - %s | FileCheck -check-prefixes CHECK,CHECK-GNUSTEP %s
-// RUN: %clang_cc1 -no-opaque-pointers -std=c++11 -triple i686-unknown-windows-msvc -fobjc-runtime=macosx -fobjc-arc -Wno-objc-root-class -emit-llvm -o - %s | FileCheck -check-prefixes CHECK,CHECK-DARWIN %s
-// RUN: %clang_cc1 -no-opaque-pointers -std=c++11 -triple i686-unknown-windows-msvc -fobjc-runtime=ios -fobjc-arc -Wno-objc-root-class -emit-llvm -o - %s | FileCheck -check-prefixes CHECK,CHECK-DARWIN %s
+// RUN: %clang_cc1 -std=c++11 -triple i686-unknown-windows-msvc -fobjc-runtime=gnustep -fobjc-arc -Wno-objc-root-class -emit-llvm -o - %s | FileCheck -check-prefixes CHECK,CHECK-GNUSTEP %s
+// RUN: %clang_cc1 -std=c++11 -triple i686-unknown-windows-msvc -fobjc-runtime=macosx -fobjc-arc -Wno-objc-root-class -emit-llvm -o - %s | FileCheck -check-prefixes CHECK,CHECK-DARWIN %s
+// RUN: %clang_cc1 -std=c++11 -triple i686-unknown-windows-msvc -fobjc-runtime=ios -fobjc-arc -Wno-objc-root-class -emit-llvm -o - %s | FileCheck -check-prefixes CHECK,CHECK-DARWIN %s
 
 // non trivially copyable, forces inalloca
 struct S {
@@ -15,8 +15,8 @@ - (void)object:(id)obj struct:(S)s {
 }
 @end
 
-// CHECK-GNUSTEP: define internal void @_i_C__object_struct_(<{ %0*, i8*, i8*, %struct.S, [3 x i8] }>* inalloca(<{ %0*, i8*, i8*, %struct.S, [3 x i8] }>) %0)
-// CHECK-DARWIN: define internal void @"\01-[C object:struct:]"(<{ %0*, i8*, i8*, %struct.S, [3 x i8] }>* inalloca(<{ %0*, i8*, i8*, %struct.S, [3 x i8] }>) %0)
-// CHECK: %obj = getelementptr inbounds <{ %0*, i8*, i8*, %struct.S, [3 x i8] }>, <{ %0*, i8*, i8*, %struct.S, [3 x i8] }>* %0, i32 0, i32 2
-// CHECK: %[[INSTANCE:[0-9]+]] = load i8*, i8** %obj, align 4
-// CHECK: call void @llvm.objc.storeStrong(i8** %obj, i8* %[[INSTANCE]])
+// CHECK-GNUSTEP: define internal void @_i_C__object_struct_(ptr inalloca(<{ ptr, ptr, ptr, %struct.S, [3 x i8] }>) %0)
+// CHECK-DARWIN: define internal void @"\01-[C object:struct:]"(ptr inalloca(<{ ptr, ptr, ptr, %struct.S, [3 x i8] }>) %0)
+// CHECK: %obj = getelementptr inbounds <{ ptr, ptr, ptr, %struct.S, [3 x i8] }>, ptr %0, i32 0, i32 2
+// CHECK: %[[INSTANCE:[0-9]+]] = load ptr, ptr %obj, align 4
+// CHECK: call void @llvm.objc.storeStrong(ptr %obj, ptr %[[INSTANCE]])

diff  --git a/clang/test/CodeGenObjCXX/arc-mangle.mm b/clang/test/CodeGenObjCXX/arc-mangle.mm
index 73c7812dfa372..d0fb1030f43dc 100644
--- a/clang/test/CodeGenObjCXX/arc-mangle.mm
+++ b/clang/test/CodeGenObjCXX/arc-mangle.mm
@@ -1,21 +1,21 @@
-// RUN: %clang_cc1 -no-opaque-pointers -fobjc-arc -fobjc-runtime-has-weak -triple %itanium_abi_triple -emit-llvm -fblocks -o - %s | FileCheck %s
-// RUN: %clang_cc1 -no-opaque-pointers -DTEST_UNALIGNED -fms-extensions -fobjc-arc -fobjc-runtime-has-weak -triple %itanium_abi_triple -emit-llvm -fblocks -o - %s | FileCheck %s --check-prefix=UNALIGNED
+// RUN: %clang_cc1 -fobjc-arc -fobjc-runtime-has-weak -triple %itanium_abi_triple -emit-llvm -fblocks -o - %s | FileCheck %s
+// RUN: %clang_cc1 -DTEST_UNALIGNED -fms-extensions -fobjc-arc -fobjc-runtime-has-weak -triple %itanium_abi_triple -emit-llvm -fblocks -o - %s | FileCheck %s --check-prefix=UNALIGNED
 
-// CHECK-LABEL: define {{.*}}void @_Z1fPU8__strongP11objc_object(i8** noundef %0)
+// CHECK-LABEL: define {{.*}}void @_Z1fPU8__strongP11objc_object(ptr noundef %0)
 void f(__strong id *) {}
-// CHECK-LABEL: define {{.*}}void @_Z1fPU6__weakP11objc_object(i8** noundef %0)
+// CHECK-LABEL: define {{.*}}void @_Z1fPU6__weakP11objc_object(ptr noundef %0)
 void f(__weak id *) {}
-// CHECK-LABEL: define {{.*}}void @_Z1fPU15__autoreleasingP11objc_object(i8** noundef %0)
+// CHECK-LABEL: define {{.*}}void @_Z1fPU15__autoreleasingP11objc_object(ptr noundef %0)
 void f(__autoreleasing id *) {}
-// CHECK-LABEL: define {{.*}}void @_Z1fPP11objc_object(i8** noundef %0)
+// CHECK-LABEL: define {{.*}}void @_Z1fPP11objc_object(ptr noundef %0)
 void f(__unsafe_unretained id *) {}
-// CHECK-LABEL: define {{.*}}void @_Z1fPU8__strongKP11objc_object(i8** noundef %0)
+// CHECK-LABEL: define {{.*}}void @_Z1fPU8__strongKP11objc_object(ptr noundef %0)
 void f(const __strong id *) {}
-// CHECK-LABEL: define {{.*}}void @_Z1fPU6__weakKP11objc_object(i8** noundef %0)
+// CHECK-LABEL: define {{.*}}void @_Z1fPU6__weakKP11objc_object(ptr noundef %0)
 void f(const __weak id *) {}
-// CHECK-LABEL: define {{.*}}void @_Z1fPU15__autoreleasingKP11objc_object(i8** noundef %0)
+// CHECK-LABEL: define {{.*}}void @_Z1fPU15__autoreleasingKP11objc_object(ptr noundef %0)
 void f(const __autoreleasing id *) {}
-// CHECK-LABEL: define {{.*}}void @_Z1fPKP11objc_object(i8** noundef %0)
+// CHECK-LABEL: define {{.*}}void @_Z1fPKP11objc_object(ptr noundef %0)
 void f(const __unsafe_unretained id *) {}
 // CHECK-LABEL: define {{.*}}void @_Z1fPFU19ns_returns_retainedP11objc_objectvE
 void f(__attribute__((ns_returns_retained)) id (*fn)()) {}
@@ -35,10 +35,10 @@ void f(void (^)(__attribute__((ns_consumed)) id)) {}
 template void g<const void>(unsigned_c<sizeof(id) + 1> *);
 
 #if TEST_UNALIGNED
-// UNALIGNED-LABEL: define {{.*}}void @_Z1gPU6__weakU11__unalignedP11objc_object(i8** noundef %0)
+// UNALIGNED-LABEL: define {{.*}}void @_Z1gPU6__weakU11__unalignedP11objc_object(ptr noundef %0)
 void g(__weak __unaligned id *) {}
-// UNALIGNED-LABEL: define {{.*}}void @_Z1gPU11__unalignedU8__strongP11objc_object(i8** noundef %0)
+// UNALIGNED-LABEL: define {{.*}}void @_Z1gPU11__unalignedU8__strongP11objc_object(ptr noundef %0)
 void g(__strong __unaligned id *) {}
-// UNALIGNED-LABEL: define {{.*}}void @_Z1gPU11__unalignedU15__autoreleasingP11objc_object(i8** noundef %0)
+// UNALIGNED-LABEL: define {{.*}}void @_Z1gPU11__unalignedU15__autoreleasingP11objc_object(ptr noundef %0)
 void g(__autoreleasing __unaligned id *) {}
 #endif // TEST_UNALIGNED

diff  --git a/clang/test/CodeGenObjCXX/arc-marker-funclet.mm b/clang/test/CodeGenObjCXX/arc-marker-funclet.mm
index 7ab2146a1779f..58d6b90a36565 100644
--- a/clang/test/CodeGenObjCXX/arc-marker-funclet.mm
+++ b/clang/test/CodeGenObjCXX/arc-marker-funclet.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-msvc -fobjc-runtime=ios-6.0 -fobjc-arc \
+// RUN: %clang_cc1 -triple i686-unknown-windows-msvc -fobjc-runtime=ios-6.0 -fobjc-arc \
 // RUN:   -fexceptions -fcxx-exceptions -emit-llvm -o - %s | FileCheck %s
 
 id f();
@@ -10,7 +10,7 @@ void g() {
   }
 }
 
-// CHECK: call noundef i8* @"?f@@YAPAUobjc_object@@XZ"() [ "funclet"(token %1) ]
+// CHECK: call noundef ptr @"?f@@YAPAUobjc_object@@XZ"() [ "funclet"(token %1) ]
 // CHECK-NEXT: call void asm sideeffect "movl{{.*}}%ebp, %ebp{{.*}}", ""() [ "funclet"(token %1) ]
 
 // The corresponding f() call was invoked from the entry basic block.

diff  --git a/clang/test/CodeGenObjCXX/arc-pseudo-destructors.mm b/clang/test/CodeGenObjCXX/arc-pseudo-destructors.mm
index 900588901b138..c64236800f4b1 100644
--- a/clang/test/CodeGenObjCXX/arc-pseudo-destructors.mm
+++ b/clang/test/CodeGenObjCXX/arc-pseudo-destructors.mm
@@ -1,21 +1,21 @@
-// RUN: %clang_cc1 -no-opaque-pointers -fobjc-arc -fobjc-runtime-has-weak -fblocks -triple x86_64-apple-darwin10.0.0 -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -fobjc-arc -fobjc-runtime-has-weak -fblocks -triple x86_64-apple-darwin10.0.0 -emit-llvm -o - %s | FileCheck %s
 
 // CHECK-LABEL: define{{.*}} void @_Z28test_objc_object_pseudo_dtorPU8__strongP11objc_objectPU6__weakS0_
 void test_objc_object_pseudo_dtor(__strong id *ptr, __weak id *wptr) {
-  // CHECK: load i8**, i8***
-  // CHECK-NEXT: load i8*, i8**
+  // CHECK: load ptr, ptr
+  // CHECK-NEXT: load ptr, ptr
   // CHECK-NEXT: call void @llvm.objc.release
   ptr->~id();
 
-  // CHECK: call void @llvm.objc.destroyWeak(i8** {{%.*}})
+  // CHECK: call void @llvm.objc.destroyWeak(ptr {{%.*}})
   wptr->~id();
 
-  // CHECK: load i8**, i8***
-  // CHECK-NEXT: load i8*, i8**
+  // CHECK: load ptr, ptr
+  // CHECK-NEXT: load ptr, ptr
   // CHECK-NEXT: call void @llvm.objc.release
   (*ptr).~id();
 
-  // CHECK: call void @llvm.objc.destroyWeak(i8** {{%.*}})
+  // CHECK: call void @llvm.objc.destroyWeak(ptr {{%.*}})
   (*wptr).~id();
   // CHECK: ret void
 }

diff  --git a/clang/test/CodeGenObjCXX/arc-rv-attr.mm b/clang/test/CodeGenObjCXX/arc-rv-attr.mm
index 4e40afd0798fc..c6a2ffb06bf34 100644
--- a/clang/test/CodeGenObjCXX/arc-rv-attr.mm
+++ b/clang/test/CodeGenObjCXX/arc-rv-attr.mm
@@ -1,10 +1,10 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple arm64-apple-ios9 -fobjc-runtime=ios-9.0 -fobjc-arc -std=c++11 -O -disable-llvm-passes -emit-llvm -o - %s | FileCheck %s -check-prefix=CHECK
+// RUN: %clang_cc1 -triple arm64-apple-ios9 -fobjc-runtime=ios-9.0 -fobjc-arc -std=c++11 -O -disable-llvm-passes -emit-llvm -o - %s | FileCheck %s -check-prefix=CHECK
 
 id foo(void);
 
 // CHECK-LABEL: define{{.*}} void @_Z14test_list_initv(
-// CHECK: %[[CALL1:.*]] = call noundef i8* @_Z3foov() [ "clang.arc.attachedcall"(i8* (i8*)* @llvm.objc.retainAutoreleasedReturnValue) ]
-// CHECK: call i8* @llvm.objc.retain(i8* %[[CALL1]])
+// CHECK: %[[CALL1:.*]] = call noundef ptr @_Z3foov() [ "clang.arc.attachedcall"(ptr @llvm.objc.retainAutoreleasedReturnValue) ]
+// CHECK: call ptr @llvm.objc.retain(ptr %[[CALL1]])
 
 void test_list_init() {
   auto t = id{foo()};

diff  --git a/clang/test/CodeGenObjCXX/arc-special-member-functions.mm b/clang/test/CodeGenObjCXX/arc-special-member-functions.mm
index b95789eea162e..2cda52e7f7ee9 100644
--- a/clang/test/CodeGenObjCXX/arc-special-member-functions.mm
+++ b/clang/test/CodeGenObjCXX/arc-special-member-functions.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -std=c++11 -fobjc-arc -fblocks -triple x86_64-apple-darwin10.0.0 -fobjc-runtime-has-weak -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -std=c++11 -fobjc-arc -fblocks -triple x86_64-apple-darwin10.0.0 -fobjc-runtime-has-weak -emit-llvm -o - %s | FileCheck %s
 
 struct ObjCMember {
   id member;
@@ -13,7 +13,7 @@
 };
 
 // CHECK: %[[STRUCT_CONTAINSWEAK:.*]] = type { %[[STRUCT_WEAK:.*]] }
-// CHECK: %[[STRUCT_WEAK]] = type { i8* }
+// CHECK: %[[STRUCT_WEAK]] = type { ptr }
 
 // The Weak object that is passed is destructed in this constructor.
 
@@ -151,17 +151,14 @@ void test_ObjCBlockMember_copy_assign(ObjCBlockMember m1, ObjCBlockMember m2) {
 }
 
 // Implicitly-generated copy assignment operator for ObjCBlockMember
-// CHECK:    define linkonce_odr noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) {{%.*}}* @_ZN15ObjCBlockMemberaSERKS_(
-// CHECK:      [[T0:%.*]] = getelementptr inbounds [[T:%.*]], [[T:%.*]]* {{%.*}}, i32 0, i32 0
-// CHECK-NEXT: [[T1:%.*]] = load i32 (i32)*, i32 (i32)** [[T0]], align 8
-// CHECK-NEXT: [[T2:%.*]] = bitcast i32 (i32)* [[T1]] to i8*
-// CHECK-NEXT: [[T3:%.*]] = call i8* @llvm.objc.retainBlock(i8* [[T2]])
-// CHECK-NEXT: [[T4:%.*]] = bitcast i8* [[T3]] to i32 (i32)*
-// CHECK-NEXT: [[T5:%.*]] = getelementptr inbounds [[T]], [[T]]* {{%.*}}, i32 0, i32 0
-// CHECK-NEXT: [[T6:%.*]] = load i32 (i32)*, i32 (i32)** [[T5]], align 8
-// CHECK-NEXT: store i32 (i32)* [[T4]], i32 (i32)** [[T5]]
-// CHECK-NEXT: [[T7:%.*]] = bitcast i32 (i32)* [[T6]] to i8*
-// CHECK-NEXT: call void @llvm.objc.release(i8* [[T7]])
+// CHECK:    define linkonce_odr noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) ptr @_ZN15ObjCBlockMemberaSERKS_(
+// CHECK:      [[T0:%.*]] = getelementptr inbounds [[T:%.*]], ptr {{%.*}}, i32 0, i32 0
+// CHECK-NEXT: [[T1:%.*]] = load ptr, ptr [[T0]], align 8
+// CHECK-NEXT: [[T3:%.*]] = call ptr @llvm.objc.retainBlock(ptr [[T1]])
+// CHECK-NEXT: [[T5:%.*]] = getelementptr inbounds [[T]], ptr {{%.*}}, i32 0, i32 0
+// CHECK-NEXT: [[T6:%.*]] = load ptr, ptr [[T5]], align 8
+// CHECK-NEXT: store ptr [[T3]], ptr [[T5]]
+// CHECK-NEXT: call void @llvm.objc.release(ptr [[T6]])
 // CHECK-NEXT: ret
 
 // Check that the Weak object passed to this constructor is not destructed after
@@ -174,7 +171,7 @@ void test_ObjCBlockMember_copy_assign(ObjCBlockMember m1, ObjCBlockMember m2) {
 // Implicitly-generated default constructor for ObjCMember
 // CHECK-LABEL: define linkonce_odr void @_ZN10ObjCMemberC2Ev
 // CHECK-NOT: objc_release
-// CHECK: store i8* null
+// CHECK: store ptr null
 // CHECK-NEXT: ret void
 
 // Implicitly-generated destructor for ObjCMember
@@ -185,31 +182,31 @@ void test_ObjCBlockMember_copy_assign(ObjCBlockMember m1, ObjCBlockMember m2) {
 // Implicitly-generated copy constructor for ObjCMember
 // CHECK-LABEL: define linkonce_odr void @_ZN10ObjCMemberC2ERKS_
 // CHECK-NOT: objc_release
-// CHECK: call i8* @llvm.objc.retain
-// CHECK-NEXT: store i8*
+// CHECK: call ptr @llvm.objc.retain
+// CHECK-NEXT: store ptr
 // CHECK-NEXT: ret void
 
 // Implicitly-generated default constructor for ObjCArrayMember
 // CHECK-LABEL: define linkonce_odr void @_ZN15ObjCArrayMemberC2Ev
-// CHECK: call void @llvm.memset.p0i8.i64
+// CHECK: call void @llvm.memset.p0.i64
 // CHECK: ret
 
 // Implicitly-generated destructor for ObjCArrayMember
 // CHECK-LABEL:    define linkonce_odr void @_ZN15ObjCArrayMemberD2Ev
-// CHECK:      [[BEGIN:%.*]] = getelementptr inbounds [2 x [3 x i8*]], [2 x [3 x i8*]]*
-// CHECK-NEXT: [[END:%.*]] = getelementptr inbounds i8*, i8** [[BEGIN]], i64 6
+// CHECK:      [[BEGIN:%.*]] = getelementptr inbounds [2 x [3 x ptr]], ptr
+// CHECK-NEXT: [[END:%.*]] = getelementptr inbounds ptr, ptr [[BEGIN]], i64 6
 // CHECK-NEXT: br label
-// CHECK:      [[PAST:%.*]] = phi i8** [ [[END]], {{%.*}} ], [ [[CUR:%.*]], {{%.*}} ]
-// CHECK-NEXT: [[CUR]] = getelementptr inbounds i8*, i8** [[PAST]], i64 -1
-// CHECK-NEXT: call void @llvm.objc.storeStrong(i8** [[CUR]], i8* null)
-// CHECK-NEXT: [[T1:%.*]] = icmp eq i8** [[CUR]], [[BEGIN]]
+// CHECK:      [[PAST:%.*]] = phi ptr [ [[END]], {{%.*}} ], [ [[CUR:%.*]], {{%.*}} ]
+// CHECK-NEXT: [[CUR]] = getelementptr inbounds ptr, ptr [[PAST]], i64 -1
+// CHECK-NEXT: call void @llvm.objc.storeStrong(ptr [[CUR]], ptr null)
+// CHECK-NEXT: [[T1:%.*]] = icmp eq ptr [[CUR]], [[BEGIN]]
 // CHECK-NEXT: br i1 [[T1]],
 // CHECK:      ret void
 
 // Implicitly-generated copy constructor for ObjCArrayMember
 // CHECK-LABEL: define linkonce_odr void @_ZN15ObjCArrayMemberC2ERKS_
-// CHECK: call i8* @llvm.objc.retain
-// CHECK-NEXT: store i8*
+// CHECK: call ptr @llvm.objc.retain
+// CHECK-NEXT: store ptr
 // CHECK: br i1
 // CHECK: ret
 
@@ -220,11 +217,11 @@ void test_ObjCBlockMember_copy_assign(ObjCBlockMember m1, ObjCBlockMember m2) {
 
 // Implicitly-generated destructor for ObjCBlockMember
 // CHECK-LABEL: define linkonce_odr void @_ZN15ObjCBlockMemberD2Ev
-// CHECK: call void @llvm.objc.storeStrong(i8*
+// CHECK: call void @llvm.objc.storeStrong(ptr
 // CHECK: ret
 
 // Implicitly-generated copy constructor for ObjCBlockMember
 // CHECK-LABEL: define linkonce_odr void @_ZN15ObjCBlockMemberC2ERKS_
-// CHECK: call i8* @llvm.objc.retainBlock
+// CHECK: call ptr @llvm.objc.retainBlock
 // CHECK: ret
 

diff  --git a/clang/test/CodeGenObjCXX/arc-weak.mm b/clang/test/CodeGenObjCXX/arc-weak.mm
index b95d434d69df2..07c4e6e71f20b 100644
--- a/clang/test/CodeGenObjCXX/arc-weak.mm
+++ b/clang/test/CodeGenObjCXX/arc-weak.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -fblocks -fobjc-arc -fobjc-runtime-has-weak -std=c++11 -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -fblocks -fobjc-arc -fobjc-runtime-has-weak -std=c++11 -o - %s | FileCheck %s
 
 __attribute((objc_root_class)) @interface A @end
 @interface B : A @end
@@ -9,26 +9,18 @@ void test0(__weak B **src) {
   __weak A *dest = *src;
 }
 // CHECK-LABEL: define{{.*}} void @_Z5test0PU6__weakP1B(
-// CHECK:       [[SRC:%.*]] = alloca [[B:%.*]]**, align 8
-// CHECK:       [[DEST:%.*]] = alloca [[A:%.*]]*, align 8
-// CHECK:       [[T0:%.*]] = load [[B]]**, [[B]]*** [[SRC]], align 8
-// CHECK-NEXT:  [[T1:%.*]] = bitcast [[B]]** [[T0]] to [[A]]**
-// CHECK-NEXT:  [[T2:%.*]] = bitcast [[A]]** [[DEST]] to i8**
-// CHECK-NEXT:  [[T3:%.*]] = bitcast [[A]]** [[T1]] to i8**
-// CHECK-NEXT:  call void @llvm.objc.copyWeak(i8** [[T2]], i8** [[T3]])
-// CHECK-NEXT:  [[T0:%.*]] = bitcast [[A]]** [[DEST]] to i8**
-// CHECK:       call void @llvm.objc.destroyWeak(i8** [[T0]])
+// CHECK:       [[SRC:%.*]] = alloca ptr, align 8
+// CHECK:       [[DEST:%.*]] = alloca ptr, align 8
+// CHECK:       [[T0:%.*]] = load ptr, ptr [[SRC]], align 8
+// CHECK-NEXT:  call void @llvm.objc.copyWeak(ptr [[DEST]], ptr [[T0]])
+// CHECK:       call void @llvm.objc.destroyWeak(ptr [[DEST]])
 
 void test1(__weak B **src) {
   __weak A *dest = static_cast<__weak B*&&>(*src);
 }
 // CHECK-LABEL: define{{.*}} void @_Z5test1PU6__weakP1B(
-// CHECK:       [[SRC:%.*]] = alloca [[B:%.*]]**, align 8
-// CHECK:       [[DEST:%.*]] = alloca [[A:%.*]]*, align 8
-// CHECK:       [[T0:%.*]] = load [[B]]**, [[B]]*** [[SRC]], align 8
-// CHECK-NEXT:  [[T1:%.*]] = bitcast [[B]]** [[T0]] to [[A]]**
-// CHECK-NEXT:  [[T2:%.*]] = bitcast [[A]]** [[DEST]] to i8**
-// CHECK-NEXT:  [[T3:%.*]] = bitcast [[A]]** [[T1]] to i8**
-// CHECK-NEXT:  call void @llvm.objc.moveWeak(i8** [[T2]], i8** [[T3]])
-// CHECK-NEXT:  [[T0:%.*]] = bitcast [[A]]** [[DEST]] to i8**
-// CHECK:       call void @llvm.objc.destroyWeak(i8** [[T0]])
+// CHECK:       [[SRC:%.*]] = alloca ptr, align 8
+// CHECK:       [[DEST:%.*]] = alloca ptr, align 8
+// CHECK:       [[T0:%.*]] = load ptr, ptr [[SRC]], align 8
+// CHECK-NEXT:  call void @llvm.objc.moveWeak(ptr [[DEST]], ptr [[T0]])
+// CHECK:       call void @llvm.objc.destroyWeak(ptr [[DEST]])

diff  --git a/clang/test/CodeGenObjCXX/auto-release-result-assert.mm b/clang/test/CodeGenObjCXX/auto-release-result-assert.mm
index 96f45e4e1f84a..f3d5f4d413f70 100644
--- a/clang/test/CodeGenObjCXX/auto-release-result-assert.mm
+++ b/clang/test/CodeGenObjCXX/auto-release-result-assert.mm
@@ -1,16 +1,16 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -fblocks -fobjc-arc -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -fblocks -fobjc-arc -o - %s | FileCheck %s
 
-// CHECK-LABEL: define{{.*}} %struct.S1* @_Z4foo1i(
-// CHECK: %[[CALL:[a-z0-9]+]] = call noundef %struct.S1* @_Z4foo0i
-// CHECK: ret %struct.S1* %[[CALL]]
+// CHECK-LABEL: define{{.*}} ptr @_Z4foo1i(
+// CHECK: %[[CALL:[a-z0-9]+]] = call noundef ptr @_Z4foo0i
+// CHECK: ret ptr %[[CALL]]
 
-// CHECK-LABEL: define{{.*}} %struct.S1* @_ZN2S22m1Ev(
-// CHECK: %[[CALL:[a-z0-9]+]] = call noundef %struct.S1* @_Z4foo0i
-// CHECK: ret %struct.S1* %[[CALL]]
+// CHECK-LABEL: define{{.*}} ptr @_ZN2S22m1Ev(
+// CHECK: %[[CALL:[a-z0-9]+]] = call noundef ptr @_Z4foo0i
+// CHECK: ret ptr %[[CALL]]
 
-// CHECK-LABEL: define internal noundef %struct.S1* @Block1_block_invoke(
-// CHECK: %[[CALL:[a-z0-9]+]] = call noundef %struct.S1* @_Z4foo0i
-// CHECK: ret %struct.S1* %[[CALL]]
+// CHECK-LABEL: define internal noundef ptr @Block1_block_invoke(
+// CHECK: %[[CALL:[a-z0-9]+]] = call noundef ptr @_Z4foo0i
+// CHECK: ret ptr %[[CALL]]
 
 struct S1;
 

diff  --git a/clang/test/CodeGenObjCXX/block-default-arg.mm b/clang/test/CodeGenObjCXX/block-default-arg.mm
index 228e1d7880018..99c0ac2ea4200 100644
--- a/clang/test/CodeGenObjCXX/block-default-arg.mm
+++ b/clang/test/CodeGenObjCXX/block-default-arg.mm
@@ -1,9 +1,8 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10.0.0 -emit-llvm -o - %s -std=c++11 -fblocks -fobjc-arc | FileCheck  %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10.0.0 -emit-llvm -o - %s -std=c++11 -fblocks -fobjc-arc | FileCheck  %s
 
-// CHECK: define internal void @___Z16test_default_argi_block_invoke(i8* noundef %[[BLOCK_DESCRIPTOR:.*]])
-// CHECK: %[[BLOCK:.*]] = bitcast i8* %[[BLOCK_DESCRIPTOR]] to <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i32 }>*
-// CHECK: %[[BLOCK_CAPTURE_ADDR:.*]] = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i32 }>, <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i32 }>* %[[BLOCK]], i32 0, i32 5
-// CHECK: %[[V0:.*]] = load i32, i32* %[[BLOCK_CAPTURE_ADDR]]
+// CHECK: define internal void @___Z16test_default_argi_block_invoke(ptr noundef %[[BLOCK_DESCRIPTOR:.*]])
+// CHECK: %[[BLOCK_CAPTURE_ADDR:.*]] = getelementptr inbounds <{ ptr, i32, i32, ptr, ptr, i32 }>, ptr %[[BLOCK_DESCRIPTOR]], i32 0, i32 5
+// CHECK: %[[V0:.*]] = load i32, ptr %[[BLOCK_CAPTURE_ADDR]]
 // CHECK: call void @_Z4foo1i(i32 noundef %[[V0]])
 
 void foo1(int);

diff  --git a/clang/test/CodeGenObjCXX/boxing.mm b/clang/test/CodeGenObjCXX/boxing.mm
index 1e477d97490bd..b9b9636daade4 100644
--- a/clang/test/CodeGenObjCXX/boxing.mm
+++ b/clang/test/CodeGenObjCXX/boxing.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -o - %s | FileCheck %s
 
 @interface NSNumber
 + (id)numberWithInt:(int)n;
@@ -8,5 +8,5 @@ + (id)numberWithInt:(int)n;
 int m = (@(n++), 0);
 
 // CHECK: define {{.*}} @__cxx_global_var_init()
-// CHECK: load i32, i32* @n
-// CHECK: store i32 %{{.*}}, i32* @n
+// CHECK: load i32, ptr @n
+// CHECK: store i32 %{{.*}}, ptr @n

diff  --git a/clang/test/CodeGenObjCXX/catch-id-type.mm b/clang/test/CodeGenObjCXX/catch-id-type.mm
index bddd0c18114d7..910ccfba6f7e9 100644
--- a/clang/test/CodeGenObjCXX/catch-id-type.mm
+++ b/clang/test/CodeGenObjCXX/catch-id-type.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple i386-apple-macosx10.6.6 -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -fobjc-exceptions -fcxx-exceptions -fexceptions -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple i386-apple-macosx10.6.6 -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -fobjc-exceptions -fcxx-exceptions -fexceptions -o - %s | FileCheck %s
 // rdar://8940528
 
 @interface ns_array
@@ -30,10 +30,10 @@ id FUNC() {
     }
     catch( id error )
     { 
-      // CHECK:      landingpad { i8*, i32 }
-      // CHECK-NEXT:   catch i8* bitcast ({ i8*, i8*, i32, i8* }* @_ZTIPU11objcproto1P4INTF to i8*)
-      // CHECK-NEXT:   catch i8* bitcast ({ i8*, i8*, i32, i8* }* @_ZTIP11objc_object to i8*)
-      // CHECK-NEXT:   catch i8* bitcast ({ i8*, i8*, i32, i8* }* @_ZTIP10objc_class to i8*)
+      // CHECK:      landingpad { ptr, i32 }
+      // CHECK-NEXT:   catch ptr @_ZTIPU11objcproto1P4INTF
+      // CHECK-NEXT:   catch ptr @_ZTIP11objc_object
+      // CHECK-NEXT:   catch ptr @_ZTIP10objc_class
       error = error; 
       groups = [ns_array array]; 
     }

diff  --git a/clang/test/CodeGenObjCXX/copy.mm b/clang/test/CodeGenObjCXX/copy.mm
index 7d1fc70d14a2c..966c210fe12b1 100644
--- a/clang/test/CodeGenObjCXX/copy.mm
+++ b/clang/test/CodeGenObjCXX/copy.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -o - %s | FileCheck %s
 
 // rdar://problem/9158302
 // This should not use a memmove_collectable in non-GC mode.
@@ -7,15 +7,12 @@
     id x;
   };
 
-  // CHECK:    define{{.*}} [[A:%.*]]* @_ZN5test04testENS_1AE(
+  // CHECK:    define{{.*}} ptr @_ZN5test04testENS_1AE(
   // CHECK:      alloca
   // CHECK-NEXT: getelementptr
   // CHECK-NEXT: store
-  // CHECK-NEXT: [[CALL:%.*]] = call noalias noundef nonnull i8* @_Znwm(
-  // CHECK-NEXT: bitcast
-  // CHECK-NEXT: bitcast
-  // CHECK-NEXT: bitcast
-  // CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i64(
+  // CHECK-NEXT: [[CALL:%.*]] = call noalias noundef nonnull ptr @_Znwm(
+  // CHECK-NEXT: call void @llvm.memcpy.p0.p0.i64(
   // CHECK-NEXT: ret
   A *test(A a) {
     return new A(a);

diff  --git a/clang/test/CodeGenObjCXX/debug-info-line.mm b/clang/test/CodeGenObjCXX/debug-info-line.mm
index 0d5505e981830..bb6eaa50d7cca 100644
--- a/clang/test/CodeGenObjCXX/debug-info-line.mm
+++ b/clang/test/CodeGenObjCXX/debug-info-line.mm
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-windows-gnu -fcxx-exceptions -fexceptions -debug-info-kind=line-tables-only -fblocks -emit-llvm %s -o - | FileCheck %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-windows-gnu -fcxx-exceptions -fexceptions -debug-info-kind=line-directives-only -fblocks -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-unknown-windows-gnu -fcxx-exceptions -fexceptions -debug-info-kind=line-tables-only -fblocks -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-unknown-windows-gnu -fcxx-exceptions -fexceptions -debug-info-kind=line-directives-only -fblocks -emit-llvm %s -o - | FileCheck %s
 
 void fn();
 
@@ -16,7 +16,7 @@ void f1() {
   }();
 }
 
-// CHECK-LABEL: define internal {{.*}}i8* @"\01-[TNSObject init]"
+// CHECK-LABEL: define internal {{.*}}ptr @"\01-[TNSObject init]"
 @implementation TNSObject
 - (id)init
 {

diff  --git a/clang/test/CodeGenObjCXX/destroy.mm b/clang/test/CodeGenObjCXX/destroy.mm
index 9e33bb66a6d74..d9f62e64848ec 100644
--- a/clang/test/CodeGenObjCXX/destroy.mm
+++ b/clang/test/CodeGenObjCXX/destroy.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -std=c++11 -triple x86_64-apple-darwin10 -emit-llvm -fobjc-arc -disable-llvm-passes -o - %s | FileCheck %s
+// RUN: %clang_cc1 -std=c++11 -triple x86_64-apple-darwin10 -emit-llvm -fobjc-arc -disable-llvm-passes -o - %s | FileCheck %s
 // rdar://18249673
 
 @class MyObject;
@@ -44,7 +44,7 @@ void test4() {
 // CHECK: call void @_ZN8tderivedIiED1Ev
 
 // CHECK-LABEL: define linkonce_odr void @_ZN7derivedD2Ev
-// CHECK: call void @llvm.objc.storeStrong(i8** {{.*}}, i8* null)
+// CHECK: call void @llvm.objc.storeStrong(ptr {{.*}}, ptr null)
 
 // CHECK-LABEL: define linkonce_odr void @_ZN8tderivedIiED2Ev
-// CHECK: call void @llvm.objc.storeStrong(i8** {{.*}}, i8* null)
+// CHECK: call void @llvm.objc.storeStrong(ptr {{.*}}, ptr null)

diff  --git a/clang/test/CodeGenObjCXX/exception-cxx.mm b/clang/test/CodeGenObjCXX/exception-cxx.mm
index 076d562f26f35..61f3d32127132 100644
--- a/clang/test/CodeGenObjCXX/exception-cxx.mm
+++ b/clang/test/CodeGenObjCXX/exception-cxx.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -x objective-c++ -triple x86_64-apple-darwin10 -emit-llvm -fcxx-exceptions -fexceptions -fobjc-exceptions -o - %s | FileCheck %s
+// RUN: %clang_cc1 -x objective-c++ -triple x86_64-apple-darwin10 -emit-llvm -fcxx-exceptions -fexceptions -fobjc-exceptions -o - %s | FileCheck %s
 
 // rdar://problem/22155434
 namespace test0 {
@@ -9,5 +9,5 @@ void foo() {
       return;
     }
   }
-// CHECK: define{{.*}} void @_ZN5test03fooEv() #0 personality i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*)
+// CHECK: define{{.*}} void @_ZN5test03fooEv() #0 personality ptr @__gxx_personality_v0
 }

diff  --git a/clang/test/CodeGenObjCXX/exceptions.mm b/clang/test/CodeGenObjCXX/exceptions.mm
index ae4e37ae71ae7..26cc48d2c2238 100644
--- a/clang/test/CodeGenObjCXX/exceptions.mm
+++ b/clang/test/CodeGenObjCXX/exceptions.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -x objective-c++ -triple x86_64-apple-darwin10 -emit-llvm -fcxx-exceptions -fexceptions -fobjc-exceptions -o - %s | FileCheck %s
+// RUN: %clang_cc1 -x objective-c++ -triple x86_64-apple-darwin10 -emit-llvm -fcxx-exceptions -fexceptions -fobjc-exceptions -o - %s | FileCheck %s
 
 @interface OCType @end
 void opaque();
@@ -6,14 +6,14 @@ @interface OCType @end
 namespace test0 {
 
   // CHECK-LABEL: define{{.*}} void @_ZN5test03fooEv
-  // CHECK-SAME:  personality i8* bitcast (i32 (...)* @__objc_personality_v0 to i8*)
+  // CHECK-SAME:  personality ptr @__objc_personality_v0
   void foo() {
     try {
       // CHECK: invoke void @_Z6opaquev
       opaque();
     } catch (OCType *T) {
-      // CHECK:      landingpad { i8*, i32 }
-      // CHECK-NEXT:   catch %struct._objc_typeinfo* @"OBJC_EHTYPE_$_OCType"
+      // CHECK:      landingpad { ptr, i32 }
+      // CHECK-NEXT:   catch ptr @"OBJC_EHTYPE_$_OCType"
     }
   }
 }
@@ -30,7 +30,7 @@ void bar() {
     } @catch (id i) {
     }
   }
-// CHECK: invoke void @objc_exception_throw(i8* [[CALL:%.*]]) [[NR:#[0-9]+]]
+// CHECK: invoke void @objc_exception_throw(ptr [[CALL:%.*]]) [[NR:#[0-9]+]]
 // CHECK:          to label [[INVOKECONT1:%.*]] unwind label [[LPAD:%.*]]
 }
 

diff  --git a/clang/test/CodeGenObjCXX/implicit-copy-assign-operator.mm b/clang/test/CodeGenObjCXX/implicit-copy-assign-operator.mm
index 7ddb9273f17ce..78520fa08a589 100644
--- a/clang/test/CodeGenObjCXX/implicit-copy-assign-operator.mm
+++ b/clang/test/CodeGenObjCXX/implicit-copy-assign-operator.mm
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -fobjc-gc -emit-llvm -triple x86_64-apple-darwin10.0.0 -fobjc-runtime=macosx-fragile-10.5 -o - %s | FileCheck %s -check-prefix=CHECK-OBJ
-// RUN: %clang_cc1 -no-opaque-pointers -x c++    -emit-llvm -triple x86_64-apple-darwin10.0.0                                    -o - %s | FileCheck %s -check-prefix=CHECK-CPP
+// RUN: %clang_cc1 -fobjc-gc -emit-llvm -triple x86_64-apple-darwin10.0.0 -fobjc-runtime=macosx-fragile-10.5 -o - %s | FileCheck %s -check-prefix=CHECK-OBJ
+// RUN: %clang_cc1 -x c++    -emit-llvm -triple x86_64-apple-darwin10.0.0                                    -o - %s | FileCheck %s -check-prefix=CHECK-CPP
 #ifdef __OBJC__
 struct A { 
   A &operator=(const A&);
@@ -43,16 +43,16 @@ void test_D(D d1, D d2) {
   d1 = d2;
 }
 
-// CHECK-OBJ-LABEL: define linkonce_odr noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) %struct.D* @_ZN1DaSERS_
+// CHECK-OBJ-LABEL: define linkonce_odr noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) ptr @_ZN1DaSERS_
 // CHECK-OBJ: {{call.*_ZN1AaSERS_}}
 // CHECK-OBJ: {{call.*_ZN1BaSERS_}}
 // CHECK-OBJ: {{call.*_ZN1CaSERKS_}}
-// CHECK-OBJ: {{call void @llvm.memcpy.p0i8.p0i8.i64.*i64 24}}
+// CHECK-OBJ: {{call void @llvm.memcpy.p0.p0.i64.*i64 24}}
 // CHECK-OBJ: {{call.*_ZN1BaSERS_}}
 // CHECK-OBJ: br
 // CHECK-OBJ: {{call.*_ZN1CaSERKS_}}
 // CHECK-OBJ: {{call.*@objc_memmove_collectable}}
-// CHECK-OBJ: {{call void @llvm.memcpy.p0i8.p0i8.i64.*i64 12}}
+// CHECK-OBJ: {{call void @llvm.memcpy.p0.p0.i64.*i64 12}}
 // CHECK-OBJ: call void @_ZN11CopyByValueC1ERKS_
 // CHECK-OBJ: {{call.*_ZN11CopyByValueaSES_}}
 // CHECK-OBJ: ret

diff  --git a/clang/test/CodeGenObjCXX/implicit-copy-constructor.mm b/clang/test/CodeGenObjCXX/implicit-copy-constructor.mm
index c6fcef167e7e2..10e3c85abfc75 100644
--- a/clang/test/CodeGenObjCXX/implicit-copy-constructor.mm
+++ b/clang/test/CodeGenObjCXX/implicit-copy-constructor.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -fobjc-gc -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -fobjc-gc -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 -emit-llvm -o - %s | FileCheck %s
 
 struct A { 
   A();
@@ -41,13 +41,13 @@ void f(D d) {
   D d2(d);
 }
 
-// CHECK-LABEL: define linkonce_odr void @_ZN1DC1ERS_(%struct.D* {{[^,]*}} %this, %struct.D* noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) %0) unnamed_addr
+// CHECK-LABEL: define linkonce_odr void @_ZN1DC1ERS_(ptr {{[^,]*}} %this, ptr noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) %0) unnamed_addr
 // CHECK: call void @_ZN1AC1Ev
 // CHECK: call void @_ZN1CC2ERS_1A
 // CHECK: call void @_ZN1AD1Ev
 // CHECK: call void @_ZN1AC2ERS_
 // CHECK: call void @_ZN1BC2ERS_
-// CHECK: {{call void @llvm.memcpy.p0i8.p0i8.i64.*i64 24}}
+// CHECK: {{call void @llvm.memcpy.p0.p0.i64.*i64 24}}
 // CHECK: call void @_ZN1BC1ERS_
 // CHECK: br label
 // CHECK: call void @_ZN1AC1Ev
@@ -58,7 +58,7 @@ void f(D d) {
 // CHECK: {{icmp eq.*, 2}}
 // CHECK: br i1
 // CHECK: {{call.*@objc_memmove_collectable}}
-// CHECK: {{call void @llvm.memcpy.p0i8.p0i8.i64.*i64 12}}
+// CHECK: {{call void @llvm.memcpy.p0.p0.i64.*i64 12}}
 // CHECK: ret void
 
 

diff  --git a/clang/test/CodeGenObjCXX/inheriting-constructor-cleanup.mm b/clang/test/CodeGenObjCXX/inheriting-constructor-cleanup.mm
index a4ebace4794ad..6756653bd95b8 100644
--- a/clang/test/CodeGenObjCXX/inheriting-constructor-cleanup.mm
+++ b/clang/test/CodeGenObjCXX/inheriting-constructor-cleanup.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-darwin -std=c++11 -fobjc-arc -emit-llvm -o - %s | FileCheck %s --implicit-check-not "call\ "
+// RUN: %clang_cc1 -triple x86_64-darwin -std=c++11 -fobjc-arc -emit-llvm -o - %s | FileCheck %s --implicit-check-not "call\ "
 // rdar://problem/45805151
 
 struct Strong {
@@ -22,22 +22,22 @@ void f() {
   Inheritor({g()});
 }
 // CHECK-LABEL: define{{.*}} void @_Z1fv
-// CHECK:       %[[TMP:.*]] = call noundef i8* @_Z1gv()
-// CHECK:       {{.*}} = notail call i8* @llvm.objc.retainAutoreleasedReturnValue(i8* %[[TMP]])
-// CHECK:       call void (%struct.Base*, i8*, ...) @_ZN4BaseC2E6Strongz(%struct.Base* {{.*}}, i8* {{.*}})
-// CHECK-NEXT:  call void @_ZN9InheritorD1Ev(%struct.Inheritor* {{.*}})
+// CHECK:       %[[TMP:.*]] = call noundef ptr @_Z1gv()
+// CHECK:       {{.*}} = notail call ptr @llvm.objc.retainAutoreleasedReturnValue(ptr %[[TMP]])
+// CHECK:       call void (ptr, ptr, ...) @_ZN4BaseC2E6Strongz(ptr {{.*}}, ptr {{.*}})
+// CHECK-NEXT:  call void @_ZN9InheritorD1Ev(ptr {{.*}})
 
-// CHECK-LABEL: define linkonce_odr void @_ZN4BaseC2E6Strongz(%struct.Base* {{.*}}, i8* {{.*}}, ...)
-// CHECK:       call void @_ZN6StrongD1Ev(%struct.Strong* {{.*}})
+// CHECK-LABEL: define linkonce_odr void @_ZN4BaseC2E6Strongz(ptr {{.*}}, ptr {{.*}}, ...)
+// CHECK:       call void @_ZN6StrongD1Ev(ptr {{.*}})
 
-// CHECK-LABEL: define linkonce_odr void @_ZN9InheritorD1Ev(%struct.Inheritor* {{.*}})
-// CHECK:       call void @_ZN9InheritorD2Ev(%struct.Inheritor* {{.*}})
+// CHECK-LABEL: define linkonce_odr void @_ZN9InheritorD1Ev(ptr {{.*}})
+// CHECK:       call void @_ZN9InheritorD2Ev(ptr {{.*}})
 
-// CHECK-LABEL: define linkonce_odr void @_ZN6StrongD1Ev(%struct.Strong* {{.*}})
-// CHECK:       call void @_ZN6StrongD2Ev(%struct.Strong* {{.*}})
+// CHECK-LABEL: define linkonce_odr void @_ZN6StrongD1Ev(ptr {{.*}})
+// CHECK:       call void @_ZN6StrongD2Ev(ptr {{.*}})
 
-// CHECK-LABEL: define linkonce_odr void @_ZN6StrongD2Ev(%struct.Strong* {{.*}})
-// CHECK:       call void @llvm.objc.storeStrong(i8** {{.*}}, i8* null)
+// CHECK-LABEL: define linkonce_odr void @_ZN6StrongD2Ev(ptr {{.*}})
+// CHECK:       call void @llvm.objc.storeStrong(ptr {{.*}}, ptr null)
 
-// CHECK-LABEL: define linkonce_odr void @_ZN9InheritorD2Ev(%struct.Inheritor* {{.*}})
-// CHECK:       call void @_ZN14NonTrivialDtorD2Ev(%struct.NonTrivialDtor* {{.*}})
+// CHECK-LABEL: define linkonce_odr void @_ZN9InheritorD2Ev(ptr {{.*}})
+// CHECK:       call void @_ZN14NonTrivialDtorD2Ev(ptr {{.*}})

diff  --git a/clang/test/CodeGenObjCXX/lambda-expressions.mm b/clang/test/CodeGenObjCXX/lambda-expressions.mm
index d81603d7b9b12..d50607ca217f2 100644
--- a/clang/test/CodeGenObjCXX/lambda-expressions.mm
+++ b/clang/test/CodeGenObjCXX/lambda-expressions.mm
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10.0.0 -emit-llvm -o - %s -fexceptions -std=c++11 -fblocks -fobjc-arc -fobjc-runtime-has-weak -DWEAK_SUPPORTED | FileCheck -check-prefix=ARC %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10.0.0 -emit-llvm -o - %s -fexceptions -std=c++11 -fblocks | FileCheck -check-prefix=MRC %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10.0.0 -emit-llvm -o - %s -fexceptions -std=c++11 -fblocks -fobjc-arc -fobjc-runtime-has-weak -DWEAK_SUPPORTED | FileCheck -check-prefix=ARC %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10.0.0 -emit-llvm -o - %s -fexceptions -std=c++11 -fblocks | FileCheck -check-prefix=MRC %s
 
 typedef int (^fp)();
 fp f() { auto x = []{ return 3; }; return x; }
@@ -8,34 +8,34 @@
 
 // MRC: @OBJC_METH_VAR_NAME{{.*}} = private unnamed_addr constant [5 x i8] c"copy\00"
 // MRC: @OBJC_METH_VAR_NAME{{.*}} = private unnamed_addr constant [12 x i8] c"autorelease\00"
-// MRC-LABEL: define{{.*}} i32 ()* @_Z1fv(
-// MRC-LABEL: define internal noundef i32 ()* @"_ZZ1fvENK3$_0cvU13block_pointerFivEEv"
-// MRC: store i8* bitcast (i8** @_NSConcreteStackBlock to i8*)
-// MRC: store i8* bitcast (i32 (i8*)* @"___ZZ1fvENK3$_0cvU13block_pointerFivEEv_block_invoke" to i8*)
-// MRC: call noundef i32 ()* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i32 ()* (i8*, i8*)*)
-// MRC: call noundef i32 ()* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i32 ()* (i8*, i8*)*)
-// MRC: ret i32 ()*
-
-// ARC-LABEL: define{{.*}} i32 ()* @_Z1fv(
-// ARC-LABEL: define internal noundef i32 ()* @"_ZZ1fvENK3$_0cvU13block_pointerFivEEv"
-// ARC: store i8* bitcast (i8** @_NSConcreteStackBlock to i8*)
-// ARC: store i8* bitcast (i32 (i8*)* @"___ZZ1fvENK3$_0cvU13block_pointerFivEEv_block_invoke" to i8*)
-// ARC: call i8* @llvm.objc.retainBlock
-// ARC: call i8* @llvm.objc.autoreleaseReturnValue
+// MRC-LABEL: define{{.*}} ptr @_Z1fv(
+// MRC-LABEL: define internal noundef ptr @"_ZZ1fvENK3$_0cvU13block_pointerFivEEv"
+// MRC: store ptr @_NSConcreteStackBlock
+// MRC: store ptr @"___ZZ1fvENK3$_0cvU13block_pointerFivEEv_block_invoke"
+// MRC: call noundef ptr @objc_msgSend
+// MRC: call noundef ptr @objc_msgSend
+// MRC: ret ptr
+
+// ARC-LABEL: define{{.*}} ptr @_Z1fv(
+// ARC-LABEL: define internal noundef ptr @"_ZZ1fvENK3$_0cvU13block_pointerFivEEv"
+// ARC: store ptr @_NSConcreteStackBlock
+// ARC: store ptr @"___ZZ1fvENK3$_0cvU13block_pointerFivEEv_block_invoke"
+// ARC: call ptr @llvm.objc.retainBlock
+// ARC: call ptr @llvm.objc.autoreleaseReturnValue
 
 typedef int (^fp)();
 fp global;
 void f2() { global = []{ return 3; }; }
 
 // MRC: define{{.*}} void @_Z2f2v() [[NUW:#[0-9]+]] {
-// MRC: store i8* bitcast (i32 (i8*)* @___Z2f2v_block_invoke to i8*),
+// MRC: store ptr @___Z2f2v_block_invoke,
 // MRC-NOT: call
 // MRC: ret void
 // ("global" contains a dangling pointer after this function runs.)
 
 // ARC: define{{.*}} void @_Z2f2v() [[NUW:#[0-9]+]] {
-// ARC: store i8* bitcast (i32 (i8*)* @___Z2f2v_block_invoke to i8*),
-// ARC: call i8* @llvm.objc.retainBlock
+// ARC: store ptr @___Z2f2v_block_invoke,
+// ARC: call ptr @llvm.objc.retainBlock
 // ARC: call void @llvm.objc.release
 // ARC-LABEL: define internal noundef i32 @___Z2f2v_block_invoke
 // ARC: call noundef i32 @"_ZZ2f2vENK3$_0clEv
@@ -62,27 +62,27 @@ - (void) test {
 }
 @end
 
-// ARC: define{{.*}} void @_ZN13LambdaCapture4foo1ERi(i32* noundef nonnull align 4 dereferenceable(4) %{{.*}})
-// ARC:   %[[CAPTURE0:.*]] = getelementptr inbounds %[[LAMBDACLASS]], %[[LAMBDACLASS]]* %{{.*}}, i32 0, i32 0
-// ARC:   store i32 %{{.*}}, i32* %[[CAPTURE0]]
+// ARC: define{{.*}} void @_ZN13LambdaCapture4foo1ERi(ptr noundef nonnull align 4 dereferenceable(4) %{{.*}})
+// ARC:   %[[CAPTURE0:.*]] = getelementptr inbounds %[[LAMBDACLASS]], ptr %{{.*}}, i32 0, i32 0
+// ARC:   store i32 %{{.*}}, ptr %[[CAPTURE0]]
 
-// ARC: define internal void @"_ZZN13LambdaCapture4foo1ERiENK3$_0clEv"(%[[LAMBDACLASS]]* {{[^,]*}} %{{.*}})
-// ARC:   %[[BLOCK:.*]] = alloca <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i32 }>
-// ARC:   %[[CAPTURE1:.*]] = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i32 }>, <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i32 }>* %[[BLOCK]], i32 0, i32 5
-// ARC:   store i32 %{{.*}}, i32* %[[CAPTURE1]]
+// ARC: define internal void @"_ZZN13LambdaCapture4foo1ERiENK3$_0clEv"(ptr {{[^,]*}} %{{.*}})
+// ARC:   %[[BLOCK:.*]] = alloca <{ ptr, i32, i32, ptr, ptr, i32 }>
+// ARC:   %[[CAPTURE1:.*]] = getelementptr inbounds <{ ptr, i32, i32, ptr, ptr, i32 }>, ptr %[[BLOCK]], i32 0, i32 5
+// ARC:   store i32 %{{.*}}, ptr %[[CAPTURE1]]
 
 // ARC-LABEL: define internal void @"_ZZ10-[Foo foo]ENK3$_4clEv"(
 // ARC-NOT: @llvm.objc.storeStrong(
 // ARC: ret void
 
 // ARC: define internal void @"___ZZN13LambdaCapture4foo1ERiENK3$_0clEv_block_invoke"
-// ARC:   %[[CAPTURE2:.*]] = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i32 }>, <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i32 }>* %{{.*}}, i32 0, i32 5
-// ARC:   store i32 %{{.*}}, i32* %[[CAPTURE2]]
+// ARC:   %[[CAPTURE2:.*]] = getelementptr inbounds <{ ptr, i32, i32, ptr, ptr, i32 }>, ptr %{{.*}}, i32 0, i32 5
+// ARC:   store i32 %{{.*}}, ptr %[[CAPTURE2]]
 
-// ARC: define internal void @"___ZZN13LambdaCapture4foo1ERiENK3$_0clEv_block_invoke_2"(i8* noundef %{{.*}})
-// ARC:   %[[CAPTURE3:.*]] = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i32 }>, <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i32 }>* %{{.*}}, i32 0, i32 5
-// ARC:   %[[V1:.*]] = load i32, i32* %[[CAPTURE3]]
-// ARC:   store i32 %[[V1]], i32* @_ZN13LambdaCapture1iE
+// ARC: define internal void @"___ZZN13LambdaCapture4foo1ERiENK3$_0clEv_block_invoke_2"(ptr noundef %{{.*}})
+// ARC:   %[[CAPTURE3:.*]] = getelementptr inbounds <{ ptr, i32, i32, ptr, ptr, i32 }>, ptr %{{.*}}, i32 0, i32 5
+// ARC:   %[[V1:.*]] = load i32, ptr %[[CAPTURE3]]
+// ARC:   store i32 %[[V1]], ptr @_ZN13LambdaCapture1iE
 
 namespace LambdaCapture {
   int i;
@@ -100,12 +100,12 @@ void foo1(int &a) {
   }
 }
 
-// ARC-LABEL: define linkonce_odr noundef i32 ()* @_ZZNK13StaticMembersIfE1fMUlvE_clEvENKUlvE_cvU13block_pointerFivEEv
+// ARC-LABEL: define linkonce_odr noundef ptr @_ZZNK13StaticMembersIfE1fMUlvE_clEvENKUlvE_cvU13block_pointerFivEEv
 
 // Check lines for BlockInLambda test below
 // ARC-LABEL: define internal noundef i32 @___ZZN13BlockInLambda1X1fEvENKUlvE_clEv_block_invoke
-// ARC: [[Y:%.*]] = getelementptr inbounds %"struct.BlockInLambda::X", %"struct.BlockInLambda::X"* {{.*}}, i32 0, i32 1
-// ARC-NEXT: [[YVAL:%.*]] = load i32, i32* [[Y]], align 4
+// ARC: [[Y:%.*]] = getelementptr inbounds %"struct.BlockInLambda::X", ptr {{.*}}, i32 0, i32 1
+// ARC-NEXT: [[YVAL:%.*]] = load i32, ptr [[Y]], align 4
 // ARC-NEXT: ret i32 [[YVAL]]
 
 typedef int (^fptr)();

diff  --git a/clang/test/CodeGenObjCXX/mangle-blocks.mm b/clang/test/CodeGenObjCXX/mangle-blocks.mm
index 65ca4a65e0ec1..3110ae7d9b7c3 100644
--- a/clang/test/CodeGenObjCXX/mangle-blocks.mm
+++ b/clang/test/CodeGenObjCXX/mangle-blocks.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -fblocks -o - -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 %s | FileCheck %s
+// RUN: %clang_cc1 -emit-llvm -fblocks -o - -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-fragile-10.5 %s | FileCheck %s
 
 // CHECK-DAG: @_ZZZN26externally_visible_statics1S3fooEiEd_Ub_E1k = linkonce_odr global i32 0
 // CHECK-DAG: @_ZZZN26externally_visible_statics10inlinefuncEvEUb_E1i = linkonce_odr global i32 0
@@ -9,7 +9,7 @@
 
 void foo() {
   // CHECK-LABEL: define internal noundef i32 @___Z3foov_block_invoke
-  // CHECK: call i32 @__cxa_guard_acquire(i64* @_ZGVZZ3foovEUb_E5value
+  // CHECK: call i32 @__cxa_guard_acquire(ptr @_ZGVZZ3foovEUb_E5value
   (void)^(int x) { 
     static int value = f();
     return x + value;

diff  --git a/clang/test/CodeGenObjCXX/message.mm b/clang/test/CodeGenObjCXX/message.mm
index 8b3b35877e77a..0be92f6521609 100644
--- a/clang/test/CodeGenObjCXX/message.mm
+++ b/clang/test/CodeGenObjCXX/message.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -x objective-c++ -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-10.7 -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -x objective-c++ -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-10.7 -emit-llvm -o - %s | FileCheck %s
 
 // Properly instantiate a non-dependent message expression which
 // requires a contextual conversion to ObjC pointer type.
@@ -16,9 +16,8 @@ - (void) foo;
   }
   template void foo<int>();
   // CHECK-LABEL:    define weak_odr void @_ZN5test03fooIiEEvv()
-  // CHECK:      [[T0:%.*]] = call noundef [[TEST0:%.*]]* @_ZN5test01AcvP5Test0Ev(
-  // CHECK-NEXT: [[T2:%.*]] = bitcast [[TEST0]]* [[T0]] to i8*
-  // CHECK-NEXT: [[T1:%.*]] = load i8*, i8**
-  // CHECK-NEXT: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*)*)(i8* noundef [[T2]], i8* noundef [[T1]])
+  // CHECK:      [[T0:%.*]] = call noundef ptr @_ZN5test01AcvP5Test0Ev(
+  // CHECK-NEXT: [[T1:%.*]] = load ptr, ptr
+  // CHECK-NEXT: call void @objc_msgSend(ptr noundef [[T0]], ptr noundef [[T1]])
   // CHECK-NEXT: ret void
 }

diff  --git a/clang/test/CodeGenObjCXX/microsoft-abi-arc-param-order.mm b/clang/test/CodeGenObjCXX/microsoft-abi-arc-param-order.mm
index a97b296e989af..37fe30544c59c 100644
--- a/clang/test/CodeGenObjCXX/microsoft-abi-arc-param-order.mm
+++ b/clang/test/CodeGenObjCXX/microsoft-abi-arc-param-order.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -mconstructor-aliases -fobjc-arc -triple i686-pc-win32 -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -mconstructor-aliases -fobjc-arc -triple i686-pc-win32 -emit-llvm -o - %s | FileCheck %s
 
 struct A {
   A();
@@ -10,11 +10,11 @@
 // Verify that we destruct things from left to right in the MS C++ ABI: a, b, c, d.
 //
 // CHECK-LABEL: define dso_local void @"?test_arc_order@@YAXUA@@PAUobjc_object@@01 at Z"
-// CHECK:                       (<{ %struct.A, i8*, %struct.A, i8* }>* inalloca(<{ %struct.A, i8*, %struct.A, i8* }>) %0)
+// CHECK:                       (ptr inalloca(<{ %struct.A, ptr, %struct.A, ptr }>) %0)
 void test_arc_order(A a, id __attribute__((ns_consumed)) b , A c, id __attribute__((ns_consumed)) d) {
-  // CHECK: call x86_thiscallcc void @"??1A@@QAE at XZ"(%struct.A* {{[^,]*}} %{{.*}})
-  // CHECK: call void @llvm.objc.storeStrong(i8** %{{.*}}, i8* null)
-  // CHECK: call x86_thiscallcc void @"??1A@@QAE at XZ"(%struct.A* {{[^,]*}} %{{.*}})
-  // CHECK: call void @llvm.objc.storeStrong(i8** %{{.*}}, i8* null)
+  // CHECK: call x86_thiscallcc void @"??1A@@QAE at XZ"(ptr {{[^,]*}} %{{.*}})
+  // CHECK: call void @llvm.objc.storeStrong(ptr %{{.*}}, ptr null)
+  // CHECK: call x86_thiscallcc void @"??1A@@QAE at XZ"(ptr {{[^,]*}} %{{.*}})
+  // CHECK: call void @llvm.objc.storeStrong(ptr %{{.*}}, ptr null)
   // CHECK: ret void
 }

diff  --git a/clang/test/CodeGenObjCXX/mrc-weak.mm b/clang/test/CodeGenObjCXX/mrc-weak.mm
index 9ae8fb3c7c638..9e22885541c64 100644
--- a/clang/test/CodeGenObjCXX/mrc-weak.mm
+++ b/clang/test/CodeGenObjCXX/mrc-weak.mm
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-10.10 -emit-llvm -fblocks -fobjc-weak -o - %s | FileCheck %s -check-prefix=CHECK -check-prefix=CHECK-MODERN
-// RUN: %clang_cc1 -no-opaque-pointers -triple i386-apple-darwin10 -fobjc-runtime=macosx-fragile-10.10 -emit-llvm -fblocks -fobjc-weak -o - %s | FileCheck %s -check-prefix=CHECK -check-prefix=CHECK-FRAGILE
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-10.10 -emit-llvm -fblocks -fobjc-weak -o - %s | FileCheck %s -check-prefix=CHECK -check-prefix=CHECK-MODERN
+// RUN: %clang_cc1 -triple i386-apple-darwin10 -fobjc-runtime=macosx-fragile-10.10 -emit-llvm -fblocks -fobjc-weak -o - %s | FileCheck %s -check-prefix=CHECK -check-prefix=CHECK-FRAGILE
 
 @interface Object
 - (instancetype) retain;
@@ -32,7 +32,7 @@ @implementation Foo
 
 void test1(__weak id x) {}
 // CHECK-LABEL: define{{.*}} void @_Z5test1P11objc_object(
-// CHECK:      [[X:%.*]] = alloca i8*,
+// CHECK:      [[X:%.*]] = alloca ptr,
 // CHECK-NEXT: llvm.objc.initWeak
 // CHECK-NEXT: llvm.objc.destroyWeak
 // CHECK-NEXT: ret void
@@ -41,12 +41,12 @@ void test2(id y) {
   __weak id z = y;
 }
 // CHECK-LABEL: define{{.*}} void @_Z5test2P11objc_object(
-// CHECK:      [[Y:%.*]] = alloca i8*,
-// CHECK-NEXT: [[Z:%.*]] = alloca i8*,
+// CHECK:      [[Y:%.*]] = alloca ptr,
+// CHECK-NEXT: [[Z:%.*]] = alloca ptr,
 // CHECK-NEXT: store
-// CHECK-NEXT: [[T0:%.*]] = load i8*, i8** [[Y]]
-// CHECK-NEXT: call i8* @llvm.objc.initWeak(i8** [[Z]], i8* [[T0]])
-// CHECK-NEXT: call void @llvm.objc.destroyWeak(i8** [[Z]])
+// CHECK-NEXT: [[T0:%.*]] = load ptr, ptr [[Y]]
+// CHECK-NEXT: call ptr @llvm.objc.initWeak(ptr [[Z]], ptr [[T0]])
+// CHECK-NEXT: call void @llvm.objc.destroyWeak(ptr [[Z]])
 // CHECK-NEXT: ret void
 
 void test3(id y) {
@@ -54,51 +54,49 @@ void test3(id y) {
   z = y;
 }
 // CHECK-LABEL: define{{.*}} void @_Z5test3P11objc_object(
-// CHECK:      [[Y:%.*]] = alloca i8*,
-// CHECK-NEXT: [[Z:%.*]] = alloca i8*,
+// CHECK:      [[Y:%.*]] = alloca ptr,
+// CHECK-NEXT: [[Z:%.*]] = alloca ptr,
 // CHECK-NEXT: store
-// CHECK-NEXT: store i8* null, i8** [[Z]]
-// CHECK-NEXT: [[T0:%.*]] = load i8*, i8** [[Y]]
-// CHECK-NEXT: call i8* @llvm.objc.storeWeak(i8** [[Z]], i8* [[T0]])
-// CHECK-NEXT: call void @llvm.objc.destroyWeak(i8** [[Z]])
+// CHECK-NEXT: store ptr null, ptr [[Z]]
+// CHECK-NEXT: [[T0:%.*]] = load ptr, ptr [[Y]]
+// CHECK-NEXT: call ptr @llvm.objc.storeWeak(ptr [[Z]], ptr [[T0]])
+// CHECK-NEXT: call void @llvm.objc.destroyWeak(ptr [[Z]])
 // CHECK-NEXT: ret void
 
 void test4(__weak id *p) {
   id y = *p;
 }
 // CHECK-LABEL: define{{.*}} void @_Z5test4PU6__weakP11objc_object(
-// CHECK:      [[P:%.*]] = alloca i8**,
-// CHECK-NEXT: [[Y:%.*]] = alloca i8*,
+// CHECK:      [[P:%.*]] = alloca ptr,
+// CHECK-NEXT: [[Y:%.*]] = alloca ptr,
 // CHECK-NEXT: store
-// CHECK-NEXT: [[T0:%.*]] = load i8**, i8*** [[P]]
-// CHECK-NEXT: [[T1:%.*]] = call i8* @llvm.objc.loadWeak(i8** [[T0]])
-// CHECK-NEXT: store i8* [[T1]], i8** [[Y]]
+// CHECK-NEXT: [[T0:%.*]] = load ptr, ptr [[P]]
+// CHECK-NEXT: [[T1:%.*]] = call ptr @llvm.objc.loadWeak(ptr [[T0]])
+// CHECK-NEXT: store ptr [[T1]], ptr [[Y]]
 // CHECK-NEXT: ret void
 
 void test5(__weak id *p) {
   id y = [*p retain];
 }
 // CHECK-LABEL: define{{.*}} void @_Z5test5PU6__weakP11objc_object
-// CHECK:      [[P:%.*]] = alloca i8**,
-// CHECK-NEXT: [[Y:%.*]] = alloca i8*,
+// CHECK:      [[P:%.*]] = alloca ptr,
+// CHECK-NEXT: [[Y:%.*]] = alloca ptr,
 // CHECK-NEXT: store
-// CHECK-NEXT: [[T0:%.*]] = load i8**, i8*** [[P]]
-// CHECK-NEXT: [[T1:%.*]] = call i8* @llvm.objc.loadWeakRetained(i8** [[T0]])
-// CHECK-NEXT: store i8* [[T1]], i8** [[Y]]
+// CHECK-NEXT: [[T0:%.*]] = load ptr, ptr [[P]]
+// CHECK-NEXT: [[T1:%.*]] = call ptr @llvm.objc.loadWeakRetained(ptr [[T0]])
+// CHECK-NEXT: store ptr [[T1]], ptr [[Y]]
 // CHECK-NEXT: ret void
 
 void test6(__weak Foo **p) {
   Foo *y = [*p retain];
 }
 // CHECK-LABEL: define{{.*}} void @_Z5test6PU6__weakP3Foo
-// CHECK:      [[P:%.*]] = alloca [[FOO:%.*]]**,
-// CHECK-NEXT: [[Y:%.*]] = alloca [[FOO]]*,
+// CHECK:      [[P:%.*]] = alloca ptr,
+// CHECK-NEXT: [[Y:%.*]] = alloca ptr,
 // CHECK-NEXT: store
-// CHECK-NEXT: [[T0:%.*]] = load [[FOO]]**, [[FOO]]*** [[P]]
-// CHECK-NEXT: [[T1:%.*]] = bitcast [[FOO]]** [[T0]] to i8**
-// CHECK-NEXT: [[T2:%.*]] = call i8* @llvm.objc.loadWeakRetained(i8** [[T1]])
-// CHECK-NEXT: [[T3:%.*]] = bitcast i8* [[T2]] to [[FOO]]*
-// CHECK-NEXT: store [[FOO]]* [[T3]], [[FOO]]** [[Y]]
+// CHECK-NEXT: [[T0:%.*]] = load ptr, ptr [[P]]
+// CHECK-NEXT: [[T2:%.*]] = call ptr @llvm.objc.loadWeakRetained(ptr [[T0]])
+// CHECK-NEXT: store ptr [[T2]], ptr [[Y]]
 // CHECK-NEXT: ret void
 
 extern "C" id get_object(void);
@@ -109,12 +107,9 @@ void test7(void) {
   use_block(^{ [p run ]; });
 }
 // CHECK-LABEL: define{{.*}} void @_Z5test7v
-// CHECK:       [[P:%.*]] = alloca [[FOO]]*,
-// CHECK:       [[T0:%.*]] = call i8* @get_object()
-// CHECK-NEXT:  [[T1:%.*]] = bitcast i8* [[T0]] to [[FOO]]*
-// CHECK-NEXT:  [[T2:%.*]] = bitcast [[FOO]]** [[P]] to i8**
-// CHECK-NEXT:  [[T3:%.*]] = bitcast [[FOO]]* [[T1]] to i8*
-// CHECK-NEXT:  call i8* @llvm.objc.initWeak(i8** [[T2]], i8* [[T3]])
+// CHECK:       [[P:%.*]] = alloca ptr,
+// CHECK:       [[T0:%.*]] = call ptr @get_object()
+// CHECK-NEXT:  call ptr @llvm.objc.initWeak(ptr [[P]], ptr [[T0]])
 // CHECK:       call void @llvm.objc.copyWeak
 // CHECK:       call void @use_block
 // CHECK:       call void @llvm.objc.destroyWeak
@@ -130,7 +125,7 @@ void test8(void) {
   use_block(^{ [p run ]; });
 }
 // CHECK-LABEL: define{{.*}} void @_Z5test8v
-// CHECK:       call i8* @llvm.objc.initWeak
+// CHECK:       call ptr @llvm.objc.initWeak
 // CHECK-NOT:   call void @llvm.objc.copyWeak
 // CHECK:       call void @use_block
 // CHECK:       call void @llvm.objc.destroyWeak

diff  --git a/clang/test/CodeGenObjCXX/msabi-stret.mm b/clang/test/CodeGenObjCXX/msabi-stret.mm
index 577a75923edf1..fbed6f7bb8cb9 100644
--- a/clang/test/CodeGenObjCXX/msabi-stret.mm
+++ b/clang/test/CodeGenObjCXX/msabi-stret.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-msvc -fobjc-runtime=ios-6.0 -Os -S -emit-llvm -o - %s -mframe-pointer=all | FileCheck %s
+// RUN: %clang_cc1 -triple i686-unknown-windows-msvc -fobjc-runtime=ios-6.0 -Os -S -emit-llvm -o - %s -mframe-pointer=all | FileCheck %s
 
 struct S {
   S() = default;
@@ -13,5 +13,5 @@ S f() {
   return [I m:S()];
 }
 
-// CHECK: declare dso_local void @objc_msgSend_stret(i8*, i8*, ...)
-// CHECK-NOT: declare dllimport void @objc_msgSend(i8*, i8*, ...)
+// CHECK: declare dso_local void @objc_msgSend_stret(ptr, ptr, ...)
+// CHECK-NOT: declare dllimport void @objc_msgSend(ptr, ptr, ...)

diff  --git a/clang/test/CodeGenObjCXX/objfw-exceptions.mm b/clang/test/CodeGenObjCXX/objfw-exceptions.mm
index 610a87f7bd44a..256d05d448d81 100644
--- a/clang/test/CodeGenObjCXX/objfw-exceptions.mm
+++ b/clang/test/CodeGenObjCXX/objfw-exceptions.mm
@@ -1,17 +1,17 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-pc-linux-gnu -emit-llvm -fexceptions -fobjc-exceptions -fobjc-runtime=objfw -fcxx-exceptions -o - %s | FileCheck %s -check-prefix=CHECK -check-prefix=CHECK-DWARF
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-pc-linux-gnu -emit-llvm -fexceptions -fobjc-exceptions -fobjc-runtime=objfw -fcxx-exceptions -exception-model=sjlj -o - %s | FileCheck %s -check-prefix=CHECK -check-prefix=CHECK-SJLJ
+// RUN: %clang_cc1 -triple x86_64-pc-linux-gnu -emit-llvm -fexceptions -fobjc-exceptions -fobjc-runtime=objfw -fcxx-exceptions -o - %s | FileCheck %s -check-prefix=CHECK -check-prefix=CHECK-DWARF
+// RUN: %clang_cc1 -triple x86_64-pc-linux-gnu -emit-llvm -fexceptions -fobjc-exceptions -fobjc-runtime=objfw -fcxx-exceptions -exception-model=sjlj -o - %s | FileCheck %s -check-prefix=CHECK -check-prefix=CHECK-SJLJ
 
 @interface OCType @end
 void opaque();
 
 // CHECK: define{{.*}} void @_Z3foov()
-// CHECK-DWARF-SAME: personality i8* bitcast (i32 (...)* @__gnu_objc_personality_v0 to i8*)
-// CHECK-SJLJ-SAME: personality i8* bitcast (i32 (...)* @__gnu_objc_personality_sj0 to i8*)
+// CHECK-DWARF-SAME: personality ptr @__gnu_objc_personality_v0
+// CHECK-SJLJ-SAME: personality ptr @__gnu_objc_personality_sj0
 void foo() {
 try {
 // CHECK: invoke void @_Z6opaquev
 opaque();
 } catch (OCType *T) {
-// CHECK:      landingpad { i8*, i32 }
+// CHECK:      landingpad { ptr, i32 }
 }
 }

diff  --git a/clang/test/CodeGenObjCXX/personality-abuse.mm b/clang/test/CodeGenObjCXX/personality-abuse.mm
index 672f64ce4ca95..d82a4e3d685d7 100644
--- a/clang/test/CodeGenObjCXX/personality-abuse.mm
+++ b/clang/test/CodeGenObjCXX/personality-abuse.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -x objective-c++ -triple x86_64-apple-darwin10 -emit-llvm -fcxx-exceptions -fexceptions -fobjc-exceptions -o - %s | FileCheck %s
+// RUN: %clang_cc1 -x objective-c++ -triple x86_64-apple-darwin10 -emit-llvm -fcxx-exceptions -fexceptions -fobjc-exceptions -o - %s | FileCheck %s
 
 extern "C" {
   int __objc_personality_v0();
@@ -16,4 +16,4 @@ void foo() {
   }
 }
 
-// CHECK: define{{.*}} void @_Z3foov() {{#[0-9]+}} personality i8* bitcast (i32 ()* @__objc_personality_v0 to i8*)
+// CHECK: define{{.*}} void @_Z3foov() {{#[0-9]+}} personality ptr @__objc_personality_v0

diff  --git a/clang/test/CodeGenObjCXX/personality.mm b/clang/test/CodeGenObjCXX/personality.mm
index b2481987153d8..c6debe6f60c2a 100644
--- a/clang/test/CodeGenObjCXX/personality.mm
+++ b/clang/test/CodeGenObjCXX/personality.mm
@@ -1,65 +1,65 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE-SEH
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE-SJLJ
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gnustep-1.7 -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GNUSTEP-1_7
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gnustep -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GNUSTEP
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC-SEH
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC-SJLJ
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW-SEH
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-linux-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW-SJLJ
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE-SEH
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE-SJLJ
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gnustep-1.7 -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GNUSTEP-1_7
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gnustep -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GNUSTEP
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC-SEH
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC-SJLJ
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW-SEH
+// RUN: %clang_cc1 -triple i686-unknown-linux-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW-SJLJ
 
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gnustep-1.7 -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gnustep -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
+// RUN: %clang_cc1 -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
+// RUN: %clang_cc1 -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
+// RUN: %clang_cc1 -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
+// RUN: %clang_cc1 -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
+// RUN: %clang_cc1 -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gnustep-1.7 -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
+// RUN: %clang_cc1 -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gnustep -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
+// RUN: %clang_cc1 -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
+// RUN: %clang_cc1 -triple i686-unknown-windows-msvc -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-WIN-MSVC
 
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE-SEH
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE-SJLJ
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gnustep-1.7 -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GNUSTEP-1_7
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gnustep -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GNUSTEP
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC-SEH
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC-SJLJ
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW-SEH
-// RUN: %clang_cc1 -no-opaque-pointers -triple i686-unknown-windows-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW-SJLJ
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE-SEH
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx-fragile -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-MACOSX-FRAGILE-SJLJ
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=macosx -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=ios -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=watchos -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-NS
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gnustep-1.7 -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GNUSTEP-1_7
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gnustep -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GNUSTEP
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC-SEH
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=gcc -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-GCC-SJLJ
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=dwarf -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=seh -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW-SEH
+// RUN: %clang_cc1 -triple i686-unknown-windows-gnu -fexceptions -exception-model=sjlj -fobjc-exceptions -fcxx-exceptions -fobjc-runtime=objfw -S -emit-llvm %s -o - | FileCheck %s -check-prefix CHECK-OBJFW-SJLJ
 
 void g(void);
 
@@ -67,20 +67,20 @@ @interface use
 + (void)e:(id)xception;
 @end
 
-// CHECK-MACOSX-FRAGILE: personality i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*)
-// CHECK-MACOSX-FRAGILE-SEH: personality i8* bitcast (i32 (...)* @__gxx_personality_seh0 to i8*)
-// CHECK-MACOSX-FRAGILE-SJLJ: personality i8* bitcast (i32 (...)* @__gxx_personality_sj0 to i8*)
-// CHECK-NS: personality i8* bitcast (i32 (...)* @__objc_personality_v0 to i8*)
-// CHECK-GNUSTEP-1_7: personality i8* bitcast (i32 (...)* @__gnustep_objcxx_personality_v0 to i8*)
-// CHECK-GNUSTEP: personality i8* bitcast (i32 (...)* @__gnustep_objcxx_personality_v0 to i8*)
-// CHECK-GCC: personality i8* bitcast (i32 (...)* @__gnu_objc_personality_v0 to i8*)
-// CHECK-GCC-SEH: personality i8* bitcast (i32 (...)* @__gnu_objc_personality_seh0 to i8*)
-// CHECK-GCC-SJLJ: personality i8* bitcast (i32 (...)* @__gnu_objc_personality_sj0 to i8*)
-// CHECK-OBJFW: personality i8* bitcast (i32 (...)* @__gnu_objc_personality_v0 to i8*)
-// CHECK-OBJFW-SEH: personality i8* bitcast (i32 (...)* @__gnu_objc_personality_seh0 to i8*)
-// CHECK-OBJFW-SJLJ: personality i8* bitcast (i32 (...)* @__gnu_objc_personality_sj0 to i8*)
+// CHECK-MACOSX-FRAGILE: personality ptr @__gxx_personality_v0
+// CHECK-MACOSX-FRAGILE-SEH: personality ptr @__gxx_personality_seh0
+// CHECK-MACOSX-FRAGILE-SJLJ: personality ptr @__gxx_personality_sj0
+// CHECK-NS: personality ptr @__objc_personality_v0
+// CHECK-GNUSTEP-1_7: personality ptr @__gnustep_objcxx_personality_v0
+// CHECK-GNUSTEP: personality ptr @__gnustep_objcxx_personality_v0
+// CHECK-GCC: personality ptr @__gnu_objc_personality_v0
+// CHECK-GCC-SEH: personality ptr @__gnu_objc_personality_seh0
+// CHECK-GCC-SJLJ: personality ptr @__gnu_objc_personality_sj0
+// CHECK-OBJFW: personality ptr @__gnu_objc_personality_v0
+// CHECK-OBJFW-SEH: personality ptr @__gnu_objc_personality_seh0
+// CHECK-OBJFW-SJLJ: personality ptr @__gnu_objc_personality_sj0
 
-// CHECK-WIN-MSVC: personality i8* bitcast (i32 (...)* @__CxxFrameHandler3  to i8*)
+// CHECK-WIN-MSVC: personality ptr @__CxxFrameHandler3
 
 void f(void) {
   @try {
@@ -91,8 +91,8 @@ void f(void) {
 }
 
 #if defined(__SEH_EXCEPTIONS__)
-// CHECK-WIN-SEH-X86: personality i8* bitcast (i32 (...)* @_except_handler3 to i8*)
-// CHECK-WIN-SEH-X64: personality i8* bitcast (i32 (...)* @__C_specific_handler to i8*)
+// CHECK-WIN-SEH-X86: personality ptr @_except_handler3
+// CHECK-WIN-SEH-X64: personality ptr @__C_specific_handler
 
 void h(void) {
   __try {

diff  --git a/clang/test/CodeGenObjCXX/property-lvalue-capture.mm b/clang/test/CodeGenObjCXX/property-lvalue-capture.mm
index cbbb6f5b42b0b..76c218bcff05d 100644
--- a/clang/test/CodeGenObjCXX/property-lvalue-capture.mm
+++ b/clang/test/CodeGenObjCXX/property-lvalue-capture.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -emit-llvm -o - %s | FileCheck %s
 // rdar://15118128
 
 template <typename T> struct Quad2 {
@@ -24,12 +24,10 @@ - (void)transformFrame:(PAGeometryFrame *)frame {
 }
 @end
 
-// CHECK:   [[OBJ:%.*]] = bitcast [[ONET:%.*]]* [[ONE:%.*]] to i8*
-// CHECK:   [[SEL:%.*]] = load i8*, i8** @OBJC_SELECTOR_REFERENCES_, align 8, !invariant.load ![[MD_NUM:[0-9]+]]
-// CHECK:   [[CALL:%.*]] = call noundef nonnull align 1 %struct.Quad2* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to %struct.Quad2* (i8*, i8*)*)(i8* noundef [[OBJ]], i8* noundef [[SEL]])
-// CHECK:   [[OBJ2:%.*]] = bitcast [[ONET]]* [[ZERO:%.*]] to i8*
-// CHECK:   [[SEL2:%.*]] = load i8*, i8** @OBJC_SELECTOR_REFERENCES_.2, align 8, !invariant.load ![[MD_NUM]]
-// CHECK:   call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*, %struct.Quad2*)*)(i8* noundef [[OBJ2]], i8* noundef [[SEL2]], %struct.Quad2* noundef nonnull align 1 [[CALL]])
+// CHECK:   [[SEL:%.*]] = load ptr, ptr @OBJC_SELECTOR_REFERENCES_, align 8, !invariant.load ![[MD_NUM:[0-9]+]]
+// CHECK:   [[CALL:%.*]] = call noundef nonnull align 1 ptr @objc_msgSend(ptr noundef [[ONE:%.*]], ptr noundef [[SEL]])
+// CHECK:   [[SEL2:%.*]] = load ptr, ptr @OBJC_SELECTOR_REFERENCES_.2, align 8, !invariant.load ![[MD_NUM]]
+// CHECK:   call void @objc_msgSend(ptr noundef [[ZERO:%.*]], ptr noundef [[SEL2]], ptr noundef nonnull align 1 [[CALL]])
 
 
 struct A {
@@ -46,8 +44,7 @@ void test(C *c, const A &a) {
  const A &result = c.prop = a;
 }
 
-// CHECK:   [[ONE1:%.*]] = load %struct.A*, %struct.A** [[AADDR:%.*]], align 8
-// CHECK:   [[OBJ3:%.*]] = bitcast [[TWOT:%.*]]* [[ZERO1:%.*]] to i8*
-// CHECK:   [[SEL3:%.*]] = load i8*, i8** @OBJC_SELECTOR_REFERENCES_.5, align 8, !invariant.load ![[MD_NUM]]
-// CHECK:   call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*, %struct.A*)*)(i8* noundef [[OBJ3]], i8* noundef [[SEL3]], %struct.A* noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) [[ONE1]])
-// CHECK:   store %struct.A* [[ONE1]], %struct.A** [[RESULT:%.*]], align 8
+// CHECK:   [[ONE1:%.*]] = load ptr, ptr [[AADDR:%.*]], align 8
+// CHECK:   [[SEL3:%.*]] = load ptr, ptr @OBJC_SELECTOR_REFERENCES_.5, align 8, !invariant.load ![[MD_NUM]]
+// CHECK:   call void @objc_msgSend(ptr noundef [[ZERO1:%.*]], ptr noundef [[SEL3]], ptr noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) [[ONE1]])
+// CHECK:   store ptr [[ONE1]], ptr [[RESULT:%.*]], align 8

diff  --git a/clang/test/CodeGenObjCXX/property-object-reference-2.mm b/clang/test/CodeGenObjCXX/property-object-reference-2.mm
index 19d1806fb8a69..a29d11466fdb8 100644
--- a/clang/test/CodeGenObjCXX/property-object-reference-2.mm
+++ b/clang/test/CodeGenObjCXX/property-object-reference-2.mm
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers %s -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-10.7 -emit-llvm -o - | FileCheck %s
-// RUN: %clang_cc1 -no-opaque-pointers %s -triple x86_64-unknown-freebsd -fobjc-runtime=gnustep-1.7 -emit-llvm -o - | FileCheck -check-prefix=CHECK-GNUSTEP %s
+// RUN: %clang_cc1 %s -triple x86_64-apple-darwin10 -fobjc-runtime=macosx-10.7 -emit-llvm -o - | FileCheck %s
+// RUN: %clang_cc1 %s -triple x86_64-unknown-freebsd -fobjc-runtime=gnustep-1.7 -emit-llvm -o - | FileCheck -check-prefix=CHECK-GNUSTEP %s
 // rdar://6137845
 
 extern int DEFAULT();
@@ -29,31 +29,25 @@ @implementation MyDocument
   @synthesize MyProperty1 = _cppObject1;
 @end
 
-// CHECK-LABEL: define internal void @__copy_helper_atomic_property_(%struct.TCPPObject* noundef %0, %struct.TCPPObject* noundef %1) #
-// CHECK: [[TWO:%.*]] = load %struct.TCPPObject*, %struct.TCPPObject** [[ADDR:%.*]], align 8
-// CHECK: [[THREE:%.*]] = load %struct.TCPPObject*, %struct.TCPPObject** [[ADDR1:%.*]], align 8
+// CHECK-LABEL: define internal void @__copy_helper_atomic_property_(ptr noundef %0, ptr noundef %1) #
+// CHECK: [[TWO:%.*]] = load ptr, ptr [[ADDR:%.*]], align 8
+// CHECK: [[THREE:%.*]] = load ptr, ptr [[ADDR1:%.*]], align 8
 // CHECK: [[CALL:%.*]] = call noundef i32 @_Z7DEFAULTv()
-// CHECK:  call void @_ZN10TCPPObjectC1ERKS_i(%struct.TCPPObject* {{[^,]*}} [[TWO]], %struct.TCPPObject* noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) [[THREE]], i32 noundef [[CALL]])
+// CHECK:  call void @_ZN10TCPPObjectC1ERKS_i(ptr {{[^,]*}} [[TWO]], ptr noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) [[THREE]], i32 noundef [[CALL]])
 // CHECK:  ret void
 
 // CHECK: define internal void @"\01-[MyDocument MyProperty]"(
-// CHECK: [[ONE:%.*]] = bitcast i8* [[ADDPTR:%.*]] to %struct.TCPPObject*
-// CHECK: [[TWO:%.*]] = bitcast %struct.TCPPObject* [[ONE]] to i8*
-// CHECK: [[THREE:%.*]] = bitcast %struct.TCPPObject* [[AGGRESULT:%.*]] to i8*
-// CHECK: call void @objc_copyCppObjectAtomic(i8* noundef [[THREE]], i8* noundef [[TWO]], i8* noundef bitcast (void (%struct.TCPPObject*, %struct.TCPPObject*)* @__copy_helper_atomic_property_ to i8*))
+// CHECK: call void @objc_copyCppObjectAtomic(ptr noundef [[AGGRESULT:%.*]], ptr noundef [[ADDPTR:%.*]], ptr noundef @__copy_helper_atomic_property_)
 // CHECK: ret void
 
-// CHECK-LABEL: define internal void @__assign_helper_atomic_property_(%struct.TCPPObject* noundef %0, %struct.TCPPObject* noundef %1) #
-// CHECK: [[THREE:%.*]] = load %struct.TCPPObject*, %struct.TCPPObject** [[ADDR1:%.*]], align 8
-// CHECK: [[TWO:%.*]] = load %struct.TCPPObject*, %struct.TCPPObject** [[ADDR:%.*]], align 8
-// CHECK: [[CALL:%.*]] = call noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) %struct.TCPPObject* @_ZN10TCPPObjectaSERKS_(%struct.TCPPObject* {{[^,]*}} [[TWO]], %struct.TCPPObject* noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) [[THREE]])
+// CHECK-LABEL: define internal void @__assign_helper_atomic_property_(ptr noundef %0, ptr noundef %1) #
+// CHECK: [[THREE:%.*]] = load ptr, ptr [[ADDR1:%.*]], align 8
+// CHECK: [[TWO:%.*]] = load ptr, ptr [[ADDR:%.*]], align 8
+// CHECK: [[CALL:%.*]] = call noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) ptr @_ZN10TCPPObjectaSERKS_(ptr {{[^,]*}} [[TWO]], ptr noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) [[THREE]])
 // CHECK:  ret void
 
 // CHECK: define internal void @"\01-[MyDocument setMyProperty:]"(
-// CHECK: [[ONE:%.*]] = bitcast i8* [[ADDRPTR:%.*]] to %struct.TCPPObject*
-// CHECK: [[TWO:%.*]] = bitcast %struct.TCPPObject* [[ONE]] to i8*
-// CHECK: [[THREE:%.*]] = bitcast %struct.TCPPObject* [[MYPROPERTY:%.*]] to i8*
-// CHECK: call void @objc_copyCppObjectAtomic(i8* noundef [[TWO]], i8* noundef [[THREE]], i8* noundef bitcast (void (%struct.TCPPObject*, %struct.TCPPObject*)* @__assign_helper_atomic_property_ to i8*))
+// CHECK: call void @objc_copyCppObjectAtomic(ptr noundef [[ADDRPTR:%.*]], ptr noundef [[MYPROPERTY:%.*]], ptr noundef @__assign_helper_atomic_property_)
 // CHECK: ret void
 
 // CHECK-GNUSTEP: objc_getCppObjectAtomic

diff  --git a/clang/test/CodeGenObjCXX/synthesized-property-cleanup.mm b/clang/test/CodeGenObjCXX/synthesized-property-cleanup.mm
index 32206c159a24a..6d9c617ac4193 100644
--- a/clang/test/CodeGenObjCXX/synthesized-property-cleanup.mm
+++ b/clang/test/CodeGenObjCXX/synthesized-property-cleanup.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple arm64e-apple-ios13.0 -debug-info-kind=standalone -fobjc-arc -fsanitize=nullability-return \
+// RUN: %clang_cc1 -triple arm64e-apple-ios13.0 -debug-info-kind=standalone -fobjc-arc -fsanitize=nullability-return \
 // RUN:   %s -emit-llvm -o - | FileCheck %s
 
 @interface NSObject
@@ -10,7 +10,7 @@ @interface NSString : NSObject
 
 // CHECK: define {{.*}}@"\01-[MyData setData:]"
 // CHECK: [[DATA:%.*]] = alloca %struct.Data
-// CHECK: call noundef %struct.Data* @_ZN4DataD1Ev(%struct.Data* {{[^,]*}} [[DATA]]){{.*}}, !dbg [[DATA_PROPERTY_LOC:![0-9]+]]
+// CHECK: call noundef ptr @_ZN4DataD1Ev(ptr {{[^,]*}} [[DATA]]){{.*}}, !dbg [[DATA_PROPERTY_LOC:![0-9]+]]
 // CHECK-NEXT: ret void
 
 // CHECK: define {{.*}}@"\01-[MyData string]"

diff  --git a/clang/test/CodeGenObjCXX/ubsan-nullability-return-notypeloc.mm b/clang/test/CodeGenObjCXX/ubsan-nullability-return-notypeloc.mm
index 93da391c8f7c1..3a4bb81cfc4c4 100644
--- a/clang/test/CodeGenObjCXX/ubsan-nullability-return-notypeloc.mm
+++ b/clang/test/CodeGenObjCXX/ubsan-nullability-return-notypeloc.mm
@@ -1,10 +1,10 @@
-// RUN: %clang_cc1 -no-opaque-pointers -fsanitize=nullability-return -emit-llvm %s -o - -triple x86_64-apple-macosx10.10.0 | FileCheck %s
+// RUN: %clang_cc1 -fsanitize=nullability-return -emit-llvm %s -o - -triple x86_64-apple-macosx10.10.0 | FileCheck %s
 
 // CHECK: [[ATTR_LOC:@[0-9]+]] = {{.*}} global { {{.*}} i32 15, i32 38
 
-// CHECK-LABEL: define{{.*}} i8* @_Z3foov()
-// CHECK: [[CALL:%.*]] = call noundef i8* @_Z6helperv()
-// CHECK: icmp ne i8* [[CALL]]
+// CHECK-LABEL: define{{.*}} ptr @_Z3foov()
+// CHECK: [[CALL:%.*]] = call noundef ptr @_Z6helperv()
+// CHECK: icmp ne ptr [[CALL]]
 // CHECK: call void @__ubsan_handle_nullability_return_v1_abort({{.*}}[[ATTR_LOC]]
 
 struct S {

diff  --git a/clang/test/CodeGenObjCXX/unknown-anytype.mm b/clang/test/CodeGenObjCXX/unknown-anytype.mm
index dee93c6fe1a6a..71b49e38c0da0 100644
--- a/clang/test/CodeGenObjCXX/unknown-anytype.mm
+++ b/clang/test/CodeGenObjCXX/unknown-anytype.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -fdebugger-support -funknown-anytype -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -fdebugger-support -funknown-anytype -emit-llvm -o - %s | FileCheck %s
 
 // rdar://13025708
 
@@ -7,7 +7,7 @@ void test0(A *a) {
   (void) [a test0: (float) 2.0];
 }
 // CHECK-LABEL: define{{.*}} void @_Z5test0P1A(
-// CHECK: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*, float)*)(
+// CHECK: call void @objc_msgSend(
 
 @interface B
 - (void) test1: (__unknown_anytype) x;
@@ -16,5 +16,5 @@ void test1(B *b) {
   (void) [b test1: (float) 2.0];
 }
 // CHECK-LABEL: define{{.*}} void @_Z5test1P1B(
-// CHECK: call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*, float)*)(
+// CHECK: call void @objc_msgSend(
 

diff  --git a/clang/test/CodeGenOpenCL/backend-unsupported-warning.ll b/clang/test/CodeGenOpenCL/backend-unsupported-warning.ll
index 597fe94885e34..8d253c76e4a83 100644
--- a/clang/test/CodeGenOpenCL/backend-unsupported-warning.ll
+++ b/clang/test/CodeGenOpenCL/backend-unsupported-warning.ll
@@ -11,7 +11,7 @@ target triple = "amdgcn-amd-amdhsa"
 @lds = external addrspace(3) global i32, align 4
 
 define i32 @use_lds_global_in_func() !dbg !5 {
-  %load = load i32, i32 addrspace(3)* @lds, !dbg !9
+  %load = load i32, ptr addrspace(3) @lds, !dbg !9
   ret i32 %load, !dbg !10
 }
 

diff  --git a/clang/test/CodeGenOpenCLCXX/address-space-castoperators.cpp b/clang/test/CodeGenOpenCLCXX/address-space-castoperators.cpp
index 093309e3d55fe..4aaa05e260649 100644
--- a/clang/test/CodeGenOpenCLCXX/address-space-castoperators.cpp
+++ b/clang/test/CodeGenOpenCLCXX/address-space-castoperators.cpp
@@ -1,13 +1,13 @@
-// RUN: %clang_cc1 -no-opaque-pointers %s -triple spir-unknown-unknown -cl-std=clc++ -emit-llvm -O0 -o - | FileCheck %s
+// RUN: %clang_cc1 %s -triple spir-unknown-unknown -cl-std=clc++ -emit-llvm -O0 -o - | FileCheck %s
 
 void test_reinterpret_cast(){
 __private float x;
 __private float& y = x; 
 // We don't need bitcast to cast pointer type and
 // address space at the same time.
-//CHECK: addrspacecast float* %x to i32 addrspace(4)*
-//CHECK: [[REG:%[0-9]+]] = load float*, float** %y
-//CHECK: addrspacecast float* [[REG]] to i32 addrspace(4)*
+//CHECK: addrspacecast ptr %x to ptr addrspace(4)
+//CHECK: [[REG:%[0-9]+]] = load ptr, ptr %y
+//CHECK: addrspacecast ptr [[REG]] to ptr addrspace(4)
 //CHECK-NOT: bitcast 
 __generic int& rc1 = reinterpret_cast<__generic int&>(x);
 __generic int& rc2 = reinterpret_cast<__generic int&>(y);

diff  --git a/clang/test/CodeGenSYCL/address-space-conversions.cpp b/clang/test/CodeGenSYCL/address-space-conversions.cpp
index dd9f53cf7f495..3933ad375412d 100644
--- a/clang/test/CodeGenSYCL/address-space-conversions.cpp
+++ b/clang/test/CodeGenSYCL/address-space-conversions.cpp
@@ -1,138 +1,138 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple spir64 -fsycl-is-device -disable-llvm-passes -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -triple spir64 -fsycl-is-device -disable-llvm-passes -emit-llvm %s -o - | FileCheck %s
 void bar(int &Data) {}
-// CHECK-DAG: define{{.*}} spir_func void @[[RAW_REF:[a-zA-Z0-9_]+]](i32 addrspace(4)* noundef align 4 dereferenceable(4) %
+// CHECK-DAG: define{{.*}} spir_func void @[[RAW_REF:[a-zA-Z0-9_]+]](ptr addrspace(4) noundef align 4 dereferenceable(4) %
 void bar2(int &Data) {}
-// CHECK-DAG: define{{.*}} spir_func void @[[RAW_REF2:[a-zA-Z0-9_]+]](i32 addrspace(4)* noundef align 4 dereferenceable(4) %
+// CHECK-DAG: define{{.*}} spir_func void @[[RAW_REF2:[a-zA-Z0-9_]+]](ptr addrspace(4) noundef align 4 dereferenceable(4) %
 void bar(__attribute__((opencl_local)) int &Data) {}
-// CHECK-DAG: define{{.*}} spir_func void [[LOC_REF:@[a-zA-Z0-9_]+]](i32 addrspace(3)* noundef align 4 dereferenceable(4) %
+// CHECK-DAG: define{{.*}} spir_func void [[LOC_REF:@[a-zA-Z0-9_]+]](ptr addrspace(3) noundef align 4 dereferenceable(4) %
 void foo(int *Data) {}
-// CHECK-DAG: define{{.*}} spir_func void @[[RAW_PTR:[a-zA-Z0-9_]+]](i32 addrspace(4)* noundef %
+// CHECK-DAG: define{{.*}} spir_func void @[[RAW_PTR:[a-zA-Z0-9_]+]](ptr addrspace(4) noundef %
 void foo2(int *Data) {}
-// CHECK-DAG: define{{.*}} spir_func void @[[RAW_PTR2:[a-zA-Z0-9_]+]](i32 addrspace(4)* noundef %
+// CHECK-DAG: define{{.*}} spir_func void @[[RAW_PTR2:[a-zA-Z0-9_]+]](ptr addrspace(4) noundef %
 void foo(__attribute__((opencl_local)) int *Data) {}
-// CHECK-DAG: define{{.*}} spir_func void [[LOC_PTR:@[a-zA-Z0-9_]+]](i32 addrspace(3)* noundef %
+// CHECK-DAG: define{{.*}} spir_func void [[LOC_PTR:@[a-zA-Z0-9_]+]](ptr addrspace(3) noundef %
 
 template <typename T>
 void tmpl(T t) {}
 // See Check Lines below.
 
 void usages() {
-  // CHECK-DAG: [[GLOB:%[a-zA-Z0-9]+]] = alloca i32 addrspace(1)*
-  // CHECK-DAG: [[GLOB]].ascast = addrspacecast i32 addrspace(1)** [[GLOB]] to i32 addrspace(1)* addrspace(4)*
+  // CHECK-DAG: [[GLOB:%[a-zA-Z0-9]+]] = alloca ptr addrspace(1)
+  // CHECK-DAG: [[GLOB]].ascast = addrspacecast ptr [[GLOB]] to ptr addrspace(4)
   __attribute__((opencl_global)) int *GLOB;
-  // CHECK-DAG: [[LOC:%[a-zA-Z0-9]+]] = alloca i32 addrspace(3)*
-  // CHECK-DAG: [[LOC]].ascast = addrspacecast i32 addrspace(3)** [[LOC]] to i32 addrspace(3)* addrspace(4)*
+  // CHECK-DAG: [[LOC:%[a-zA-Z0-9]+]] = alloca ptr addrspace(3)
+  // CHECK-DAG: [[LOC]].ascast = addrspacecast ptr [[LOC]] to ptr addrspace(4)
   __attribute__((opencl_local)) int *LOC;
-  // CHECK-DAG: [[NoAS:%[a-zA-Z0-9]+]] = alloca i32 addrspace(4)*
-  // CHECK-DAG: [[NoAS]].ascast = addrspacecast i32 addrspace(4)** [[NoAS]] to i32 addrspace(4)* addrspace(4)*
+  // CHECK-DAG: [[NoAS:%[a-zA-Z0-9]+]] = alloca ptr addrspace(4)
+  // CHECK-DAG: [[NoAS]].ascast = addrspacecast ptr [[NoAS]] to ptr addrspace(4)
   int *NoAS;
-  // CHECK-DAG: [[PRIV:%[a-zA-Z0-9]+]] = alloca i32*
-  // CHECK-DAG: [[PRIV]].ascast = addrspacecast i32** [[PRIV]] to i32* addrspace(4)*
+  // CHECK-DAG: [[PRIV:%[a-zA-Z0-9]+]] = alloca ptr
+  // CHECK-DAG: [[PRIV]].ascast = addrspacecast ptr [[PRIV]] to ptr addrspace(4)
   __attribute__((opencl_private)) int *PRIV;
-  // CHECK-DAG: [[GLOB_DEVICE:%[a-zA-Z0-9]+]] = alloca i32 addrspace(5)*
+  // CHECK-DAG: [[GLOB_DEVICE:%[a-zA-Z0-9]+]] = alloca ptr addrspace(5)
   __attribute__((opencl_global_device)) int *GLOBDEVICE;
-  // CHECK-DAG: [[GLOB_HOST:%[a-zA-Z0-9]+]] = alloca i32 addrspace(6)*
+  // CHECK-DAG: [[GLOB_HOST:%[a-zA-Z0-9]+]] = alloca ptr addrspace(6)
   __attribute__((opencl_global_host)) int *GLOBHOST;
 
   // Explicit conversions
   // From named address spaces to default address space
-  // CHECK-DAG: [[GLOB_LOAD:%[a-zA-Z0-9]+]] = load i32 addrspace(1)*, i32 addrspace(1)* addrspace(4)* [[GLOB]].ascast
-  // CHECK-DAG: [[GLOB_CAST:%[a-zA-Z0-9]+]] = addrspacecast i32 addrspace(1)* [[GLOB_LOAD]] to i32 addrspace(4)*
-  // CHECK-DAG: store i32 addrspace(4)* [[GLOB_CAST]], i32 addrspace(4)* addrspace(4)* [[NoAS]].ascast
+  // CHECK-DAG: [[GLOB_LOAD:%[a-zA-Z0-9]+]] = load ptr addrspace(1), ptr addrspace(4) [[GLOB]].ascast
+  // CHECK-DAG: [[GLOB_CAST:%[a-zA-Z0-9]+]] = addrspacecast ptr addrspace(1) [[GLOB_LOAD]] to ptr addrspace(4)
+  // CHECK-DAG: store ptr addrspace(4) [[GLOB_CAST]], ptr addrspace(4) [[NoAS]].ascast
   NoAS = (int *)GLOB;
-  // CHECK-DAG: [[LOC_LOAD:%[a-zA-Z0-9]+]] = load i32 addrspace(3)*, i32 addrspace(3)* addrspace(4)* [[LOC]].ascast
-  // CHECK-DAG: [[LOC_CAST:%[a-zA-Z0-9]+]] = addrspacecast i32 addrspace(3)* [[LOC_LOAD]] to i32 addrspace(4)*
-  // CHECK-DAG: store i32 addrspace(4)* [[LOC_CAST]], i32 addrspace(4)* addrspace(4)* [[NoAS]].ascast
+  // CHECK-DAG: [[LOC_LOAD:%[a-zA-Z0-9]+]] = load ptr addrspace(3), ptr addrspace(4) [[LOC]].ascast
+  // CHECK-DAG: [[LOC_CAST:%[a-zA-Z0-9]+]] = addrspacecast ptr addrspace(3) [[LOC_LOAD]] to ptr addrspace(4)
+  // CHECK-DAG: store ptr addrspace(4) [[LOC_CAST]], ptr addrspace(4) [[NoAS]].ascast
   NoAS = (int *)LOC;
-  // CHECK-DAG: [[PRIV_LOAD:%[a-zA-Z0-9]+]] = load i32*, i32* addrspace(4)* [[PRIV]].ascast
-  // CHECK-DAG: [[PRIV_CAST:%[a-zA-Z0-9]+]] = addrspacecast i32* [[PRIV_LOAD]] to i32 addrspace(4)*
-  // CHECK-DAG: store i32 addrspace(4)* [[PRIV_CAST]], i32 addrspace(4)* addrspace(4)* [[NoAS]].ascast
+  // CHECK-DAG: [[PRIV_LOAD:%[a-zA-Z0-9]+]] = load ptr, ptr addrspace(4) [[PRIV]].ascast
+  // CHECK-DAG: [[PRIV_CAST:%[a-zA-Z0-9]+]] = addrspacecast ptr [[PRIV_LOAD]] to ptr addrspace(4)
+  // CHECK-DAG: store ptr addrspace(4) [[PRIV_CAST]], ptr addrspace(4) [[NoAS]].ascast
   NoAS = (int *)PRIV;
   // From default address space to named address space
-  // CHECK-DAG: [[NoAS_LOAD:%[a-zA-Z0-9]+]] = load i32 addrspace(4)*, i32 addrspace(4)* addrspace(4)* [[NoAS]].ascast
-  // CHECK-DAG: [[NoAS_CAST:%[a-zA-Z0-9]+]] = addrspacecast i32 addrspace(4)* [[NoAS_LOAD]] to i32 addrspace(1)*
-  // CHECK-DAG: store i32 addrspace(1)* [[NoAS_CAST]], i32 addrspace(1)* addrspace(4)* [[GLOB]].ascast
+  // CHECK-DAG: [[NoAS_LOAD:%[a-zA-Z0-9]+]] = load ptr addrspace(4), ptr addrspace(4) [[NoAS]].ascast
+  // CHECK-DAG: [[NoAS_CAST:%[a-zA-Z0-9]+]] = addrspacecast ptr addrspace(4) [[NoAS_LOAD]] to ptr addrspace(1)
+  // CHECK-DAG: store ptr addrspace(1) [[NoAS_CAST]], ptr addrspace(4) [[GLOB]].ascast
   GLOB = (__attribute__((opencl_global)) int *)NoAS;
-  // CHECK-DAG: [[NoAS_LOAD:%[a-zA-Z0-9]+]] = load i32 addrspace(4)*, i32 addrspace(4)* addrspace(4)* [[NoAS]].ascast
-  // CHECK-DAG: [[NoAS_CAST:%[a-zA-Z0-9]+]] = addrspacecast i32 addrspace(4)* [[NoAS_LOAD]] to i32 addrspace(3)*
-  // CHECK-DAG: store i32 addrspace(3)* [[NoAS_CAST]], i32 addrspace(3)* addrspace(4)* [[LOC]].ascast
+  // CHECK-DAG: [[NoAS_LOAD:%[a-zA-Z0-9]+]] = load ptr addrspace(4), ptr addrspace(4) [[NoAS]].ascast
+  // CHECK-DAG: [[NoAS_CAST:%[a-zA-Z0-9]+]] = addrspacecast ptr addrspace(4) [[NoAS_LOAD]] to ptr addrspace(3)
+  // CHECK-DAG: store ptr addrspace(3) [[NoAS_CAST]], ptr addrspace(4) [[LOC]].ascast
   LOC = (__attribute__((opencl_local)) int *)NoAS;
-  // CHECK-DAG: [[NoAS_LOAD:%[a-zA-Z0-9]+]] = load i32 addrspace(4)*, i32 addrspace(4)* addrspace(4)* [[NoAS]].ascast
-  // CHECK-DAG: [[NoAS_CAST:%[a-zA-Z0-9]+]] = addrspacecast i32 addrspace(4)* [[NoAS_LOAD]] to i32*
-  // CHECK-DAG: store i32* [[NoAS_CAST]], i32* addrspace(4)* [[PRIV]].ascast
+  // CHECK-DAG: [[NoAS_LOAD:%[a-zA-Z0-9]+]] = load ptr addrspace(4), ptr addrspace(4) [[NoAS]].ascast
+  // CHECK-DAG: [[NoAS_CAST:%[a-zA-Z0-9]+]] = addrspacecast ptr addrspace(4) [[NoAS_LOAD]] to ptr
+  // CHECK-DAG: store ptr [[NoAS_CAST]], ptr addrspace(4) [[PRIV]].ascast
   PRIV = (__attribute__((opencl_private)) int *)NoAS;
   // From opencl_global_[host/device] address spaces to opencl_global
-  // CHECK-DAG: [[GLOBDEVICE_LOAD:%[a-zA-Z0-9]+]] = load i32 addrspace(5)*, i32 addrspace(5)* addrspace(4)* [[GLOB_DEVICE]].ascast
-  // CHECK-DAG: [[GLOBDEVICE_CAST:%[a-zA-Z0-9]+]] = addrspacecast i32 addrspace(5)* [[GLOBDEVICE_LOAD]] to i32 addrspace(1)*
-  // CHECK-DAG: store i32 addrspace(1)* [[GLOBDEVICE_CAST]], i32 addrspace(1)* addrspace(4)* [[GLOB]].ascast
+  // CHECK-DAG: [[GLOBDEVICE_LOAD:%[a-zA-Z0-9]+]] = load ptr addrspace(5), ptr addrspace(4) [[GLOB_DEVICE]].ascast
+  // CHECK-DAG: [[GLOBDEVICE_CAST:%[a-zA-Z0-9]+]] = addrspacecast ptr addrspace(5) [[GLOBDEVICE_LOAD]] to ptr addrspace(1)
+  // CHECK-DAG: store ptr addrspace(1) [[GLOBDEVICE_CAST]], ptr addrspace(4) [[GLOB]].ascast
   GLOB = (__attribute__((opencl_global)) int *)GLOBDEVICE;
-  // CHECK-DAG: [[GLOBHOST_LOAD:%[a-zA-Z0-9]+]] = load i32 addrspace(6)*, i32 addrspace(6)* addrspace(4)* [[GLOB_HOST]].ascast
-  // CHECK-DAG: [[GLOBHOST_CAST:%[a-zA-Z0-9]+]] = addrspacecast i32 addrspace(6)* [[GLOBHOST_LOAD]] to i32 addrspace(1)*
-  // CHECK-DAG: store i32 addrspace(1)* [[GLOBHOST_CAST]], i32 addrspace(1)* addrspace(4)* [[GLOB]].ascast
+  // CHECK-DAG: [[GLOBHOST_LOAD:%[a-zA-Z0-9]+]] = load ptr addrspace(6), ptr addrspace(4) [[GLOB_HOST]].ascast
+  // CHECK-DAG: [[GLOBHOST_CAST:%[a-zA-Z0-9]+]] = addrspacecast ptr addrspace(6) [[GLOBHOST_LOAD]] to ptr addrspace(1)
+  // CHECK-DAG: store ptr addrspace(1) [[GLOBHOST_CAST]], ptr addrspace(4) [[GLOB]].ascast
   GLOB = (__attribute__((opencl_global)) int *)GLOBHOST;
 
   bar(*GLOB);
-  // CHECK-DAG: [[GLOB_LOAD:%[a-zA-Z0-9]+]] = load i32 addrspace(1)*, i32 addrspace(1)* addrspace(4)* [[GLOB]].ascast
-  // CHECK-DAG: [[GLOB_CAST:%[a-zA-Z0-9]+]] = addrspacecast i32 addrspace(1)* [[GLOB_LOAD]] to i32 addrspace(4)*
-  // CHECK-DAG: call spir_func void @[[RAW_REF]](i32 addrspace(4)* noundef align 4 dereferenceable(4) [[GLOB_CAST]])
+  // CHECK-DAG: [[GLOB_LOAD:%[a-zA-Z0-9]+]] = load ptr addrspace(1), ptr addrspace(4) [[GLOB]].ascast
+  // CHECK-DAG: [[GLOB_CAST:%[a-zA-Z0-9]+]] = addrspacecast ptr addrspace(1) [[GLOB_LOAD]] to ptr addrspace(4)
+  // CHECK-DAG: call spir_func void @[[RAW_REF]](ptr addrspace(4) noundef align 4 dereferenceable(4) [[GLOB_CAST]])
   bar2(*GLOB);
-  // CHECK-DAG: [[GLOB_LOAD2:%[a-zA-Z0-9]+]] = load i32 addrspace(1)*, i32 addrspace(1)* addrspace(4)* [[GLOB]].ascast
-  // CHECK-DAG: [[GLOB_CAST2:%[a-zA-Z0-9]+]] = addrspacecast i32 addrspace(1)* [[GLOB_LOAD2]] to i32 addrspace(4)*
-  // CHECK-DAG: call spir_func void @[[RAW_REF2]](i32 addrspace(4)* noundef align 4 dereferenceable(4) [[GLOB_CAST2]])
+  // CHECK-DAG: [[GLOB_LOAD2:%[a-zA-Z0-9]+]] = load ptr addrspace(1), ptr addrspace(4) [[GLOB]].ascast
+  // CHECK-DAG: [[GLOB_CAST2:%[a-zA-Z0-9]+]] = addrspacecast ptr addrspace(1) [[GLOB_LOAD2]] to ptr addrspace(4)
+  // CHECK-DAG: call spir_func void @[[RAW_REF2]](ptr addrspace(4) noundef align 4 dereferenceable(4) [[GLOB_CAST2]])
 
   bar(*LOC);
-  // CHECK-DAG: [[LOC_LOAD:%[a-zA-Z0-9]+]] = load i32 addrspace(3)*, i32 addrspace(3)* addrspace(4)* [[LOC]].ascast
-  // CHECK-DAG: call spir_func void [[LOC_REF]](i32 addrspace(3)* noundef align 4 dereferenceable(4) [[LOC_LOAD]])
+  // CHECK-DAG: [[LOC_LOAD:%[a-zA-Z0-9]+]] = load ptr addrspace(3), ptr addrspace(4) [[LOC]].ascast
+  // CHECK-DAG: call spir_func void [[LOC_REF]](ptr addrspace(3) noundef align 4 dereferenceable(4) [[LOC_LOAD]])
   bar2(*LOC);
-  // CHECK-DAG: [[LOC_LOAD2:%[a-zA-Z0-9]+]] = load i32 addrspace(3)*, i32 addrspace(3)* addrspace(4)* [[LOC]].ascast
-  // CHECK-DAG: [[LOC_CAST2:%[a-zA-Z0-9]+]] = addrspacecast i32 addrspace(3)* [[LOC_LOAD2]] to i32 addrspace(4)*
-  // CHECK-DAG: call spir_func void @[[RAW_REF2]](i32 addrspace(4)* noundef align 4 dereferenceable(4) [[LOC_CAST2]])
+  // CHECK-DAG: [[LOC_LOAD2:%[a-zA-Z0-9]+]] = load ptr addrspace(3), ptr addrspace(4) [[LOC]].ascast
+  // CHECK-DAG: [[LOC_CAST2:%[a-zA-Z0-9]+]] = addrspacecast ptr addrspace(3) [[LOC_LOAD2]] to ptr addrspace(4)
+  // CHECK-DAG: call spir_func void @[[RAW_REF2]](ptr addrspace(4) noundef align 4 dereferenceable(4) [[LOC_CAST2]])
 
   bar(*NoAS);
-  // CHECK-DAG: [[NoAS_LOAD:%[a-zA-Z0-9]+]] = load i32 addrspace(4)*, i32 addrspace(4)* addrspace(4)* [[NoAS]].ascast
-  // CHECK-DAG: call spir_func void @[[RAW_REF]](i32 addrspace(4)* noundef align 4 dereferenceable(4) [[NoAS_LOAD]])
+  // CHECK-DAG: [[NoAS_LOAD:%[a-zA-Z0-9]+]] = load ptr addrspace(4), ptr addrspace(4) [[NoAS]].ascast
+  // CHECK-DAG: call spir_func void @[[RAW_REF]](ptr addrspace(4) noundef align 4 dereferenceable(4) [[NoAS_LOAD]])
   bar2(*NoAS);
-  // CHECK-DAG: [[NoAS_LOAD2:%[a-zA-Z0-9]+]] = load i32 addrspace(4)*, i32 addrspace(4)* addrspace(4)* [[NoAS]].ascast
-  // CHECK-DAG: call spir_func void @[[RAW_REF2]](i32 addrspace(4)* noundef align 4 dereferenceable(4) [[NoAS_LOAD2]])
+  // CHECK-DAG: [[NoAS_LOAD2:%[a-zA-Z0-9]+]] = load ptr addrspace(4), ptr addrspace(4) [[NoAS]].ascast
+  // CHECK-DAG: call spir_func void @[[RAW_REF2]](ptr addrspace(4) noundef align 4 dereferenceable(4) [[NoAS_LOAD2]])
 
   foo(GLOB);
-  // CHECK-DAG: [[GLOB_LOAD3:%[a-zA-Z0-9]+]] = load i32 addrspace(1)*, i32 addrspace(1)* addrspace(4)* [[GLOB]].ascast
-  // CHECK-DAG: [[GLOB_CAST3:%[a-zA-Z0-9]+]] = addrspacecast i32 addrspace(1)* [[GLOB_LOAD3]] to i32 addrspace(4)*
-  // CHECK-DAG: call spir_func void @[[RAW_PTR]](i32 addrspace(4)* noundef [[GLOB_CAST3]])
+  // CHECK-DAG: [[GLOB_LOAD3:%[a-zA-Z0-9]+]] = load ptr addrspace(1), ptr addrspace(4) [[GLOB]].ascast
+  // CHECK-DAG: [[GLOB_CAST3:%[a-zA-Z0-9]+]] = addrspacecast ptr addrspace(1) [[GLOB_LOAD3]] to ptr addrspace(4)
+  // CHECK-DAG: call spir_func void @[[RAW_PTR]](ptr addrspace(4) noundef [[GLOB_CAST3]])
   foo2(GLOB);
-  // CHECK-DAG: [[GLOB_LOAD4:%[a-zA-Z0-9]+]] = load i32 addrspace(1)*, i32 addrspace(1)* addrspace(4)* [[GLOB]].ascast
-  // CHECK-DAG: [[GLOB_CAST4:%[a-zA-Z0-9]+]] = addrspacecast i32 addrspace(1)* [[GLOB_LOAD4]] to i32 addrspace(4)*
-  // CHECK-DAG: call spir_func void @[[RAW_PTR2]](i32 addrspace(4)* noundef [[GLOB_CAST4]])
+  // CHECK-DAG: [[GLOB_LOAD4:%[a-zA-Z0-9]+]] = load ptr addrspace(1), ptr addrspace(4) [[GLOB]].ascast
+  // CHECK-DAG: [[GLOB_CAST4:%[a-zA-Z0-9]+]] = addrspacecast ptr addrspace(1) [[GLOB_LOAD4]] to ptr addrspace(4)
+  // CHECK-DAG: call spir_func void @[[RAW_PTR2]](ptr addrspace(4) noundef [[GLOB_CAST4]])
   foo(LOC);
-  // CHECK-DAG: [[LOC_LOAD3:%[a-zA-Z0-9]+]] = load i32 addrspace(3)*, i32 addrspace(3)* addrspace(4)* [[LOC]].ascast
-  // CHECK-DAG: call spir_func void [[LOC_PTR]](i32 addrspace(3)* noundef [[LOC_LOAD3]])
+  // CHECK-DAG: [[LOC_LOAD3:%[a-zA-Z0-9]+]] = load ptr addrspace(3), ptr addrspace(4) [[LOC]].ascast
+  // CHECK-DAG: call spir_func void [[LOC_PTR]](ptr addrspace(3) noundef [[LOC_LOAD3]])
   foo2(LOC);
-  // CHECK-DAG: [[LOC_LOAD4:%[a-zA-Z0-9]+]] = load i32 addrspace(3)*, i32 addrspace(3)* addrspace(4)* [[LOC]].ascast
-  // CHECK-DAG: [[LOC_CAST4:%[a-zA-Z0-9]+]] = addrspacecast i32 addrspace(3)* [[LOC_LOAD4]] to i32 addrspace(4)*
-  // CHECK-DAG: call spir_func void @[[RAW_PTR2]](i32 addrspace(4)* noundef [[LOC_CAST4]])
+  // CHECK-DAG: [[LOC_LOAD4:%[a-zA-Z0-9]+]] = load ptr addrspace(3), ptr addrspace(4) [[LOC]].ascast
+  // CHECK-DAG: [[LOC_CAST4:%[a-zA-Z0-9]+]] = addrspacecast ptr addrspace(3) [[LOC_LOAD4]] to ptr addrspace(4)
+  // CHECK-DAG: call spir_func void @[[RAW_PTR2]](ptr addrspace(4) noundef [[LOC_CAST4]])
   foo(NoAS);
-  // CHECK-DAG: [[NoAS_LOAD3:%[a-zA-Z0-9]+]] = load i32 addrspace(4)*, i32 addrspace(4)* addrspace(4)* [[NoAS]].ascast
-  // CHECK-DAG: call spir_func void @[[RAW_PTR]](i32 addrspace(4)* noundef [[NoAS_LOAD3]])
+  // CHECK-DAG: [[NoAS_LOAD3:%[a-zA-Z0-9]+]] = load ptr addrspace(4), ptr addrspace(4) [[NoAS]].ascast
+  // CHECK-DAG: call spir_func void @[[RAW_PTR]](ptr addrspace(4) noundef [[NoAS_LOAD3]])
   foo2(NoAS);
-  // CHECK-DAG: [[NoAS_LOAD4:%[a-zA-Z0-9]+]] = load i32 addrspace(4)*, i32 addrspace(4)* addrspace(4)* [[NoAS]].ascast
-  // CHECK-DAG: call spir_func void @[[RAW_PTR2]](i32 addrspace(4)* noundef [[NoAS_LOAD4]])
+  // CHECK-DAG: [[NoAS_LOAD4:%[a-zA-Z0-9]+]] = load ptr addrspace(4), ptr addrspace(4) [[NoAS]].ascast
+  // CHECK-DAG: call spir_func void @[[RAW_PTR2]](ptr addrspace(4) noundef [[NoAS_LOAD4]])
 
   // Ensure that we still get 3 
diff erent template instantiations.
   tmpl(GLOB);
-  // CHECK-DAG: [[GLOB_LOAD4:%[a-zA-Z0-9]+]] = load i32 addrspace(1)*, i32 addrspace(1)* addrspace(4)* [[GLOB]].ascast
-  // CHECK-DAG: call spir_func void @_Z4tmplIPU3AS1iEvT_(i32 addrspace(1)* noundef [[GLOB_LOAD4]])
+  // CHECK-DAG: [[GLOB_LOAD4:%[a-zA-Z0-9]+]] = load ptr addrspace(1), ptr addrspace(4) [[GLOB]].ascast
+  // CHECK-DAG: call spir_func void @_Z4tmplIPU3AS1iEvT_(ptr addrspace(1) noundef [[GLOB_LOAD4]])
   tmpl(LOC);
-  // CHECK-DAG: [[LOC_LOAD5:%[a-zA-Z0-9]+]] = load i32 addrspace(3)*, i32 addrspace(3)* addrspace(4)* [[LOC]].ascast
-  // CHECK-DAG: call spir_func void @_Z4tmplIPU3AS3iEvT_(i32 addrspace(3)* noundef [[LOC_LOAD5]])
+  // CHECK-DAG: [[LOC_LOAD5:%[a-zA-Z0-9]+]] = load ptr addrspace(3), ptr addrspace(4) [[LOC]].ascast
+  // CHECK-DAG: call spir_func void @_Z4tmplIPU3AS3iEvT_(ptr addrspace(3) noundef [[LOC_LOAD5]])
   tmpl(PRIV);
-  // CHECK-DAG: [[PRIV_LOAD5:%[a-zA-Z0-9]+]] = load i32*, i32* addrspace(4)* [[PRIV]].ascast
-  // CHECK-DAG: call spir_func void @_Z4tmplIPU3AS0iEvT_(i32* noundef [[PRIV_LOAD5]])
+  // CHECK-DAG: [[PRIV_LOAD5:%[a-zA-Z0-9]+]] = load ptr, ptr addrspace(4) [[PRIV]].ascast
+  // CHECK-DAG: call spir_func void @_Z4tmplIPU3AS0iEvT_(ptr noundef [[PRIV_LOAD5]])
   tmpl(NoAS);
-  // CHECK-DAG: [[NoAS_LOAD5:%[a-zA-Z0-9]+]] = load i32 addrspace(4)*, i32 addrspace(4)* addrspace(4)* [[NoAS]].ascast
-  // CHECK-DAG: call spir_func void @_Z4tmplIPiEvT_(i32 addrspace(4)* noundef [[NoAS_LOAD5]])
+  // CHECK-DAG: [[NoAS_LOAD5:%[a-zA-Z0-9]+]] = load ptr addrspace(4), ptr addrspace(4) [[NoAS]].ascast
+  // CHECK-DAG: call spir_func void @_Z4tmplIPiEvT_(ptr addrspace(4) noundef [[NoAS_LOAD5]])
 }
 
-// CHECK-DAG: define linkonce_odr spir_func void @_Z4tmplIPU3AS1iEvT_(i32 addrspace(1)* noundef %
-// CHECK-DAG: define linkonce_odr spir_func void @_Z4tmplIPU3AS3iEvT_(i32 addrspace(3)* noundef %
-// CHECK-DAG: define linkonce_odr spir_func void @_Z4tmplIPU3AS0iEvT_(i32* noundef %
-// CHECK-DAG: define linkonce_odr spir_func void @_Z4tmplIPiEvT_(i32 addrspace(4)* noundef %
+// CHECK-DAG: define linkonce_odr spir_func void @_Z4tmplIPU3AS1iEvT_(ptr addrspace(1) noundef %
+// CHECK-DAG: define linkonce_odr spir_func void @_Z4tmplIPU3AS3iEvT_(ptr addrspace(3) noundef %
+// CHECK-DAG: define linkonce_odr spir_func void @_Z4tmplIPU3AS0iEvT_(ptr noundef %
+// CHECK-DAG: define linkonce_odr spir_func void @_Z4tmplIPiEvT_(ptr addrspace(4) noundef %

diff  --git a/clang/test/CodeGenSYCL/address-space-deduction.cpp b/clang/test/CodeGenSYCL/address-space-deduction.cpp
index 730bcae4e4f79..96075a47343fe 100644
--- a/clang/test/CodeGenSYCL/address-space-deduction.cpp
+++ b/clang/test/CodeGenSYCL/address-space-deduction.cpp
@@ -1,90 +1,88 @@
 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
-// RUN: %clang_cc1 -no-opaque-pointers -triple spir64 -fsycl-is-device -disable-llvm-passes -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -triple spir64 -fsycl-is-device -disable-llvm-passes -emit-llvm %s -o - | FileCheck %s
 
 // CHECK-LABEL: @_Z4testv(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[I:%.*]] = alloca i32, align 4
-// CHECK-NEXT:    [[PPTR:%.*]] = alloca i32 addrspace(4)*, align 8
+// CHECK-NEXT:    [[PPTR:%.*]] = alloca ptr addrspace(4), align 8
 // CHECK-NEXT:    [[IS_I_PTR:%.*]] = alloca i8, align 1
 // CHECK-NEXT:    [[VAR23:%.*]] = alloca i32, align 4
-// CHECK-NEXT:    [[CP:%.*]] = alloca i8 addrspace(4)*, align 8
+// CHECK-NEXT:    [[CP:%.*]] = alloca ptr addrspace(4), align 8
 // CHECK-NEXT:    [[ARR:%.*]] = alloca [42 x i32], align 4
-// CHECK-NEXT:    [[CPP:%.*]] = alloca i8 addrspace(4)*, align 8
-// CHECK-NEXT:    [[APTR:%.*]] = alloca i32 addrspace(4)*, align 8
-// CHECK-NEXT:    [[STR:%.*]] = alloca i8 addrspace(4)*, align 8
-// CHECK-NEXT:    [[PHI_STR:%.*]] = alloca i8 addrspace(4)*, align 8
-// CHECK-NEXT:    [[SELECT_NULL:%.*]] = alloca i8 addrspace(4)*, align 8
-// CHECK-NEXT:    [[SELECT_STR_TRIVIAL1:%.*]] = alloca i8 addrspace(4)*, align 8
-// CHECK-NEXT:    [[SELECT_STR_TRIVIAL2:%.*]] = alloca i8 addrspace(4)*, align 8
-// CHECK-NEXT:    [[I_ASCAST:%.*]] = addrspacecast i32* [[I]] to i32 addrspace(4)*
-// CHECK-NEXT:    [[PPTR_ASCAST:%.*]] = addrspacecast i32 addrspace(4)** [[PPTR]] to i32 addrspace(4)* addrspace(4)*
-// CHECK-NEXT:    [[IS_I_PTR_ASCAST:%.*]] = addrspacecast i8* [[IS_I_PTR]] to i8 addrspace(4)*
-// CHECK-NEXT:    [[VAR23_ASCAST:%.*]] = addrspacecast i32* [[VAR23]] to i32 addrspace(4)*
-// CHECK-NEXT:    [[CP_ASCAST:%.*]] = addrspacecast i8 addrspace(4)** [[CP]] to i8 addrspace(4)* addrspace(4)*
-// CHECK-NEXT:    [[ARR_ASCAST:%.*]] = addrspacecast [42 x i32]* [[ARR]] to [42 x i32] addrspace(4)*
-// CHECK-NEXT:    [[CPP_ASCAST:%.*]] = addrspacecast i8 addrspace(4)** [[CPP]] to i8 addrspace(4)* addrspace(4)*
-// CHECK-NEXT:    [[APTR_ASCAST:%.*]] = addrspacecast i32 addrspace(4)** [[APTR]] to i32 addrspace(4)* addrspace(4)*
-// CHECK-NEXT:    [[STR_ASCAST:%.*]] = addrspacecast i8 addrspace(4)** [[STR]] to i8 addrspace(4)* addrspace(4)*
-// CHECK-NEXT:    [[PHI_STR_ASCAST:%.*]] = addrspacecast i8 addrspace(4)** [[PHI_STR]] to i8 addrspace(4)* addrspace(4)*
-// CHECK-NEXT:    [[SELECT_NULL_ASCAST:%.*]] = addrspacecast i8 addrspace(4)** [[SELECT_NULL]] to i8 addrspace(4)* addrspace(4)*
-// CHECK-NEXT:    [[SELECT_STR_TRIVIAL1_ASCAST:%.*]] = addrspacecast i8 addrspace(4)** [[SELECT_STR_TRIVIAL1]] to i8 addrspace(4)* addrspace(4)*
-// CHECK-NEXT:    [[SELECT_STR_TRIVIAL2_ASCAST:%.*]] = addrspacecast i8 addrspace(4)** [[SELECT_STR_TRIVIAL2]] to i8 addrspace(4)* addrspace(4)*
-// CHECK-NEXT:    store i32 0, i32 addrspace(4)* [[I_ASCAST]], align 4
-// CHECK-NEXT:    store i32 addrspace(4)* [[I_ASCAST]], i32 addrspace(4)* addrspace(4)* [[PPTR_ASCAST]], align 8
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32 addrspace(4)*, i32 addrspace(4)* addrspace(4)* [[PPTR_ASCAST]], align 8
-// CHECK-NEXT:    [[CMP:%.*]] = icmp eq i32 addrspace(4)* [[TMP0]], [[I_ASCAST]]
+// CHECK-NEXT:    [[CPP:%.*]] = alloca ptr addrspace(4), align 8
+// CHECK-NEXT:    [[APTR:%.*]] = alloca ptr addrspace(4), align 8
+// CHECK-NEXT:    [[STR:%.*]] = alloca ptr addrspace(4), align 8
+// CHECK-NEXT:    [[PHI_STR:%.*]] = alloca ptr addrspace(4), align 8
+// CHECK-NEXT:    [[SELECT_NULL:%.*]] = alloca ptr addrspace(4), align 8
+// CHECK-NEXT:    [[SELECT_STR_TRIVIAL1:%.*]] = alloca ptr addrspace(4), align 8
+// CHECK-NEXT:    [[SELECT_STR_TRIVIAL2:%.*]] = alloca ptr addrspace(4), align 8
+// CHECK-NEXT:    [[I_ASCAST:%.*]] = addrspacecast ptr [[I]] to ptr addrspace(4)
+// CHECK-NEXT:    [[PPTR_ASCAST:%.*]] = addrspacecast ptr [[PPTR]] to ptr addrspace(4)
+// CHECK-NEXT:    [[IS_I_PTR_ASCAST:%.*]] = addrspacecast ptr [[IS_I_PTR]] to ptr addrspace(4)
+// CHECK-NEXT:    [[VAR23_ASCAST:%.*]] = addrspacecast ptr [[VAR23]] to ptr addrspace(4)
+// CHECK-NEXT:    [[CP_ASCAST:%.*]] = addrspacecast ptr [[CP]] to ptr addrspace(4)
+// CHECK-NEXT:    [[ARR_ASCAST:%.*]] = addrspacecast ptr [[ARR]] to ptr addrspace(4)
+// CHECK-NEXT:    [[CPP_ASCAST:%.*]] = addrspacecast ptr [[CPP]] to ptr addrspace(4)
+// CHECK-NEXT:    [[APTR_ASCAST:%.*]] = addrspacecast ptr [[APTR]] to ptr addrspace(4)
+// CHECK-NEXT:    [[STR_ASCAST:%.*]] = addrspacecast ptr [[STR]] to ptr addrspace(4)
+// CHECK-NEXT:    [[PHI_STR_ASCAST:%.*]] = addrspacecast ptr [[PHI_STR]] to ptr addrspace(4)
+// CHECK-NEXT:    [[SELECT_NULL_ASCAST:%.*]] = addrspacecast ptr [[SELECT_NULL]] to ptr addrspace(4)
+// CHECK-NEXT:    [[SELECT_STR_TRIVIAL1_ASCAST:%.*]] = addrspacecast ptr [[SELECT_STR_TRIVIAL1]] to ptr addrspace(4)
+// CHECK-NEXT:    [[SELECT_STR_TRIVIAL2_ASCAST:%.*]] = addrspacecast ptr [[SELECT_STR_TRIVIAL2]] to ptr addrspace(4)
+// CHECK-NEXT:    store i32 0, ptr addrspace(4) [[I_ASCAST]], align 4
+// CHECK-NEXT:    store ptr addrspace(4) [[I_ASCAST]], ptr addrspace(4) [[PPTR_ASCAST]], align 8
+// CHECK-NEXT:    [[TMP0:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[PPTR_ASCAST]], align 8
+// CHECK-NEXT:    [[CMP:%.*]] = icmp eq ptr addrspace(4) [[TMP0]], [[I_ASCAST]]
 // CHECK-NEXT:    [[FROMBOOL:%.*]] = zext i1 [[CMP]] to i8
-// CHECK-NEXT:    store i8 [[FROMBOOL]], i8 addrspace(4)* [[IS_I_PTR_ASCAST]], align 1
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32 addrspace(4)*, i32 addrspace(4)* addrspace(4)* [[PPTR_ASCAST]], align 8
-// CHECK-NEXT:    store i32 66, i32 addrspace(4)* [[TMP1]], align 4
-// CHECK-NEXT:    store i32 23, i32 addrspace(4)* [[VAR23_ASCAST]], align 4
-// CHECK-NEXT:    [[TMP2:%.*]] = bitcast i32 addrspace(4)* [[VAR23_ASCAST]] to i8 addrspace(4)*
-// CHECK-NEXT:    store i8 addrspace(4)* [[TMP2]], i8 addrspace(4)* addrspace(4)* [[CP_ASCAST]], align 8
-// CHECK-NEXT:    [[TMP3:%.*]] = load i8 addrspace(4)*, i8 addrspace(4)* addrspace(4)* [[CP_ASCAST]], align 8
-// CHECK-NEXT:    store i8 41, i8 addrspace(4)* [[TMP3]], align 1
-// CHECK-NEXT:    [[ARRAYDECAY:%.*]] = getelementptr inbounds [42 x i32], [42 x i32] addrspace(4)* [[ARR_ASCAST]], i64 0, i64 0
-// CHECK-NEXT:    [[TMP4:%.*]] = bitcast i32 addrspace(4)* [[ARRAYDECAY]] to i8 addrspace(4)*
-// CHECK-NEXT:    store i8 addrspace(4)* [[TMP4]], i8 addrspace(4)* addrspace(4)* [[CPP_ASCAST]], align 8
-// CHECK-NEXT:    [[TMP5:%.*]] = load i8 addrspace(4)*, i8 addrspace(4)* addrspace(4)* [[CPP_ASCAST]], align 8
-// CHECK-NEXT:    store i8 43, i8 addrspace(4)* [[TMP5]], align 1
-// CHECK-NEXT:    [[ARRAYDECAY1:%.*]] = getelementptr inbounds [42 x i32], [42 x i32] addrspace(4)* [[ARR_ASCAST]], i64 0, i64 0
-// CHECK-NEXT:    [[ADD_PTR:%.*]] = getelementptr inbounds i32, i32 addrspace(4)* [[ARRAYDECAY1]], i64 10
-// CHECK-NEXT:    store i32 addrspace(4)* [[ADD_PTR]], i32 addrspace(4)* addrspace(4)* [[APTR_ASCAST]], align 8
-// CHECK-NEXT:    [[TMP6:%.*]] = load i32 addrspace(4)*, i32 addrspace(4)* addrspace(4)* [[APTR_ASCAST]], align 8
-// CHECK-NEXT:    [[ARRAYDECAY2:%.*]] = getelementptr inbounds [42 x i32], [42 x i32] addrspace(4)* [[ARR_ASCAST]], i64 0, i64 0
-// CHECK-NEXT:    [[ADD_PTR3:%.*]] = getelementptr inbounds i32, i32 addrspace(4)* [[ARRAYDECAY2]], i64 168
-// CHECK-NEXT:    [[CMP4:%.*]] = icmp ult i32 addrspace(4)* [[TMP6]], [[ADD_PTR3]]
+// CHECK-NEXT:    store i8 [[FROMBOOL]], ptr addrspace(4) [[IS_I_PTR_ASCAST]], align 1
+// CHECK-NEXT:    [[TMP1:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[PPTR_ASCAST]], align 8
+// CHECK-NEXT:    store i32 66, ptr addrspace(4) [[TMP1]], align 4
+// CHECK-NEXT:    store i32 23, ptr addrspace(4) [[VAR23_ASCAST]], align 4
+// CHECK-NEXT:    store ptr addrspace(4) [[VAR23_ASCAST]], ptr addrspace(4) [[CP_ASCAST]], align 8
+// CHECK-NEXT:    [[TMP3:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[CP_ASCAST]], align 8
+// CHECK-NEXT:    store i8 41, ptr addrspace(4) [[TMP3]], align 1
+// CHECK-NEXT:    [[ARRAYDECAY:%.*]] = getelementptr inbounds [42 x i32], ptr addrspace(4) [[ARR_ASCAST]], i64 0, i64 0
+// CHECK-NEXT:    store ptr addrspace(4) [[ARRAYDECAY]], ptr addrspace(4) [[CPP_ASCAST]], align 8
+// CHECK-NEXT:    [[TMP5:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[CPP_ASCAST]], align 8
+// CHECK-NEXT:    store i8 43, ptr addrspace(4) [[TMP5]], align 1
+// CHECK-NEXT:    [[ARRAYDECAY1:%.*]] = getelementptr inbounds [42 x i32], ptr addrspace(4) [[ARR_ASCAST]], i64 0, i64 0
+// CHECK-NEXT:    [[ADD_PTR:%.*]] = getelementptr inbounds i32, ptr addrspace(4) [[ARRAYDECAY1]], i64 10
+// CHECK-NEXT:    store ptr addrspace(4) [[ADD_PTR]], ptr addrspace(4) [[APTR_ASCAST]], align 8
+// CHECK-NEXT:    [[TMP6:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[APTR_ASCAST]], align 8
+// CHECK-NEXT:    [[ARRAYDECAY2:%.*]] = getelementptr inbounds [42 x i32], ptr addrspace(4) [[ARR_ASCAST]], i64 0, i64 0
+// CHECK-NEXT:    [[ADD_PTR3:%.*]] = getelementptr inbounds i32, ptr addrspace(4) [[ARRAYDECAY2]], i64 168
+// CHECK-NEXT:    [[CMP4:%.*]] = icmp ult ptr addrspace(4) [[TMP6]], [[ADD_PTR3]]
 // CHECK-NEXT:    br i1 [[CMP4]], label [[IF_THEN:%.*]], label [[IF_END:%.*]]
 // CHECK:       if.then:
-// CHECK-NEXT:    [[TMP7:%.*]] = load i32 addrspace(4)*, i32 addrspace(4)* addrspace(4)* [[APTR_ASCAST]], align 8
-// CHECK-NEXT:    store i32 44, i32 addrspace(4)* [[TMP7]], align 4
+// CHECK-NEXT:    [[TMP7:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[APTR_ASCAST]], align 8
+// CHECK-NEXT:    store i32 44, ptr addrspace(4) [[TMP7]], align 4
 // CHECK-NEXT:    br label [[IF_END]]
 // CHECK:       if.end:
-// CHECK-NEXT:    store i8 addrspace(4)* getelementptr inbounds ([14 x i8], [14 x i8] addrspace(4)* addrspacecast ([14 x i8] addrspace(1)* @.str to [14 x i8] addrspace(4)*), i64 0, i64 0), i8 addrspace(4)* addrspace(4)* [[STR_ASCAST]], align 8
-// CHECK-NEXT:    [[TMP8:%.*]] = load i8 addrspace(4)*, i8 addrspace(4)* addrspace(4)* [[STR_ASCAST]], align 8
-// CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i8, i8 addrspace(4)* [[TMP8]], i64 0
-// CHECK-NEXT:    [[TMP9:%.*]] = load i8, i8 addrspace(4)* [[ARRAYIDX]], align 1
+// CHECK-NEXT:    store ptr addrspace(4) addrspacecast (ptr addrspace(1) @.str to ptr addrspace(4)), ptr addrspace(4) [[STR_ASCAST]], align 8
+// CHECK-NEXT:    [[TMP8:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[STR_ASCAST]], align 8
+// CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i8, ptr addrspace(4) [[TMP8]], i64 0
+// CHECK-NEXT:    [[TMP9:%.*]] = load i8, ptr addrspace(4) [[ARRAYIDX]], align 1
 // CHECK-NEXT:    [[CONV:%.*]] = sext i8 [[TMP9]] to i32
-// CHECK-NEXT:    store i32 [[CONV]], i32 addrspace(4)* [[I_ASCAST]], align 4
-// CHECK-NEXT:    [[TMP10:%.*]] = load i32, i32 addrspace(4)* [[I_ASCAST]], align 4
+// CHECK-NEXT:    store i32 [[CONV]], ptr addrspace(4) [[I_ASCAST]], align 4
+// CHECK-NEXT:    [[TMP10:%.*]] = load i32, ptr addrspace(4) [[I_ASCAST]], align 4
 // CHECK-NEXT:    [[CMP5:%.*]] = icmp sgt i32 [[TMP10]], 2
 // CHECK-NEXT:    br i1 [[CMP5]], label [[COND_TRUE:%.*]], label [[COND_FALSE:%.*]]
 // CHECK:       cond.true:
-// CHECK-NEXT:    [[TMP11:%.*]] = load i8 addrspace(4)*, i8 addrspace(4)* addrspace(4)* [[STR_ASCAST]], align 8
+// CHECK-NEXT:    [[TMP11:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[STR_ASCAST]], align 8
 // CHECK-NEXT:    br label [[COND_END:%.*]]
 // CHECK:       cond.false:
 // CHECK-NEXT:    br label [[COND_END]]
 // CHECK:       cond.end:
-// CHECK-NEXT:    [[COND:%.*]] = phi i8 addrspace(4)* [ [[TMP11]], [[COND_TRUE]] ], [ getelementptr inbounds ([21 x i8], [21 x i8] addrspace(4)* addrspacecast ([21 x i8] addrspace(1)* @.str.1 to [21 x i8] addrspace(4)*), i64 0, i64 0), [[COND_FALSE]] ]
-// CHECK-NEXT:    store i8 addrspace(4)* [[COND]], i8 addrspace(4)* addrspace(4)* [[PHI_STR_ASCAST]], align 8
-// CHECK-NEXT:    [[TMP12:%.*]] = load i32, i32 addrspace(4)* [[I_ASCAST]], align 4
+// CHECK-NEXT:    [[COND:%.*]] = phi ptr addrspace(4) [ [[TMP11]], [[COND_TRUE]] ], [ addrspacecast (ptr addrspace(1) @.str.1 to ptr addrspace(4)), [[COND_FALSE]] ]
+// CHECK-NEXT:    store ptr addrspace(4) [[COND]], ptr addrspace(4) [[PHI_STR_ASCAST]], align 8
+// CHECK-NEXT:    [[TMP12:%.*]] = load i32, ptr addrspace(4) [[I_ASCAST]], align 4
 // CHECK-NEXT:    [[CMP6:%.*]] = icmp sgt i32 [[TMP12]], 2
 // CHECK-NEXT:    [[TMP13:%.*]] = zext i1 [[CMP6]] to i64
-// CHECK-NEXT:    [[COND7:%.*]] = select i1 [[CMP6]], i8 addrspace(4)* getelementptr inbounds ([24 x i8], [24 x i8] addrspace(4)* addrspacecast ([24 x i8] addrspace(1)* @.str.2 to [24 x i8] addrspace(4)*), i64 0, i64 0), i8 addrspace(4)* null
-// CHECK-NEXT:    store i8 addrspace(4)* [[COND7]], i8 addrspace(4)* addrspace(4)* [[SELECT_NULL_ASCAST]], align 8
-// CHECK-NEXT:    [[TMP14:%.*]] = load i8 addrspace(4)*, i8 addrspace(4)* addrspace(4)* [[STR_ASCAST]], align 8
-// CHECK-NEXT:    store i8 addrspace(4)* [[TMP14]], i8 addrspace(4)* addrspace(4)* [[SELECT_STR_TRIVIAL1_ASCAST]], align 8
-// CHECK-NEXT:    store i8 addrspace(4)* getelementptr inbounds ([21 x i8], [21 x i8] addrspace(4)* addrspacecast ([21 x i8] addrspace(1)* @.str.1 to [21 x i8] addrspace(4)*), i64 0, i64 0), i8 addrspace(4)* addrspace(4)* [[SELECT_STR_TRIVIAL2_ASCAST]], align 8
+// CHECK-NEXT:    [[COND7:%.*]] = select i1 [[CMP6]], ptr addrspace(4) addrspacecast (ptr addrspace(1) @.str.2 to ptr addrspace(4)), ptr addrspace(4) null
+// CHECK-NEXT:    store ptr addrspace(4) [[COND7]], ptr addrspace(4) [[SELECT_NULL_ASCAST]], align 8
+// CHECK-NEXT:    [[TMP14:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[STR_ASCAST]], align 8
+// CHECK-NEXT:    store ptr addrspace(4) [[TMP14]], ptr addrspace(4) [[SELECT_STR_TRIVIAL1_ASCAST]], align 8
+// CHECK-NEXT:    store ptr addrspace(4) addrspacecast (ptr addrspace(1) @.str.1 to ptr addrspace(4)), ptr addrspace(4) [[SELECT_STR_TRIVIAL2_ASCAST]], align 8
 // CHECK-NEXT:    ret void
 //
 void test() {

diff  --git a/clang/test/CodeGenSYCL/address-space-mangling.cpp b/clang/test/CodeGenSYCL/address-space-mangling.cpp
index be0e1edff6bd0..868bf8ccbdcf8 100644
--- a/clang/test/CodeGenSYCL/address-space-mangling.cpp
+++ b/clang/test/CodeGenSYCL/address-space-mangling.cpp
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple spir64 -fsycl-is-device -disable-llvm-passes -emit-llvm %s -o - | FileCheck %s --check-prefix=SPIR
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64 -fsycl-is-device -disable-llvm-passes -emit-llvm %s -o - | FileCheck %s --check-prefix=X86
+// RUN: %clang_cc1 -triple spir64 -fsycl-is-device -disable-llvm-passes -emit-llvm %s -o - | FileCheck %s --check-prefix=SPIR
+// RUN: %clang_cc1 -triple x86_64 -fsycl-is-device -disable-llvm-passes -emit-llvm %s -o - | FileCheck %s --check-prefix=X86
 
 // REQUIRES: x86-registered-target
 
@@ -8,15 +8,15 @@ void foo(__attribute__((opencl_local)) int *);
 void foo(__attribute__((opencl_private)) int *);
 void foo(int *);
 
-// SPIR: declare spir_func void @_Z3fooPU3AS1i(i32 addrspace(1)* noundef) #1
-// SPIR: declare spir_func void @_Z3fooPU3AS3i(i32 addrspace(3)* noundef) #1
-// SPIR: declare spir_func void @_Z3fooPU3AS0i(i32* noundef) #1
-// SPIR: declare spir_func void @_Z3fooPi(i32 addrspace(4)* noundef) #1
+// SPIR: declare spir_func void @_Z3fooPU3AS1i(ptr addrspace(1) noundef) #1
+// SPIR: declare spir_func void @_Z3fooPU3AS3i(ptr addrspace(3) noundef) #1
+// SPIR: declare spir_func void @_Z3fooPU3AS0i(ptr noundef) #1
+// SPIR: declare spir_func void @_Z3fooPi(ptr addrspace(4) noundef) #1
 
-// X86: declare void @_Z3fooPU8SYglobali(i32* noundef) #1
-// X86: declare void @_Z3fooPU7SYlocali(i32* noundef) #1
-// X86: declare void @_Z3fooPU9SYprivatei(i32* noundef) #1
-// X86: declare void @_Z3fooPi(i32* noundef) #1
+// X86: declare void @_Z3fooPU8SYglobali(ptr noundef) #1
+// X86: declare void @_Z3fooPU7SYlocali(ptr noundef) #1
+// X86: declare void @_Z3fooPU9SYprivatei(ptr noundef) #1
+// X86: declare void @_Z3fooPi(ptr noundef) #1
 
 void test() {
   __attribute__((opencl_global)) int *glob;

diff  --git a/clang/test/CodeGenSYCL/debug-info-kernel-variables.cpp b/clang/test/CodeGenSYCL/debug-info-kernel-variables.cpp
index 44a574a546fb0..361ce0abb1044 100644
--- a/clang/test/CodeGenSYCL/debug-info-kernel-variables.cpp
+++ b/clang/test/CodeGenSYCL/debug-info-kernel-variables.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers %s -o - -O0 -emit-llvm                                     \
+// RUN: %clang_cc1 %s -o - -O0 -emit-llvm                                     \
 // RUN:            -triple spir64-unknown-unknown                             \
 // RUN:            -aux-triple x86_64-unknown-linux-gnu                       \
 // RUN:            -fsycl-is-device                                           \
@@ -35,12 +35,12 @@ int my_host() {
 // CHECK:        %my_param.addr = alloca i32, align 4
 // CHECK:        %my_local = alloca i32, align 4
 // CHECK:        call void @llvm.dbg.declare(
-// CHECK-SAME:     metadata i32* %my_param.addr,
+// CHECK-SAME:     metadata ptr %my_param.addr,
 // CHECK-SAME:     metadata [[MY_PARAM:![0-9]+]],
 // CHECK-SAME:     metadata !DIExpression(DW_OP_constu, 4, DW_OP_swap, DW_OP_xderef)
 // CHECK-SAME:     )
 // CHECK:        call void @llvm.dbg.declare(
-// CHECK-SAME:     metadata i32* %my_local,
+// CHECK-SAME:     metadata ptr %my_local,
 // CHECK-SAME:     metadata [[MY_LOCAL:![0-9]+]],
 // CHECK-SAME:     metadata !DIExpression(DW_OP_constu, 4, DW_OP_swap, DW_OP_xderef)
 // CHECK-SAME:     )

diff  --git a/clang/test/CodeGenSYCL/functionptr-addrspace.cpp b/clang/test/CodeGenSYCL/functionptr-addrspace.cpp
index 6fc2542794472..340caa1e62c59 100644
--- a/clang/test/CodeGenSYCL/functionptr-addrspace.cpp
+++ b/clang/test/CodeGenSYCL/functionptr-addrspace.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -fsycl-is-device -emit-llvm -triple spir64 -verify -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -fsycl-is-device -emit-llvm -triple spir64 -verify -emit-llvm %s -o - | FileCheck %s
 
 // expected-no-diagnostics
 
@@ -7,7 +7,7 @@ __attribute__((sycl_kernel)) void kernel_single_task(const Func &kernelFunc) {
   kernelFunc();
 }
 
-// CHECK: define dso_local spir_func{{.*}}invoke_function{{.*}}(i32 ()* noundef %fptr, i32 addrspace(4)* noundef %ptr)
+// CHECK: define dso_local spir_func{{.*}}invoke_function{{.*}}(ptr noundef %fptr, ptr addrspace(4) noundef %ptr)
 void invoke_function(int (*fptr)(), int *ptr) {}
 
 int f() { return 0; }

diff  --git a/clang/test/Coverage/codegen-gnu.m b/clang/test/Coverage/codegen-gnu.m
index 10be6fbabca27..67f99064a6ab4 100644
--- a/clang/test/Coverage/codegen-gnu.m
+++ b/clang/test/Coverage/codegen-gnu.m
@@ -1,3 +1,3 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple i386-unknown-unknown -fobjc-exceptions -fobjc-runtime=gcc -emit-llvm -o %t %s
+// RUN: %clang_cc1 -triple i386-unknown-unknown -fobjc-exceptions -fobjc-runtime=gcc -emit-llvm -o %t %s
 
 #include "objc-language-features.inc"

diff  --git a/clang/test/Coverage/codegen-next.m b/clang/test/Coverage/codegen-next.m
index d9f012d602064..5430cf28daf06 100644
--- a/clang/test/Coverage/codegen-next.m
+++ b/clang/test/Coverage/codegen-next.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -fobjc-exceptions -triple x86_64-apple-darwin -o %t %s
-// RUN: %clang_cc1 -no-opaque-pointers -debug-info-kind=limited -emit-llvm -fobjc-exceptions -triple x86_64-apple-darwin -o %t %s
+// RUN: %clang_cc1 -emit-llvm -fobjc-exceptions -triple x86_64-apple-darwin -o %t %s
+// RUN: %clang_cc1 -debug-info-kind=limited -emit-llvm -fobjc-exceptions -triple x86_64-apple-darwin -o %t %s
 
 // An error could be seen for targeting x86_64-win32;
 //

diff  --git a/clang/test/CoverageMapping/ir.c b/clang/test/CoverageMapping/ir.c
index 62276189b98e9..6fa9281ab70df 100644
--- a/clang/test/CoverageMapping/ir.c
+++ b/clang/test/CoverageMapping/ir.c
@@ -1,5 +1,5 @@
 // Check the data structures emitted by coverage mapping
-// RUN: %clang_cc1 -mllvm -emptyline-comment-coverage=false -triple x86_64-apple-macosx10.9 -main-file-name ir.c %s -o - -emit-llvm -fprofile-instrument=clang -fcoverage-mapping -mllvm -enable-name-compression=false -no-opaque-pointers | FileCheck %s -check-prefixes=COMMON,DARWIN
+// RUN: %clang_cc1 -mllvm -emptyline-comment-coverage=false -triple x86_64-apple-macosx10.9 -main-file-name ir.c %s -o - -emit-llvm -fprofile-instrument=clang -fcoverage-mapping -mllvm -enable-name-compression=false | FileCheck %s -check-prefixes=COMMON,DARWIN
 // RUN: %clang_cc1 -mllvm -emptyline-comment-coverage=false -triple x86_64-apple-macosx10.9 -main-file-name ir.c %s -o - -emit-llvm -fprofile-instrument=clang -fcoverage-mapping -mllvm -enable-name-compression=false -opaque-pointers | FileCheck %s -check-prefixes=COMMON,DARWIN
 // RUN: %clang_cc1 -mllvm -emptyline-comment-coverage=false -triple x86_64--windows-msvc -main-file-name ir.c %s -o - -emit-llvm -fprofile-instrument=clang -fcoverage-mapping -mllvm -enable-name-compression=false | FileCheck %s -check-prefixes=COMMON,WINDOWS
 

diff  --git a/clang/test/Frontend/fixed_point_add.c b/clang/test/Frontend/fixed_point_add.c
index 5096d8fc44765..1f1b8d642707b 100644
--- a/clang/test/Frontend/fixed_point_add.c
+++ b/clang/test/Frontend/fixed_point_add.c
@@ -1,6 +1,6 @@
 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -triple x86_64-unknown-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,SIGNED
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -triple x86_64-unknown-linux-gnu -fpadding-on-unsigned-fixed-point -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,UNSIGNED
+// RUN: %clang_cc1 -ffixed-point -triple x86_64-unknown-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,SIGNED
+// RUN: %clang_cc1 -ffixed-point -triple x86_64-unknown-linux-gnu -fpadding-on-unsigned-fixed-point -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,UNSIGNED
 
 short _Accum sa;
 _Accum a, a2, a3, a4;
@@ -30,10 +30,10 @@ _Bool b;
 
 // CHECK-LABEL: @sadd_sasasa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i16, i16* @sa, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i16, ptr @sa, align 2
 // CHECK-NEXT:    [[TMP2:%.*]] = add i16 [[TMP0]], [[TMP1]]
-// CHECK-NEXT:    store i16 [[TMP2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[TMP2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void sadd_sasasa(void) {
@@ -42,12 +42,12 @@ void sadd_sasasa(void) {
 
 // CHECK-LABEL: @sadd_asaa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i32
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // CHECK-NEXT:    [[TMP2:%.*]] = add i32 [[UPSCALE]], [[TMP1]]
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void sadd_asaa(void) {
@@ -56,11 +56,11 @@ void sadd_asaa(void) {
 
 // CHECK-LABEL: @sadd_sasasf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* @sf, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr @sf, align 1
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i8 [[TMP1]] to i16
 // CHECK-NEXT:    [[TMP2:%.*]] = add i16 [[TMP0]], [[RESIZE]]
-// CHECK-NEXT:    store i16 [[TMP2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[TMP2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void sadd_sasasf(void) {
@@ -69,15 +69,15 @@ void sadd_sasasf(void) {
 
 // CHECK-LABEL: @sadd_sasaf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i16, i16* @f, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i16, ptr @f, align 2
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i24
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i24 [[RESIZE]], 8
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i16 [[TMP1]] to i24
 // CHECK-NEXT:    [[TMP2:%.*]] = add i24 [[UPSCALE]], [[RESIZE1]]
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i24 [[TMP2]], 8
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i24 [[DOWNSCALE]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void sadd_sasaf(void) {
@@ -86,12 +86,12 @@ void sadd_sasaf(void) {
 
 // CHECK-LABEL: @sadd_aasf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* @sf, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr @sf, align 1
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i8 [[TMP1]] to i32
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // CHECK-NEXT:    [[TMP2:%.*]] = add i32 [[TMP0]], [[UPSCALE]]
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void sadd_aasf(void) {
@@ -100,15 +100,15 @@ void sadd_aasf(void) {
 
 // CHECK-LABEL: @sadd_aalf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @lf, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @lf, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i48
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i48 [[RESIZE]], 16
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i48
 // CHECK-NEXT:    [[TMP2:%.*]] = add i48 [[UPSCALE]], [[RESIZE1]]
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i48 [[TMP2]], 16
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i48 [[DOWNSCALE]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void sadd_aalf(void) {
@@ -117,23 +117,23 @@ void sadd_aalf(void) {
 
 // SIGNED-LABEL: @sadd_sasausa(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa, align 2
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i17
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i17 [[RESIZE]], 1
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i16 [[TMP1]] to i17
 // SIGNED-NEXT:    [[TMP2:%.*]] = add i17 [[UPSCALE]], [[RESIZE1]]
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i17 [[TMP2]], 1
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i17 [[DOWNSCALE]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sadd_sasausa(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa, align 2
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = add i16 [[TMP0]], [[TMP1]]
-// UNSIGNED-NEXT:    store i16 [[TMP2]], i16* @sa, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP2]], ptr @sa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void sadd_sasausa(void) {
@@ -142,25 +142,25 @@ void sadd_sasausa(void) {
 
 // SIGNED-LABEL: @sadd_asaua(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ua, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ua, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i33
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i33 [[RESIZE]], 9
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i33
 // SIGNED-NEXT:    [[TMP2:%.*]] = add i33 [[UPSCALE]], [[RESIZE1]]
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i33 [[TMP2]], 1
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i33 [[DOWNSCALE]] to i32
-// SIGNED-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// SIGNED-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sadd_asaua(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ua, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ua, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i32
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = add i32 [[UPSCALE]], [[TMP1]]
-// UNSIGNED-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// UNSIGNED-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void sadd_asaua(void) {
@@ -169,24 +169,24 @@ void sadd_asaua(void) {
 
 // SIGNED-LABEL: @sadd_sasausf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i8, i8* @usf, align 1
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i8, ptr @usf, align 1
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i17
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i17 [[RESIZE]], 1
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i8 [[TMP1]] to i17
 // SIGNED-NEXT:    [[TMP2:%.*]] = add i17 [[UPSCALE]], [[RESIZE1]]
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i17 [[TMP2]], 1
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i17 [[DOWNSCALE]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sadd_sasausf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i8, i8* @usf, align 1
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i8, ptr @usf, align 1
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i8 [[TMP1]] to i16
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = add i16 [[TMP0]], [[RESIZE]]
-// UNSIGNED-NEXT:    store i16 [[TMP2]], i16* @sa, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP2]], ptr @sa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void sadd_sasausf(void) {
@@ -195,28 +195,28 @@ void sadd_sasausf(void) {
 
 // SIGNED-LABEL: @sadd_sasaulf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ulf, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ulf, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i41
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i41 [[RESIZE]], 25
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i41
 // SIGNED-NEXT:    [[TMP2:%.*]] = add i41 [[UPSCALE]], [[RESIZE1]]
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i41 [[TMP2]], 25
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i41 [[DOWNSCALE]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sadd_sasaulf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ulf, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ulf, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i40
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE]], 24
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i40
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = add i40 [[UPSCALE]], [[RESIZE1]]
 // UNSIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i40 [[TMP2]], 24
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[DOWNSCALE]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void sadd_sasaulf(void) {
@@ -225,14 +225,14 @@ void sadd_sasaulf(void) {
 
 // CHECK-LABEL: @sadd_aaaaa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a2, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a2, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = add i32 [[TMP0]], [[TMP1]]
-// CHECK-NEXT:    [[TMP3:%.*]] = load i32, i32* @a3, align 4
+// CHECK-NEXT:    [[TMP3:%.*]] = load i32, ptr @a3, align 4
 // CHECK-NEXT:    [[TMP4:%.*]] = add i32 [[TMP2]], [[TMP3]]
-// CHECK-NEXT:    [[TMP5:%.*]] = load i32, i32* @a4, align 4
+// CHECK-NEXT:    [[TMP5:%.*]] = load i32, ptr @a4, align 4
 // CHECK-NEXT:    [[TMP6:%.*]] = add i32 [[TMP4]], [[TMP5]]
-// CHECK-NEXT:    store i32 [[TMP6]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP6]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void sadd_aaaaa(void) {
@@ -242,10 +242,10 @@ void sadd_aaaaa(void) {
 
 // CHECK-LABEL: @uadd_usausausa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa, align 2
 // CHECK-NEXT:    [[TMP2:%.*]] = add i16 [[TMP0]], [[TMP1]]
-// CHECK-NEXT:    store i16 [[TMP2]], i16* @usa, align 2
+// CHECK-NEXT:    store i16 [[TMP2]], ptr @usa, align 2
 // CHECK-NEXT:    ret void
 //
 void uadd_usausausa(void) {
@@ -254,12 +254,12 @@ void uadd_usausausa(void) {
 
 // CHECK-LABEL: @uadd_uausaua(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @ua, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @ua, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i32
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // CHECK-NEXT:    [[TMP2:%.*]] = add i32 [[UPSCALE]], [[TMP1]]
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @ua, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @ua, align 4
 // CHECK-NEXT:    ret void
 //
 void uadd_uausaua(void) {
@@ -268,11 +268,11 @@ void uadd_uausaua(void) {
 
 // CHECK-LABEL: @uadd_usausausf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* @usf, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr @usf, align 1
 // CHECK-NEXT:    [[RESIZE:%.*]] = zext i8 [[TMP1]] to i16
 // CHECK-NEXT:    [[TMP2:%.*]] = add i16 [[TMP0]], [[RESIZE]]
-// CHECK-NEXT:    store i16 [[TMP2]], i16* @usa, align 2
+// CHECK-NEXT:    store i16 [[TMP2]], ptr @usa, align 2
 // CHECK-NEXT:    ret void
 //
 void uadd_usausausf(void) {
@@ -281,15 +281,15 @@ void uadd_usausausf(void) {
 
 // CHECK-LABEL: @uadd_usausauf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i16, i16* @uf, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i16, ptr @uf, align 2
 // CHECK-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i24
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i24 [[RESIZE]], 8
 // CHECK-NEXT:    [[RESIZE1:%.*]] = zext i16 [[TMP1]] to i24
 // CHECK-NEXT:    [[TMP2:%.*]] = add i24 [[UPSCALE]], [[RESIZE1]]
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = lshr i24 [[TMP2]], 8
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i24 [[DOWNSCALE]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // CHECK-NEXT:    ret void
 //
 void uadd_usausauf(void) {
@@ -299,14 +299,14 @@ void uadd_usausauf(void) {
 
 // CHECK-LABEL: @int_sasai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i39
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i39
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE1]], 7
 // CHECK-NEXT:    [[TMP2:%.*]] = add i39 [[RESIZE]], [[UPSCALE]]
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[TMP2]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void int_sasai(void) {
@@ -315,14 +315,14 @@ void int_sasai(void) {
 
 // CHECK-LABEL: @int_sasaui(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @ui, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @ui, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i40
 // CHECK-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i40
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE1]], 7
 // CHECK-NEXT:    [[TMP2:%.*]] = add i40 [[RESIZE]], [[UPSCALE]]
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[TMP2]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void int_sasaui(void) {
@@ -331,26 +331,26 @@ void int_sasaui(void) {
 
 // SIGNED-LABEL: @int_usausai(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i40
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i40
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE1]], 8
 // SIGNED-NEXT:    [[TMP2:%.*]] = add i40 [[RESIZE]], [[UPSCALE]]
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[TMP2]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @int_usausai(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i39
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i39
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE1]], 7
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = add i39 [[RESIZE]], [[UPSCALE]]
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[TMP2]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void int_usausai(void) {
@@ -359,26 +359,26 @@ void int_usausai(void) {
 
 // SIGNED-LABEL: @int_usausaui(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ui, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ui, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i40
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i40
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE1]], 8
 // SIGNED-NEXT:    [[TMP2:%.*]] = add i40 [[RESIZE]], [[UPSCALE]]
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[TMP2]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @int_usausaui(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ui, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ui, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i39
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i39
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE1]], 7
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = add i39 [[RESIZE]], [[UPSCALE]]
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[TMP2]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void int_usausaui(void) {
@@ -387,14 +387,14 @@ void int_usausaui(void) {
 
 // CHECK-LABEL: @int_lflfui(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @lf, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @ui, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @lf, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @ui, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i64
 // CHECK-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i64
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i64 [[RESIZE1]], 31
 // CHECK-NEXT:    [[TMP2:%.*]] = add i64 [[RESIZE]], [[UPSCALE]]
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i64 [[TMP2]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE2]], i32* @lf, align 4
+// CHECK-NEXT:    store i32 [[RESIZE2]], ptr @lf, align 4
 // CHECK-NEXT:    ret void
 //
 void int_lflfui(void) {
@@ -403,8 +403,8 @@ void int_lflfui(void) {
 
 // CHECK-LABEL: @int_aab(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* @b, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr @b, align 1
 // CHECK-NEXT:    [[TOBOOL:%.*]] = trunc i8 [[TMP1]] to i1
 // CHECK-NEXT:    [[CONV:%.*]] = zext i1 [[TOBOOL]] to i32
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i47
@@ -412,7 +412,7 @@ void int_lflfui(void) {
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i47 [[RESIZE1]], 15
 // CHECK-NEXT:    [[TMP2:%.*]] = add i47 [[RESIZE]], [[UPSCALE]]
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i47 [[TMP2]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void int_aab(void) {
@@ -422,10 +422,10 @@ void int_aab(void) {
 
 // CHECK-LABEL: @sat_sassasas(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i16, i16* @sa_sat, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i16, ptr @sa_sat, align 2
 // CHECK-NEXT:    [[TMP2:%.*]] = call i16 @llvm.sadd.sat.i16(i16 [[TMP0]], i16 [[TMP1]])
-// CHECK-NEXT:    store i16 [[TMP2]], i16* @sa_sat, align 2
+// CHECK-NEXT:    store i16 [[TMP2]], ptr @sa_sat, align 2
 // CHECK-NEXT:    ret void
 //
 void sat_sassasas(void) {
@@ -434,20 +434,20 @@ void sat_sassasas(void) {
 
 // SIGNED-LABEL: @sat_usasusausas(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa_sat, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa_sat, align 2
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.uadd.sat.i16(i16 [[TMP0]], i16 [[TMP1]])
-// SIGNED-NEXT:    store i16 [[TMP2]], i16* @usa_sat, align 2
+// SIGNED-NEXT:    store i16 [[TMP2]], ptr @usa_sat, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sat_usasusausas(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa_sat, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa_sat, align 2
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.sadd.sat.i16(i16 [[TMP0]], i16 [[TMP1]])
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = trunc i16 [[TMP2]] to i15
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i15 [[RESIZE]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE1]], i16* @usa_sat, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE1]], ptr @usa_sat, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void sat_usasusausas(void) {
@@ -456,24 +456,24 @@ void sat_usasusausas(void) {
 
 // SIGNED-LABEL: @sat_uasuausas(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa_sat, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa_sat, align 2
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP1]] to i32
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i32 @llvm.uadd.sat.i32(i32 [[TMP0]], i32 [[UPSCALE]])
-// SIGNED-NEXT:    store i32 [[TMP2]], i32* @ua_sat, align 4
+// SIGNED-NEXT:    store i32 [[TMP2]], ptr @ua_sat, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sat_uasuausas(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa_sat, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa_sat, align 2
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP1]] to i32
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i32 @llvm.sadd.sat.i32(i32 [[TMP0]], i32 [[UPSCALE]])
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = trunc i32 [[TMP2]] to i31
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = zext i31 [[RESIZE1]] to i32
-// UNSIGNED-NEXT:    store i32 [[RESIZE2]], i32* @ua_sat, align 4
+// UNSIGNED-NEXT:    store i32 [[RESIZE2]], ptr @ua_sat, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void sat_uasuausas(void) {
@@ -482,8 +482,8 @@ void sat_uasuausas(void) {
 
 // CHECK-LABEL: @sat_sassasi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa_sat, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa_sat, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i39
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i39
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE1]], 7
@@ -493,7 +493,7 @@ void sat_uasuausas(void) {
 // CHECK-NEXT:    [[TMP4:%.*]] = icmp slt i39 [[SATMAX]], -32768
 // CHECK-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP4]], i39 -32768, i39 [[SATMAX]]
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[SATMIN]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa_sat, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa_sat, align 2
 // CHECK-NEXT:    ret void
 //
 void sat_sassasi(void) {
@@ -502,8 +502,8 @@ void sat_sassasi(void) {
 
 // CHECK-LABEL: @sat_sassasui(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa_sat, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @ui, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa_sat, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @ui, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i40
 // CHECK-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i40
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE1]], 7
@@ -513,7 +513,7 @@ void sat_sassasi(void) {
 // CHECK-NEXT:    [[TMP4:%.*]] = icmp slt i40 [[SATMAX]], -32768
 // CHECK-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP4]], i40 -32768, i40 [[SATMAX]]
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[SATMIN]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa_sat, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa_sat, align 2
 // CHECK-NEXT:    ret void
 //
 void sat_sassasui(void) {
@@ -522,20 +522,20 @@ void sat_sassasui(void) {
 
 // SIGNED-LABEL: @sat_ufsufsufs(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf_sat, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @uf_sat, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf_sat, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @uf_sat, align 2
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.uadd.sat.i16(i16 [[TMP0]], i16 [[TMP1]])
-// SIGNED-NEXT:    store i16 [[TMP2]], i16* @uf_sat, align 2
+// SIGNED-NEXT:    store i16 [[TMP2]], ptr @uf_sat, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sat_ufsufsufs(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf_sat, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @uf_sat, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf_sat, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @uf_sat, align 2
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.sadd.sat.i16(i16 [[TMP0]], i16 [[TMP1]])
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = trunc i16 [[TMP2]] to i15
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i15 [[RESIZE]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE1]], i16* @uf_sat, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE1]], ptr @uf_sat, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void sat_ufsufsufs(void) {
@@ -544,8 +544,8 @@ void sat_ufsufsufs(void) {
 
 // SIGNED-LABEL: @sat_usasusasi(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa_sat, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa_sat, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i40
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i40
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE1]], 8
@@ -555,13 +555,13 @@ void sat_ufsufsufs(void) {
 // SIGNED-NEXT:    [[TMP4:%.*]] = icmp slt i40 [[SATMAX]], 0
 // SIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP4]], i40 0, i40 [[SATMAX]]
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[SATMIN]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa_sat, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa_sat, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sat_usasusasi(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa_sat, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa_sat, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i39
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i39
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE1]], 7
@@ -571,7 +571,7 @@ void sat_ufsufsufs(void) {
 // UNSIGNED-NEXT:    [[TMP4:%.*]] = icmp slt i39 [[SATMAX]], 0
 // UNSIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP4]], i39 0, i39 [[SATMAX]]
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[SATMIN]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa_sat, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa_sat, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void sat_usasusasi(void) {

diff  --git a/clang/test/Frontend/fixed_point_comparisons.c b/clang/test/Frontend/fixed_point_comparisons.c
index b920e77fb0753..a59f06b9c9420 100644
--- a/clang/test/Frontend/fixed_point_comparisons.c
+++ b/clang/test/Frontend/fixed_point_comparisons.c
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -triple x86_64-unknown-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,UNPADDED
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -triple x86_64-unknown-linux-gnu -fpadding-on-unsigned-fixed-point -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,PADDED
+// RUN: %clang_cc1 -ffixed-point -triple x86_64-unknown-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,UNPADDED
+// RUN: %clang_cc1 -ffixed-point -triple x86_64-unknown-linux-gnu -fpadding-on-unsigned-fixed-point -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,PADDED
 
 // Fixed point against other fixed point
 _Bool b_eq_true = 2.5hk == 2.5uhk;  // CHECK-DAG: @b_eq_true  = {{.*}}global i8 1, align 1
@@ -62,71 +62,71 @@ void TestComparisons(void) {
   // Each of these should be a fixed point conversion followed by the actual
   // comparison operation.
   sa == a;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sa, align 2
-  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, i32* %a, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sa, align 2
+  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, ptr %a, align 4
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i32
   // CHECK-NEXT: [[UPSCALE_A:%[a-z0-9]+]] = shl i32 [[RESIZE_A]], 8
   // CHECK-NEXT: {{.*}} = icmp eq i32 [[UPSCALE_A]], [[A2]]
 
   sa != a;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sa, align 2
-  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, i32* %a, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sa, align 2
+  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, ptr %a, align 4
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i32
   // CHECK-NEXT: [[UPSCALE_A:%[a-z0-9]+]] = shl i32 [[RESIZE_A]], 8
   // CHECK-NEXT: {{.*}} = icmp ne i32 [[UPSCALE_A]], [[A2]]
 
   sa > a;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sa, align 2
-  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, i32* %a, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sa, align 2
+  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, ptr %a, align 4
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i32
   // CHECK-NEXT: [[UPSCALE_A:%[a-z0-9]+]] = shl i32 [[RESIZE_A]], 8
   // CHECK-NEXT: {{.*}} = icmp sgt i32 [[UPSCALE_A]], [[A2]]
 
   sa >= a;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sa, align 2
-  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, i32* %a, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sa, align 2
+  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, ptr %a, align 4
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i32
   // CHECK-NEXT: [[UPSCALE_A:%[a-z0-9]+]] = shl i32 [[RESIZE_A]], 8
   // CHECK-NEXT: {{.*}} = icmp sge i32 [[UPSCALE_A]], [[A2]]
 
   sa < a;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sa, align 2
-  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, i32* %a, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sa, align 2
+  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, ptr %a, align 4
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i32
   // CHECK-NEXT: [[UPSCALE_A:%[a-z0-9]+]] = shl i32 [[RESIZE_A]], 8
   // CHECK-NEXT: {{.*}} = icmp slt i32 [[UPSCALE_A]], [[A2]]
 
   sa <= a;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sa, align 2
-  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, i32* %a, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sa, align 2
+  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, ptr %a, align 4
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i32
   // CHECK-NEXT: [[UPSCALE_A:%[a-z0-9]+]] = shl i32 [[RESIZE_A]], 8
   // CHECK-NEXT: {{.*}} = icmp sle i32 [[UPSCALE_A]], [[A2]]
 
   usa > ua;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %usa, align 2
-  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, i32* %ua, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %usa, align 2
+  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, ptr %ua, align 4
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = zext i16 [[A]] to i32
   // CHECK-NEXT: [[UPSCALE_A:%[a-z0-9]+]] = shl i32 [[RESIZE_A]], 8
   // CHECK-NEXT: {{.*}} = icmp ugt i32 [[UPSCALE_A]], [[A2]]
 
   usa >= ua;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %usa, align 2
-  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, i32* %ua, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %usa, align 2
+  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, ptr %ua, align 4
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = zext i16 [[A]] to i32
   // CHECK-NEXT: [[UPSCALE_A:%[a-z0-9]+]] = shl i32 [[RESIZE_A]], 8
   // CHECK-NEXT: {{.*}} = icmp uge i32 [[UPSCALE_A]], [[A2]]
 
   usa < ua;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %usa, align 2
-  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, i32* %ua, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %usa, align 2
+  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, ptr %ua, align 4
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = zext i16 [[A]] to i32
   // CHECK-NEXT: [[UPSCALE_A:%[a-z0-9]+]] = shl i32 [[RESIZE_A]], 8
   // CHECK-NEXT: {{.*}} = icmp ult i32 [[UPSCALE_A]], [[A2]]
 
   usa <= ua;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %usa, align 2
-  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, i32* %ua, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %usa, align 2
+  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, ptr %ua, align 4
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = zext i16 [[A]] to i32
   // CHECK-NEXT: [[UPSCALE_A:%[a-z0-9]+]] = shl i32 [[RESIZE_A]], 8
   // CHECK-NEXT: {{.*}} = icmp ule i32 [[UPSCALE_A]], [[A2]]
@@ -148,56 +148,56 @@ void TestIntComparisons(void) {
   // These comparisons shouldn't be that 
diff erent from comparing against fixed
   // point types with other fixed point types.
   sa == i;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sa, align 2
-  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, i32* %i, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sa, align 2
+  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, ptr %i, align 4
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i39
   // CHECK-NEXT: [[RESIZE_I:%[a-z0-9]+]] = sext i32 [[I]] to i39
   // CHECK-NEXT: [[UPSCALE_I:%[a-z0-9]+]] = shl i39 [[RESIZE_I]], 7
   // CHECK-NEXT: {{.*}} = icmp eq i39 [[RESIZE_A]], [[UPSCALE_I]]
 
   sa != i;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sa, align 2
-  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, i32* %i, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sa, align 2
+  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, ptr %i, align 4
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i39
   // CHECK-NEXT: [[RESIZE_I:%[a-z0-9]+]] = sext i32 [[I]] to i39
   // CHECK-NEXT: [[UPSCALE_I:%[a-z0-9]+]] = shl i39 [[RESIZE_I]], 7
   // CHECK-NEXT: {{.*}} = icmp ne i39 [[RESIZE_A]], [[UPSCALE_I]]
 
   sa > i;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sa, align 2
-  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, i32* %i, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sa, align 2
+  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, ptr %i, align 4
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i39
   // CHECK-NEXT: [[RESIZE_I:%[a-z0-9]+]] = sext i32 [[I]] to i39
   // CHECK-NEXT: [[UPSCALE_I:%[a-z0-9]+]] = shl i39 [[RESIZE_I]], 7
   // CHECK-NEXT: {{.*}} = icmp sgt i39 [[RESIZE_A]], [[UPSCALE_I]]
 
   sa >= i;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sa, align 2
-  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, i32* %i, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sa, align 2
+  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, ptr %i, align 4
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i39
   // CHECK-NEXT: [[RESIZE_I:%[a-z0-9]+]] = sext i32 [[I]] to i39
   // CHECK-NEXT: [[UPSCALE_I:%[a-z0-9]+]] = shl i39 [[RESIZE_I]], 7
   // CHECK-NEXT: {{.*}} = icmp sge i39 [[RESIZE_A]], [[UPSCALE_I]]
 
   sa < i;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sa, align 2
-  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, i32* %i, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sa, align 2
+  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, ptr %i, align 4
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i39
   // CHECK-NEXT: [[RESIZE_I:%[a-z0-9]+]] = sext i32 [[I]] to i39
   // CHECK-NEXT: [[UPSCALE_I:%[a-z0-9]+]] = shl i39 [[RESIZE_I]], 7
   // CHECK-NEXT: {{.*}} = icmp slt i39 [[RESIZE_A]], [[UPSCALE_I]]
 
   sa <= i;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sa, align 2
-  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, i32* %i, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sa, align 2
+  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, ptr %i, align 4
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i39
   // CHECK-NEXT: [[RESIZE_I:%[a-z0-9]+]] = sext i32 [[I]] to i39
   // CHECK-NEXT: [[UPSCALE_I:%[a-z0-9]+]] = shl i39 [[RESIZE_I]], 7
   // CHECK-NEXT: {{.*}} = icmp sle i39 [[RESIZE_A]], [[UPSCALE_I]]
 
   usa > ui;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %usa, align 2
-  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, i32* %ui, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %usa, align 2
+  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, ptr %ui, align 4
   // UNPADDED-NEXT: [[RESIZE_A:%[a-z0-9]+]] = zext i16 [[A]] to i40
   // UNPADDED-NEXT: [[RESIZE_I:%[a-z0-9]+]] = zext i32 [[I]] to i40
   // UNPADDED-NEXT: [[UPSCALE_I:%[a-z0-9]+]] = shl i40 [[RESIZE_I]], 8
@@ -208,8 +208,8 @@ void TestIntComparisons(void) {
   // PADDED-NEXT: {{.*}} = icmp ugt i39 [[RESIZE_A]], [[UPSCALE_I]]
 
   usa >= ui;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %usa, align 2
-  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, i32* %ui, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %usa, align 2
+  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, ptr %ui, align 4
   // UNPADDED-NEXT: [[RESIZE_A:%[a-z0-9]+]] = zext i16 [[A]] to i40
   // UNPADDED-NEXT: [[RESIZE_I:%[a-z0-9]+]] = zext i32 [[I]] to i40
   // UNPADDED-NEXT: [[UPSCALE_I:%[a-z0-9]+]] = shl i40 [[RESIZE_I]], 8
@@ -220,8 +220,8 @@ void TestIntComparisons(void) {
   // PADDED-NEXT: {{.*}} = icmp uge i39 [[RESIZE_A]], [[UPSCALE_I]]
 
   usa < ui;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %usa, align 2
-  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, i32* %ui, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %usa, align 2
+  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, ptr %ui, align 4
   // UNPADDED-NEXT: [[RESIZE_A:%[a-z0-9]+]] = zext i16 [[A]] to i40
   // UNPADDED-NEXT: [[RESIZE_I:%[a-z0-9]+]] = zext i32 [[I]] to i40
   // UNPADDED-NEXT: [[UPSCALE_I:%[a-z0-9]+]] = shl i40 [[RESIZE_I]], 8
@@ -232,8 +232,8 @@ void TestIntComparisons(void) {
   // PADDED-NEXT: {{.*}} = icmp ult i39 [[RESIZE_A]], [[UPSCALE_I]]
 
   usa <= ui;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %usa, align 2
-  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, i32* %ui, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %usa, align 2
+  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, ptr %ui, align 4
   // UNPADDED-NEXT: [[RESIZE_A:%[a-z0-9]+]] = zext i16 [[A]] to i40
   // UNPADDED-NEXT: [[RESIZE_I:%[a-z0-9]+]] = zext i32 [[I]] to i40
   // UNPADDED-NEXT: [[UPSCALE_I:%[a-z0-9]+]] = shl i40 [[RESIZE_I]], 8
@@ -247,8 +247,8 @@ void TestIntComparisons(void) {
   // from comparing to an int other than varying sizes. The integer types are
   // still converted to ints or unsigned ints from UsualUnaryConversions().
   sa == b;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sa, align 2
-  // CHECK-NEXT: [[B:%[0-9]+]] = load i8, i8* %b, align 1
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sa, align 2
+  // CHECK-NEXT: [[B:%[0-9]+]] = load i8, ptr %b, align 1
   // CHECK-NEXT: %tobool = trunc i8 [[B]] to i1
   // CHECK-NEXT: [[CONV_B:%[a-z0-9]+]] = zext i1 %tobool to i32
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i39
@@ -257,8 +257,8 @@ void TestIntComparisons(void) {
   // CHECK-NEXT: {{.*}} = icmp eq i39 [[RESIZE_A]], [[UPSCALE_B]]
 
   sa == c;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sa, align 2
-  // CHECK-NEXT: [[C:%[0-9]+]] = load i8, i8* %c, align 1
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sa, align 2
+  // CHECK-NEXT: [[C:%[0-9]+]] = load i8, ptr %c, align 1
   // CHECK-NEXT: [[CONV_C:%[a-z0-9]+]] = sext i8 [[C]] to i32
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i39
   // CHECK-NEXT: [[RESIZE_C:%[a-z0-9]+]] = sext i32 [[CONV_C]] to i39
@@ -266,8 +266,8 @@ void TestIntComparisons(void) {
   // CHECK-NEXT: {{.*}} = icmp eq i39 [[RESIZE_A]], [[UPSCALE_C]]
 
   sa == s;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sa, align 2
-  // CHECK-NEXT: [[S:%[0-9]+]] = load i16, i16* %s, align 2
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sa, align 2
+  // CHECK-NEXT: [[S:%[0-9]+]] = load i16, ptr %s, align 2
   // CHECK-NEXT: [[CONV_S:%[a-z0-9]+]] = sext i16 [[S]] to i32
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i39
   // CHECK-NEXT: [[RESIZE_S:%[a-z0-9]+]] = sext i32 [[CONV_S]] to i39
@@ -276,8 +276,8 @@ void TestIntComparisons(void) {
 
   // An enum value is IntegralCast to an unsigned int.
   usa == e;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %usa, align 2
-  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, i32* %e, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %usa, align 2
+  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, ptr %e, align 4
   // UNPADDED-NEXT: [[RESIZE_A:%[a-z0-9]+]] = zext i16 [[A]] to i40
   // UNPADDED-NEXT: [[RESIZE_I:%[a-z0-9]+]] = zext i32 [[I]] to i40
   // UNPADDED-NEXT: [[UPSCALE_I:%[a-z0-9]+]] = shl i40 [[RESIZE_I]], 8
@@ -296,8 +296,8 @@ void TestComparisonSignage(void) {
 
   // Signed vs unsigned fixed point comparison
   sa == usa;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sa, align 2
-  // CHECK-NEXT: [[A2:%[0-9]+]] = load i16, i16* %usa, align 2
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sa, align 2
+  // CHECK-NEXT: [[A2:%[0-9]+]] = load i16, ptr %usa, align 2
   // UNPADDED-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i17
   // UNPADDED-NEXT: [[UPSCALE_A:%[a-z0-9]+]] = shl i17 [[RESIZE_A]], 1
   // UNPADDED-NEXT: [[RESIZE_A2:%[a-z0-9]+]] = zext i16 [[A2]] to i17
@@ -306,8 +306,8 @@ void TestComparisonSignage(void) {
 
   // Signed int vs unsigned fixed point
   sa == ui;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sa, align 2
-  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, i32* %ui, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sa, align 2
+  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, ptr %ui, align 4
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i40
   // CHECK-NEXT: [[RESIZE_I:%[a-z0-9]+]] = zext i32 [[I]] to i40
   // CHECK-NEXT: [[UPSCALE_I:%[a-z0-9]+]] = shl i40 [[RESIZE_I]], 7
@@ -315,8 +315,8 @@ void TestComparisonSignage(void) {
 
   // Signed fixed point vs unsigned int
   usa == i;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %usa, align 2
-  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, i32* %i, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %usa, align 2
+  // CHECK-NEXT: [[I:%[0-9]+]] = load i32, ptr %i, align 4
   // UNPADDED-NEXT: [[RESIZE_A:%[a-z0-9]+]] = zext i16 [[A]] to i40
   // UNPADDED-NEXT: [[RESIZE_I:%[a-z0-9]+]] = sext i32 [[I]] to i40
   // UNPADDED-NEXT: [[UPSCALE_I:%[a-z0-9]+]] = shl i40 [[RESIZE_I]], 8
@@ -338,22 +338,22 @@ void TestSaturationComparisons(void) {
   // counterparts since when comparing, we convert both operands to a common
   // type that should be able to hold both values.
   sat_sa == sat_a;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sat_sa, align 2
-  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, i32* %sat_a, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sat_sa, align 2
+  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, ptr %sat_a, align 4
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i32
   // CHECK-NEXT: [[UPSCALE_A:%[a-z0-9]+]] = shl i32 [[RESIZE_A]], 8
   // CHECK-NEXT: {{.*}} = icmp eq i32 [[UPSCALE_A]], [[A2]]
 
   sat_sa == a;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sat_sa, align 2
-  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, i32* %a, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sat_sa, align 2
+  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, ptr %a, align 4
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i32
   // CHECK-NEXT: [[UPSCALE_A:%[a-z0-9]+]] = shl i32 [[RESIZE_A]], 8
   // CHECK-NEXT: {{.*}} = icmp eq i32 [[UPSCALE_A]], [[A2]]
 
   sat_sa == sat_usa;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sat_sa, align 2
-  // CHECK-NEXT: [[A2:%[0-9]+]] = load i16, i16* %sat_usa, align 2
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sat_sa, align 2
+  // CHECK-NEXT: [[A2:%[0-9]+]] = load i16, ptr %sat_usa, align 2
   // UNPADDED-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i17
   // UNPADDED-NEXT: [[UPSCALE_A:%[a-z0-9]+]] = shl i17 [[RESIZE_A]], 1
   // UNPADDED-NEXT: [[RESIZE_A2:%[a-z0-9]+]] = zext i16 [[A2]] to i17
@@ -368,11 +368,11 @@ void StoreBooleanResult(void) {
 
   // Check that the result can properly be stored as an int.
   res = sa == a;
-  // CHECK:      [[A:%[0-9]+]] = load i16, i16* %sa, align 2
-  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, i32* %a, align 4
+  // CHECK:      [[A:%[0-9]+]] = load i16, ptr %sa, align 2
+  // CHECK-NEXT: [[A2:%[0-9]+]] = load i32, ptr %a, align 4
   // CHECK-NEXT: [[RESIZE_A:%[a-z0-9]+]] = sext i16 [[A]] to i32
   // CHECK-NEXT: [[UPSCALE_A:%[a-z0-9]+]] = shl i32 [[RESIZE_A]], 8
   // CHECK-NEXT: [[RES:%[0-9]+]] = icmp eq i32 [[UPSCALE_A]], [[A2]]
   // CHECK-NEXT: %conv = zext i1 [[RES]] to i32
-  // CHECK-NEXT: store i32 %conv, i32* %res, align 4
+  // CHECK-NEXT: store i32 %conv, ptr %res, align 4
 }

diff  --git a/clang/test/Frontend/fixed_point_compound.c b/clang/test/Frontend/fixed_point_compound.c
index 264e6442355b8..a1d6ff19912bf 100644
--- a/clang/test/Frontend/fixed_point_compound.c
+++ b/clang/test/Frontend/fixed_point_compound.c
@@ -1,6 +1,6 @@
 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -triple x86_64-unknown-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,SIGNED
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -triple x86_64-unknown-linux-gnu -fpadding-on-unsigned-fixed-point -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,UNSIGNED
+// RUN: %clang_cc1 -ffixed-point -triple x86_64-unknown-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,SIGNED
+// RUN: %clang_cc1 -ffixed-point -triple x86_64-unknown-linux-gnu -fpadding-on-unsigned-fixed-point -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,UNSIGNED
 
 short _Fract shf;
 _Accum a;
@@ -21,14 +21,14 @@ float fl;
 
 // CHECK-LABEL: @add_shfa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* @shf, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr @shf, align 1
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i8 [[TMP1]] to i32
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // CHECK-NEXT:    [[TMP2:%.*]] = add i32 [[UPSCALE]], [[TMP0]]
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i32 [[TMP2]], 8
 // CHECK-NEXT:    [[RESIZE1:%.*]] = trunc i32 [[DOWNSCALE]] to i8
-// CHECK-NEXT:    store i8 [[RESIZE1]], i8* @shf, align 1
+// CHECK-NEXT:    store i8 [[RESIZE1]], ptr @shf, align 1
 // CHECK-NEXT:    ret void
 //
 void add_shfa(void) {
@@ -37,24 +37,24 @@ void add_shfa(void) {
 
 // SIGNED-LABEL: @add_auf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP1]] to i33
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i33 [[RESIZE]], 1
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i16 [[TMP0]] to i33
 // SIGNED-NEXT:    [[TMP2:%.*]] = add i33 [[UPSCALE]], [[RESIZE1]]
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i33 [[TMP2]], 1
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i33 [[DOWNSCALE]] to i32
-// SIGNED-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// SIGNED-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @add_auf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i32
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = add i32 [[TMP1]], [[RESIZE]]
-// UNSIGNED-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// UNSIGNED-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void add_auf(void) {
@@ -63,14 +63,14 @@ void add_auf(void) {
 
 // CHECK-LABEL: @add_ufula(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i64, i64* @ula, align 8
-// CHECK-NEXT:    [[TMP1:%.*]] = load i16, i16* @uf, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i64, ptr @ula, align 8
+// CHECK-NEXT:    [[TMP1:%.*]] = load i16, ptr @uf, align 2
 // CHECK-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP1]] to i64
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i64 [[RESIZE]], 16
 // CHECK-NEXT:    [[TMP2:%.*]] = add i64 [[UPSCALE]], [[TMP0]]
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = lshr i64 [[TMP2]], 16
 // CHECK-NEXT:    [[RESIZE1:%.*]] = trunc i64 [[DOWNSCALE]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE1]], i16* @uf, align 2
+// CHECK-NEXT:    store i16 [[RESIZE1]], ptr @uf, align 2
 // CHECK-NEXT:    ret void
 //
 void add_ufula(void) {
@@ -79,8 +79,8 @@ void add_ufula(void) {
 
 // SIGNED-LABEL: @add_ulashf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i8, i8* @shf, align 1
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i64, i64* @ula, align 8
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i8, ptr @shf, align 1
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i64, ptr @ula, align 8
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i64 [[TMP1]] to i65
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = sext i8 [[TMP0]] to i65
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i65 [[RESIZE1]], 25
@@ -88,17 +88,17 @@ void add_ufula(void) {
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i65 [[TMP2]], 1
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i65 [[DOWNSCALE]] to i64
 // SIGNED-NEXT:    [[UPSCALE3:%.*]] = shl i64 [[RESIZE2]], 1
-// SIGNED-NEXT:    store i64 [[UPSCALE3]], i64* @ula, align 8
+// SIGNED-NEXT:    store i64 [[UPSCALE3]], ptr @ula, align 8
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @add_ulashf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i8, i8* @shf, align 1
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i64, i64* @ula, align 8
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i8, ptr @shf, align 1
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i64, ptr @ula, align 8
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = sext i8 [[TMP0]] to i64
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i64 [[RESIZE]], 24
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = add i64 [[TMP1]], [[UPSCALE]]
-// UNSIGNED-NEXT:    store i64 [[TMP2]], i64* @ula, align 8
+// UNSIGNED-NEXT:    store i64 [[TMP2]], ptr @ula, align 8
 // UNSIGNED-NEXT:    ret void
 //
 void add_ulashf(void) {
@@ -107,8 +107,8 @@ void add_ulashf(void) {
 
 // SIGNED-LABEL: @add_ufshf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i8, i8* @shf, align 1
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @uf, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i8, ptr @shf, align 1
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @uf, align 2
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP1]] to i17
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = sext i8 [[TMP0]] to i17
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i17 [[RESIZE1]], 9
@@ -116,17 +116,17 @@ void add_ulashf(void) {
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i17 [[TMP2]], 1
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i17 [[DOWNSCALE]] to i16
 // SIGNED-NEXT:    [[UPSCALE3:%.*]] = shl i16 [[RESIZE2]], 1
-// SIGNED-NEXT:    store i16 [[UPSCALE3]], i16* @uf, align 2
+// SIGNED-NEXT:    store i16 [[UPSCALE3]], ptr @uf, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @add_ufshf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i8, i8* @shf, align 1
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @uf, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i8, ptr @shf, align 1
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @uf, align 2
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = sext i8 [[TMP0]] to i16
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i16 [[RESIZE]], 8
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = add i16 [[TMP1]], [[UPSCALE]]
-// UNSIGNED-NEXT:    store i16 [[TMP2]], i16* @uf, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP2]], ptr @uf, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void add_ufshf(void) {
@@ -135,12 +135,12 @@ void add_ufshf(void) {
 
 // CHECK-LABEL: @add_ashf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i8, i8* @shf, align 1
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i8, ptr @shf, align 1
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i8 [[TMP0]] to i32
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // CHECK-NEXT:    [[TMP2:%.*]] = add i32 [[TMP1]], [[UPSCALE]]
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void add_ashf(void) {
@@ -149,14 +149,14 @@ void add_ashf(void) {
 
 // CHECK-LABEL: @add_ai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @i, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @i, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP1]] to i47
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP0]] to i47
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i47 [[RESIZE1]], 15
 // CHECK-NEXT:    [[TMP2:%.*]] = add i47 [[RESIZE]], [[UPSCALE]]
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i47 [[TMP2]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void add_ai(void) {
@@ -165,14 +165,14 @@ void add_ai(void) {
 
 // CHECK-LABEL: @add_au(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @u, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @u, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP1]] to i48
 // CHECK-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP0]] to i48
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i48 [[RESIZE1]], 15
 // CHECK-NEXT:    [[TMP2:%.*]] = add i48 [[RESIZE]], [[UPSCALE]]
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i48 [[TMP2]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void add_au(void) {
@@ -181,24 +181,24 @@ void add_au(void) {
 
 // SIGNED-LABEL: @add_ulai(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @i, align 4
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i64, i64* @ula, align 8
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @i, align 4
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i64, ptr @ula, align 8
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i64 [[TMP1]] to i65
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP0]] to i65
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i65 [[RESIZE1]], 32
 // SIGNED-NEXT:    [[TMP2:%.*]] = add i65 [[RESIZE]], [[UPSCALE]]
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i65 [[TMP2]] to i64
-// SIGNED-NEXT:    store i64 [[RESIZE2]], i64* @ula, align 8
+// SIGNED-NEXT:    store i64 [[RESIZE2]], ptr @ula, align 8
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @add_ulai(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @i, align 4
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i64, i64* @ula, align 8
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @i, align 4
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i64, ptr @ula, align 8
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i64
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i64 [[RESIZE]], 31
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = add i64 [[TMP1]], [[UPSCALE]]
-// UNSIGNED-NEXT:    store i64 [[TMP2]], i64* @ula, align 8
+// UNSIGNED-NEXT:    store i64 [[TMP2]], ptr @ula, align 8
 // UNSIGNED-NEXT:    ret void
 //
 void add_ulai(void) {
@@ -207,8 +207,8 @@ void add_ulai(void) {
 
 // SIGNED-LABEL: @add_iula(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i64, i64* @ula, align 8
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i64, ptr @ula, align 8
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP1]] to i65
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i65 [[RESIZE]], 32
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i64 [[TMP0]] to i65
@@ -216,19 +216,19 @@ void add_ulai(void) {
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i65 [[TMP2]] to i64
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = lshr i64 [[RESIZE2]], 32
 // SIGNED-NEXT:    [[RESIZE3:%.*]] = trunc i64 [[DOWNSCALE]] to i32
-// SIGNED-NEXT:    store i32 [[RESIZE3]], i32* @i, align 4
+// SIGNED-NEXT:    store i32 [[RESIZE3]], ptr @i, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @add_iula(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i64, i64* @ula, align 8
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i64, ptr @ula, align 8
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP1]] to i64
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i64 [[RESIZE]], 31
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = add i64 [[UPSCALE]], [[TMP0]]
 // UNSIGNED-NEXT:    [[DOWNSCALE:%.*]] = lshr i64 [[TMP2]], 31
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = trunc i64 [[DOWNSCALE]] to i32
-// UNSIGNED-NEXT:    store i32 [[RESIZE1]], i32* @i, align 4
+// UNSIGNED-NEXT:    store i32 [[RESIZE1]], ptr @i, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void add_iula(void) {
@@ -237,8 +237,8 @@ void add_iula(void) {
 
 // CHECK-LABEL: @add_ca(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* @c, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr @c, align 1
 // CHECK-NEXT:    [[CONV:%.*]] = sext i8 [[TMP1]] to i32
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[CONV]] to i47
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i47 [[RESIZE]], 15
@@ -250,7 +250,7 @@ void add_iula(void) {
 // CHECK-NEXT:    [[TMP5:%.*]] = select i1 [[TMP3]], i32 [[TMP4]], i32 [[RESIZE2]]
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i32 [[TMP5]], 15
 // CHECK-NEXT:    [[RESIZE3:%.*]] = trunc i32 [[DOWNSCALE]] to i8
-// CHECK-NEXT:    store i8 [[RESIZE3]], i8* @c, align 1
+// CHECK-NEXT:    store i8 [[RESIZE3]], ptr @c, align 1
 // CHECK-NEXT:    ret void
 //
 void add_ca(void) {
@@ -259,8 +259,8 @@ void add_ca(void) {
 
 // CHECK-LABEL: @add_sai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @i, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @sa, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @i, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @sa, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP1]] to i47
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP0]] to i47
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i47 [[RESIZE1]], 15
@@ -270,7 +270,7 @@ void add_ca(void) {
 // CHECK-NEXT:    [[TMP4:%.*]] = icmp slt i47 [[SATMAX]], -2147483648
 // CHECK-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP4]], i47 -2147483648, i47 [[SATMAX]]
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i47 [[SATMIN]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE2]], i32* @sa, align 4
+// CHECK-NEXT:    store i32 [[RESIZE2]], ptr @sa, align 4
 // CHECK-NEXT:    ret void
 //
 void add_sai(void) {
@@ -279,8 +279,8 @@ void add_sai(void) {
 
 // CHECK-LABEL: @add_csa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @sa, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* @c, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @sa, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr @c, align 1
 // CHECK-NEXT:    [[CONV:%.*]] = sext i8 [[TMP1]] to i32
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[CONV]] to i47
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i47 [[RESIZE]], 15
@@ -296,7 +296,7 @@ void add_sai(void) {
 // CHECK-NEXT:    [[TMP7:%.*]] = select i1 [[TMP5]], i32 [[TMP6]], i32 [[RESIZE2]]
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i32 [[TMP7]], 15
 // CHECK-NEXT:    [[RESIZE3:%.*]] = trunc i32 [[DOWNSCALE]] to i8
-// CHECK-NEXT:    store i8 [[RESIZE3]], i8* @c, align 1
+// CHECK-NEXT:    store i8 [[RESIZE3]], ptr @c, align 1
 // CHECK-NEXT:    ret void
 //
 void add_csa(void) {
@@ -305,24 +305,24 @@ void add_csa(void) {
 
 // SIGNED-LABEL: @add_sulau(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @u, align 4
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i64, i64* @sula, align 8
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @u, align 4
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i64, ptr @sula, align 8
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i32 [[TMP0]] to i64
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i64 [[RESIZE]], 32
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i64 @llvm.uadd.sat.i64(i64 [[TMP1]], i64 [[UPSCALE]])
-// SIGNED-NEXT:    store i64 [[TMP2]], i64* @sula, align 8
+// SIGNED-NEXT:    store i64 [[TMP2]], ptr @sula, align 8
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @add_sulau(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @u, align 4
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i64, i64* @sula, align 8
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @u, align 4
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i64, ptr @sula, align 8
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = trunc i64 [[TMP1]] to i63
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP0]] to i63
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i63 [[RESIZE1]], 31
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i63 @llvm.uadd.sat.i63(i63 [[RESIZE]], i63 [[UPSCALE]])
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = zext i63 [[TMP2]] to i64
-// UNSIGNED-NEXT:    store i64 [[RESIZE2]], i64* @sula, align 8
+// UNSIGNED-NEXT:    store i64 [[RESIZE2]], ptr @sula, align 8
 // UNSIGNED-NEXT:    ret void
 //
 void add_sulau(void) {
@@ -331,8 +331,8 @@ void add_sulau(void) {
 
 // SIGNED-LABEL: @add_sshsuf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @suf, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i8, i8* @sshf, align 1
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @suf, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i8, ptr @sshf, align 1
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i8 [[TMP1]] to i17
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i17 [[RESIZE]], 9
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i16 [[TMP0]] to i17
@@ -341,19 +341,19 @@ void add_sulau(void) {
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i17 [[DOWNSCALE]] to i16
 // SIGNED-NEXT:    [[DOWNSCALE3:%.*]] = ashr i16 [[RESIZE2]], 8
 // SIGNED-NEXT:    [[RESIZE4:%.*]] = trunc i16 [[DOWNSCALE3]] to i8
-// SIGNED-NEXT:    store i8 [[RESIZE4]], i8* @sshf, align 1
+// SIGNED-NEXT:    store i8 [[RESIZE4]], ptr @sshf, align 1
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @add_sshsuf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @suf, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i8, i8* @sshf, align 1
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @suf, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i8, ptr @sshf, align 1
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = sext i8 [[TMP1]] to i16
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i16 [[RESIZE]], 8
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.sadd.sat.i16(i16 [[UPSCALE]], i16 [[TMP0]])
 // UNSIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i16 [[TMP2]], 8
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = trunc i16 [[DOWNSCALE]] to i8
-// UNSIGNED-NEXT:    store i8 [[RESIZE1]], i8* @sshf, align 1
+// UNSIGNED-NEXT:    store i8 [[RESIZE1]], ptr @sshf, align 1
 // UNSIGNED-NEXT:    ret void
 //
 void add_sshsuf(void) {
@@ -362,14 +362,14 @@ void add_sshsuf(void) {
 
 // CHECK-LABEL: @add_afl(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = sitofp i32 [[TMP1]] to float
 // CHECK-NEXT:    [[TMP3:%.*]] = fmul float [[TMP2]], 0x3F00000000000000
 // CHECK-NEXT:    [[ADD:%.*]] = fadd float [[TMP3]], [[TMP0]]
 // CHECK-NEXT:    [[TMP4:%.*]] = fmul float [[ADD]], 3.276800e+04
 // CHECK-NEXT:    [[TMP5:%.*]] = fptosi float [[TMP4]] to i32
-// CHECK-NEXT:    store i32 [[TMP5]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP5]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void add_afl(void) {
@@ -378,12 +378,12 @@ void add_afl(void) {
 
 // CHECK-LABEL: @add_fla(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = sitofp i32 [[TMP0]] to float
 // CHECK-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x3F00000000000000
-// CHECK-NEXT:    [[TMP3:%.*]] = load float, float* @fl, align 4
+// CHECK-NEXT:    [[TMP3:%.*]] = load float, ptr @fl, align 4
 // CHECK-NEXT:    [[ADD:%.*]] = fadd float [[TMP3]], [[TMP2]]
-// CHECK-NEXT:    store float [[ADD]], float* @fl, align 4
+// CHECK-NEXT:    store float [[ADD]], ptr @fl, align 4
 // CHECK-NEXT:    ret void
 //
 void add_fla(void) {
@@ -392,14 +392,14 @@ void add_fla(void) {
 
 // CHECK-LABEL: @add_safl(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @sa, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @sa, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = sitofp i32 [[TMP1]] to float
 // CHECK-NEXT:    [[TMP3:%.*]] = fmul float [[TMP2]], 0x3F00000000000000
 // CHECK-NEXT:    [[ADD:%.*]] = fadd float [[TMP3]], [[TMP0]]
 // CHECK-NEXT:    [[TMP4:%.*]] = fmul float [[ADD]], 3.276800e+04
 // CHECK-NEXT:    [[TMP5:%.*]] = call i32 @llvm.fptosi.sat.i32.f32(float [[TMP4]])
-// CHECK-NEXT:    store i32 [[TMP5]], i32* @sa, align 4
+// CHECK-NEXT:    store i32 [[TMP5]], ptr @sa, align 4
 // CHECK-NEXT:    ret void
 //
 void add_safl(void) {
@@ -408,12 +408,12 @@ void add_safl(void) {
 
 // CHECK-LABEL: @add_flsa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @sa, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @sa, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = sitofp i32 [[TMP0]] to float
 // CHECK-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x3F00000000000000
-// CHECK-NEXT:    [[TMP3:%.*]] = load float, float* @fl, align 4
+// CHECK-NEXT:    [[TMP3:%.*]] = load float, ptr @fl, align 4
 // CHECK-NEXT:    [[ADD:%.*]] = fadd float [[TMP3]], [[TMP2]]
-// CHECK-NEXT:    store float [[ADD]], float* @fl, align 4
+// CHECK-NEXT:    store float [[ADD]], ptr @fl, align 4
 // CHECK-NEXT:    ret void
 //
 void add_flsa(void) {
@@ -425,24 +425,24 @@ void add_flsa(void) {
 
 // SIGNED-LABEL: @sub_auf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP1]] to i33
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i33 [[RESIZE]], 1
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i16 [[TMP0]] to i33
 // SIGNED-NEXT:    [[TMP2:%.*]] = sub i33 [[UPSCALE]], [[RESIZE1]]
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i33 [[TMP2]], 1
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i33 [[DOWNSCALE]] to i32
-// SIGNED-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// SIGNED-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sub_auf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i32
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = sub i32 [[TMP1]], [[RESIZE]]
-// UNSIGNED-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// UNSIGNED-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void sub_auf(void) {
@@ -451,14 +451,14 @@ void sub_auf(void) {
 
 // CHECK-LABEL: @sub_ai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @i, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @i, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP1]] to i47
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP0]] to i47
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i47 [[RESIZE1]], 15
 // CHECK-NEXT:    [[TMP2:%.*]] = sub i47 [[RESIZE]], [[UPSCALE]]
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i47 [[TMP2]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void sub_ai(void) {
@@ -467,8 +467,8 @@ void sub_ai(void) {
 
 // CHECK-LABEL: @sub_csa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @sa, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* @c, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @sa, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr @c, align 1
 // CHECK-NEXT:    [[CONV:%.*]] = sext i8 [[TMP1]] to i32
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[CONV]] to i47
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i47 [[RESIZE]], 15
@@ -484,7 +484,7 @@ void sub_ai(void) {
 // CHECK-NEXT:    [[TMP7:%.*]] = select i1 [[TMP5]], i32 [[TMP6]], i32 [[RESIZE2]]
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i32 [[TMP7]], 15
 // CHECK-NEXT:    [[RESIZE3:%.*]] = trunc i32 [[DOWNSCALE]] to i8
-// CHECK-NEXT:    store i8 [[RESIZE3]], i8* @c, align 1
+// CHECK-NEXT:    store i8 [[RESIZE3]], ptr @c, align 1
 // CHECK-NEXT:    ret void
 //
 void sub_csa(void) {
@@ -493,14 +493,14 @@ void sub_csa(void) {
 
 // CHECK-LABEL: @sub_afl(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = sitofp i32 [[TMP1]] to float
 // CHECK-NEXT:    [[TMP3:%.*]] = fmul float [[TMP2]], 0x3F00000000000000
 // CHECK-NEXT:    [[SUB:%.*]] = fsub float [[TMP3]], [[TMP0]]
 // CHECK-NEXT:    [[TMP4:%.*]] = fmul float [[SUB]], 3.276800e+04
 // CHECK-NEXT:    [[TMP5:%.*]] = fptosi float [[TMP4]] to i32
-// CHECK-NEXT:    store i32 [[TMP5]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP5]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void sub_afl(void) {
@@ -510,24 +510,24 @@ void sub_afl(void) {
 
 // SIGNED-LABEL: @mul_auf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP1]] to i33
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i33 [[RESIZE]], 1
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i16 [[TMP0]] to i33
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i33 @llvm.smul.fix.i33(i33 [[UPSCALE]], i33 [[RESIZE1]], i32 16)
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i33 [[TMP2]], 1
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i33 [[DOWNSCALE]] to i32
-// SIGNED-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// SIGNED-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @mul_auf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i32
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i32 @llvm.smul.fix.i32(i32 [[TMP1]], i32 [[RESIZE]], i32 15)
-// UNSIGNED-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// UNSIGNED-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void mul_auf(void) {
@@ -536,14 +536,14 @@ void mul_auf(void) {
 
 // CHECK-LABEL: @mul_ai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @i, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @i, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP1]] to i47
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP0]] to i47
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i47 [[RESIZE1]], 15
 // CHECK-NEXT:    [[TMP2:%.*]] = call i47 @llvm.smul.fix.i47(i47 [[RESIZE]], i47 [[UPSCALE]], i32 15)
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i47 [[TMP2]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void mul_ai(void) {
@@ -552,8 +552,8 @@ void mul_ai(void) {
 
 // CHECK-LABEL: @mul_csa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @sa, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* @c, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @sa, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr @c, align 1
 // CHECK-NEXT:    [[CONV:%.*]] = sext i8 [[TMP1]] to i32
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[CONV]] to i47
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i47 [[RESIZE]], 15
@@ -569,7 +569,7 @@ void mul_ai(void) {
 // CHECK-NEXT:    [[TMP7:%.*]] = select i1 [[TMP5]], i32 [[TMP6]], i32 [[RESIZE2]]
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i32 [[TMP7]], 15
 // CHECK-NEXT:    [[RESIZE3:%.*]] = trunc i32 [[DOWNSCALE]] to i8
-// CHECK-NEXT:    store i8 [[RESIZE3]], i8* @c, align 1
+// CHECK-NEXT:    store i8 [[RESIZE3]], ptr @c, align 1
 // CHECK-NEXT:    ret void
 //
 void mul_csa(void) {
@@ -578,14 +578,14 @@ void mul_csa(void) {
 
 // CHECK-LABEL: @mul_afl(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = sitofp i32 [[TMP1]] to float
 // CHECK-NEXT:    [[TMP3:%.*]] = fmul float [[TMP2]], 0x3F00000000000000
 // CHECK-NEXT:    [[MUL:%.*]] = fmul float [[TMP3]], [[TMP0]]
 // CHECK-NEXT:    [[TMP4:%.*]] = fmul float [[MUL]], 3.276800e+04
 // CHECK-NEXT:    [[TMP5:%.*]] = fptosi float [[TMP4]] to i32
-// CHECK-NEXT:    store i32 [[TMP5]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP5]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void mul_afl(void) {
@@ -595,24 +595,24 @@ void mul_afl(void) {
 
 // SIGNED-LABEL: @div_auf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP1]] to i33
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i33 [[RESIZE]], 1
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i16 [[TMP0]] to i33
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i33 @llvm.sdiv.fix.i33(i33 [[UPSCALE]], i33 [[RESIZE1]], i32 16)
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i33 [[TMP2]], 1
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i33 [[DOWNSCALE]] to i32
-// SIGNED-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// SIGNED-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @div_auf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i32
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i32 @llvm.sdiv.fix.i32(i32 [[TMP1]], i32 [[RESIZE]], i32 15)
-// UNSIGNED-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// UNSIGNED-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void div_auf(void) {
@@ -621,14 +621,14 @@ void div_auf(void) {
 
 // CHECK-LABEL: @div_ai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @i, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @i, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP1]] to i47
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP0]] to i47
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i47 [[RESIZE1]], 15
 // CHECK-NEXT:    [[TMP2:%.*]] = call i47 @llvm.sdiv.fix.i47(i47 [[RESIZE]], i47 [[UPSCALE]], i32 15)
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i47 [[TMP2]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void div_ai(void) {
@@ -637,8 +637,8 @@ void div_ai(void) {
 
 // CHECK-LABEL: @div_csa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @sa, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* @c, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @sa, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr @c, align 1
 // CHECK-NEXT:    [[CONV:%.*]] = sext i8 [[TMP1]] to i32
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[CONV]] to i47
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i47 [[RESIZE]], 15
@@ -654,7 +654,7 @@ void div_ai(void) {
 // CHECK-NEXT:    [[TMP7:%.*]] = select i1 [[TMP5]], i32 [[TMP6]], i32 [[RESIZE2]]
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i32 [[TMP7]], 15
 // CHECK-NEXT:    [[RESIZE3:%.*]] = trunc i32 [[DOWNSCALE]] to i8
-// CHECK-NEXT:    store i8 [[RESIZE3]], i8* @c, align 1
+// CHECK-NEXT:    store i8 [[RESIZE3]], ptr @c, align 1
 // CHECK-NEXT:    ret void
 //
 void div_csa(void) {
@@ -663,14 +663,14 @@ void div_csa(void) {
 
 // CHECK-LABEL: @div_afl(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = sitofp i32 [[TMP1]] to float
 // CHECK-NEXT:    [[TMP3:%.*]] = fmul float [[TMP2]], 0x3F00000000000000
 // CHECK-NEXT:    [[DIV:%.*]] = fdiv float [[TMP3]], [[TMP0]]
 // CHECK-NEXT:    [[TMP4:%.*]] = fmul float [[DIV]], 3.276800e+04
 // CHECK-NEXT:    [[TMP5:%.*]] = fptosi float [[TMP4]] to i32
-// CHECK-NEXT:    store i32 [[TMP5]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP5]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void div_afl(void) {
@@ -680,10 +680,10 @@ void div_afl(void) {
 
 // CHECK-LABEL: @shft_ai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @i, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @i, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = shl i32 [[TMP1]], [[TMP0]]
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void shft_ai(void) {
@@ -692,20 +692,20 @@ void shft_ai(void) {
 
 // SIGNED-LABEL: @shft_sufi(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @i, align 4
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @suf, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @i, align 4
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @suf, align 2
 // SIGNED-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP0]] to i16
 // SIGNED-NEXT:    [[TMP3:%.*]] = call i16 @llvm.ushl.sat.i16(i16 [[TMP1]], i16 [[TMP2]])
-// SIGNED-NEXT:    store i16 [[TMP3]], i16* @suf, align 2
+// SIGNED-NEXT:    store i16 [[TMP3]], ptr @suf, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @shft_sufi(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @i, align 4
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @suf, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @i, align 4
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @suf, align 2
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP0]] to i16
 // UNSIGNED-NEXT:    [[TMP3:%.*]] = call i16 @llvm.sshl.sat.i16(i16 [[TMP1]], i16 [[TMP2]])
-// UNSIGNED-NEXT:    store i16 [[TMP3]], i16* @suf, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP3]], ptr @suf, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void shft_sufi(void) {
@@ -714,11 +714,11 @@ void shft_sufi(void) {
 
 // CHECK-LABEL: @shft_ulai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @i, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i64, i64* @ula, align 8
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @i, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i64, ptr @ula, align 8
 // CHECK-NEXT:    [[TMP2:%.*]] = zext i32 [[TMP0]] to i64
 // CHECK-NEXT:    [[TMP3:%.*]] = lshr i64 [[TMP1]], [[TMP2]]
-// CHECK-NEXT:    store i64 [[TMP3]], i64* @ula, align 8
+// CHECK-NEXT:    store i64 [[TMP3]], ptr @ula, align 8
 // CHECK-NEXT:    ret void
 //
 void shft_ulai(void) {

diff  --git a/clang/test/Frontend/fixed_point_conversions.c b/clang/test/Frontend/fixed_point_conversions.c
index f362f036c6761..ebd1d7e521df4 100644
--- a/clang/test/Frontend/fixed_point_conversions.c
+++ b/clang/test/Frontend/fixed_point_conversions.c
@@ -1,6 +1,6 @@
 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -triple x86_64-unknown-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,SIGNED
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -triple x86_64-unknown-linux-gnu -S -emit-llvm %s -o - -fpadding-on-unsigned-fixed-point | FileCheck %s --check-prefixes=CHECK,UNSIGNED
+// RUN: %clang_cc1 -ffixed-point -triple x86_64-unknown-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,SIGNED
+// RUN: %clang_cc1 -ffixed-point -triple x86_64-unknown-linux-gnu -S -emit-llvm %s -o - -fpadding-on-unsigned-fixed-point | FileCheck %s --check-prefixes=CHECK,UNSIGNED
 
 short _Accum sa;
 _Accum a, a2;
@@ -37,8 +37,8 @@ double d;
 
 // CHECK-LABEL: @fix_same1(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    store i32 [[TMP0]], i32* @a2, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    store i32 [[TMP0]], ptr @a2, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_same1(void) {
@@ -47,8 +47,8 @@ void fix_same1(void) {
 
 // CHECK-LABEL: @fix_same2(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    store i32 [[TMP0]], i32* @a2, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    store i32 [[TMP0]], ptr @a2, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_same2(void) {
@@ -58,10 +58,10 @@ void fix_same2(void) {
 
 // CHECK-LABEL: @fix_castdown1(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i64, i64* @la, align 8
+// CHECK-NEXT:    [[TMP0:%.*]] = load i64, ptr @la, align 8
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i64 [[TMP0]], 16
 // CHECK-NEXT:    [[RESIZE:%.*]] = trunc i64 [[DOWNSCALE]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_castdown1(void) {
@@ -70,10 +70,10 @@ void fix_castdown1(void) {
 
 // CHECK-LABEL: @fix_castdown2(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i64, i64* @la, align 8
+// CHECK-NEXT:    [[TMP0:%.*]] = load i64, ptr @la, align 8
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i64 [[TMP0]], 16
 // CHECK-NEXT:    [[RESIZE:%.*]] = trunc i64 [[DOWNSCALE]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_castdown2(void) {
@@ -82,10 +82,10 @@ void fix_castdown2(void) {
 
 // CHECK-LABEL: @fix_castdown3(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i32 [[TMP0]], 8
 // CHECK-NEXT:    [[RESIZE:%.*]] = trunc i32 [[DOWNSCALE]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void fix_castdown3(void) {
@@ -94,10 +94,10 @@ void fix_castdown3(void) {
 
 // CHECK-LABEL: @fix_castdown4(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i32 [[TMP0]], 8
 // CHECK-NEXT:    [[RESIZE:%.*]] = trunc i32 [[DOWNSCALE]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void fix_castdown4(void) {
@@ -107,10 +107,10 @@ void fix_castdown4(void) {
 
 // CHECK-LABEL: @fix_castup1(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i32
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
-// CHECK-NEXT:    store i32 [[UPSCALE]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[UPSCALE]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_castup1(void) {
@@ -119,10 +119,10 @@ void fix_castup1(void) {
 
 // CHECK-LABEL: @fix_castup2(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i32
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
-// CHECK-NEXT:    store i32 [[UPSCALE]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[UPSCALE]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_castup2(void) {
@@ -131,10 +131,10 @@ void fix_castup2(void) {
 
 // CHECK-LABEL: @fix_castup3(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i64, i64* @la, align 8
+// CHECK-NEXT:    [[TMP0:%.*]] = load i64, ptr @la, align 8
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i64 [[TMP0]], 16
 // CHECK-NEXT:    [[RESIZE:%.*]] = trunc i64 [[DOWNSCALE]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_castup3(void) {
@@ -143,10 +143,10 @@ void fix_castup3(void) {
 
 // CHECK-LABEL: @fix_castup4(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i64, i64* @la, align 8
+// CHECK-NEXT:    [[TMP0:%.*]] = load i64, ptr @la, align 8
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i64 [[TMP0]], 16
 // CHECK-NEXT:    [[RESIZE:%.*]] = trunc i64 [[DOWNSCALE]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_castup4(void) {
@@ -156,15 +156,15 @@ void fix_castup4(void) {
 
 // SIGNED-LABEL: @fix_sign1(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i32 [[TMP0]], 1
-// SIGNED-NEXT:    store i32 [[UPSCALE]], i32* @ua, align 4
+// SIGNED-NEXT:    store i32 [[UPSCALE]], ptr @ua, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @fix_sign1(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// UNSIGNED-NEXT:    store i32 [[TMP0]], i32* @ua, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// UNSIGNED-NEXT:    store i32 [[TMP0]], ptr @ua, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void fix_sign1(void) {
@@ -173,15 +173,15 @@ void fix_sign1(void) {
 
 // SIGNED-LABEL: @fix_sign2(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = lshr i32 [[TMP0]], 1
-// SIGNED-NEXT:    store i32 [[DOWNSCALE]], i32* @a, align 4
+// SIGNED-NEXT:    store i32 [[DOWNSCALE]], ptr @a, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @fix_sign2(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
-// UNSIGNED-NEXT:    store i32 [[TMP0]], i32* @a, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
+// UNSIGNED-NEXT:    store i32 [[TMP0]], ptr @a, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void fix_sign2(void) {
@@ -190,15 +190,15 @@ void fix_sign2(void) {
 
 // SIGNED-LABEL: @fix_sign3(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i32 [[TMP0]], 1
-// SIGNED-NEXT:    store i32 [[UPSCALE]], i32* @ua, align 4
+// SIGNED-NEXT:    store i32 [[UPSCALE]], ptr @ua, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @fix_sign3(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// UNSIGNED-NEXT:    store i32 [[TMP0]], i32* @ua, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// UNSIGNED-NEXT:    store i32 [[TMP0]], ptr @ua, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void fix_sign3(void) {
@@ -207,15 +207,15 @@ void fix_sign3(void) {
 
 // SIGNED-LABEL: @fix_sign4(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = lshr i32 [[TMP0]], 1
-// SIGNED-NEXT:    store i32 [[DOWNSCALE]], i32* @a, align 4
+// SIGNED-NEXT:    store i32 [[DOWNSCALE]], ptr @a, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @fix_sign4(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
-// UNSIGNED-NEXT:    store i32 [[TMP0]], i32* @a, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
+// UNSIGNED-NEXT:    store i32 [[TMP0]], ptr @a, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void fix_sign4(void) {
@@ -224,18 +224,18 @@ void fix_sign4(void) {
 
 // SIGNED-LABEL: @fix_sign5(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i64
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i64 [[RESIZE]], 17
-// SIGNED-NEXT:    store i64 [[UPSCALE]], i64* @ula, align 8
+// SIGNED-NEXT:    store i64 [[UPSCALE]], ptr @ula, align 8
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @fix_sign5(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i64
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i64 [[RESIZE]], 16
-// UNSIGNED-NEXT:    store i64 [[UPSCALE]], i64* @ula, align 8
+// UNSIGNED-NEXT:    store i64 [[UPSCALE]], ptr @ula, align 8
 // UNSIGNED-NEXT:    ret void
 //
 void fix_sign5(void) {
@@ -245,14 +245,14 @@ void fix_sign5(void) {
 
 // CHECK-LABEL: @fix_sat1(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @sat_a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @sat_a, align 4
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i32 [[TMP0]], 8
 // CHECK-NEXT:    [[TMP1:%.*]] = icmp sgt i32 [[DOWNSCALE]], 32767
 // CHECK-NEXT:    [[SATMAX:%.*]] = select i1 [[TMP1]], i32 32767, i32 [[DOWNSCALE]]
 // CHECK-NEXT:    [[TMP2:%.*]] = icmp slt i32 [[SATMAX]], -32768
 // CHECK-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP2]], i32 -32768, i32 [[SATMAX]]
 // CHECK-NEXT:    [[RESIZE:%.*]] = trunc i32 [[SATMIN]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE]], i16* @sat_sa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE]], ptr @sat_sa, align 2
 // CHECK-NEXT:    ret void
 //
 void fix_sat1(void) {
@@ -262,14 +262,14 @@ void fix_sat1(void) {
 
 // CHECK-LABEL: @fix_sat2(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @sat_a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @sat_a, align 4
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i32 [[TMP0]], 8
 // CHECK-NEXT:    [[TMP1:%.*]] = icmp sgt i32 [[DOWNSCALE]], 127
 // CHECK-NEXT:    [[SATMAX:%.*]] = select i1 [[TMP1]], i32 127, i32 [[DOWNSCALE]]
 // CHECK-NEXT:    [[TMP2:%.*]] = icmp slt i32 [[SATMAX]], -128
 // CHECK-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP2]], i32 -128, i32 [[SATMAX]]
 // CHECK-NEXT:    [[RESIZE:%.*]] = trunc i32 [[SATMIN]] to i8
-// CHECK-NEXT:    store i8 [[RESIZE]], i8* @sat_sf, align 1
+// CHECK-NEXT:    store i8 [[RESIZE]], ptr @sat_sf, align 1
 // CHECK-NEXT:    ret void
 //
 void fix_sat2(void) {
@@ -279,13 +279,13 @@ void fix_sat2(void) {
 
 // CHECK-LABEL: @fix_sat3(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = icmp sgt i32 [[TMP0]], 32767
 // CHECK-NEXT:    [[SATMAX:%.*]] = select i1 [[TMP1]], i32 32767, i32 [[TMP0]]
 // CHECK-NEXT:    [[TMP2:%.*]] = icmp slt i32 [[SATMAX]], -32768
 // CHECK-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP2]], i32 -32768, i32 [[SATMAX]]
 // CHECK-NEXT:    [[RESIZE:%.*]] = trunc i32 [[SATMIN]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE]], i16* @sat_f, align 2
+// CHECK-NEXT:    store i16 [[RESIZE]], ptr @sat_f, align 2
 // CHECK-NEXT:    ret void
 //
 void fix_sat3(void) {
@@ -295,7 +295,7 @@ void fix_sat3(void) {
 
 // CHECK-LABEL: @fix_sat4(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @sat_a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @sat_a, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i48
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i48 [[RESIZE]], 16
 // CHECK-NEXT:    [[TMP1:%.*]] = icmp sgt i48 [[UPSCALE]], 2147483647
@@ -303,7 +303,7 @@ void fix_sat3(void) {
 // CHECK-NEXT:    [[TMP2:%.*]] = icmp slt i48 [[SATMAX]], -2147483648
 // CHECK-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP2]], i48 -2147483648, i48 [[SATMAX]]
 // CHECK-NEXT:    [[RESIZE1:%.*]] = trunc i48 [[SATMIN]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE1]], i32* @sat_lf, align 4
+// CHECK-NEXT:    store i32 [[RESIZE1]], ptr @sat_lf, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_sat4(void) {
@@ -313,26 +313,26 @@ void fix_sat4(void) {
 
 // SIGNED-LABEL: @fix_sat5(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @sat_a, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @sat_a, align 4
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i32 [[TMP0]], 7
 // SIGNED-NEXT:    [[TMP1:%.*]] = icmp sgt i32 [[DOWNSCALE]], 65535
 // SIGNED-NEXT:    [[SATMAX:%.*]] = select i1 [[TMP1]], i32 65535, i32 [[DOWNSCALE]]
 // SIGNED-NEXT:    [[TMP2:%.*]] = icmp slt i32 [[SATMAX]], 0
 // SIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP2]], i32 0, i32 [[SATMAX]]
 // SIGNED-NEXT:    [[RESIZE:%.*]] = trunc i32 [[SATMIN]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE]], i16* @sat_usa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE]], ptr @sat_usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @fix_sat5(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @sat_a, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @sat_a, align 4
 // UNSIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i32 [[TMP0]], 8
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = icmp sgt i32 [[DOWNSCALE]], 32767
 // UNSIGNED-NEXT:    [[SATMAX:%.*]] = select i1 [[TMP1]], i32 32767, i32 [[DOWNSCALE]]
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = icmp slt i32 [[SATMAX]], 0
 // UNSIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP2]], i32 0, i32 [[SATMAX]]
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = trunc i32 [[SATMIN]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE]], i16* @sat_usa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE]], ptr @sat_usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void fix_sat5(void) {
@@ -342,21 +342,21 @@ void fix_sat5(void) {
 
 // SIGNED-LABEL: @fix_sat6(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @sat_a, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @sat_a, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i33
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i33 [[RESIZE]], 1
 // SIGNED-NEXT:    [[TMP1:%.*]] = icmp slt i33 [[UPSCALE]], 0
 // SIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP1]], i33 0, i33 [[UPSCALE]]
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = trunc i33 [[SATMIN]] to i32
-// SIGNED-NEXT:    store i32 [[RESIZE1]], i32* @sat_ua, align 4
+// SIGNED-NEXT:    store i32 [[RESIZE1]], ptr @sat_ua, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @fix_sat6(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @sat_a, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @sat_a, align 4
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = icmp slt i32 [[TMP0]], 0
 // UNSIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP1]], i32 0, i32 [[TMP0]]
-// UNSIGNED-NEXT:    store i32 [[SATMIN]], i32* @sat_ua, align 4
+// UNSIGNED-NEXT:    store i32 [[SATMIN]], ptr @sat_ua, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void fix_sat6(void) {
@@ -366,8 +366,8 @@ void fix_sat6(void) {
 
 // CHECK-LABEL: @fix_sat7(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    store i32 [[TMP0]], i32* @sat_a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    store i32 [[TMP0]], ptr @sat_a, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_sat7(void) {
@@ -377,8 +377,8 @@ void fix_sat7(void) {
 
 // CHECK-LABEL: @fix_sat8(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @sat_a, align 4
-// CHECK-NEXT:    store i32 [[TMP0]], i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @sat_a, align 4
+// CHECK-NEXT:    store i32 [[TMP0]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_sat8(void) {
@@ -388,9 +388,9 @@ void fix_sat8(void) {
 
 // CHECK-LABEL: @fix_sat9(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sat_f, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sat_f, align 2
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE]], i32* @sat_a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE]], ptr @sat_a, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_sat9(void) {
@@ -400,22 +400,22 @@ void fix_sat9(void) {
 
 // SIGNED-LABEL: @fix_sat10(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i8, i8* @sat_sf, align 1
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i8, ptr @sat_sf, align 1
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i8 [[TMP0]] to i32
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 9
 // SIGNED-NEXT:    [[TMP1:%.*]] = icmp slt i32 [[UPSCALE]], 0
 // SIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP1]], i32 0, i32 [[UPSCALE]]
-// SIGNED-NEXT:    store i32 [[SATMIN]], i32* @sat_ua, align 4
+// SIGNED-NEXT:    store i32 [[SATMIN]], ptr @sat_ua, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @fix_sat10(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i8, i8* @sat_sf, align 1
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i8, ptr @sat_sf, align 1
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = sext i8 [[TMP0]] to i32
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = icmp slt i32 [[UPSCALE]], 0
 // UNSIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP1]], i32 0, i32 [[UPSCALE]]
-// UNSIGNED-NEXT:    store i32 [[SATMIN]], i32* @sat_ua, align 4
+// UNSIGNED-NEXT:    store i32 [[SATMIN]], ptr @sat_ua, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void fix_sat10(void) {
@@ -426,10 +426,10 @@ void fix_sat10(void) {
 
 // CHECK-LABEL: @fix_fract1(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i32 [[TMP0]], 8
 // CHECK-NEXT:    [[RESIZE:%.*]] = trunc i32 [[DOWNSCALE]] to i8
-// CHECK-NEXT:    store i8 [[RESIZE]], i8* @sf, align 1
+// CHECK-NEXT:    store i8 [[RESIZE]], ptr @sf, align 1
 // CHECK-NEXT:    ret void
 //
 void fix_fract1(void) {
@@ -439,10 +439,10 @@ void fix_fract1(void) {
 
 // CHECK-LABEL: @fix_fract2(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i8, i8* @sf, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i8, ptr @sf, align 1
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i8 [[TMP0]] to i32
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
-// CHECK-NEXT:    store i32 [[UPSCALE]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[UPSCALE]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_fract2(void) {
@@ -452,9 +452,9 @@ void fix_fract2(void) {
 
 // CHECK-LABEL: @fix_fract3(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = trunc i32 [[TMP0]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE]], i16* @f, align 2
+// CHECK-NEXT:    store i16 [[RESIZE]], ptr @f, align 2
 // CHECK-NEXT:    ret void
 //
 void fix_fract3(void) {
@@ -464,9 +464,9 @@ void fix_fract3(void) {
 
 // CHECK-LABEL: @fix_fract4(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @f, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @f, align 2
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_fract4(void) {
@@ -475,9 +475,9 @@ void fix_fract4(void) {
 
 // CHECK-LABEL: @fix_fract5(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
 // CHECK-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE]], i32* @ua, align 4
+// CHECK-NEXT:    store i32 [[RESIZE]], ptr @ua, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_fract5(void) {
@@ -487,9 +487,9 @@ void fix_fract5(void) {
 
 // CHECK-LABEL: @fix_fract6(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = trunc i32 [[TMP0]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE]], i16* @uf, align 2
+// CHECK-NEXT:    store i16 [[RESIZE]], ptr @uf, align 2
 // CHECK-NEXT:    ret void
 //
 void fix_fract6(void) {
@@ -499,13 +499,13 @@ void fix_fract6(void) {
 
 // CHECK-LABEL: @fix_int1(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = icmp slt i16 [[TMP0]], 0
 // CHECK-NEXT:    [[TMP2:%.*]] = add i16 [[TMP0]], 127
 // CHECK-NEXT:    [[TMP3:%.*]] = select i1 [[TMP1]], i16 [[TMP2]], i16 [[TMP0]]
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i16 [[TMP3]], 7
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[DOWNSCALE]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE]], i32* @i, align 4
+// CHECK-NEXT:    store i32 [[RESIZE]], ptr @i, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_int1(void) {
@@ -515,18 +515,18 @@ void fix_int1(void) {
 
 // SIGNED-LABEL: @fix_int2(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = lshr i16 [[TMP0]], 8
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[DOWNSCALE]] to i32
-// SIGNED-NEXT:    store i32 [[RESIZE]], i32* @i, align 4
+// SIGNED-NEXT:    store i32 [[RESIZE]], ptr @i, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @fix_int2(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
 // UNSIGNED-NEXT:    [[DOWNSCALE:%.*]] = lshr i16 [[TMP0]], 7
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[DOWNSCALE]] to i32
-// UNSIGNED-NEXT:    store i32 [[RESIZE]], i32* @i, align 4
+// UNSIGNED-NEXT:    store i32 [[RESIZE]], ptr @i, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void fix_int2(void) {
@@ -537,10 +537,10 @@ void fix_int2(void) {
 
 // CHECK-LABEL: @int_fix1(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = trunc i32 [[TMP0]] to i16
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i16 [[RESIZE]], 7
-// CHECK-NEXT:    store i16 [[UPSCALE]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[UPSCALE]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void int_fix1(void) {
@@ -549,10 +549,10 @@ void int_fix1(void) {
 
 // CHECK-LABEL: @int_fix2(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @ui, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @ui, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = trunc i32 [[TMP0]] to i16
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i16 [[RESIZE]], 7
-// CHECK-NEXT:    store i16 [[UPSCALE]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[UPSCALE]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void int_fix2(void) {
@@ -561,18 +561,18 @@ void int_fix2(void) {
 
 // SIGNED-LABEL: @int_fix3(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @i, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @i, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = trunc i32 [[TMP0]] to i16
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i16 [[RESIZE]], 8
-// SIGNED-NEXT:    store i16 [[UPSCALE]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[UPSCALE]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @int_fix3(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @i, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @i, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = trunc i32 [[TMP0]] to i16
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i16 [[RESIZE]], 7
-// UNSIGNED-NEXT:    store i16 [[UPSCALE]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[UPSCALE]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void int_fix3(void) {
@@ -581,18 +581,18 @@ void int_fix3(void) {
 
 // SIGNED-LABEL: @int_fix4(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ui, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ui, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = trunc i32 [[TMP0]] to i16
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i16 [[RESIZE]], 8
-// SIGNED-NEXT:    store i16 [[UPSCALE]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[UPSCALE]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @int_fix4(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ui, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ui, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = trunc i32 [[TMP0]] to i16
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i16 [[RESIZE]], 7
-// UNSIGNED-NEXT:    store i16 [[UPSCALE]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[UPSCALE]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void int_fix4(void) {
@@ -601,10 +601,10 @@ void int_fix4(void) {
 
 // CHECK-LABEL: @int_fix5(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @s, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @s, align 2
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i64
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i64 [[RESIZE]], 31
-// CHECK-NEXT:    store i64 [[UPSCALE]], i64* @la, align 8
+// CHECK-NEXT:    store i64 [[UPSCALE]], ptr @la, align 8
 // CHECK-NEXT:    ret void
 //
 void int_fix5(void) {
@@ -614,7 +614,7 @@ void int_fix5(void) {
 
 // CHECK-LABEL: @int_sat1(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i39
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE]], 7
 // CHECK-NEXT:    [[TMP1:%.*]] = icmp sgt i39 [[UPSCALE]], 32767
@@ -622,7 +622,7 @@ void int_fix5(void) {
 // CHECK-NEXT:    [[TMP2:%.*]] = icmp slt i39 [[SATMAX]], -32768
 // CHECK-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP2]], i39 -32768, i39 [[SATMAX]]
 // CHECK-NEXT:    [[RESIZE1:%.*]] = trunc i39 [[SATMIN]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE1]], i16* @sat_sa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE1]], ptr @sat_sa, align 2
 // CHECK-NEXT:    ret void
 //
 void int_sat1(void) {
@@ -631,13 +631,13 @@ void int_sat1(void) {
 
 // CHECK-LABEL: @int_sat2(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @ui, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @ui, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = zext i32 [[TMP0]] to i39
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE]], 7
 // CHECK-NEXT:    [[TMP1:%.*]] = icmp ugt i39 [[UPSCALE]], 32767
 // CHECK-NEXT:    [[SATMAX:%.*]] = select i1 [[TMP1]], i39 32767, i39 [[UPSCALE]]
 // CHECK-NEXT:    [[RESIZE1:%.*]] = trunc i39 [[SATMAX]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE1]], i16* @sat_sa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE1]], ptr @sat_sa, align 2
 // CHECK-NEXT:    ret void
 //
 void int_sat2(void) {
@@ -646,7 +646,7 @@ void int_sat2(void) {
 
 // SIGNED-LABEL: @int_sat3(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @i, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @i, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i40
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE]], 8
 // SIGNED-NEXT:    [[TMP1:%.*]] = icmp sgt i40 [[UPSCALE]], 65535
@@ -654,12 +654,12 @@ void int_sat2(void) {
 // SIGNED-NEXT:    [[TMP2:%.*]] = icmp slt i40 [[SATMAX]], 0
 // SIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP2]], i40 0, i40 [[SATMAX]]
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = trunc i40 [[SATMIN]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE1]], i16* @sat_usa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE1]], ptr @sat_usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @int_sat3(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @i, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @i, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i39
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE]], 7
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = icmp sgt i39 [[UPSCALE]], 32767
@@ -667,7 +667,7 @@ void int_sat2(void) {
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = icmp slt i39 [[SATMAX]], 0
 // UNSIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP2]], i39 0, i39 [[SATMAX]]
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = trunc i39 [[SATMIN]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE1]], i16* @sat_usa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE1]], ptr @sat_usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void int_sat3(void) {
@@ -676,24 +676,24 @@ void int_sat3(void) {
 
 // SIGNED-LABEL: @int_sat4(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ui, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ui, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i32 [[TMP0]] to i40
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE]], 8
 // SIGNED-NEXT:    [[TMP1:%.*]] = icmp ugt i40 [[UPSCALE]], 65535
 // SIGNED-NEXT:    [[SATMAX:%.*]] = select i1 [[TMP1]], i40 65535, i40 [[UPSCALE]]
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = trunc i40 [[SATMAX]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE1]], i16* @sat_usa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE1]], ptr @sat_usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @int_sat4(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ui, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ui, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i32 [[TMP0]] to i39
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE]], 7
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = icmp ugt i39 [[UPSCALE]], 32767
 // UNSIGNED-NEXT:    [[SATMAX:%.*]] = select i1 [[TMP1]], i39 32767, i39 [[UPSCALE]]
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = trunc i39 [[SATMAX]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE1]], i16* @sat_usa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE1]], ptr @sat_usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void int_sat4(void) {
@@ -703,10 +703,10 @@ void int_sat4(void) {
 
 // CHECK-LABEL: @float_fix1(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = fmul float [[TMP0]], 1.280000e+02
 // CHECK-NEXT:    [[TMP2:%.*]] = fptosi float [[TMP1]] to i16
-// CHECK-NEXT:    store i16 [[TMP2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[TMP2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void float_fix1(void) {
@@ -715,10 +715,10 @@ void float_fix1(void) {
 
 // CHECK-LABEL: @float_fix2(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = fmul float [[TMP0]], 3.276800e+04
 // CHECK-NEXT:    [[TMP2:%.*]] = fptosi float [[TMP1]] to i32
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void float_fix2(void) {
@@ -727,10 +727,10 @@ void float_fix2(void) {
 
 // CHECK-LABEL: @float_fix3(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = fmul float [[TMP0]], 0x41E0000000000000
 // CHECK-NEXT:    [[TMP2:%.*]] = fptosi float [[TMP1]] to i64
-// CHECK-NEXT:    store i64 [[TMP2]], i64* @la, align 8
+// CHECK-NEXT:    store i64 [[TMP2]], ptr @la, align 8
 // CHECK-NEXT:    ret void
 //
 void float_fix3(void) {
@@ -739,10 +739,10 @@ void float_fix3(void) {
 
 // CHECK-LABEL: @float_fix4(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = fmul float [[TMP0]], 1.280000e+02
 // CHECK-NEXT:    [[TMP2:%.*]] = fptosi float [[TMP1]] to i8
-// CHECK-NEXT:    store i8 [[TMP2]], i8* @sf, align 1
+// CHECK-NEXT:    store i8 [[TMP2]], ptr @sf, align 1
 // CHECK-NEXT:    ret void
 //
 void float_fix4(void) {
@@ -751,10 +751,10 @@ void float_fix4(void) {
 
 // CHECK-LABEL: @float_fix5(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = fmul float [[TMP0]], 0x41E0000000000000
 // CHECK-NEXT:    [[TMP2:%.*]] = fptosi float [[TMP1]] to i32
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @lf, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @lf, align 4
 // CHECK-NEXT:    ret void
 //
 void float_fix5(void) {
@@ -763,18 +763,18 @@ void float_fix5(void) {
 
 // SIGNED-LABEL: @float_fix6(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
 // SIGNED-NEXT:    [[TMP1:%.*]] = fmul float [[TMP0]], 6.553600e+04
 // SIGNED-NEXT:    [[TMP2:%.*]] = fptoui float [[TMP1]] to i32
-// SIGNED-NEXT:    store i32 [[TMP2]], i32* @ua, align 4
+// SIGNED-NEXT:    store i32 [[TMP2]], ptr @ua, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @float_fix6(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = fmul float [[TMP0]], 3.276800e+04
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = fptosi float [[TMP1]] to i32
-// UNSIGNED-NEXT:    store i32 [[TMP2]], i32* @ua, align 4
+// UNSIGNED-NEXT:    store i32 [[TMP2]], ptr @ua, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void float_fix6(void) {
@@ -783,18 +783,18 @@ void float_fix6(void) {
 
 // SIGNED-LABEL: @float_fix7(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
 // SIGNED-NEXT:    [[TMP1:%.*]] = fmul float [[TMP0]], 6.553600e+04
 // SIGNED-NEXT:    [[TMP2:%.*]] = fptoui float [[TMP1]] to i16
-// SIGNED-NEXT:    store i16 [[TMP2]], i16* @uf, align 2
+// SIGNED-NEXT:    store i16 [[TMP2]], ptr @uf, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @float_fix7(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = fmul float [[TMP0]], 3.276800e+04
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = fptosi float [[TMP1]] to i16
-// UNSIGNED-NEXT:    store i16 [[TMP2]], i16* @uf, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP2]], ptr @uf, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void float_fix7(void) {
@@ -804,10 +804,10 @@ void float_fix7(void) {
 
 // CHECK-LABEL: @fix_float1(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = sitofp i16 [[TMP0]] to float
 // CHECK-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 7.812500e-03
-// CHECK-NEXT:    store float [[TMP2]], float* @fl, align 4
+// CHECK-NEXT:    store float [[TMP2]], ptr @fl, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_float1(void) {
@@ -816,10 +816,10 @@ void fix_float1(void) {
 
 // CHECK-LABEL: @fix_float2(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = sitofp i32 [[TMP0]] to float
 // CHECK-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x3F00000000000000
-// CHECK-NEXT:    store float [[TMP2]], float* @fl, align 4
+// CHECK-NEXT:    store float [[TMP2]], ptr @fl, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_float2(void) {
@@ -828,10 +828,10 @@ void fix_float2(void) {
 
 // CHECK-LABEL: @fix_float3(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i64, i64* @la, align 8
+// CHECK-NEXT:    [[TMP0:%.*]] = load i64, ptr @la, align 8
 // CHECK-NEXT:    [[TMP1:%.*]] = sitofp i64 [[TMP0]] to float
 // CHECK-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x3E00000000000000
-// CHECK-NEXT:    store float [[TMP2]], float* @fl, align 4
+// CHECK-NEXT:    store float [[TMP2]], ptr @fl, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_float3(void) {
@@ -840,10 +840,10 @@ void fix_float3(void) {
 
 // CHECK-LABEL: @fix_float4(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i8, i8* @sf, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i8, ptr @sf, align 1
 // CHECK-NEXT:    [[TMP1:%.*]] = sitofp i8 [[TMP0]] to float
 // CHECK-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 7.812500e-03
-// CHECK-NEXT:    store float [[TMP2]], float* @fl, align 4
+// CHECK-NEXT:    store float [[TMP2]], ptr @fl, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_float4(void) {
@@ -852,10 +852,10 @@ void fix_float4(void) {
 
 // CHECK-LABEL: @fix_float5(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @lf, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @lf, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = sitofp i32 [[TMP0]] to float
 // CHECK-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x3E00000000000000
-// CHECK-NEXT:    store float [[TMP2]], float* @fl, align 4
+// CHECK-NEXT:    store float [[TMP2]], ptr @fl, align 4
 // CHECK-NEXT:    ret void
 //
 void fix_float5(void) {
@@ -864,18 +864,18 @@ void fix_float5(void) {
 
 // SIGNED-LABEL: @fix_float6(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
 // SIGNED-NEXT:    [[TMP1:%.*]] = uitofp i32 [[TMP0]] to float
 // SIGNED-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x3EF0000000000000
-// SIGNED-NEXT:    store float [[TMP2]], float* @fl, align 4
+// SIGNED-NEXT:    store float [[TMP2]], ptr @fl, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @fix_float6(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = uitofp i32 [[TMP0]] to float
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x3F00000000000000
-// UNSIGNED-NEXT:    store float [[TMP2]], float* @fl, align 4
+// UNSIGNED-NEXT:    store float [[TMP2]], ptr @fl, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void fix_float6(void) {
@@ -884,18 +884,18 @@ void fix_float6(void) {
 
 // SIGNED-LABEL: @fix_float7(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
 // SIGNED-NEXT:    [[TMP1:%.*]] = uitofp i16 [[TMP0]] to float
 // SIGNED-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x3EF0000000000000
-// SIGNED-NEXT:    store float [[TMP2]], float* @fl, align 4
+// SIGNED-NEXT:    store float [[TMP2]], ptr @fl, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @fix_float7(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = uitofp i16 [[TMP0]] to float
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x3F00000000000000
-// UNSIGNED-NEXT:    store float [[TMP2]], float* @fl, align 4
+// UNSIGNED-NEXT:    store float [[TMP2]], ptr @fl, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void fix_float7(void) {
@@ -905,10 +905,10 @@ void fix_float7(void) {
 
 // CHECK-LABEL: @float_sat1(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = fmul float [[TMP0]], 1.280000e+02
 // CHECK-NEXT:    [[TMP2:%.*]] = call i16 @llvm.fptosi.sat.i16.f32(float [[TMP1]])
-// CHECK-NEXT:    store i16 [[TMP2]], i16* @sat_sa, align 2
+// CHECK-NEXT:    store i16 [[TMP2]], ptr @sat_sa, align 2
 // CHECK-NEXT:    ret void
 //
 void float_sat1(void) {
@@ -917,10 +917,10 @@ void float_sat1(void) {
 
 // CHECK-LABEL: @float_sat2(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = fmul float [[TMP0]], 3.276800e+04
 // CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.fptosi.sat.i32.f32(float [[TMP1]])
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @sat_a, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @sat_a, align 4
 // CHECK-NEXT:    ret void
 //
 void float_sat2(void) {
@@ -929,10 +929,10 @@ void float_sat2(void) {
 
 // CHECK-LABEL: @float_sat3(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = fmul float [[TMP0]], 0x41E0000000000000
 // CHECK-NEXT:    [[TMP2:%.*]] = call i64 @llvm.fptosi.sat.i64.f32(float [[TMP1]])
-// CHECK-NEXT:    store i64 [[TMP2]], i64* @sat_la, align 8
+// CHECK-NEXT:    store i64 [[TMP2]], ptr @sat_la, align 8
 // CHECK-NEXT:    ret void
 //
 void float_sat3(void) {
@@ -941,10 +941,10 @@ void float_sat3(void) {
 
 // CHECK-LABEL: @float_sat4(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = fmul float [[TMP0]], 1.280000e+02
 // CHECK-NEXT:    [[TMP2:%.*]] = call i8 @llvm.fptosi.sat.i8.f32(float [[TMP1]])
-// CHECK-NEXT:    store i8 [[TMP2]], i8* @sat_sf, align 1
+// CHECK-NEXT:    store i8 [[TMP2]], ptr @sat_sf, align 1
 // CHECK-NEXT:    ret void
 //
 void float_sat4(void) {
@@ -953,20 +953,20 @@ void float_sat4(void) {
 
 // SIGNED-LABEL: @float_sat5(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
 // SIGNED-NEXT:    [[TMP1:%.*]] = fmul float [[TMP0]], 6.553600e+04
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i32 @llvm.fptoui.sat.i32.f32(float [[TMP1]])
-// SIGNED-NEXT:    store i32 [[TMP2]], i32* @sat_ua, align 4
+// SIGNED-NEXT:    store i32 [[TMP2]], ptr @sat_ua, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @float_sat5(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = fmul float [[TMP0]], 3.276800e+04
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i32 @llvm.fptosi.sat.i32.f32(float [[TMP1]])
 // UNSIGNED-NEXT:    [[TMP3:%.*]] = icmp slt i32 [[TMP2]], 0
 // UNSIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP3]], i32 0, i32 [[TMP2]]
-// UNSIGNED-NEXT:    store i32 [[SATMIN]], i32* @sat_ua, align 4
+// UNSIGNED-NEXT:    store i32 [[SATMIN]], ptr @sat_ua, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void float_sat5(void) {
@@ -975,20 +975,20 @@ void float_sat5(void) {
 
 // SIGNED-LABEL: @float_sat6(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
 // SIGNED-NEXT:    [[TMP1:%.*]] = fmul float [[TMP0]], 6.553600e+04
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.fptoui.sat.i16.f32(float [[TMP1]])
-// SIGNED-NEXT:    store i16 [[TMP2]], i16* @sat_uf, align 2
+// SIGNED-NEXT:    store i16 [[TMP2]], ptr @sat_uf, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @float_sat6(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load float, float* @fl, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load float, ptr @fl, align 4
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = fmul float [[TMP0]], 3.276800e+04
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.fptosi.sat.i16.f32(float [[TMP1]])
 // UNSIGNED-NEXT:    [[TMP3:%.*]] = icmp slt i16 [[TMP2]], 0
 // UNSIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP3]], i16 0, i16 [[TMP2]]
-// UNSIGNED-NEXT:    store i16 [[SATMIN]], i16* @sat_uf, align 2
+// UNSIGNED-NEXT:    store i16 [[SATMIN]], ptr @sat_uf, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void float_sat6(void) {

diff  --git a/clang/test/Frontend/fixed_point_conversions_half.c b/clang/test/Frontend/fixed_point_conversions_half.c
index a636593ed8687..4094b6af5dee6 100644
--- a/clang/test/Frontend/fixed_point_conversions_half.c
+++ b/clang/test/Frontend/fixed_point_conversions_half.c
@@ -1,6 +1,6 @@
 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -triple arm64-unknown-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,SIGNED
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -triple arm64-unknown-linux-gnu -S -emit-llvm %s -o - -fpadding-on-unsigned-fixed-point | FileCheck %s --check-prefixes=CHECK,UNSIGNED
+// RUN: %clang_cc1 -ffixed-point -triple arm64-unknown-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,SIGNED
+// RUN: %clang_cc1 -ffixed-point -triple arm64-unknown-linux-gnu -S -emit-llvm %s -o - -fpadding-on-unsigned-fixed-point | FileCheck %s --check-prefixes=CHECK,UNSIGNED
 
 short _Fract sf;
 long _Fract lf;
@@ -25,10 +25,10 @@ _Float16 h;
 
 // CHECK-LABEL: @half_fix1(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load half, half* @h, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load half, ptr @h, align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = fmul half [[TMP0]], 0xH5800
 // CHECK-NEXT:    [[TMP2:%.*]] = fptosi half [[TMP1]] to i8
-// CHECK-NEXT:    store i8 [[TMP2]], i8* @sf, align 1
+// CHECK-NEXT:    store i8 [[TMP2]], ptr @sf, align 1
 // CHECK-NEXT:    ret void
 //
 void half_fix1(void) {
@@ -37,11 +37,11 @@ void half_fix1(void) {
 
 // CHECK-LABEL: @half_fix2(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load half, half* @h, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load half, ptr @h, align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = fpext half [[TMP0]] to float
 // CHECK-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x41E0000000000000
 // CHECK-NEXT:    [[TMP3:%.*]] = fptosi float [[TMP2]] to i32
-// CHECK-NEXT:    store i32 [[TMP3]], i32* @lf, align 4
+// CHECK-NEXT:    store i32 [[TMP3]], ptr @lf, align 4
 // CHECK-NEXT:    ret void
 //
 void half_fix2(void) {
@@ -50,10 +50,10 @@ void half_fix2(void) {
 
 // CHECK-LABEL: @half_fix3(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load half, half* @h, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load half, ptr @h, align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = fmul half [[TMP0]], 0xH5800
 // CHECK-NEXT:    [[TMP2:%.*]] = fptosi half [[TMP1]] to i16
-// CHECK-NEXT:    store i16 [[TMP2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[TMP2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void half_fix3(void) {
@@ -62,11 +62,11 @@ void half_fix3(void) {
 
 // CHECK-LABEL: @half_fix4(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load half, half* @h, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load half, ptr @h, align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = fpext half [[TMP0]] to float
 // CHECK-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x41E0000000000000
 // CHECK-NEXT:    [[TMP3:%.*]] = fptosi float [[TMP2]] to i64
-// CHECK-NEXT:    store i64 [[TMP3]], i64* @la, align 8
+// CHECK-NEXT:    store i64 [[TMP3]], ptr @la, align 8
 // CHECK-NEXT:    ret void
 //
 void half_fix4(void) {
@@ -75,19 +75,19 @@ void half_fix4(void) {
 
 // SIGNED-LABEL: @half_fix5(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load half, half* @h, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load half, ptr @h, align 2
 // SIGNED-NEXT:    [[TMP1:%.*]] = fpext half [[TMP0]] to float
 // SIGNED-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 2.560000e+02
 // SIGNED-NEXT:    [[TMP3:%.*]] = fptoui float [[TMP2]] to i16
-// SIGNED-NEXT:    store i16 [[TMP3]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[TMP3]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @half_fix5(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load half, half* @h, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load half, ptr @h, align 2
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = fmul half [[TMP0]], 0xH5800
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = fptosi half [[TMP1]] to i16
-// UNSIGNED-NEXT:    store i16 [[TMP2]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP2]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void half_fix5(void) {
@@ -96,20 +96,20 @@ void half_fix5(void) {
 
 // SIGNED-LABEL: @half_fix6(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load half, half* @h, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load half, ptr @h, align 2
 // SIGNED-NEXT:    [[TMP1:%.*]] = fpext half [[TMP0]] to float
 // SIGNED-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x41F0000000000000
 // SIGNED-NEXT:    [[TMP3:%.*]] = fptoui float [[TMP2]] to i64
-// SIGNED-NEXT:    store i64 [[TMP3]], i64* @ula, align 8
+// SIGNED-NEXT:    store i64 [[TMP3]], ptr @ula, align 8
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @half_fix6(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load half, half* @h, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load half, ptr @h, align 2
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = fpext half [[TMP0]] to float
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x41E0000000000000
 // UNSIGNED-NEXT:    [[TMP3:%.*]] = fptosi float [[TMP2]] to i64
-// UNSIGNED-NEXT:    store i64 [[TMP3]], i64* @ula, align 8
+// UNSIGNED-NEXT:    store i64 [[TMP3]], ptr @ula, align 8
 // UNSIGNED-NEXT:    ret void
 //
 void half_fix6(void) {
@@ -119,10 +119,10 @@ void half_fix6(void) {
 
 // CHECK-LABEL: @half_sat1(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load half, half* @h, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load half, ptr @h, align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = fmul half [[TMP0]], 0xH5800
 // CHECK-NEXT:    [[TMP2:%.*]] = call i8 @llvm.fptosi.sat.i8.f16(half [[TMP1]])
-// CHECK-NEXT:    store i8 [[TMP2]], i8* @sf_sat, align 1
+// CHECK-NEXT:    store i8 [[TMP2]], ptr @sf_sat, align 1
 // CHECK-NEXT:    ret void
 //
 void half_sat1(void) {
@@ -131,11 +131,11 @@ void half_sat1(void) {
 
 // CHECK-LABEL: @half_sat2(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load half, half* @h, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load half, ptr @h, align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = fpext half [[TMP0]] to float
 // CHECK-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x41E0000000000000
 // CHECK-NEXT:    [[TMP3:%.*]] = call i32 @llvm.fptosi.sat.i32.f32(float [[TMP2]])
-// CHECK-NEXT:    store i32 [[TMP3]], i32* @lf_sat, align 4
+// CHECK-NEXT:    store i32 [[TMP3]], ptr @lf_sat, align 4
 // CHECK-NEXT:    ret void
 //
 void half_sat2(void) {
@@ -144,10 +144,10 @@ void half_sat2(void) {
 
 // CHECK-LABEL: @half_sat3(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load half, half* @h, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load half, ptr @h, align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = fmul half [[TMP0]], 0xH5800
 // CHECK-NEXT:    [[TMP2:%.*]] = call i16 @llvm.fptosi.sat.i16.f16(half [[TMP1]])
-// CHECK-NEXT:    store i16 [[TMP2]], i16* @sa_sat, align 2
+// CHECK-NEXT:    store i16 [[TMP2]], ptr @sa_sat, align 2
 // CHECK-NEXT:    ret void
 //
 void half_sat3(void) {
@@ -156,11 +156,11 @@ void half_sat3(void) {
 
 // CHECK-LABEL: @half_sat4(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load half, half* @h, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load half, ptr @h, align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = fpext half [[TMP0]] to float
 // CHECK-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x41E0000000000000
 // CHECK-NEXT:    [[TMP3:%.*]] = call i64 @llvm.fptosi.sat.i64.f32(float [[TMP2]])
-// CHECK-NEXT:    store i64 [[TMP3]], i64* @la_sat, align 8
+// CHECK-NEXT:    store i64 [[TMP3]], ptr @la_sat, align 8
 // CHECK-NEXT:    ret void
 //
 void half_sat4(void) {
@@ -169,21 +169,21 @@ void half_sat4(void) {
 
 // SIGNED-LABEL: @half_sat5(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load half, half* @h, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load half, ptr @h, align 2
 // SIGNED-NEXT:    [[TMP1:%.*]] = fpext half [[TMP0]] to float
 // SIGNED-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 2.560000e+02
 // SIGNED-NEXT:    [[TMP3:%.*]] = call i16 @llvm.fptoui.sat.i16.f32(float [[TMP2]])
-// SIGNED-NEXT:    store i16 [[TMP3]], i16* @usa_sat, align 2
+// SIGNED-NEXT:    store i16 [[TMP3]], ptr @usa_sat, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @half_sat5(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load half, half* @h, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load half, ptr @h, align 2
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = fmul half [[TMP0]], 0xH5800
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.fptosi.sat.i16.f16(half [[TMP1]])
 // UNSIGNED-NEXT:    [[TMP3:%.*]] = icmp slt i16 [[TMP2]], 0
 // UNSIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP3]], i16 0, i16 [[TMP2]]
-// UNSIGNED-NEXT:    store i16 [[SATMIN]], i16* @usa_sat, align 2
+// UNSIGNED-NEXT:    store i16 [[SATMIN]], ptr @usa_sat, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void half_sat5(void) {
@@ -192,22 +192,22 @@ void half_sat5(void) {
 
 // SIGNED-LABEL: @half_sat6(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load half, half* @h, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load half, ptr @h, align 2
 // SIGNED-NEXT:    [[TMP1:%.*]] = fpext half [[TMP0]] to float
 // SIGNED-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x41F0000000000000
 // SIGNED-NEXT:    [[TMP3:%.*]] = call i64 @llvm.fptoui.sat.i64.f32(float [[TMP2]])
-// SIGNED-NEXT:    store i64 [[TMP3]], i64* @ula_sat, align 8
+// SIGNED-NEXT:    store i64 [[TMP3]], ptr @ula_sat, align 8
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @half_sat6(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load half, half* @h, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load half, ptr @h, align 2
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = fpext half [[TMP0]] to float
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x41E0000000000000
 // UNSIGNED-NEXT:    [[TMP3:%.*]] = call i64 @llvm.fptosi.sat.i64.f32(float [[TMP2]])
 // UNSIGNED-NEXT:    [[TMP4:%.*]] = icmp slt i64 [[TMP3]], 0
 // UNSIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP4]], i64 0, i64 [[TMP3]]
-// UNSIGNED-NEXT:    store i64 [[SATMIN]], i64* @ula_sat, align 8
+// UNSIGNED-NEXT:    store i64 [[SATMIN]], ptr @ula_sat, align 8
 // UNSIGNED-NEXT:    ret void
 //
 void half_sat6(void) {
@@ -217,10 +217,10 @@ void half_sat6(void) {
 
 // CHECK-LABEL: @fix_half1(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i8, i8* @sf, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i8, ptr @sf, align 1
 // CHECK-NEXT:    [[TMP1:%.*]] = sitofp i8 [[TMP0]] to half
 // CHECK-NEXT:    [[TMP2:%.*]] = fmul half [[TMP1]], 0xH2000
-// CHECK-NEXT:    store half [[TMP2]], half* @h, align 2
+// CHECK-NEXT:    store half [[TMP2]], ptr @h, align 2
 // CHECK-NEXT:    ret void
 //
 void fix_half1(void) {
@@ -229,11 +229,11 @@ void fix_half1(void) {
 
 // CHECK-LABEL: @fix_half2(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @lf, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @lf, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = sitofp i32 [[TMP0]] to float
 // CHECK-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x3E00000000000000
 // CHECK-NEXT:    [[TMP3:%.*]] = fptrunc float [[TMP2]] to half
-// CHECK-NEXT:    store half [[TMP3]], half* @h, align 2
+// CHECK-NEXT:    store half [[TMP3]], ptr @h, align 2
 // CHECK-NEXT:    ret void
 //
 void fix_half2(void) {
@@ -242,10 +242,10 @@ void fix_half2(void) {
 
 // CHECK-LABEL: @fix_half3(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = sitofp i16 [[TMP0]] to half
 // CHECK-NEXT:    [[TMP2:%.*]] = fmul half [[TMP1]], 0xH2000
-// CHECK-NEXT:    store half [[TMP2]], half* @h, align 2
+// CHECK-NEXT:    store half [[TMP2]], ptr @h, align 2
 // CHECK-NEXT:    ret void
 //
 void fix_half3(void) {
@@ -254,11 +254,11 @@ void fix_half3(void) {
 
 // CHECK-LABEL: @fix_half4(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i64, i64* @la, align 8
+// CHECK-NEXT:    [[TMP0:%.*]] = load i64, ptr @la, align 8
 // CHECK-NEXT:    [[TMP1:%.*]] = sitofp i64 [[TMP0]] to float
 // CHECK-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x3E00000000000000
 // CHECK-NEXT:    [[TMP3:%.*]] = fptrunc float [[TMP2]] to half
-// CHECK-NEXT:    store half [[TMP3]], half* @h, align 2
+// CHECK-NEXT:    store half [[TMP3]], ptr @h, align 2
 // CHECK-NEXT:    ret void
 //
 void fix_half4(void) {
@@ -267,19 +267,19 @@ void fix_half4(void) {
 
 // SIGNED-LABEL: @fix_half5(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
 // SIGNED-NEXT:    [[TMP1:%.*]] = uitofp i16 [[TMP0]] to float
 // SIGNED-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 3.906250e-03
 // SIGNED-NEXT:    [[TMP3:%.*]] = fptrunc float [[TMP2]] to half
-// SIGNED-NEXT:    store half [[TMP3]], half* @h, align 2
+// SIGNED-NEXT:    store half [[TMP3]], ptr @h, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @fix_half5(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = uitofp i16 [[TMP0]] to half
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = fmul half [[TMP1]], 0xH2000
-// UNSIGNED-NEXT:    store half [[TMP2]], half* @h, align 2
+// UNSIGNED-NEXT:    store half [[TMP2]], ptr @h, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void fix_half5(void) {
@@ -288,20 +288,20 @@ void fix_half5(void) {
 
 // SIGNED-LABEL: @fix_half6(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i64, i64* @ula, align 8
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i64, ptr @ula, align 8
 // SIGNED-NEXT:    [[TMP1:%.*]] = uitofp i64 [[TMP0]] to float
 // SIGNED-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x3DF0000000000000
 // SIGNED-NEXT:    [[TMP3:%.*]] = fptrunc float [[TMP2]] to half
-// SIGNED-NEXT:    store half [[TMP3]], half* @h, align 2
+// SIGNED-NEXT:    store half [[TMP3]], ptr @h, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @fix_half6(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i64, i64* @ula, align 8
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i64, ptr @ula, align 8
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = uitofp i64 [[TMP0]] to float
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = fmul float [[TMP1]], 0x3E00000000000000
 // UNSIGNED-NEXT:    [[TMP3:%.*]] = fptrunc float [[TMP2]] to half
-// UNSIGNED-NEXT:    store half [[TMP3]], half* @h, align 2
+// UNSIGNED-NEXT:    store half [[TMP3]], ptr @h, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void fix_half6(void) {

diff  --git a/clang/test/Frontend/fixed_point_div.c b/clang/test/Frontend/fixed_point_div.c
index ef33c1a7170dd..cf11b75233c8d 100644
--- a/clang/test/Frontend/fixed_point_div.c
+++ b/clang/test/Frontend/fixed_point_div.c
@@ -1,6 +1,6 @@
 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -triple x86_64-unknown-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,SIGNED
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -triple x86_64-unknown-linux-gnu -fpadding-on-unsigned-fixed-point -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,UNSIGNED
+// RUN: %clang_cc1 -ffixed-point -triple x86_64-unknown-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,SIGNED
+// RUN: %clang_cc1 -ffixed-point -triple x86_64-unknown-linux-gnu -fpadding-on-unsigned-fixed-point -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,UNSIGNED
 
 short _Accum sa;
 _Accum a, a2, a3, a4;
@@ -30,10 +30,10 @@ _Bool b;
 
 // CHECK-LABEL: @sdiv_sasasa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i16, i16* @sa, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i16, ptr @sa, align 2
 // CHECK-NEXT:    [[TMP2:%.*]] = call i16 @llvm.sdiv.fix.i16(i16 [[TMP0]], i16 [[TMP1]], i32 7)
-// CHECK-NEXT:    store i16 [[TMP2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[TMP2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void sdiv_sasasa(void) {
@@ -42,12 +42,12 @@ void sdiv_sasasa(void) {
 
 // CHECK-LABEL: @sdiv_asaa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i32
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.sdiv.fix.i32(i32 [[UPSCALE]], i32 [[TMP1]], i32 15)
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void sdiv_asaa(void) {
@@ -56,11 +56,11 @@ void sdiv_asaa(void) {
 
 // CHECK-LABEL: @sdiv_sasasf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* @sf, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr @sf, align 1
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i8 [[TMP1]] to i16
 // CHECK-NEXT:    [[TMP2:%.*]] = call i16 @llvm.sdiv.fix.i16(i16 [[TMP0]], i16 [[RESIZE]], i32 7)
-// CHECK-NEXT:    store i16 [[TMP2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[TMP2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void sdiv_sasasf(void) {
@@ -69,15 +69,15 @@ void sdiv_sasasf(void) {
 
 // CHECK-LABEL: @sdiv_sasaf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i16, i16* @f, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i16, ptr @f, align 2
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i24
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i24 [[RESIZE]], 8
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i16 [[TMP1]] to i24
 // CHECK-NEXT:    [[TMP2:%.*]] = call i24 @llvm.sdiv.fix.i24(i24 [[UPSCALE]], i24 [[RESIZE1]], i32 15)
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i24 [[TMP2]], 8
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i24 [[DOWNSCALE]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void sdiv_sasaf(void) {
@@ -86,12 +86,12 @@ void sdiv_sasaf(void) {
 
 // CHECK-LABEL: @sdiv_aasf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* @sf, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr @sf, align 1
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i8 [[TMP1]] to i32
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.sdiv.fix.i32(i32 [[TMP0]], i32 [[UPSCALE]], i32 15)
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void sdiv_aasf(void) {
@@ -100,15 +100,15 @@ void sdiv_aasf(void) {
 
 // CHECK-LABEL: @sdiv_aalf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @lf, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @lf, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i48
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i48 [[RESIZE]], 16
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i48
 // CHECK-NEXT:    [[TMP2:%.*]] = call i48 @llvm.sdiv.fix.i48(i48 [[UPSCALE]], i48 [[RESIZE1]], i32 31)
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i48 [[TMP2]], 16
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i48 [[DOWNSCALE]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void sdiv_aalf(void) {
@@ -117,23 +117,23 @@ void sdiv_aalf(void) {
 
 // SIGNED-LABEL: @sdiv_sasausa(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa, align 2
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i17
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i17 [[RESIZE]], 1
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i16 [[TMP1]] to i17
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i17 @llvm.sdiv.fix.i17(i17 [[UPSCALE]], i17 [[RESIZE1]], i32 8)
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i17 [[TMP2]], 1
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i17 [[DOWNSCALE]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sdiv_sasausa(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa, align 2
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.sdiv.fix.i16(i16 [[TMP0]], i16 [[TMP1]], i32 7)
-// UNSIGNED-NEXT:    store i16 [[TMP2]], i16* @sa, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP2]], ptr @sa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void sdiv_sasausa(void) {
@@ -142,25 +142,25 @@ void sdiv_sasausa(void) {
 
 // SIGNED-LABEL: @sdiv_asaua(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ua, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ua, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i33
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i33 [[RESIZE]], 9
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i33
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i33 @llvm.sdiv.fix.i33(i33 [[UPSCALE]], i33 [[RESIZE1]], i32 16)
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i33 [[TMP2]], 1
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i33 [[DOWNSCALE]] to i32
-// SIGNED-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// SIGNED-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sdiv_asaua(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ua, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ua, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i32
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i32 @llvm.sdiv.fix.i32(i32 [[UPSCALE]], i32 [[TMP1]], i32 15)
-// UNSIGNED-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// UNSIGNED-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void sdiv_asaua(void) {
@@ -169,24 +169,24 @@ void sdiv_asaua(void) {
 
 // SIGNED-LABEL: @sdiv_sasausf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i8, i8* @usf, align 1
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i8, ptr @usf, align 1
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i17
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i17 [[RESIZE]], 1
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i8 [[TMP1]] to i17
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i17 @llvm.sdiv.fix.i17(i17 [[UPSCALE]], i17 [[RESIZE1]], i32 8)
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i17 [[TMP2]], 1
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i17 [[DOWNSCALE]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sdiv_sasausf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i8, i8* @usf, align 1
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i8, ptr @usf, align 1
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i8 [[TMP1]] to i16
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.sdiv.fix.i16(i16 [[TMP0]], i16 [[RESIZE]], i32 7)
-// UNSIGNED-NEXT:    store i16 [[TMP2]], i16* @sa, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP2]], ptr @sa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void sdiv_sasausf(void) {
@@ -195,28 +195,28 @@ void sdiv_sasausf(void) {
 
 // SIGNED-LABEL: @sdiv_sasaulf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ulf, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ulf, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i41
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i41 [[RESIZE]], 25
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i41
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i41 @llvm.sdiv.fix.i41(i41 [[UPSCALE]], i41 [[RESIZE1]], i32 32)
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i41 [[TMP2]], 25
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i41 [[DOWNSCALE]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sdiv_sasaulf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ulf, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ulf, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i40
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE]], 24
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i40
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i40 @llvm.sdiv.fix.i40(i40 [[UPSCALE]], i40 [[RESIZE1]], i32 31)
 // UNSIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i40 [[TMP2]], 24
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[DOWNSCALE]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void sdiv_sasaulf(void) {
@@ -225,14 +225,14 @@ void sdiv_sasaulf(void) {
 
 // CHECK-LABEL: @sdiv_aaaaa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a2, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a2, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.sdiv.fix.i32(i32 [[TMP0]], i32 [[TMP1]], i32 15)
-// CHECK-NEXT:    [[TMP3:%.*]] = load i32, i32* @a3, align 4
+// CHECK-NEXT:    [[TMP3:%.*]] = load i32, ptr @a3, align 4
 // CHECK-NEXT:    [[TMP4:%.*]] = call i32 @llvm.sdiv.fix.i32(i32 [[TMP2]], i32 [[TMP3]], i32 15)
-// CHECK-NEXT:    [[TMP5:%.*]] = load i32, i32* @a4, align 4
+// CHECK-NEXT:    [[TMP5:%.*]] = load i32, ptr @a4, align 4
 // CHECK-NEXT:    [[TMP6:%.*]] = call i32 @llvm.sdiv.fix.i32(i32 [[TMP4]], i32 [[TMP5]], i32 15)
-// CHECK-NEXT:    store i32 [[TMP6]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP6]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void sdiv_aaaaa(void) {
@@ -242,18 +242,18 @@ void sdiv_aaaaa(void) {
 
 // SIGNED-LABEL: @udiv_usausausa(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa, align 2
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.udiv.fix.i16(i16 [[TMP0]], i16 [[TMP1]], i32 8)
-// SIGNED-NEXT:    store i16 [[TMP2]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[TMP2]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @udiv_usausausa(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa, align 2
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.sdiv.fix.i16(i16 [[TMP0]], i16 [[TMP1]], i32 7)
-// UNSIGNED-NEXT:    store i16 [[TMP2]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP2]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void udiv_usausausa(void) {
@@ -262,22 +262,22 @@ void udiv_usausausa(void) {
 
 // SIGNED-LABEL: @udiv_uausaua(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ua, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ua, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i32
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i32 @llvm.udiv.fix.i32(i32 [[UPSCALE]], i32 [[TMP1]], i32 16)
-// SIGNED-NEXT:    store i32 [[TMP2]], i32* @ua, align 4
+// SIGNED-NEXT:    store i32 [[TMP2]], ptr @ua, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @udiv_uausaua(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ua, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ua, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i32
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i32 @llvm.sdiv.fix.i32(i32 [[UPSCALE]], i32 [[TMP1]], i32 15)
-// UNSIGNED-NEXT:    store i32 [[TMP2]], i32* @ua, align 4
+// UNSIGNED-NEXT:    store i32 [[TMP2]], ptr @ua, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void udiv_uausaua(void) {
@@ -286,20 +286,20 @@ void udiv_uausaua(void) {
 
 // SIGNED-LABEL: @udiv_usausausf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i8, i8* @usf, align 1
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i8, ptr @usf, align 1
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i8 [[TMP1]] to i16
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.udiv.fix.i16(i16 [[TMP0]], i16 [[RESIZE]], i32 8)
-// SIGNED-NEXT:    store i16 [[TMP2]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[TMP2]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @udiv_usausausf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i8, i8* @usf, align 1
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i8, ptr @usf, align 1
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i8 [[TMP1]] to i16
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.sdiv.fix.i16(i16 [[TMP0]], i16 [[RESIZE]], i32 7)
-// UNSIGNED-NEXT:    store i16 [[TMP2]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP2]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void udiv_usausausf(void) {
@@ -308,28 +308,28 @@ void udiv_usausausf(void) {
 
 // SIGNED-LABEL: @udiv_usausauf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @uf, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @uf, align 2
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i24
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i24 [[RESIZE]], 8
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i16 [[TMP1]] to i24
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i24 @llvm.udiv.fix.i24(i24 [[UPSCALE]], i24 [[RESIZE1]], i32 16)
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = lshr i24 [[TMP2]], 8
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i24 [[DOWNSCALE]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @udiv_usausauf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @uf, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @uf, align 2
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i24
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i24 [[RESIZE]], 8
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i16 [[TMP1]] to i24
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i24 @llvm.sdiv.fix.i24(i24 [[UPSCALE]], i24 [[RESIZE1]], i32 15)
 // UNSIGNED-NEXT:    [[DOWNSCALE:%.*]] = lshr i24 [[TMP2]], 8
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i24 [[DOWNSCALE]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void udiv_usausauf(void) {
@@ -339,14 +339,14 @@ void udiv_usausauf(void) {
 
 // CHECK-LABEL: @int_sasai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i39
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i39
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE1]], 7
 // CHECK-NEXT:    [[TMP2:%.*]] = call i39 @llvm.sdiv.fix.i39(i39 [[RESIZE]], i39 [[UPSCALE]], i32 7)
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[TMP2]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void int_sasai(void) {
@@ -355,14 +355,14 @@ void int_sasai(void) {
 
 // CHECK-LABEL: @int_sasaui(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @ui, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @ui, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i40
 // CHECK-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i40
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE1]], 7
 // CHECK-NEXT:    [[TMP2:%.*]] = call i40 @llvm.sdiv.fix.i40(i40 [[RESIZE]], i40 [[UPSCALE]], i32 7)
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[TMP2]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void int_sasaui(void) {
@@ -371,26 +371,26 @@ void int_sasaui(void) {
 
 // SIGNED-LABEL: @int_usausai(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i40
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i40
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE1]], 8
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i40 @llvm.sdiv.fix.i40(i40 [[RESIZE]], i40 [[UPSCALE]], i32 8)
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[TMP2]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @int_usausai(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i39
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i39
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE1]], 7
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i39 @llvm.sdiv.fix.i39(i39 [[RESIZE]], i39 [[UPSCALE]], i32 7)
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[TMP2]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void int_usausai(void) {
@@ -399,26 +399,26 @@ void int_usausai(void) {
 
 // SIGNED-LABEL: @int_usausaui(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ui, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ui, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i40
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i40
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE1]], 8
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i40 @llvm.udiv.fix.i40(i40 [[RESIZE]], i40 [[UPSCALE]], i32 8)
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[TMP2]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @int_usausaui(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ui, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ui, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i39
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i39
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE1]], 7
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i39 @llvm.udiv.fix.i39(i39 [[RESIZE]], i39 [[UPSCALE]], i32 7)
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[TMP2]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void int_usausaui(void) {
@@ -427,14 +427,14 @@ void int_usausaui(void) {
 
 // CHECK-LABEL: @int_lflfui(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @lf, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @ui, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @lf, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @ui, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i64
 // CHECK-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i64
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i64 [[RESIZE1]], 31
 // CHECK-NEXT:    [[TMP2:%.*]] = call i64 @llvm.sdiv.fix.i64(i64 [[RESIZE]], i64 [[UPSCALE]], i32 31)
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i64 [[TMP2]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE2]], i32* @lf, align 4
+// CHECK-NEXT:    store i32 [[RESIZE2]], ptr @lf, align 4
 // CHECK-NEXT:    ret void
 //
 void int_lflfui(void) {
@@ -443,8 +443,8 @@ void int_lflfui(void) {
 
 // CHECK-LABEL: @int_aab(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* @b, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr @b, align 1
 // CHECK-NEXT:    [[TOBOOL:%.*]] = trunc i8 [[TMP1]] to i1
 // CHECK-NEXT:    [[CONV:%.*]] = zext i1 [[TOBOOL]] to i32
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i47
@@ -452,7 +452,7 @@ void int_lflfui(void) {
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i47 [[RESIZE1]], 15
 // CHECK-NEXT:    [[TMP2:%.*]] = call i47 @llvm.sdiv.fix.i47(i47 [[RESIZE]], i47 [[UPSCALE]], i32 15)
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i47 [[TMP2]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void int_aab(void) {
@@ -461,14 +461,14 @@ void int_aab(void) {
 
 // CHECK-LABEL: @int_aia(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @i, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @i, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i47
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i47 [[RESIZE]], 15
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i47
 // CHECK-NEXT:    [[TMP2:%.*]] = call i47 @llvm.sdiv.fix.i47(i47 [[UPSCALE]], i47 [[RESIZE1]], i32 15)
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i47 [[TMP2]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void int_aia(void) {
@@ -477,26 +477,26 @@ void int_aia(void) {
 
 // SIGNED-LABEL: @int_usauiusa(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ui, align 4
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ui, align 4
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa, align 2
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i32 [[TMP0]] to i40
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE]], 8
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i16 [[TMP1]] to i40
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i40 @llvm.udiv.fix.i40(i40 [[UPSCALE]], i40 [[RESIZE1]], i32 8)
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[TMP2]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @int_usauiusa(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ui, align 4
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ui, align 4
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa, align 2
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i32 [[TMP0]] to i39
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE]], 7
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i16 [[TMP1]] to i39
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i39 @llvm.udiv.fix.i39(i39 [[UPSCALE]], i39 [[RESIZE1]], i32 7)
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[TMP2]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void int_usauiusa(void) {
@@ -505,14 +505,14 @@ void int_usauiusa(void) {
 
 // CHECK-LABEL: @int_sauisa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @ui, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i16, i16* @sa, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @ui, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i16, ptr @sa, align 2
 // CHECK-NEXT:    [[RESIZE:%.*]] = zext i32 [[TMP0]] to i40
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE]], 7
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i16 [[TMP1]] to i40
 // CHECK-NEXT:    [[TMP2:%.*]] = call i40 @llvm.sdiv.fix.i40(i40 [[UPSCALE]], i40 [[RESIZE1]], i32 7)
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[TMP2]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void int_sauisa(void) {
@@ -522,10 +522,10 @@ void int_sauisa(void) {
 
 // CHECK-LABEL: @sat_sassasas(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i16, i16* @sa_sat, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i16, ptr @sa_sat, align 2
 // CHECK-NEXT:    [[TMP2:%.*]] = call i16 @llvm.sdiv.fix.sat.i16(i16 [[TMP0]], i16 [[TMP1]], i32 7)
-// CHECK-NEXT:    store i16 [[TMP2]], i16* @sa_sat, align 2
+// CHECK-NEXT:    store i16 [[TMP2]], ptr @sa_sat, align 2
 // CHECK-NEXT:    ret void
 //
 void sat_sassasas(void) {
@@ -534,20 +534,20 @@ void sat_sassasas(void) {
 
 // SIGNED-LABEL: @sat_usasusausas(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa_sat, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa_sat, align 2
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.udiv.fix.sat.i16(i16 [[TMP0]], i16 [[TMP1]], i32 8)
-// SIGNED-NEXT:    store i16 [[TMP2]], i16* @usa_sat, align 2
+// SIGNED-NEXT:    store i16 [[TMP2]], ptr @usa_sat, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sat_usasusausas(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa_sat, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa_sat, align 2
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.sdiv.fix.sat.i16(i16 [[TMP0]], i16 [[TMP1]], i32 7)
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = trunc i16 [[TMP2]] to i15
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i15 [[RESIZE]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE1]], i16* @usa_sat, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE1]], ptr @usa_sat, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void sat_usasusausas(void) {
@@ -556,24 +556,24 @@ void sat_usasusausas(void) {
 
 // SIGNED-LABEL: @sat_uasuausas(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa_sat, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa_sat, align 2
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP1]] to i32
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i32 @llvm.udiv.fix.sat.i32(i32 [[TMP0]], i32 [[UPSCALE]], i32 16)
-// SIGNED-NEXT:    store i32 [[TMP2]], i32* @ua_sat, align 4
+// SIGNED-NEXT:    store i32 [[TMP2]], ptr @ua_sat, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sat_uasuausas(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa_sat, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa_sat, align 2
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP1]] to i32
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i32 @llvm.sdiv.fix.sat.i32(i32 [[TMP0]], i32 [[UPSCALE]], i32 15)
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = trunc i32 [[TMP2]] to i31
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = zext i31 [[RESIZE1]] to i32
-// UNSIGNED-NEXT:    store i32 [[RESIZE2]], i32* @ua_sat, align 4
+// UNSIGNED-NEXT:    store i32 [[RESIZE2]], ptr @ua_sat, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void sat_uasuausas(void) {
@@ -582,8 +582,8 @@ void sat_uasuausas(void) {
 
 // CHECK-LABEL: @sat_sassasi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa_sat, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa_sat, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i39
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i39
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE1]], 7
@@ -593,7 +593,7 @@ void sat_uasuausas(void) {
 // CHECK-NEXT:    [[TMP4:%.*]] = icmp slt i39 [[SATMAX]], -32768
 // CHECK-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP4]], i39 -32768, i39 [[SATMAX]]
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[SATMIN]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa_sat, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa_sat, align 2
 // CHECK-NEXT:    ret void
 //
 void sat_sassasi(void) {
@@ -602,8 +602,8 @@ void sat_sassasi(void) {
 
 // CHECK-LABEL: @sat_sassasui(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa_sat, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @ui, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa_sat, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @ui, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i40
 // CHECK-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i40
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE1]], 7
@@ -613,7 +613,7 @@ void sat_sassasi(void) {
 // CHECK-NEXT:    [[TMP4:%.*]] = icmp slt i40 [[SATMAX]], -32768
 // CHECK-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP4]], i40 -32768, i40 [[SATMAX]]
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[SATMIN]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa_sat, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa_sat, align 2
 // CHECK-NEXT:    ret void
 //
 void sat_sassasui(void) {
@@ -622,20 +622,20 @@ void sat_sassasui(void) {
 
 // SIGNED-LABEL: @sat_ufsufsufs(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf_sat, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @uf_sat, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf_sat, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @uf_sat, align 2
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.udiv.fix.sat.i16(i16 [[TMP0]], i16 [[TMP1]], i32 16)
-// SIGNED-NEXT:    store i16 [[TMP2]], i16* @uf_sat, align 2
+// SIGNED-NEXT:    store i16 [[TMP2]], ptr @uf_sat, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sat_ufsufsufs(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf_sat, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @uf_sat, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf_sat, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @uf_sat, align 2
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.sdiv.fix.sat.i16(i16 [[TMP0]], i16 [[TMP1]], i32 15)
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = trunc i16 [[TMP2]] to i15
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i15 [[RESIZE]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE1]], i16* @uf_sat, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE1]], ptr @uf_sat, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void sat_ufsufsufs(void) {
@@ -644,8 +644,8 @@ void sat_ufsufsufs(void) {
 
 // SIGNED-LABEL: @sat_usasusasi(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa_sat, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa_sat, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i40
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i40
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE1]], 8
@@ -655,13 +655,13 @@ void sat_ufsufsufs(void) {
 // SIGNED-NEXT:    [[TMP4:%.*]] = icmp slt i40 [[SATMAX]], 0
 // SIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP4]], i40 0, i40 [[SATMAX]]
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[SATMIN]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa_sat, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa_sat, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sat_usasusasi(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa_sat, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa_sat, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i39
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i39
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE1]], 7
@@ -671,7 +671,7 @@ void sat_ufsufsufs(void) {
 // UNSIGNED-NEXT:    [[TMP4:%.*]] = icmp slt i39 [[SATMAX]], 0
 // UNSIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP4]], i39 0, i39 [[SATMAX]]
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[SATMIN]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa_sat, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa_sat, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void sat_usasusasi(void) {

diff  --git a/clang/test/Frontend/fixed_point_mul.c b/clang/test/Frontend/fixed_point_mul.c
index 83d586af4ec5e..fc65848710417 100644
--- a/clang/test/Frontend/fixed_point_mul.c
+++ b/clang/test/Frontend/fixed_point_mul.c
@@ -1,6 +1,6 @@
 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -triple x86_64-unknown-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,SIGNED
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -triple x86_64-unknown-linux-gnu -fpadding-on-unsigned-fixed-point -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,UNSIGNED
+// RUN: %clang_cc1 -ffixed-point -triple x86_64-unknown-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,SIGNED
+// RUN: %clang_cc1 -ffixed-point -triple x86_64-unknown-linux-gnu -fpadding-on-unsigned-fixed-point -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,UNSIGNED
 
 short _Accum sa;
 _Accum a, a2, a3, a4;
@@ -30,10 +30,10 @@ _Bool b;
 
 // CHECK-LABEL: @smul_sasasa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i16, i16* @sa, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i16, ptr @sa, align 2
 // CHECK-NEXT:    [[TMP2:%.*]] = call i16 @llvm.smul.fix.i16(i16 [[TMP0]], i16 [[TMP1]], i32 7)
-// CHECK-NEXT:    store i16 [[TMP2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[TMP2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void smul_sasasa(void) {
@@ -42,12 +42,12 @@ void smul_sasasa(void) {
 
 // CHECK-LABEL: @smul_asaa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i32
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.smul.fix.i32(i32 [[UPSCALE]], i32 [[TMP1]], i32 15)
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void smul_asaa(void) {
@@ -56,11 +56,11 @@ void smul_asaa(void) {
 
 // CHECK-LABEL: @smul_sasasf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* @sf, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr @sf, align 1
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i8 [[TMP1]] to i16
 // CHECK-NEXT:    [[TMP2:%.*]] = call i16 @llvm.smul.fix.i16(i16 [[TMP0]], i16 [[RESIZE]], i32 7)
-// CHECK-NEXT:    store i16 [[TMP2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[TMP2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void smul_sasasf(void) {
@@ -69,15 +69,15 @@ void smul_sasasf(void) {
 
 // CHECK-LABEL: @smul_sasaf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i16, i16* @f, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i16, ptr @f, align 2
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i24
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i24 [[RESIZE]], 8
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i16 [[TMP1]] to i24
 // CHECK-NEXT:    [[TMP2:%.*]] = call i24 @llvm.smul.fix.i24(i24 [[UPSCALE]], i24 [[RESIZE1]], i32 15)
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i24 [[TMP2]], 8
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i24 [[DOWNSCALE]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void smul_sasaf(void) {
@@ -86,12 +86,12 @@ void smul_sasaf(void) {
 
 // CHECK-LABEL: @smul_aasf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* @sf, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr @sf, align 1
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i8 [[TMP1]] to i32
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.smul.fix.i32(i32 [[TMP0]], i32 [[UPSCALE]], i32 15)
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void smul_aasf(void) {
@@ -100,15 +100,15 @@ void smul_aasf(void) {
 
 // CHECK-LABEL: @smul_aalf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @lf, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @lf, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i48
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i48 [[RESIZE]], 16
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i48
 // CHECK-NEXT:    [[TMP2:%.*]] = call i48 @llvm.smul.fix.i48(i48 [[UPSCALE]], i48 [[RESIZE1]], i32 31)
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i48 [[TMP2]], 16
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i48 [[DOWNSCALE]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void smul_aalf(void) {
@@ -117,23 +117,23 @@ void smul_aalf(void) {
 
 // SIGNED-LABEL: @smul_sasausa(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa, align 2
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i17
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i17 [[RESIZE]], 1
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i16 [[TMP1]] to i17
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i17 @llvm.smul.fix.i17(i17 [[UPSCALE]], i17 [[RESIZE1]], i32 8)
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i17 [[TMP2]], 1
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i17 [[DOWNSCALE]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @smul_sasausa(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa, align 2
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.smul.fix.i16(i16 [[TMP0]], i16 [[TMP1]], i32 7)
-// UNSIGNED-NEXT:    store i16 [[TMP2]], i16* @sa, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP2]], ptr @sa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void smul_sasausa(void) {
@@ -142,25 +142,25 @@ void smul_sasausa(void) {
 
 // SIGNED-LABEL: @smul_asaua(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ua, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ua, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i33
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i33 [[RESIZE]], 9
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i33
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i33 @llvm.smul.fix.i33(i33 [[UPSCALE]], i33 [[RESIZE1]], i32 16)
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i33 [[TMP2]], 1
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i33 [[DOWNSCALE]] to i32
-// SIGNED-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// SIGNED-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @smul_asaua(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ua, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ua, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i32
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i32 @llvm.smul.fix.i32(i32 [[UPSCALE]], i32 [[TMP1]], i32 15)
-// UNSIGNED-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// UNSIGNED-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void smul_asaua(void) {
@@ -169,24 +169,24 @@ void smul_asaua(void) {
 
 // SIGNED-LABEL: @smul_sasausf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i8, i8* @usf, align 1
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i8, ptr @usf, align 1
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i17
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i17 [[RESIZE]], 1
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i8 [[TMP1]] to i17
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i17 @llvm.smul.fix.i17(i17 [[UPSCALE]], i17 [[RESIZE1]], i32 8)
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i17 [[TMP2]], 1
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i17 [[DOWNSCALE]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @smul_sasausf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i8, i8* @usf, align 1
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i8, ptr @usf, align 1
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i8 [[TMP1]] to i16
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.smul.fix.i16(i16 [[TMP0]], i16 [[RESIZE]], i32 7)
-// UNSIGNED-NEXT:    store i16 [[TMP2]], i16* @sa, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP2]], ptr @sa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void smul_sasausf(void) {
@@ -195,28 +195,28 @@ void smul_sasausf(void) {
 
 // SIGNED-LABEL: @smul_sasaulf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ulf, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ulf, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i41
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i41 [[RESIZE]], 25
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i41
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i41 @llvm.smul.fix.i41(i41 [[UPSCALE]], i41 [[RESIZE1]], i32 32)
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i41 [[TMP2]], 25
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i41 [[DOWNSCALE]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @smul_sasaulf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ulf, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ulf, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i40
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE]], 24
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i40
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i40 @llvm.smul.fix.i40(i40 [[UPSCALE]], i40 [[RESIZE1]], i32 31)
 // UNSIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i40 [[TMP2]], 24
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[DOWNSCALE]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void smul_sasaulf(void) {
@@ -225,14 +225,14 @@ void smul_sasaulf(void) {
 
 // CHECK-LABEL: @smul_aaaaa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a2, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a2, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.smul.fix.i32(i32 [[TMP0]], i32 [[TMP1]], i32 15)
-// CHECK-NEXT:    [[TMP3:%.*]] = load i32, i32* @a3, align 4
+// CHECK-NEXT:    [[TMP3:%.*]] = load i32, ptr @a3, align 4
 // CHECK-NEXT:    [[TMP4:%.*]] = call i32 @llvm.smul.fix.i32(i32 [[TMP2]], i32 [[TMP3]], i32 15)
-// CHECK-NEXT:    [[TMP5:%.*]] = load i32, i32* @a4, align 4
+// CHECK-NEXT:    [[TMP5:%.*]] = load i32, ptr @a4, align 4
 // CHECK-NEXT:    [[TMP6:%.*]] = call i32 @llvm.smul.fix.i32(i32 [[TMP4]], i32 [[TMP5]], i32 15)
-// CHECK-NEXT:    store i32 [[TMP6]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP6]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void smul_aaaaa(void) {
@@ -242,18 +242,18 @@ void smul_aaaaa(void) {
 
 // SIGNED-LABEL: @umul_usausausa(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa, align 2
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.umul.fix.i16(i16 [[TMP0]], i16 [[TMP1]], i32 8)
-// SIGNED-NEXT:    store i16 [[TMP2]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[TMP2]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @umul_usausausa(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa, align 2
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.smul.fix.i16(i16 [[TMP0]], i16 [[TMP1]], i32 7)
-// UNSIGNED-NEXT:    store i16 [[TMP2]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP2]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void umul_usausausa(void) {
@@ -262,22 +262,22 @@ void umul_usausausa(void) {
 
 // SIGNED-LABEL: @umul_uausaua(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ua, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ua, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i32
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i32 @llvm.umul.fix.i32(i32 [[UPSCALE]], i32 [[TMP1]], i32 16)
-// SIGNED-NEXT:    store i32 [[TMP2]], i32* @ua, align 4
+// SIGNED-NEXT:    store i32 [[TMP2]], ptr @ua, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @umul_uausaua(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ua, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ua, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i32
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i32 @llvm.smul.fix.i32(i32 [[UPSCALE]], i32 [[TMP1]], i32 15)
-// UNSIGNED-NEXT:    store i32 [[TMP2]], i32* @ua, align 4
+// UNSIGNED-NEXT:    store i32 [[TMP2]], ptr @ua, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void umul_uausaua(void) {
@@ -286,20 +286,20 @@ void umul_uausaua(void) {
 
 // SIGNED-LABEL: @umul_usausausf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i8, i8* @usf, align 1
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i8, ptr @usf, align 1
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i8 [[TMP1]] to i16
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.umul.fix.i16(i16 [[TMP0]], i16 [[RESIZE]], i32 8)
-// SIGNED-NEXT:    store i16 [[TMP2]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[TMP2]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @umul_usausausf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i8, i8* @usf, align 1
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i8, ptr @usf, align 1
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i8 [[TMP1]] to i16
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.smul.fix.i16(i16 [[TMP0]], i16 [[RESIZE]], i32 7)
-// UNSIGNED-NEXT:    store i16 [[TMP2]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP2]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void umul_usausausf(void) {
@@ -308,28 +308,28 @@ void umul_usausausf(void) {
 
 // SIGNED-LABEL: @umul_usausauf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @uf, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @uf, align 2
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i24
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i24 [[RESIZE]], 8
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i16 [[TMP1]] to i24
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i24 @llvm.umul.fix.i24(i24 [[UPSCALE]], i24 [[RESIZE1]], i32 16)
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = lshr i24 [[TMP2]], 8
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i24 [[DOWNSCALE]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @umul_usausauf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @uf, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @uf, align 2
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i24
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i24 [[RESIZE]], 8
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i16 [[TMP1]] to i24
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i24 @llvm.smul.fix.i24(i24 [[UPSCALE]], i24 [[RESIZE1]], i32 15)
 // UNSIGNED-NEXT:    [[DOWNSCALE:%.*]] = lshr i24 [[TMP2]], 8
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i24 [[DOWNSCALE]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void umul_usausauf(void) {
@@ -339,14 +339,14 @@ void umul_usausauf(void) {
 
 // CHECK-LABEL: @int_sasai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i39
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i39
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE1]], 7
 // CHECK-NEXT:    [[TMP2:%.*]] = call i39 @llvm.smul.fix.i39(i39 [[RESIZE]], i39 [[UPSCALE]], i32 7)
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[TMP2]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void int_sasai(void) {
@@ -355,14 +355,14 @@ void int_sasai(void) {
 
 // CHECK-LABEL: @int_sasaui(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @ui, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @ui, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i40
 // CHECK-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i40
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE1]], 7
 // CHECK-NEXT:    [[TMP2:%.*]] = call i40 @llvm.smul.fix.i40(i40 [[RESIZE]], i40 [[UPSCALE]], i32 7)
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[TMP2]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void int_sasaui(void) {
@@ -371,26 +371,26 @@ void int_sasaui(void) {
 
 // SIGNED-LABEL: @int_usausai(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i40
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i40
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE1]], 8
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i40 @llvm.smul.fix.i40(i40 [[RESIZE]], i40 [[UPSCALE]], i32 8)
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[TMP2]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @int_usausai(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i39
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i39
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE1]], 7
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i39 @llvm.smul.fix.i39(i39 [[RESIZE]], i39 [[UPSCALE]], i32 7)
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[TMP2]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void int_usausai(void) {
@@ -399,26 +399,26 @@ void int_usausai(void) {
 
 // SIGNED-LABEL: @int_usausaui(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ui, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ui, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i40
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i40
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE1]], 8
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i40 @llvm.umul.fix.i40(i40 [[RESIZE]], i40 [[UPSCALE]], i32 8)
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[TMP2]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @int_usausaui(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ui, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ui, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i39
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i39
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE1]], 7
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i39 @llvm.umul.fix.i39(i39 [[RESIZE]], i39 [[UPSCALE]], i32 7)
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[TMP2]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void int_usausaui(void) {
@@ -427,14 +427,14 @@ void int_usausaui(void) {
 
 // CHECK-LABEL: @int_lflfui(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @lf, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @ui, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @lf, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @ui, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i64
 // CHECK-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i64
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i64 [[RESIZE1]], 31
 // CHECK-NEXT:    [[TMP2:%.*]] = call i64 @llvm.smul.fix.i64(i64 [[RESIZE]], i64 [[UPSCALE]], i32 31)
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i64 [[TMP2]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE2]], i32* @lf, align 4
+// CHECK-NEXT:    store i32 [[RESIZE2]], ptr @lf, align 4
 // CHECK-NEXT:    ret void
 //
 void int_lflfui(void) {
@@ -443,8 +443,8 @@ void int_lflfui(void) {
 
 // CHECK-LABEL: @int_aab(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* @b, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr @b, align 1
 // CHECK-NEXT:    [[TOBOOL:%.*]] = trunc i8 [[TMP1]] to i1
 // CHECK-NEXT:    [[CONV:%.*]] = zext i1 [[TOBOOL]] to i32
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i47
@@ -452,7 +452,7 @@ void int_lflfui(void) {
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i47 [[RESIZE1]], 15
 // CHECK-NEXT:    [[TMP2:%.*]] = call i47 @llvm.smul.fix.i47(i47 [[RESIZE]], i47 [[UPSCALE]], i32 15)
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i47 [[TMP2]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void int_aab(void) {
@@ -461,14 +461,14 @@ void int_aab(void) {
 
 // CHECK-LABEL: @int_aia(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @i, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @i, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i47
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i47 [[RESIZE]], 15
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i47
 // CHECK-NEXT:    [[TMP2:%.*]] = call i47 @llvm.smul.fix.i47(i47 [[UPSCALE]], i47 [[RESIZE1]], i32 15)
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i47 [[TMP2]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void int_aia(void) {
@@ -477,26 +477,26 @@ void int_aia(void) {
 
 // SIGNED-LABEL: @int_usauiusa(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ui, align 4
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ui, align 4
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa, align 2
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i32 [[TMP0]] to i40
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE]], 8
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i16 [[TMP1]] to i40
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i40 @llvm.umul.fix.i40(i40 [[UPSCALE]], i40 [[RESIZE1]], i32 8)
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[TMP2]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @int_usauiusa(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ui, align 4
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ui, align 4
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa, align 2
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i32 [[TMP0]] to i39
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE]], 7
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i16 [[TMP1]] to i39
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i39 @llvm.umul.fix.i39(i39 [[UPSCALE]], i39 [[RESIZE1]], i32 7)
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[TMP2]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void int_usauiusa(void) {
@@ -505,14 +505,14 @@ void int_usauiusa(void) {
 
 // CHECK-LABEL: @int_sauisa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @ui, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i16, i16* @sa, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @ui, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i16, ptr @sa, align 2
 // CHECK-NEXT:    [[RESIZE:%.*]] = zext i32 [[TMP0]] to i40
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE]], 7
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i16 [[TMP1]] to i40
 // CHECK-NEXT:    [[TMP2:%.*]] = call i40 @llvm.smul.fix.i40(i40 [[UPSCALE]], i40 [[RESIZE1]], i32 7)
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[TMP2]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void int_sauisa(void) {
@@ -522,10 +522,10 @@ void int_sauisa(void) {
 
 // CHECK-LABEL: @sat_sassasas(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i16, i16* @sa_sat, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i16, ptr @sa_sat, align 2
 // CHECK-NEXT:    [[TMP2:%.*]] = call i16 @llvm.smul.fix.sat.i16(i16 [[TMP0]], i16 [[TMP1]], i32 7)
-// CHECK-NEXT:    store i16 [[TMP2]], i16* @sa_sat, align 2
+// CHECK-NEXT:    store i16 [[TMP2]], ptr @sa_sat, align 2
 // CHECK-NEXT:    ret void
 //
 void sat_sassasas(void) {
@@ -534,20 +534,20 @@ void sat_sassasas(void) {
 
 // SIGNED-LABEL: @sat_usasusausas(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa_sat, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa_sat, align 2
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.umul.fix.sat.i16(i16 [[TMP0]], i16 [[TMP1]], i32 8)
-// SIGNED-NEXT:    store i16 [[TMP2]], i16* @usa_sat, align 2
+// SIGNED-NEXT:    store i16 [[TMP2]], ptr @usa_sat, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sat_usasusausas(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa_sat, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa_sat, align 2
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.smul.fix.sat.i16(i16 [[TMP0]], i16 [[TMP1]], i32 7)
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = trunc i16 [[TMP2]] to i15
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i15 [[RESIZE]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE1]], i16* @usa_sat, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE1]], ptr @usa_sat, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void sat_usasusausas(void) {
@@ -556,24 +556,24 @@ void sat_usasusausas(void) {
 
 // SIGNED-LABEL: @sat_uasuausas(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa_sat, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa_sat, align 2
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP1]] to i32
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i32 @llvm.umul.fix.sat.i32(i32 [[TMP0]], i32 [[UPSCALE]], i32 16)
-// SIGNED-NEXT:    store i32 [[TMP2]], i32* @ua_sat, align 4
+// SIGNED-NEXT:    store i32 [[TMP2]], ptr @ua_sat, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sat_uasuausas(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa_sat, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa_sat, align 2
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP1]] to i32
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i32 @llvm.smul.fix.sat.i32(i32 [[TMP0]], i32 [[UPSCALE]], i32 15)
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = trunc i32 [[TMP2]] to i31
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = zext i31 [[RESIZE1]] to i32
-// UNSIGNED-NEXT:    store i32 [[RESIZE2]], i32* @ua_sat, align 4
+// UNSIGNED-NEXT:    store i32 [[RESIZE2]], ptr @ua_sat, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void sat_uasuausas(void) {
@@ -582,8 +582,8 @@ void sat_uasuausas(void) {
 
 // CHECK-LABEL: @sat_sassasi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa_sat, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa_sat, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i39
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i39
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE1]], 7
@@ -593,7 +593,7 @@ void sat_uasuausas(void) {
 // CHECK-NEXT:    [[TMP4:%.*]] = icmp slt i39 [[SATMAX]], -32768
 // CHECK-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP4]], i39 -32768, i39 [[SATMAX]]
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[SATMIN]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa_sat, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa_sat, align 2
 // CHECK-NEXT:    ret void
 //
 void sat_sassasi(void) {
@@ -602,8 +602,8 @@ void sat_sassasi(void) {
 
 // CHECK-LABEL: @sat_sassasui(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa_sat, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @ui, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa_sat, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @ui, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i40
 // CHECK-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i40
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE1]], 7
@@ -613,7 +613,7 @@ void sat_sassasi(void) {
 // CHECK-NEXT:    [[TMP4:%.*]] = icmp slt i40 [[SATMAX]], -32768
 // CHECK-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP4]], i40 -32768, i40 [[SATMAX]]
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[SATMIN]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa_sat, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa_sat, align 2
 // CHECK-NEXT:    ret void
 //
 void sat_sassasui(void) {
@@ -622,20 +622,20 @@ void sat_sassasui(void) {
 
 // SIGNED-LABEL: @sat_ufsufsufs(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf_sat, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @uf_sat, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf_sat, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @uf_sat, align 2
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.umul.fix.sat.i16(i16 [[TMP0]], i16 [[TMP1]], i32 16)
-// SIGNED-NEXT:    store i16 [[TMP2]], i16* @uf_sat, align 2
+// SIGNED-NEXT:    store i16 [[TMP2]], ptr @uf_sat, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sat_ufsufsufs(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf_sat, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @uf_sat, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf_sat, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @uf_sat, align 2
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.smul.fix.sat.i16(i16 [[TMP0]], i16 [[TMP1]], i32 15)
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = trunc i16 [[TMP2]] to i15
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i15 [[RESIZE]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE1]], i16* @uf_sat, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE1]], ptr @uf_sat, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void sat_ufsufsufs(void) {
@@ -644,8 +644,8 @@ void sat_ufsufsufs(void) {
 
 // SIGNED-LABEL: @sat_usasusasi(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa_sat, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa_sat, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i40
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i40
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE1]], 8
@@ -655,13 +655,13 @@ void sat_ufsufsufs(void) {
 // SIGNED-NEXT:    [[TMP4:%.*]] = icmp slt i40 [[SATMAX]], 0
 // SIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP4]], i40 0, i40 [[SATMAX]]
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[SATMIN]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa_sat, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa_sat, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sat_usasusasi(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa_sat, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa_sat, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i39
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i39
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE1]], 7
@@ -671,7 +671,7 @@ void sat_ufsufsufs(void) {
 // UNSIGNED-NEXT:    [[TMP4:%.*]] = icmp slt i39 [[SATMAX]], 0
 // UNSIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP4]], i39 0, i39 [[SATMAX]]
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[SATMIN]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa_sat, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa_sat, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void sat_usasusasi(void) {

diff  --git a/clang/test/Frontend/fixed_point_same_fbits.c b/clang/test/Frontend/fixed_point_same_fbits.c
index 7574ee23cedc8..27762e5ca12a9 100644
--- a/clang/test/Frontend/fixed_point_same_fbits.c
+++ b/clang/test/Frontend/fixed_point_same_fbits.c
@@ -1,5 +1,5 @@
-// RUN: %clang -Xclang -no-opaque-pointers -ffixed-point -S -emit-llvm -o - %s | FileCheck %s -check-prefix=DEFAULT
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -fpadding-on-unsigned-fixed-point -S -emit-llvm -o - %s | FileCheck %s -check-prefix=SAME
+// RUN: %clang -ffixed-point -S -emit-llvm -o - %s | FileCheck %s -check-prefix=DEFAULT
+// RUN: %clang_cc1 -ffixed-point -fpadding-on-unsigned-fixed-point -S -emit-llvm -o - %s | FileCheck %s -check-prefix=SAME
 
 /* The scale for unsigned fixed point types should be the same as that of signed
  * fixed point types when -fsame-fbits is enabled. */
@@ -12,17 +12,17 @@ void func(void) {
   unsigned _Fract u_fract = 0.5ur;
   unsigned long _Fract u_long_fract = 0.5ulr;
 
-// DEFAULT: store i16 128, i16* {{.*}}, align 2
-// DEFAULT: store i32 32768, i32* {{.*}}, align 4
-// DEFAULT: store i64 2147483648, i64* {{.*}}, align 8
-// DEFAULT: store i8  -128, i8* {{.*}}, align 1
-// DEFAULT: store i16 -32768, i16* {{.*}}, align 2
-// DEFAULT: store i32 -2147483648, i32* {{.*}}, align 4
+// DEFAULT: store i16 128, ptr {{.*}}, align 2
+// DEFAULT: store i32 32768, ptr {{.*}}, align 4
+// DEFAULT: store i64 2147483648, ptr {{.*}}, align 8
+// DEFAULT: store i8  -128, ptr {{.*}}, align 1
+// DEFAULT: store i16 -32768, ptr {{.*}}, align 2
+// DEFAULT: store i32 -2147483648, ptr {{.*}}, align 4
 
-// SAME: store i16 64, i16* {{.*}}, align 2
-// SAME: store i32 16384, i32* {{.*}}, align 4
-// SAME: store i64 1073741824, i64* {{.*}}, align 8
-// SAME: store i8  64, i8* {{.*}}, align 1
-// SAME: store i16 16384, i16* {{.*}}, align 2
-// SAME: store i32 1073741824, i32* {{.*}}, align 4
+// SAME: store i16 64, ptr {{.*}}, align 2
+// SAME: store i32 16384, ptr {{.*}}, align 4
+// SAME: store i64 1073741824, ptr {{.*}}, align 8
+// SAME: store i8  64, ptr {{.*}}, align 1
+// SAME: store i16 16384, ptr {{.*}}, align 2
+// SAME: store i32 1073741824, ptr {{.*}}, align 4
 }

diff  --git a/clang/test/Frontend/fixed_point_shift.c b/clang/test/Frontend/fixed_point_shift.c
index de782d4e59ebb..cbfc9181c7627 100644
--- a/clang/test/Frontend/fixed_point_shift.c
+++ b/clang/test/Frontend/fixed_point_shift.c
@@ -1,6 +1,6 @@
 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -triple x86_64-unknown-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,SIGNED
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -triple x86_64-unknown-linux-gnu -fpadding-on-unsigned-fixed-point -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,UNSIGNED
+// RUN: %clang_cc1 -ffixed-point -triple x86_64-unknown-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,SIGNED
+// RUN: %clang_cc1 -ffixed-point -triple x86_64-unknown-linux-gnu -fpadding-on-unsigned-fixed-point -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,UNSIGNED
 
 short _Accum sa;
 _Accum a;
@@ -36,11 +36,11 @@ unsigned u;
 
 // CHECK-LABEL: @sleft_sasai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    [[TMP3:%.*]] = shl i16 [[TMP0]], [[TMP2]]
-// CHECK-NEXT:    store i16 [[TMP3]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[TMP3]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void sleft_sasai(void) {
@@ -49,10 +49,10 @@ void sleft_sasai(void) {
 
 // CHECK-LABEL: @sleft_aai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = shl i32 [[TMP0]], [[TMP1]]
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void sleft_aai(void) {
@@ -61,11 +61,11 @@ void sleft_aai(void) {
 
 // CHECK-LABEL: @sleft_lalai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i64, i64* @la, align 8
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i64, ptr @la, align 8
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = zext i32 [[TMP1]] to i64
 // CHECK-NEXT:    [[TMP3:%.*]] = shl i64 [[TMP0]], [[TMP2]]
-// CHECK-NEXT:    store i64 [[TMP3]], i64* @la, align 8
+// CHECK-NEXT:    store i64 [[TMP3]], ptr @la, align 8
 // CHECK-NEXT:    ret void
 //
 void sleft_lalai(void) {
@@ -74,11 +74,11 @@ void sleft_lalai(void) {
 
 // CHECK-LABEL: @sleft_sfsfi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i8, i8* @sf, align 1
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i8, ptr @sf, align 1
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i8
 // CHECK-NEXT:    [[TMP3:%.*]] = shl i8 [[TMP0]], [[TMP2]]
-// CHECK-NEXT:    store i8 [[TMP3]], i8* @sf, align 1
+// CHECK-NEXT:    store i8 [[TMP3]], ptr @sf, align 1
 // CHECK-NEXT:    ret void
 //
 void sleft_sfsfi(void) {
@@ -87,11 +87,11 @@ void sleft_sfsfi(void) {
 
 // CHECK-LABEL: @sleft_ffi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @f, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @f, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    [[TMP3:%.*]] = shl i16 [[TMP0]], [[TMP2]]
-// CHECK-NEXT:    store i16 [[TMP3]], i16* @f, align 2
+// CHECK-NEXT:    store i16 [[TMP3]], ptr @f, align 2
 // CHECK-NEXT:    ret void
 //
 void sleft_ffi(void) {
@@ -100,10 +100,10 @@ void sleft_ffi(void) {
 
 // CHECK-LABEL: @sleft_lflfi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @lf, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @lf, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = shl i32 [[TMP0]], [[TMP1]]
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @lf, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @lf, align 4
 // CHECK-NEXT:    ret void
 //
 void sleft_lflfi(void) {
@@ -112,10 +112,10 @@ void sleft_lflfi(void) {
 
 // CHECK-LABEL: @sleft_aau(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @u, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @u, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = shl i32 [[TMP0]], [[TMP1]]
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void sleft_aau(void) {
@@ -124,11 +124,11 @@ void sleft_aau(void) {
 
 // CHECK-LABEL: @sleft_ffu(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @f, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @u, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @f, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @u, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    [[TMP3:%.*]] = shl i16 [[TMP0]], [[TMP2]]
-// CHECK-NEXT:    store i16 [[TMP3]], i16* @f, align 2
+// CHECK-NEXT:    store i16 [[TMP3]], ptr @f, align 2
 // CHECK-NEXT:    ret void
 //
 void sleft_ffu(void) {
@@ -138,11 +138,11 @@ void sleft_ffu(void) {
 
 // CHECK-LABEL: @uleft_usausai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    [[TMP3:%.*]] = shl i16 [[TMP0]], [[TMP2]]
-// CHECK-NEXT:    store i16 [[TMP3]], i16* @usa, align 2
+// CHECK-NEXT:    store i16 [[TMP3]], ptr @usa, align 2
 // CHECK-NEXT:    ret void
 //
 void uleft_usausai(void) {
@@ -151,10 +151,10 @@ void uleft_usausai(void) {
 
 // CHECK-LABEL: @uleft_uauai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = shl i32 [[TMP0]], [[TMP1]]
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @ua, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @ua, align 4
 // CHECK-NEXT:    ret void
 //
 void uleft_uauai(void) {
@@ -163,11 +163,11 @@ void uleft_uauai(void) {
 
 // CHECK-LABEL: @uleft_ulaulai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i64, i64* @ula, align 8
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i64, ptr @ula, align 8
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = zext i32 [[TMP1]] to i64
 // CHECK-NEXT:    [[TMP3:%.*]] = shl i64 [[TMP0]], [[TMP2]]
-// CHECK-NEXT:    store i64 [[TMP3]], i64* @ula, align 8
+// CHECK-NEXT:    store i64 [[TMP3]], ptr @ula, align 8
 // CHECK-NEXT:    ret void
 //
 void uleft_ulaulai(void) {
@@ -176,11 +176,11 @@ void uleft_ulaulai(void) {
 
 // CHECK-LABEL: @uleft_usfusfi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i8, i8* @usf, align 1
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i8, ptr @usf, align 1
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i8
 // CHECK-NEXT:    [[TMP3:%.*]] = shl i8 [[TMP0]], [[TMP2]]
-// CHECK-NEXT:    store i8 [[TMP3]], i8* @usf, align 1
+// CHECK-NEXT:    store i8 [[TMP3]], ptr @usf, align 1
 // CHECK-NEXT:    ret void
 //
 void uleft_usfusfi(void) {
@@ -189,11 +189,11 @@ void uleft_usfusfi(void) {
 
 // CHECK-LABEL: @uleft_ufufi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    [[TMP3:%.*]] = shl i16 [[TMP0]], [[TMP2]]
-// CHECK-NEXT:    store i16 [[TMP3]], i16* @uf, align 2
+// CHECK-NEXT:    store i16 [[TMP3]], ptr @uf, align 2
 // CHECK-NEXT:    ret void
 //
 void uleft_ufufi(void) {
@@ -202,10 +202,10 @@ void uleft_ufufi(void) {
 
 // CHECK-LABEL: @uleft_ulfulfi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @ulf, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @ulf, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = shl i32 [[TMP0]], [[TMP1]]
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @ulf, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @ulf, align 4
 // CHECK-NEXT:    ret void
 //
 void uleft_ulfulfi(void) {
@@ -214,10 +214,10 @@ void uleft_ulfulfi(void) {
 
 // CHECK-LABEL: @uleft_uauau(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @u, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @u, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = shl i32 [[TMP0]], [[TMP1]]
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @ua, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @ua, align 4
 // CHECK-NEXT:    ret void
 //
 void uleft_uauau(void) {
@@ -226,11 +226,11 @@ void uleft_uauau(void) {
 
 // CHECK-LABEL: @uleft_ufufu(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @u, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @u, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    [[TMP3:%.*]] = shl i16 [[TMP0]], [[TMP2]]
-// CHECK-NEXT:    store i16 [[TMP3]], i16* @uf, align 2
+// CHECK-NEXT:    store i16 [[TMP3]], ptr @uf, align 2
 // CHECK-NEXT:    ret void
 //
 void uleft_ufufu(void) {
@@ -240,11 +240,11 @@ void uleft_ufufu(void) {
 
 // CHECK-LABEL: @sright_sasai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    [[TMP3:%.*]] = ashr i16 [[TMP0]], [[TMP2]]
-// CHECK-NEXT:    store i16 [[TMP3]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[TMP3]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void sright_sasai(void) {
@@ -253,10 +253,10 @@ void sright_sasai(void) {
 
 // CHECK-LABEL: @sright_aai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = ashr i32 [[TMP0]], [[TMP1]]
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void sright_aai(void) {
@@ -265,11 +265,11 @@ void sright_aai(void) {
 
 // CHECK-LABEL: @sright_lalai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i64, i64* @la, align 8
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i64, ptr @la, align 8
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = zext i32 [[TMP1]] to i64
 // CHECK-NEXT:    [[TMP3:%.*]] = ashr i64 [[TMP0]], [[TMP2]]
-// CHECK-NEXT:    store i64 [[TMP3]], i64* @la, align 8
+// CHECK-NEXT:    store i64 [[TMP3]], ptr @la, align 8
 // CHECK-NEXT:    ret void
 //
 void sright_lalai(void) {
@@ -278,11 +278,11 @@ void sright_lalai(void) {
 
 // CHECK-LABEL: @sright_sfsfi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i8, i8* @sf, align 1
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i8, ptr @sf, align 1
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i8
 // CHECK-NEXT:    [[TMP3:%.*]] = ashr i8 [[TMP0]], [[TMP2]]
-// CHECK-NEXT:    store i8 [[TMP3]], i8* @sf, align 1
+// CHECK-NEXT:    store i8 [[TMP3]], ptr @sf, align 1
 // CHECK-NEXT:    ret void
 //
 void sright_sfsfi(void) {
@@ -291,11 +291,11 @@ void sright_sfsfi(void) {
 
 // CHECK-LABEL: @sright_ffi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @f, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @f, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    [[TMP3:%.*]] = ashr i16 [[TMP0]], [[TMP2]]
-// CHECK-NEXT:    store i16 [[TMP3]], i16* @f, align 2
+// CHECK-NEXT:    store i16 [[TMP3]], ptr @f, align 2
 // CHECK-NEXT:    ret void
 //
 void sright_ffi(void) {
@@ -304,10 +304,10 @@ void sright_ffi(void) {
 
 // CHECK-LABEL: @sright_lflfi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @lf, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @lf, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = ashr i32 [[TMP0]], [[TMP1]]
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @lf, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @lf, align 4
 // CHECK-NEXT:    ret void
 //
 void sright_lflfi(void) {
@@ -316,10 +316,10 @@ void sright_lflfi(void) {
 
 // CHECK-LABEL: @sright_aau(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @u, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @u, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = ashr i32 [[TMP0]], [[TMP1]]
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void sright_aau(void) {
@@ -328,11 +328,11 @@ void sright_aau(void) {
 
 // CHECK-LABEL: @sright_ffu(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @f, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @u, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @f, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @u, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    [[TMP3:%.*]] = ashr i16 [[TMP0]], [[TMP2]]
-// CHECK-NEXT:    store i16 [[TMP3]], i16* @f, align 2
+// CHECK-NEXT:    store i16 [[TMP3]], ptr @f, align 2
 // CHECK-NEXT:    ret void
 //
 void sright_ffu(void) {
@@ -342,11 +342,11 @@ void sright_ffu(void) {
 
 // CHECK-LABEL: @uright_usausai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    [[TMP3:%.*]] = lshr i16 [[TMP0]], [[TMP2]]
-// CHECK-NEXT:    store i16 [[TMP3]], i16* @usa, align 2
+// CHECK-NEXT:    store i16 [[TMP3]], ptr @usa, align 2
 // CHECK-NEXT:    ret void
 //
 void uright_usausai(void) {
@@ -355,10 +355,10 @@ void uright_usausai(void) {
 
 // CHECK-LABEL: @uright_uauai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = lshr i32 [[TMP0]], [[TMP1]]
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @ua, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @ua, align 4
 // CHECK-NEXT:    ret void
 //
 void uright_uauai(void) {
@@ -367,11 +367,11 @@ void uright_uauai(void) {
 
 // CHECK-LABEL: @uright_ulaulai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i64, i64* @ula, align 8
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i64, ptr @ula, align 8
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = zext i32 [[TMP1]] to i64
 // CHECK-NEXT:    [[TMP3:%.*]] = lshr i64 [[TMP0]], [[TMP2]]
-// CHECK-NEXT:    store i64 [[TMP3]], i64* @ula, align 8
+// CHECK-NEXT:    store i64 [[TMP3]], ptr @ula, align 8
 // CHECK-NEXT:    ret void
 //
 void uright_ulaulai(void) {
@@ -380,11 +380,11 @@ void uright_ulaulai(void) {
 
 // CHECK-LABEL: @uright_usfusfi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i8, i8* @usf, align 1
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i8, ptr @usf, align 1
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i8
 // CHECK-NEXT:    [[TMP3:%.*]] = lshr i8 [[TMP0]], [[TMP2]]
-// CHECK-NEXT:    store i8 [[TMP3]], i8* @usf, align 1
+// CHECK-NEXT:    store i8 [[TMP3]], ptr @usf, align 1
 // CHECK-NEXT:    ret void
 //
 void uright_usfusfi(void) {
@@ -393,11 +393,11 @@ void uright_usfusfi(void) {
 
 // CHECK-LABEL: @uright_ufufi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    [[TMP3:%.*]] = lshr i16 [[TMP0]], [[TMP2]]
-// CHECK-NEXT:    store i16 [[TMP3]], i16* @uf, align 2
+// CHECK-NEXT:    store i16 [[TMP3]], ptr @uf, align 2
 // CHECK-NEXT:    ret void
 //
 void uright_ufufi(void) {
@@ -406,10 +406,10 @@ void uright_ufufi(void) {
 
 // CHECK-LABEL: @uright_ulfulfi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @ulf, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @ulf, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = lshr i32 [[TMP0]], [[TMP1]]
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @ulf, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @ulf, align 4
 // CHECK-NEXT:    ret void
 //
 void uright_ulfulfi(void) {
@@ -418,10 +418,10 @@ void uright_ulfulfi(void) {
 
 // CHECK-LABEL: @uright_uauau(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @u, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @u, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = lshr i32 [[TMP0]], [[TMP1]]
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @ua, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @ua, align 4
 // CHECK-NEXT:    ret void
 //
 void uright_uauau(void) {
@@ -430,11 +430,11 @@ void uright_uauau(void) {
 
 // CHECK-LABEL: @uright_ufufu(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @u, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @u, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    [[TMP3:%.*]] = lshr i16 [[TMP0]], [[TMP2]]
-// CHECK-NEXT:    store i16 [[TMP3]], i16* @uf, align 2
+// CHECK-NEXT:    store i16 [[TMP3]], ptr @uf, align 2
 // CHECK-NEXT:    ret void
 //
 void uright_ufufu(void) {
@@ -444,11 +444,11 @@ void uright_ufufu(void) {
 
 // CHECK-LABEL: @satleft_sassasi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa_sat, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa_sat, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    [[TMP3:%.*]] = call i16 @llvm.sshl.sat.i16(i16 [[TMP0]], i16 [[TMP2]])
-// CHECK-NEXT:    store i16 [[TMP3]], i16* @sa_sat, align 2
+// CHECK-NEXT:    store i16 [[TMP3]], ptr @sa_sat, align 2
 // CHECK-NEXT:    ret void
 //
 void satleft_sassasi(void) {
@@ -457,10 +457,10 @@ void satleft_sassasi(void) {
 
 // CHECK-LABEL: @satleft_asasi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a_sat, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a_sat, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.sshl.sat.i32(i32 [[TMP0]], i32 [[TMP1]])
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @a_sat, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @a_sat, align 4
 // CHECK-NEXT:    ret void
 //
 void satleft_asasi(void) {
@@ -469,11 +469,11 @@ void satleft_asasi(void) {
 
 // CHECK-LABEL: @satleft_sfssfsi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i8, i8* @sf_sat, align 1
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i8, ptr @sf_sat, align 1
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i8
 // CHECK-NEXT:    [[TMP3:%.*]] = call i8 @llvm.sshl.sat.i8(i8 [[TMP0]], i8 [[TMP2]])
-// CHECK-NEXT:    store i8 [[TMP3]], i8* @sf_sat, align 1
+// CHECK-NEXT:    store i8 [[TMP3]], ptr @sf_sat, align 1
 // CHECK-NEXT:    ret void
 //
 void satleft_sfssfsi(void) {
@@ -482,11 +482,11 @@ void satleft_sfssfsi(void) {
 
 // CHECK-LABEL: @satleft_fsfsi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @f_sat, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @f_sat, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // CHECK-NEXT:    [[TMP3:%.*]] = call i16 @llvm.sshl.sat.i16(i16 [[TMP0]], i16 [[TMP2]])
-// CHECK-NEXT:    store i16 [[TMP3]], i16* @f_sat, align 2
+// CHECK-NEXT:    store i16 [[TMP3]], ptr @f_sat, align 2
 // CHECK-NEXT:    ret void
 //
 void satleft_fsfsi(void) {
@@ -495,20 +495,20 @@ void satleft_fsfsi(void) {
 
 // SIGNED-LABEL: @satleft_usasusasi(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa_sat, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa_sat, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // SIGNED-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // SIGNED-NEXT:    [[TMP3:%.*]] = call i16 @llvm.ushl.sat.i16(i16 [[TMP0]], i16 [[TMP2]])
-// SIGNED-NEXT:    store i16 [[TMP3]], i16* @usa_sat, align 2
+// SIGNED-NEXT:    store i16 [[TMP3]], ptr @usa_sat, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @satleft_usasusasi(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa_sat, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa_sat, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // UNSIGNED-NEXT:    [[TMP3:%.*]] = call i16 @llvm.sshl.sat.i16(i16 [[TMP0]], i16 [[TMP2]])
-// UNSIGNED-NEXT:    store i16 [[TMP3]], i16* @usa_sat, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP3]], ptr @usa_sat, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void satleft_usasusasi(void) {
@@ -517,18 +517,18 @@ void satleft_usasusasi(void) {
 
 // SIGNED-LABEL: @satleft_uasuasi(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua_sat, align 4
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua_sat, align 4
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i32 @llvm.ushl.sat.i32(i32 [[TMP0]], i32 [[TMP1]])
-// SIGNED-NEXT:    store i32 [[TMP2]], i32* @ua_sat, align 4
+// SIGNED-NEXT:    store i32 [[TMP2]], ptr @ua_sat, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @satleft_uasuasi(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua_sat, align 4
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua_sat, align 4
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i32 @llvm.sshl.sat.i32(i32 [[TMP0]], i32 [[TMP1]])
-// UNSIGNED-NEXT:    store i32 [[TMP2]], i32* @ua_sat, align 4
+// UNSIGNED-NEXT:    store i32 [[TMP2]], ptr @ua_sat, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void satleft_uasuasi(void) {
@@ -537,20 +537,20 @@ void satleft_uasuasi(void) {
 
 // SIGNED-LABEL: @satleft_usfsusfsi(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i8, i8* @usf_sat, align 1
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i8, ptr @usf_sat, align 1
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // SIGNED-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i8
 // SIGNED-NEXT:    [[TMP3:%.*]] = call i8 @llvm.ushl.sat.i8(i8 [[TMP0]], i8 [[TMP2]])
-// SIGNED-NEXT:    store i8 [[TMP3]], i8* @usf_sat, align 1
+// SIGNED-NEXT:    store i8 [[TMP3]], ptr @usf_sat, align 1
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @satleft_usfsusfsi(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i8, i8* @usf_sat, align 1
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i8, ptr @usf_sat, align 1
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i8
 // UNSIGNED-NEXT:    [[TMP3:%.*]] = call i8 @llvm.sshl.sat.i8(i8 [[TMP0]], i8 [[TMP2]])
-// UNSIGNED-NEXT:    store i8 [[TMP3]], i8* @usf_sat, align 1
+// UNSIGNED-NEXT:    store i8 [[TMP3]], ptr @usf_sat, align 1
 // UNSIGNED-NEXT:    ret void
 //
 void satleft_usfsusfsi(void) {
@@ -559,20 +559,20 @@ void satleft_usfsusfsi(void) {
 
 // SIGNED-LABEL: @satleft_ufsufsi(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf_sat, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf_sat, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // SIGNED-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // SIGNED-NEXT:    [[TMP3:%.*]] = call i16 @llvm.ushl.sat.i16(i16 [[TMP0]], i16 [[TMP2]])
-// SIGNED-NEXT:    store i16 [[TMP3]], i16* @uf_sat, align 2
+// SIGNED-NEXT:    store i16 [[TMP3]], ptr @uf_sat, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @satleft_ufsufsi(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf_sat, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf_sat, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = trunc i32 [[TMP1]] to i16
 // UNSIGNED-NEXT:    [[TMP3:%.*]] = call i16 @llvm.sshl.sat.i16(i16 [[TMP0]], i16 [[TMP2]])
-// UNSIGNED-NEXT:    store i16 [[TMP3]], i16* @uf_sat, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP3]], ptr @uf_sat, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void satleft_ufsufsi(void) {

diff  --git a/clang/test/Frontend/fixed_point_sub.c b/clang/test/Frontend/fixed_point_sub.c
index 7098dd18a5bb3..ecadeccd8bb5d 100644
--- a/clang/test/Frontend/fixed_point_sub.c
+++ b/clang/test/Frontend/fixed_point_sub.c
@@ -1,6 +1,6 @@
 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -triple x86_64-unknown-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,SIGNED
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -triple x86_64-unknown-linux-gnu -fpadding-on-unsigned-fixed-point -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,UNSIGNED
+// RUN: %clang_cc1 -ffixed-point -triple x86_64-unknown-linux-gnu -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,SIGNED
+// RUN: %clang_cc1 -ffixed-point -triple x86_64-unknown-linux-gnu -fpadding-on-unsigned-fixed-point -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,UNSIGNED
 
 short _Accum sa;
 _Accum a, a2, a3, a4;
@@ -30,10 +30,10 @@ _Bool b;
 
 // CHECK-LABEL: @ssub_sasasa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i16, i16* @sa, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i16, ptr @sa, align 2
 // CHECK-NEXT:    [[TMP2:%.*]] = sub i16 [[TMP0]], [[TMP1]]
-// CHECK-NEXT:    store i16 [[TMP2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[TMP2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void ssub_sasasa(void) {
@@ -42,12 +42,12 @@ void ssub_sasasa(void) {
 
 // CHECK-LABEL: @ssub_asaa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i32
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // CHECK-NEXT:    [[TMP2:%.*]] = sub i32 [[UPSCALE]], [[TMP1]]
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void ssub_asaa(void) {
@@ -56,11 +56,11 @@ void ssub_asaa(void) {
 
 // CHECK-LABEL: @ssub_sasasf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* @sf, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr @sf, align 1
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i8 [[TMP1]] to i16
 // CHECK-NEXT:    [[TMP2:%.*]] = sub i16 [[TMP0]], [[RESIZE]]
-// CHECK-NEXT:    store i16 [[TMP2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[TMP2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void ssub_sasasf(void) {
@@ -69,15 +69,15 @@ void ssub_sasasf(void) {
 
 // CHECK-LABEL: @ssub_sasaf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i16, i16* @f, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i16, ptr @f, align 2
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i24
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i24 [[RESIZE]], 8
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i16 [[TMP1]] to i24
 // CHECK-NEXT:    [[TMP2:%.*]] = sub i24 [[UPSCALE]], [[RESIZE1]]
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i24 [[TMP2]], 8
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i24 [[DOWNSCALE]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void ssub_sasaf(void) {
@@ -86,12 +86,12 @@ void ssub_sasaf(void) {
 
 // CHECK-LABEL: @ssub_aasf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* @sf, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr @sf, align 1
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i8 [[TMP1]] to i32
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // CHECK-NEXT:    [[TMP2:%.*]] = sub i32 [[TMP0]], [[UPSCALE]]
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void ssub_aasf(void) {
@@ -100,15 +100,15 @@ void ssub_aasf(void) {
 
 // CHECK-LABEL: @ssub_aalf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @lf, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @lf, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i48
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i48 [[RESIZE]], 16
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i48
 // CHECK-NEXT:    [[TMP2:%.*]] = sub i48 [[UPSCALE]], [[RESIZE1]]
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = ashr i48 [[TMP2]], 16
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i48 [[DOWNSCALE]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void ssub_aalf(void) {
@@ -117,23 +117,23 @@ void ssub_aalf(void) {
 
 // SIGNED-LABEL: @ssub_sasausa(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa, align 2
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i17
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i17 [[RESIZE]], 1
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i16 [[TMP1]] to i17
 // SIGNED-NEXT:    [[TMP2:%.*]] = sub i17 [[UPSCALE]], [[RESIZE1]]
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i17 [[TMP2]], 1
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i17 [[DOWNSCALE]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @ssub_sasausa(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa, align 2
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = sub i16 [[TMP0]], [[TMP1]]
-// UNSIGNED-NEXT:    store i16 [[TMP2]], i16* @sa, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP2]], ptr @sa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void ssub_sasausa(void) {
@@ -142,25 +142,25 @@ void ssub_sasausa(void) {
 
 // SIGNED-LABEL: @ssub_asaua(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ua, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ua, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i33
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i33 [[RESIZE]], 9
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i33
 // SIGNED-NEXT:    [[TMP2:%.*]] = sub i33 [[UPSCALE]], [[RESIZE1]]
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i33 [[TMP2]], 1
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i33 [[DOWNSCALE]] to i32
-// SIGNED-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// SIGNED-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @ssub_asaua(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ua, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ua, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i32
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = sub i32 [[UPSCALE]], [[TMP1]]
-// UNSIGNED-NEXT:    store i32 [[TMP2]], i32* @a, align 4
+// UNSIGNED-NEXT:    store i32 [[TMP2]], ptr @a, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void ssub_asaua(void) {
@@ -169,24 +169,24 @@ void ssub_asaua(void) {
 
 // SIGNED-LABEL: @ssub_sasausf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i8, i8* @usf, align 1
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i8, ptr @usf, align 1
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i17
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i17 [[RESIZE]], 1
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i8 [[TMP1]] to i17
 // SIGNED-NEXT:    [[TMP2:%.*]] = sub i17 [[UPSCALE]], [[RESIZE1]]
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i17 [[TMP2]], 1
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i17 [[DOWNSCALE]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @ssub_sasausf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i8, i8* @usf, align 1
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i8, ptr @usf, align 1
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i8 [[TMP1]] to i16
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = sub i16 [[TMP0]], [[RESIZE]]
-// UNSIGNED-NEXT:    store i16 [[TMP2]], i16* @sa, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP2]], ptr @sa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void ssub_sasausf(void) {
@@ -195,28 +195,28 @@ void ssub_sasausf(void) {
 
 // SIGNED-LABEL: @ssub_sasaulf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ulf, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ulf, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i41
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i41 [[RESIZE]], 25
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i41
 // SIGNED-NEXT:    [[TMP2:%.*]] = sub i41 [[UPSCALE]], [[RESIZE1]]
 // SIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i41 [[TMP2]], 25
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i41 [[DOWNSCALE]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @ssub_sasaulf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ulf, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ulf, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i40
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE]], 24
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i40
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = sub i40 [[UPSCALE]], [[RESIZE1]]
 // UNSIGNED-NEXT:    [[DOWNSCALE:%.*]] = ashr i40 [[TMP2]], 24
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[DOWNSCALE]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void ssub_sasaulf(void) {
@@ -225,14 +225,14 @@ void ssub_sasaulf(void) {
 
 // CHECK-LABEL: @ssub_aaaaa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @a2, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @a2, align 4
 // CHECK-NEXT:    [[TMP2:%.*]] = sub i32 [[TMP0]], [[TMP1]]
-// CHECK-NEXT:    [[TMP3:%.*]] = load i32, i32* @a3, align 4
+// CHECK-NEXT:    [[TMP3:%.*]] = load i32, ptr @a3, align 4
 // CHECK-NEXT:    [[TMP4:%.*]] = sub i32 [[TMP2]], [[TMP3]]
-// CHECK-NEXT:    [[TMP5:%.*]] = load i32, i32* @a4, align 4
+// CHECK-NEXT:    [[TMP5:%.*]] = load i32, ptr @a4, align 4
 // CHECK-NEXT:    [[TMP6:%.*]] = sub i32 [[TMP4]], [[TMP5]]
-// CHECK-NEXT:    store i32 [[TMP6]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP6]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void ssub_aaaaa(void) {
@@ -242,10 +242,10 @@ void ssub_aaaaa(void) {
 
 // CHECK-LABEL: @usub_usausausa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa, align 2
 // CHECK-NEXT:    [[TMP2:%.*]] = sub i16 [[TMP0]], [[TMP1]]
-// CHECK-NEXT:    store i16 [[TMP2]], i16* @usa, align 2
+// CHECK-NEXT:    store i16 [[TMP2]], ptr @usa, align 2
 // CHECK-NEXT:    ret void
 //
 void usub_usausausa(void) {
@@ -254,12 +254,12 @@ void usub_usausausa(void) {
 
 // CHECK-LABEL: @usub_uausaua(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @ua, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @ua, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i32
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // CHECK-NEXT:    [[TMP2:%.*]] = sub i32 [[UPSCALE]], [[TMP1]]
-// CHECK-NEXT:    store i32 [[TMP2]], i32* @ua, align 4
+// CHECK-NEXT:    store i32 [[TMP2]], ptr @ua, align 4
 // CHECK-NEXT:    ret void
 //
 void usub_uausaua(void) {
@@ -268,11 +268,11 @@ void usub_uausaua(void) {
 
 // CHECK-LABEL: @usub_usausausf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* @usf, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr @usf, align 1
 // CHECK-NEXT:    [[RESIZE:%.*]] = zext i8 [[TMP1]] to i16
 // CHECK-NEXT:    [[TMP2:%.*]] = sub i16 [[TMP0]], [[RESIZE]]
-// CHECK-NEXT:    store i16 [[TMP2]], i16* @usa, align 2
+// CHECK-NEXT:    store i16 [[TMP2]], ptr @usa, align 2
 // CHECK-NEXT:    ret void
 //
 void usub_usausausf(void) {
@@ -281,15 +281,15 @@ void usub_usausausf(void) {
 
 // CHECK-LABEL: @usub_usausauf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i16, i16* @uf, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i16, ptr @uf, align 2
 // CHECK-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i24
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i24 [[RESIZE]], 8
 // CHECK-NEXT:    [[RESIZE1:%.*]] = zext i16 [[TMP1]] to i24
 // CHECK-NEXT:    [[TMP2:%.*]] = sub i24 [[UPSCALE]], [[RESIZE1]]
 // CHECK-NEXT:    [[DOWNSCALE:%.*]] = lshr i24 [[TMP2]], 8
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i24 [[DOWNSCALE]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // CHECK-NEXT:    ret void
 //
 void usub_usausauf(void) {
@@ -299,14 +299,14 @@ void usub_usausauf(void) {
 
 // CHECK-LABEL: @int_sasai(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i39
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i39
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE1]], 7
 // CHECK-NEXT:    [[TMP2:%.*]] = sub i39 [[RESIZE]], [[UPSCALE]]
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[TMP2]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void int_sasai(void) {
@@ -315,14 +315,14 @@ void int_sasai(void) {
 
 // CHECK-LABEL: @int_sasaui(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @ui, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @ui, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i40
 // CHECK-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i40
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE1]], 7
 // CHECK-NEXT:    [[TMP2:%.*]] = sub i40 [[RESIZE]], [[UPSCALE]]
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[TMP2]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa, align 2
 // CHECK-NEXT:    ret void
 //
 void int_sasaui(void) {
@@ -331,26 +331,26 @@ void int_sasaui(void) {
 
 // SIGNED-LABEL: @int_usausai(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i40
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i40
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE1]], 8
 // SIGNED-NEXT:    [[TMP2:%.*]] = sub i40 [[RESIZE]], [[UPSCALE]]
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[TMP2]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @int_usausai(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i39
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i39
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE1]], 7
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = sub i39 [[RESIZE]], [[UPSCALE]]
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[TMP2]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void int_usausai(void) {
@@ -359,26 +359,26 @@ void int_usausai(void) {
 
 // SIGNED-LABEL: @int_usausaui(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ui, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ui, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i40
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i40
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE1]], 8
 // SIGNED-NEXT:    [[TMP2:%.*]] = sub i40 [[RESIZE]], [[UPSCALE]]
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[TMP2]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @int_usausaui(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @ui, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @ui, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i39
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i39
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE1]], 7
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = sub i39 [[RESIZE]], [[UPSCALE]]
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[TMP2]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void int_usausaui(void) {
@@ -387,14 +387,14 @@ void int_usausaui(void) {
 
 // CHECK-LABEL: @int_lflfui(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @lf, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @ui, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @lf, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @ui, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i64
 // CHECK-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i64
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i64 [[RESIZE1]], 31
 // CHECK-NEXT:    [[TMP2:%.*]] = sub i64 [[RESIZE]], [[UPSCALE]]
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i64 [[TMP2]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE2]], i32* @lf, align 4
+// CHECK-NEXT:    store i32 [[RESIZE2]], ptr @lf, align 4
 // CHECK-NEXT:    ret void
 //
 void int_lflfui(void) {
@@ -403,8 +403,8 @@ void int_lflfui(void) {
 
 // CHECK-LABEL: @int_aab(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* @b, align 1
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    [[TMP1:%.*]] = load i8, ptr @b, align 1
 // CHECK-NEXT:    [[TOBOOL:%.*]] = trunc i8 [[TMP1]] to i1
 // CHECK-NEXT:    [[CONV:%.*]] = zext i1 [[TOBOOL]] to i32
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i32 [[TMP0]] to i47
@@ -412,7 +412,7 @@ void int_lflfui(void) {
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i47 [[RESIZE1]], 15
 // CHECK-NEXT:    [[TMP2:%.*]] = sub i47 [[RESIZE]], [[UPSCALE]]
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i47 [[TMP2]] to i32
-// CHECK-NEXT:    store i32 [[RESIZE2]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[RESIZE2]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void int_aab(void) {
@@ -422,10 +422,10 @@ void int_aab(void) {
 
 // CHECK-LABEL: @sat_sassasas(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i16, i16* @sa_sat, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i16, ptr @sa_sat, align 2
 // CHECK-NEXT:    [[TMP2:%.*]] = call i16 @llvm.ssub.sat.i16(i16 [[TMP0]], i16 [[TMP1]])
-// CHECK-NEXT:    store i16 [[TMP2]], i16* @sa_sat, align 2
+// CHECK-NEXT:    store i16 [[TMP2]], ptr @sa_sat, align 2
 // CHECK-NEXT:    ret void
 //
 void sat_sassasas(void) {
@@ -434,22 +434,22 @@ void sat_sassasas(void) {
 
 // SIGNED-LABEL: @sat_usasusausas(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa_sat, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa_sat, align 2
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.usub.sat.i16(i16 [[TMP0]], i16 [[TMP1]])
-// SIGNED-NEXT:    store i16 [[TMP2]], i16* @usa_sat, align 2
+// SIGNED-NEXT:    store i16 [[TMP2]], ptr @usa_sat, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sat_usasusausas(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa_sat, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa_sat, align 2
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.ssub.sat.i16(i16 [[TMP0]], i16 [[TMP1]])
 // UNSIGNED-NEXT:    [[TMP3:%.*]] = icmp slt i16 [[TMP2]], 0
 // UNSIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP3]], i16 0, i16 [[TMP2]]
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = trunc i16 [[SATMIN]] to i15
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i15 [[RESIZE]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE1]], i16* @usa_sat, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE1]], ptr @usa_sat, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void sat_usasusausas(void) {
@@ -458,18 +458,18 @@ void sat_usasusausas(void) {
 
 // SIGNED-LABEL: @sat_uasuausas(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa_sat, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa_sat, align 2
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP1]] to i32
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i32 @llvm.usub.sat.i32(i32 [[TMP0]], i32 [[UPSCALE]])
-// SIGNED-NEXT:    store i32 [[TMP2]], i32* @ua_sat, align 4
+// SIGNED-NEXT:    store i32 [[TMP2]], ptr @ua_sat, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sat_uasuausas(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @usa_sat, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @usa_sat, align 2
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP1]] to i32
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i32 [[RESIZE]], 8
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i32 @llvm.ssub.sat.i32(i32 [[TMP0]], i32 [[UPSCALE]])
@@ -477,7 +477,7 @@ void sat_usasusausas(void) {
 // UNSIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP3]], i32 0, i32 [[TMP2]]
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = trunc i32 [[SATMIN]] to i31
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = zext i31 [[RESIZE1]] to i32
-// UNSIGNED-NEXT:    store i32 [[RESIZE2]], i32* @ua_sat, align 4
+// UNSIGNED-NEXT:    store i32 [[RESIZE2]], ptr @ua_sat, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void sat_uasuausas(void) {
@@ -486,8 +486,8 @@ void sat_uasuausas(void) {
 
 // CHECK-LABEL: @sat_sassasi(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa_sat, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa_sat, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i39
 // CHECK-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i39
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE1]], 7
@@ -497,7 +497,7 @@ void sat_uasuausas(void) {
 // CHECK-NEXT:    [[TMP4:%.*]] = icmp slt i39 [[SATMAX]], -32768
 // CHECK-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP4]], i39 -32768, i39 [[SATMAX]]
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[SATMIN]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa_sat, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa_sat, align 2
 // CHECK-NEXT:    ret void
 //
 void sat_sassasi(void) {
@@ -506,8 +506,8 @@ void sat_sassasi(void) {
 
 // CHECK-LABEL: @sat_sassasui(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sa_sat, align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* @ui, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sa_sat, align 2
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr @ui, align 4
 // CHECK-NEXT:    [[RESIZE:%.*]] = sext i16 [[TMP0]] to i40
 // CHECK-NEXT:    [[RESIZE1:%.*]] = zext i32 [[TMP1]] to i40
 // CHECK-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE1]], 7
@@ -517,7 +517,7 @@ void sat_sassasi(void) {
 // CHECK-NEXT:    [[TMP4:%.*]] = icmp slt i40 [[SATMAX]], -32768
 // CHECK-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP4]], i40 -32768, i40 [[SATMAX]]
 // CHECK-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[SATMIN]] to i16
-// CHECK-NEXT:    store i16 [[RESIZE2]], i16* @sa_sat, align 2
+// CHECK-NEXT:    store i16 [[RESIZE2]], ptr @sa_sat, align 2
 // CHECK-NEXT:    ret void
 //
 void sat_sassasui(void) {
@@ -526,22 +526,22 @@ void sat_sassasui(void) {
 
 // SIGNED-LABEL: @sat_ufsufsufs(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf_sat, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @uf_sat, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf_sat, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @uf_sat, align 2
 // SIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.usub.sat.i16(i16 [[TMP0]], i16 [[TMP1]])
-// SIGNED-NEXT:    store i16 [[TMP2]], i16* @uf_sat, align 2
+// SIGNED-NEXT:    store i16 [[TMP2]], ptr @uf_sat, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sat_ufsufsufs(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf_sat, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, i16* @uf_sat, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf_sat, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i16, ptr @uf_sat, align 2
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = call i16 @llvm.ssub.sat.i16(i16 [[TMP0]], i16 [[TMP1]])
 // UNSIGNED-NEXT:    [[TMP3:%.*]] = icmp slt i16 [[TMP2]], 0
 // UNSIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP3]], i16 0, i16 [[TMP2]]
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = trunc i16 [[SATMIN]] to i15
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i15 [[RESIZE]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE1]], i16* @uf_sat, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE1]], ptr @uf_sat, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void sat_ufsufsufs(void) {
@@ -550,8 +550,8 @@ void sat_ufsufsufs(void) {
 
 // SIGNED-LABEL: @sat_usasusasi(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa_sat, align 2
-// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa_sat, align 2
+// SIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // SIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i40
 // SIGNED-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i40
 // SIGNED-NEXT:    [[UPSCALE:%.*]] = shl i40 [[RESIZE1]], 8
@@ -561,13 +561,13 @@ void sat_ufsufsufs(void) {
 // SIGNED-NEXT:    [[TMP4:%.*]] = icmp slt i40 [[SATMAX]], 0
 // SIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP4]], i40 0, i40 [[SATMAX]]
 // SIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i40 [[SATMIN]] to i16
-// SIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa_sat, align 2
+// SIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa_sat, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @sat_usasusasi(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa_sat, align 2
-// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, i32* @i, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa_sat, align 2
+// UNSIGNED-NEXT:    [[TMP1:%.*]] = load i32, ptr @i, align 4
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = zext i16 [[TMP0]] to i39
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = sext i32 [[TMP1]] to i39
 // UNSIGNED-NEXT:    [[UPSCALE:%.*]] = shl i39 [[RESIZE1]], 7
@@ -577,7 +577,7 @@ void sat_ufsufsufs(void) {
 // UNSIGNED-NEXT:    [[TMP4:%.*]] = icmp slt i39 [[SATMAX]], 0
 // UNSIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP4]], i39 0, i39 [[SATMAX]]
 // UNSIGNED-NEXT:    [[RESIZE2:%.*]] = trunc i39 [[SATMIN]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE2]], i16* @usa_sat, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE2]], ptr @usa_sat, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void sat_usasusasi(void) {

diff  --git a/clang/test/Frontend/fixed_point_to_bool.c b/clang/test/Frontend/fixed_point_to_bool.c
index e519533810728..4b9ed6cebe323 100644
--- a/clang/test/Frontend/fixed_point_to_bool.c
+++ b/clang/test/Frontend/fixed_point_to_bool.c
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -S -emit-llvm %s -o - | FileCheck %s
-// RUN: %clang_cc1 -no-opaque-pointers -ffixed-point -S -emit-llvm %s -o - -fpadding-on-unsigned-fixed-point | FileCheck %s
+// RUN: %clang_cc1 -ffixed-point -S -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -ffixed-point -S -emit-llvm %s -o - -fpadding-on-unsigned-fixed-point | FileCheck %s
 
 _Bool global_b = 1.0k;  // @global_b = {{*.}}global i8 1, align 1
 _Bool global_b2 = 0.0k; // @global_b2 = {{*.}}global i8 0, align 1
@@ -9,43 +9,43 @@ void func(void) {
   unsigned _Accum ua = 0.5uk;
   _Bool b;
 
-  // CHECK: store i8 1, i8* %b, align 1
-  // CHECK-NEXT: store i8 0, i8* %b, align 1
-  // CHECK: store i8 1, i8* %b, align 1
-  // CHECK-NEXT: store i8 0, i8* %b, align 1
+  // CHECK: store i8 1, ptr %b, align 1
+  // CHECK-NEXT: store i8 0, ptr %b, align 1
+  // CHECK: store i8 1, ptr %b, align 1
+  // CHECK-NEXT: store i8 0, ptr %b, align 1
   b = 0.5k;
   b = 0.0k;
   b = 0.5uk;
   b = 0.0uk;
 
-  // CHECK-NEXT: store i8 1, i8* %b, align 1
-  // CHECK-NEXT: store i8 0, i8* %b, align 1
-  // CHECK-NEXT: store i8 1, i8* %b, align 1
-  // CHECK-NEXT: store i8 0, i8* %b, align 1
+  // CHECK-NEXT: store i8 1, ptr %b, align 1
+  // CHECK-NEXT: store i8 0, ptr %b, align 1
+  // CHECK-NEXT: store i8 1, ptr %b, align 1
+  // CHECK-NEXT: store i8 0, ptr %b, align 1
   b = (_Bool)0.5r;
   b = (_Bool)0.0r;
   b = (_Bool)0.5ur;
   b = (_Bool)0.0ur;
 
-  // CHECK-NEXT: [[ACCUM:%[0-9a-z]+]] = load i32, i32* %a, align 4
+  // CHECK-NEXT: [[ACCUM:%[0-9a-z]+]] = load i32, ptr %a, align 4
   // CHECK-NEXT: [[NOTZERO:%[0-9a-z]+]] = icmp ne i32 [[ACCUM]], 0
   // CHECK-NEXT: [[FROMBOOL:%[0-9a-z]+]] = zext i1 [[NOTZERO]] to i8
-  // CHECK-NEXT: store i8 [[FROMBOOL]], i8* %b, align 1
+  // CHECK-NEXT: store i8 [[FROMBOOL]], ptr %b, align 1
   b = a;
 
-  // CHECK-NEXT: [[ACCUM:%[0-9a-z]+]] = load i32, i32* %ua, align 4
+  // CHECK-NEXT: [[ACCUM:%[0-9a-z]+]] = load i32, ptr %ua, align 4
   // CHECK-NEXT: [[NOTZERO:%[0-9a-z]+]] = icmp ne i32 [[ACCUM]], 0
   // CHECK-NEXT: [[FROMBOOL:%[0-9a-z]+]] = zext i1 [[NOTZERO]] to i8
-  // CHECK-NEXT: store i8 [[FROMBOOL]], i8* %b, align 1
+  // CHECK-NEXT: store i8 [[FROMBOOL]], ptr %b, align 1
   b = ua;
 
-  // CHECK-NEXT: [[ACCUM:%[0-9a-z]+]] = load i32, i32* %a, align 4
+  // CHECK-NEXT: [[ACCUM:%[0-9a-z]+]] = load i32, ptr %a, align 4
   // CHECK-NEXT: [[NOTZERO:%[0-9a-z]+]] = icmp ne i32 [[ACCUM]], 0
   // CHECK-NEXT: br i1 [[NOTZERO]], label %if.then, label %if.end
   if (a) {
   }
 
-  // CHECK:      [[ACCUM:%[0-9a-z]+]] = load i32, i32* %ua, align 4
+  // CHECK:      [[ACCUM:%[0-9a-z]+]] = load i32, ptr %ua, align 4
   // CHECK-NEXT: [[NOTZERO:%[0-9a-z]+]] = icmp ne i32 [[ACCUM]], 0
   // CHECK-NEXT: br i1 [[NOTZERO]], label %if.then{{[0-9]+}}, label %if.end{{[0-9]+}}
   if (ua) {

diff  --git a/clang/test/Frontend/fixed_point_unary.c b/clang/test/Frontend/fixed_point_unary.c
index f0ad440f367ea..a398a4be203d2 100644
--- a/clang/test/Frontend/fixed_point_unary.c
+++ b/clang/test/Frontend/fixed_point_unary.c
@@ -1,6 +1,6 @@
 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-linux-gnu -ffixed-point -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,SIGNED
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-unknown-linux-gnu -ffixed-point -fpadding-on-unsigned-fixed-point -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,UNSIGNED
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -ffixed-point -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,SIGNED
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -ffixed-point -fpadding-on-unsigned-fixed-point -S -emit-llvm %s -o - | FileCheck %s --check-prefixes=CHECK,UNSIGNED
 
 _Accum a;
 _Fract f;
@@ -20,9 +20,9 @@ int i;
 
 // CHECK-LABEL: @inc_a(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = sub i32 [[TMP0]], -32768
-// CHECK-NEXT:    store i32 [[TMP1]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP1]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void inc_a(void) {
@@ -31,9 +31,9 @@ void inc_a(void) {
 
 // CHECK-LABEL: @inc_f(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @f, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @f, align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = sub i16 [[TMP0]], -32768
-// CHECK-NEXT:    store i16 [[TMP1]], i16* @f, align 2
+// CHECK-NEXT:    store i16 [[TMP1]], ptr @f, align 2
 // CHECK-NEXT:    ret void
 //
 void inc_f(void) {
@@ -42,9 +42,9 @@ void inc_f(void) {
 
 // CHECK-LABEL: @inc_lf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @lf, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @lf, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = sub i32 [[TMP0]], -2147483648
-// CHECK-NEXT:    store i32 [[TMP1]], i32* @lf, align 4
+// CHECK-NEXT:    store i32 [[TMP1]], ptr @lf, align 4
 // CHECK-NEXT:    ret void
 //
 void inc_lf(void) {
@@ -53,16 +53,16 @@ void inc_lf(void) {
 
 // SIGNED-LABEL: @inc_ua(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
 // SIGNED-NEXT:    [[TMP1:%.*]] = add i32 [[TMP0]], 65536
-// SIGNED-NEXT:    store i32 [[TMP1]], i32* @ua, align 4
+// SIGNED-NEXT:    store i32 [[TMP1]], ptr @ua, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @inc_ua(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = add i32 [[TMP0]], 32768
-// UNSIGNED-NEXT:    store i32 [[TMP1]], i32* @ua, align 4
+// UNSIGNED-NEXT:    store i32 [[TMP1]], ptr @ua, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void inc_ua(void) {
@@ -71,16 +71,16 @@ void inc_ua(void) {
 
 // SIGNED-LABEL: @inc_usa(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
 // SIGNED-NEXT:    [[TMP1:%.*]] = add i16 [[TMP0]], 256
-// SIGNED-NEXT:    store i16 [[TMP1]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[TMP1]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @inc_usa(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = add i16 [[TMP0]], 128
-// UNSIGNED-NEXT:    store i16 [[TMP1]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP1]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void inc_usa(void) {
@@ -89,16 +89,16 @@ void inc_usa(void) {
 
 // SIGNED-LABEL: @inc_uf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
 // SIGNED-NEXT:    [[TMP1:%.*]] = add i16 [[TMP0]], poison
-// SIGNED-NEXT:    store i16 [[TMP1]], i16* @uf, align 2
+// SIGNED-NEXT:    store i16 [[TMP1]], ptr @uf, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @inc_uf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = add i16 [[TMP0]], -32768
-// UNSIGNED-NEXT:    store i16 [[TMP1]], i16* @uf, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP1]], ptr @uf, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void inc_uf(void) {
@@ -107,9 +107,9 @@ void inc_uf(void) {
 
 // CHECK-LABEL: @inc_sa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @sa, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @sa, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.ssub.sat.i32(i32 [[TMP0]], i32 -32768)
-// CHECK-NEXT:    store i32 [[TMP1]], i32* @sa, align 4
+// CHECK-NEXT:    store i32 [[TMP1]], ptr @sa, align 4
 // CHECK-NEXT:    ret void
 //
 void inc_sa(void) {
@@ -118,9 +118,9 @@ void inc_sa(void) {
 
 // CHECK-LABEL: @inc_sf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sf, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sf, align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = call i16 @llvm.ssub.sat.i16(i16 [[TMP0]], i16 -32768)
-// CHECK-NEXT:    store i16 [[TMP1]], i16* @sf, align 2
+// CHECK-NEXT:    store i16 [[TMP1]], ptr @sf, align 2
 // CHECK-NEXT:    ret void
 //
 void inc_sf(void) {
@@ -129,9 +129,9 @@ void inc_sf(void) {
 
 // CHECK-LABEL: @inc_slf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @slf, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @slf, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.ssub.sat.i32(i32 [[TMP0]], i32 -2147483648)
-// CHECK-NEXT:    store i32 [[TMP1]], i32* @slf, align 4
+// CHECK-NEXT:    store i32 [[TMP1]], ptr @slf, align 4
 // CHECK-NEXT:    ret void
 //
 void inc_slf(void) {
@@ -140,18 +140,18 @@ void inc_slf(void) {
 
 // SIGNED-LABEL: @inc_sua(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @sua, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @sua, align 4
 // SIGNED-NEXT:    [[TMP1:%.*]] = call i32 @llvm.uadd.sat.i32(i32 [[TMP0]], i32 65536)
-// SIGNED-NEXT:    store i32 [[TMP1]], i32* @sua, align 4
+// SIGNED-NEXT:    store i32 [[TMP1]], ptr @sua, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @inc_sua(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @sua, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @sua, align 4
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = call i32 @llvm.sadd.sat.i32(i32 [[TMP0]], i32 32768)
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = trunc i32 [[TMP1]] to i31
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i31 [[RESIZE]] to i32
-// UNSIGNED-NEXT:    store i32 [[RESIZE1]], i32* @sua, align 4
+// UNSIGNED-NEXT:    store i32 [[RESIZE1]], ptr @sua, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void inc_sua(void) {
@@ -160,18 +160,18 @@ void inc_sua(void) {
 
 // SIGNED-LABEL: @inc_susa(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @susa, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @susa, align 2
 // SIGNED-NEXT:    [[TMP1:%.*]] = call i16 @llvm.uadd.sat.i16(i16 [[TMP0]], i16 256)
-// SIGNED-NEXT:    store i16 [[TMP1]], i16* @susa, align 2
+// SIGNED-NEXT:    store i16 [[TMP1]], ptr @susa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @inc_susa(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @susa, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @susa, align 2
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = call i16 @llvm.sadd.sat.i16(i16 [[TMP0]], i16 128)
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = trunc i16 [[TMP1]] to i15
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i15 [[RESIZE]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE1]], i16* @susa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE1]], ptr @susa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void inc_susa(void) {
@@ -180,18 +180,18 @@ void inc_susa(void) {
 
 // SIGNED-LABEL: @inc_suf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @suf, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @suf, align 2
 // SIGNED-NEXT:    [[TMP1:%.*]] = call i16 @llvm.uadd.sat.i16(i16 [[TMP0]], i16 -1)
-// SIGNED-NEXT:    store i16 [[TMP1]], i16* @suf, align 2
+// SIGNED-NEXT:    store i16 [[TMP1]], ptr @suf, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @inc_suf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @suf, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @suf, align 2
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = call i16 @llvm.sadd.sat.i16(i16 [[TMP0]], i16 32767)
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = trunc i16 [[TMP1]] to i15
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i15 [[RESIZE]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE1]], i16* @suf, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE1]], ptr @suf, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void inc_suf(void) {
@@ -201,9 +201,9 @@ void inc_suf(void) {
 
 // CHECK-LABEL: @dec_a(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = add i32 [[TMP0]], -32768
-// CHECK-NEXT:    store i32 [[TMP1]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP1]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void dec_a(void) {
@@ -212,9 +212,9 @@ void dec_a(void) {
 
 // CHECK-LABEL: @dec_f(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @f, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @f, align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = add i16 [[TMP0]], -32768
-// CHECK-NEXT:    store i16 [[TMP1]], i16* @f, align 2
+// CHECK-NEXT:    store i16 [[TMP1]], ptr @f, align 2
 // CHECK-NEXT:    ret void
 //
 void dec_f(void) {
@@ -223,9 +223,9 @@ void dec_f(void) {
 
 // CHECK-LABEL: @dec_lf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @lf, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @lf, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = add i32 [[TMP0]], -2147483648
-// CHECK-NEXT:    store i32 [[TMP1]], i32* @lf, align 4
+// CHECK-NEXT:    store i32 [[TMP1]], ptr @lf, align 4
 // CHECK-NEXT:    ret void
 //
 void dec_lf(void) {
@@ -234,16 +234,16 @@ void dec_lf(void) {
 
 // SIGNED-LABEL: @dec_ua(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
 // SIGNED-NEXT:    [[TMP1:%.*]] = sub i32 [[TMP0]], 65536
-// SIGNED-NEXT:    store i32 [[TMP1]], i32* @ua, align 4
+// SIGNED-NEXT:    store i32 [[TMP1]], ptr @ua, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @dec_ua(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @ua, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @ua, align 4
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = sub i32 [[TMP0]], 32768
-// UNSIGNED-NEXT:    store i32 [[TMP1]], i32* @ua, align 4
+// UNSIGNED-NEXT:    store i32 [[TMP1]], ptr @ua, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void dec_ua(void) {
@@ -252,16 +252,16 @@ void dec_ua(void) {
 
 // SIGNED-LABEL: @dec_usa(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
 // SIGNED-NEXT:    [[TMP1:%.*]] = sub i16 [[TMP0]], 256
-// SIGNED-NEXT:    store i16 [[TMP1]], i16* @usa, align 2
+// SIGNED-NEXT:    store i16 [[TMP1]], ptr @usa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @dec_usa(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = sub i16 [[TMP0]], 128
-// UNSIGNED-NEXT:    store i16 [[TMP1]], i16* @usa, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP1]], ptr @usa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void dec_usa(void) {
@@ -270,16 +270,16 @@ void dec_usa(void) {
 
 // SIGNED-LABEL: @dec_uf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
 // SIGNED-NEXT:    [[TMP1:%.*]] = sub i16 [[TMP0]], poison
-// SIGNED-NEXT:    store i16 [[TMP1]], i16* @uf, align 2
+// SIGNED-NEXT:    store i16 [[TMP1]], ptr @uf, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @dec_uf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = sub i16 [[TMP0]], -32768
-// UNSIGNED-NEXT:    store i16 [[TMP1]], i16* @uf, align 2
+// UNSIGNED-NEXT:    store i16 [[TMP1]], ptr @uf, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void dec_uf(void) {
@@ -288,9 +288,9 @@ void dec_uf(void) {
 
 // CHECK-LABEL: @dec_sa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @sa, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @sa, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.sadd.sat.i32(i32 [[TMP0]], i32 -32768)
-// CHECK-NEXT:    store i32 [[TMP1]], i32* @sa, align 4
+// CHECK-NEXT:    store i32 [[TMP1]], ptr @sa, align 4
 // CHECK-NEXT:    ret void
 //
 void dec_sa(void) {
@@ -299,9 +299,9 @@ void dec_sa(void) {
 
 // CHECK-LABEL: @dec_sf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sf, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sf, align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = call i16 @llvm.sadd.sat.i16(i16 [[TMP0]], i16 -32768)
-// CHECK-NEXT:    store i16 [[TMP1]], i16* @sf, align 2
+// CHECK-NEXT:    store i16 [[TMP1]], ptr @sf, align 2
 // CHECK-NEXT:    ret void
 //
 void dec_sf(void) {
@@ -310,9 +310,9 @@ void dec_sf(void) {
 
 // CHECK-LABEL: @dec_slf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @slf, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @slf, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.sadd.sat.i32(i32 [[TMP0]], i32 -2147483648)
-// CHECK-NEXT:    store i32 [[TMP1]], i32* @slf, align 4
+// CHECK-NEXT:    store i32 [[TMP1]], ptr @slf, align 4
 // CHECK-NEXT:    ret void
 //
 void dec_slf(void) {
@@ -321,20 +321,20 @@ void dec_slf(void) {
 
 // SIGNED-LABEL: @dec_sua(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @sua, align 4
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @sua, align 4
 // SIGNED-NEXT:    [[TMP1:%.*]] = call i32 @llvm.usub.sat.i32(i32 [[TMP0]], i32 65536)
-// SIGNED-NEXT:    store i32 [[TMP1]], i32* @sua, align 4
+// SIGNED-NEXT:    store i32 [[TMP1]], ptr @sua, align 4
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @dec_sua(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, i32* @sua, align 4
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i32, ptr @sua, align 4
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = call i32 @llvm.ssub.sat.i32(i32 [[TMP0]], i32 32768)
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = icmp slt i32 [[TMP1]], 0
 // UNSIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP2]], i32 0, i32 [[TMP1]]
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = trunc i32 [[SATMIN]] to i31
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i31 [[RESIZE]] to i32
-// UNSIGNED-NEXT:    store i32 [[RESIZE1]], i32* @sua, align 4
+// UNSIGNED-NEXT:    store i32 [[RESIZE1]], ptr @sua, align 4
 // UNSIGNED-NEXT:    ret void
 //
 void dec_sua(void) {
@@ -343,20 +343,20 @@ void dec_sua(void) {
 
 // SIGNED-LABEL: @dec_susa(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @susa, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @susa, align 2
 // SIGNED-NEXT:    [[TMP1:%.*]] = call i16 @llvm.usub.sat.i16(i16 [[TMP0]], i16 256)
-// SIGNED-NEXT:    store i16 [[TMP1]], i16* @susa, align 2
+// SIGNED-NEXT:    store i16 [[TMP1]], ptr @susa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @dec_susa(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @susa, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @susa, align 2
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = call i16 @llvm.ssub.sat.i16(i16 [[TMP0]], i16 128)
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = icmp slt i16 [[TMP1]], 0
 // UNSIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP2]], i16 0, i16 [[TMP1]]
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = trunc i16 [[SATMIN]] to i15
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i15 [[RESIZE]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE1]], i16* @susa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE1]], ptr @susa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void dec_susa(void) {
@@ -365,20 +365,20 @@ void dec_susa(void) {
 
 // SIGNED-LABEL: @dec_suf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @suf, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @suf, align 2
 // SIGNED-NEXT:    [[TMP1:%.*]] = call i16 @llvm.usub.sat.i16(i16 [[TMP0]], i16 -1)
-// SIGNED-NEXT:    store i16 [[TMP1]], i16* @suf, align 2
+// SIGNED-NEXT:    store i16 [[TMP1]], ptr @suf, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @dec_suf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @suf, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @suf, align 2
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = call i16 @llvm.ssub.sat.i16(i16 [[TMP0]], i16 32767)
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = icmp slt i16 [[TMP1]], 0
 // UNSIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP2]], i16 0, i16 [[TMP1]]
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = trunc i16 [[SATMIN]] to i15
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i15 [[RESIZE]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE1]], i16* @suf, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE1]], ptr @suf, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void dec_suf(void) {
@@ -388,9 +388,9 @@ void dec_suf(void) {
 
 // CHECK-LABEL: @neg_a(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = sub i32 0, [[TMP0]]
-// CHECK-NEXT:    store i32 [[TMP1]], i32* @a, align 4
+// CHECK-NEXT:    store i32 [[TMP1]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void neg_a(void) {
@@ -399,9 +399,9 @@ void neg_a(void) {
 
 // CHECK-LABEL: @neg_f(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @f, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @f, align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = sub i16 0, [[TMP0]]
-// CHECK-NEXT:    store i16 [[TMP1]], i16* @f, align 2
+// CHECK-NEXT:    store i16 [[TMP1]], ptr @f, align 2
 // CHECK-NEXT:    ret void
 //
 void neg_f(void) {
@@ -410,9 +410,9 @@ void neg_f(void) {
 
 // CHECK-LABEL: @neg_usa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @usa, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @usa, align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = sub i16 0, [[TMP0]]
-// CHECK-NEXT:    store i16 [[TMP1]], i16* @usa, align 2
+// CHECK-NEXT:    store i16 [[TMP1]], ptr @usa, align 2
 // CHECK-NEXT:    ret void
 //
 void neg_usa(void) {
@@ -421,9 +421,9 @@ void neg_usa(void) {
 
 // CHECK-LABEL: @neg_uf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = sub i16 0, [[TMP0]]
-// CHECK-NEXT:    store i16 [[TMP1]], i16* @uf, align 2
+// CHECK-NEXT:    store i16 [[TMP1]], ptr @uf, align 2
 // CHECK-NEXT:    ret void
 //
 void neg_uf(void) {
@@ -432,9 +432,9 @@ void neg_uf(void) {
 
 // CHECK-LABEL: @neg_sa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @sa, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @sa, align 4
 // CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.ssub.sat.i32(i32 0, i32 [[TMP0]])
-// CHECK-NEXT:    store i32 [[TMP1]], i32* @sa, align 4
+// CHECK-NEXT:    store i32 [[TMP1]], ptr @sa, align 4
 // CHECK-NEXT:    ret void
 //
 void neg_sa(void) {
@@ -443,9 +443,9 @@ void neg_sa(void) {
 
 // CHECK-LABEL: @neg_sf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @sf, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @sf, align 2
 // CHECK-NEXT:    [[TMP1:%.*]] = call i16 @llvm.ssub.sat.i16(i16 0, i16 [[TMP0]])
-// CHECK-NEXT:    store i16 [[TMP1]], i16* @sf, align 2
+// CHECK-NEXT:    store i16 [[TMP1]], ptr @sf, align 2
 // CHECK-NEXT:    ret void
 //
 void neg_sf(void) {
@@ -454,20 +454,20 @@ void neg_sf(void) {
 
 // SIGNED-LABEL: @neg_susa(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @susa, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @susa, align 2
 // SIGNED-NEXT:    [[TMP1:%.*]] = call i16 @llvm.usub.sat.i16(i16 0, i16 [[TMP0]])
-// SIGNED-NEXT:    store i16 [[TMP1]], i16* @susa, align 2
+// SIGNED-NEXT:    store i16 [[TMP1]], ptr @susa, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @neg_susa(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @susa, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @susa, align 2
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = call i16 @llvm.ssub.sat.i16(i16 0, i16 [[TMP0]])
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = icmp slt i16 [[TMP1]], 0
 // UNSIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP2]], i16 0, i16 [[TMP1]]
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = trunc i16 [[SATMIN]] to i15
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i15 [[RESIZE]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE1]], i16* @susa, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE1]], ptr @susa, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void neg_susa(void) {
@@ -476,20 +476,20 @@ void neg_susa(void) {
 
 // SIGNED-LABEL: @neg_suf(
 // SIGNED-NEXT:  entry:
-// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @suf, align 2
+// SIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @suf, align 2
 // SIGNED-NEXT:    [[TMP1:%.*]] = call i16 @llvm.usub.sat.i16(i16 0, i16 [[TMP0]])
-// SIGNED-NEXT:    store i16 [[TMP1]], i16* @suf, align 2
+// SIGNED-NEXT:    store i16 [[TMP1]], ptr @suf, align 2
 // SIGNED-NEXT:    ret void
 //
 // UNSIGNED-LABEL: @neg_suf(
 // UNSIGNED-NEXT:  entry:
-// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, i16* @suf, align 2
+// UNSIGNED-NEXT:    [[TMP0:%.*]] = load i16, ptr @suf, align 2
 // UNSIGNED-NEXT:    [[TMP1:%.*]] = call i16 @llvm.ssub.sat.i16(i16 0, i16 [[TMP0]])
 // UNSIGNED-NEXT:    [[TMP2:%.*]] = icmp slt i16 [[TMP1]], 0
 // UNSIGNED-NEXT:    [[SATMIN:%.*]] = select i1 [[TMP2]], i16 0, i16 [[TMP1]]
 // UNSIGNED-NEXT:    [[RESIZE:%.*]] = trunc i16 [[SATMIN]] to i15
 // UNSIGNED-NEXT:    [[RESIZE1:%.*]] = zext i15 [[RESIZE]] to i16
-// UNSIGNED-NEXT:    store i16 [[RESIZE1]], i16* @suf, align 2
+// UNSIGNED-NEXT:    store i16 [[RESIZE1]], ptr @suf, align 2
 // UNSIGNED-NEXT:    ret void
 //
 void neg_suf(void) {
@@ -499,8 +499,8 @@ void neg_suf(void) {
 
 // CHECK-LABEL: @plus_a(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
-// CHECK-NEXT:    store i32 [[TMP0]], i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
+// CHECK-NEXT:    store i32 [[TMP0]], ptr @a, align 4
 // CHECK-NEXT:    ret void
 //
 void plus_a(void) {
@@ -509,8 +509,8 @@ void plus_a(void) {
 
 // CHECK-LABEL: @plus_uf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
-// CHECK-NEXT:    store i16 [[TMP0]], i16* @uf, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
+// CHECK-NEXT:    store i16 [[TMP0]], ptr @uf, align 2
 // CHECK-NEXT:    ret void
 //
 void plus_uf(void) {
@@ -519,8 +519,8 @@ void plus_uf(void) {
 
 // CHECK-LABEL: @plus_sa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @sa, align 4
-// CHECK-NEXT:    store i32 [[TMP0]], i32* @sa, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @sa, align 4
+// CHECK-NEXT:    store i32 [[TMP0]], ptr @sa, align 4
 // CHECK-NEXT:    ret void
 //
 void plus_sa(void) {
@@ -530,11 +530,11 @@ void plus_sa(void) {
 
 // CHECK-LABEL: @not_a(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* @a, align 4
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr @a, align 4
 // CHECK-NEXT:    [[TOBOOL:%.*]] = icmp ne i32 [[TMP0]], 0
 // CHECK-NEXT:    [[LNOT:%.*]] = xor i1 [[TOBOOL]], true
 // CHECK-NEXT:    [[LNOT_EXT:%.*]] = zext i1 [[LNOT]] to i32
-// CHECK-NEXT:    store i32 [[LNOT_EXT]], i32* @i, align 4
+// CHECK-NEXT:    store i32 [[LNOT_EXT]], ptr @i, align 4
 // CHECK-NEXT:    ret void
 //
 void not_a(void) {
@@ -543,11 +543,11 @@ void not_a(void) {
 
 // CHECK-LABEL: @not_uf(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @uf, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @uf, align 2
 // CHECK-NEXT:    [[TOBOOL:%.*]] = icmp ne i16 [[TMP0]], 0
 // CHECK-NEXT:    [[LNOT:%.*]] = xor i1 [[TOBOOL]], true
 // CHECK-NEXT:    [[LNOT_EXT:%.*]] = zext i1 [[LNOT]] to i32
-// CHECK-NEXT:    store i32 [[LNOT_EXT]], i32* @i, align 4
+// CHECK-NEXT:    store i32 [[LNOT_EXT]], ptr @i, align 4
 // CHECK-NEXT:    ret void
 //
 void not_uf(void) {
@@ -556,11 +556,11 @@ void not_uf(void) {
 
 // CHECK-LABEL: @not_susa(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* @susa, align 2
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr @susa, align 2
 // CHECK-NEXT:    [[TOBOOL:%.*]] = icmp ne i16 [[TMP0]], 0
 // CHECK-NEXT:    [[LNOT:%.*]] = xor i1 [[TOBOOL]], true
 // CHECK-NEXT:    [[LNOT_EXT:%.*]] = zext i1 [[LNOT]] to i32
-// CHECK-NEXT:    store i32 [[LNOT_EXT]], i32* @i, align 4
+// CHECK-NEXT:    store i32 [[LNOT_EXT]], ptr @i, align 4
 // CHECK-NEXT:    ret void
 //
 void not_susa(void) {

diff  --git a/clang/test/Headers/stdarg.cpp b/clang/test/Headers/stdarg.cpp
index dcbe173a4f41a..20bf17caf15f7 100644
--- a/clang/test/Headers/stdarg.cpp
+++ b/clang/test/Headers/stdarg.cpp
@@ -1,32 +1,32 @@
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -std=c99 -x c %s -triple aarch64-linux -o - | FileCheck %s --check-prefix=AARCH64-C
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -std=c++17 -x c++ %s -triple aarch64-linux -o - | FileCheck %s --check-prefix=AARCH64-CXX
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -std=c99 -x c %s -triple x86_64-linux -o - | FileCheck %s --check-prefix=X86_64-C
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -std=c++17 -x c++ %s -triple x86_64-linux -o - | FileCheck %s --check-prefix=X86_64-CXX
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -std=c99 -x c %s -triple ppc64-linux -o - | FileCheck %s --check-prefix=PPC64-C
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -std=c++17 -x c++ %s -triple ppc64-linux -o - | FileCheck %s --check-prefix=PPC64-CXX
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -std=c99 -x c %s -triple armv7-apple-darwin9 -target-abi aapcs -o - | FileCheck %s --check-prefix=AAPCS-C
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -std=c++17 -x c++ %s -triple armv7-apple-darwin9 -target-abi aapcs -o - | FileCheck %s --check-prefix=AAPCS-CXX
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -std=c99 -x c %s -triple s390x-linux -o - | FileCheck %s --check-prefix=SYSTEMZ-C
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -std=c++17 -x c++ %s -triple s390x-linux -o - | FileCheck %s --check-prefix=SYSTEMZ-CXX
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -std=c99 -x c %s -triple i686-linux -o - | FileCheck %s --check-prefix=CHARPTR-C
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -std=c++17 -x c++ %s -triple i686-linux -o - | FileCheck %s --check-prefix=CHARPTR-CXX
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -std=c99 -x c %s -triple xcore -o - | FileCheck %s --check-prefix=VOIDPTR-C
-// RUN: %clang_cc1 -no-opaque-pointers -emit-llvm -std=c++17 -x c++ %s -triple xcore -o - | FileCheck %s --check-prefix=VOIDPTR-CXX
+// RUN: %clang_cc1 -emit-llvm -std=c99 -x c %s -triple aarch64-linux -o - | FileCheck %s --check-prefix=AARCH64-C
+// RUN: %clang_cc1 -emit-llvm -std=c++17 -x c++ %s -triple aarch64-linux -o - | FileCheck %s --check-prefix=AARCH64-CXX
+// RUN: %clang_cc1 -emit-llvm -std=c99 -x c %s -triple x86_64-linux -o - | FileCheck %s --check-prefix=X86_64-C
+// RUN: %clang_cc1 -emit-llvm -std=c++17 -x c++ %s -triple x86_64-linux -o - | FileCheck %s --check-prefix=X86_64-CXX
+// RUN: %clang_cc1 -emit-llvm -std=c99 -x c %s -triple ppc64-linux -o - | FileCheck %s --check-prefix=PPC64-C
+// RUN: %clang_cc1 -emit-llvm -std=c++17 -x c++ %s -triple ppc64-linux -o - | FileCheck %s --check-prefix=PPC64-CXX
+// RUN: %clang_cc1 -emit-llvm -std=c99 -x c %s -triple armv7-apple-darwin9 -target-abi aapcs -o - | FileCheck %s --check-prefix=AAPCS-C
+// RUN: %clang_cc1 -emit-llvm -std=c++17 -x c++ %s -triple armv7-apple-darwin9 -target-abi aapcs -o - | FileCheck %s --check-prefix=AAPCS-CXX
+// RUN: %clang_cc1 -emit-llvm -std=c99 -x c %s -triple s390x-linux -o - | FileCheck %s --check-prefix=SYSTEMZ-C
+// RUN: %clang_cc1 -emit-llvm -std=c++17 -x c++ %s -triple s390x-linux -o - | FileCheck %s --check-prefix=SYSTEMZ-CXX
+// RUN: %clang_cc1 -emit-llvm -std=c99 -x c %s -triple i686-linux -o - | FileCheck %s --check-prefix=CHARPTR-C
+// RUN: %clang_cc1 -emit-llvm -std=c++17 -x c++ %s -triple i686-linux -o - | FileCheck %s --check-prefix=CHARPTR-CXX
+// RUN: %clang_cc1 -emit-llvm -std=c99 -x c %s -triple xcore -o - | FileCheck %s --check-prefix=VOIDPTR-C
+// RUN: %clang_cc1 -emit-llvm -std=c++17 -x c++ %s -triple xcore -o - | FileCheck %s --check-prefix=VOIDPTR-CXX
 
 #include <stdarg.h>
 
-// AARCH64-C: define {{.*}} @f(i32 noundef %n, %struct.__va_list* noundef %list)
-// AARCH64-CXX: define {{.*}} @_Z1fiSt9__va_list(i32 noundef %n, %"struct.std::__va_list"* noundef %list)
-// X86_64-C: define {{.*}} @f(i32 noundef %n, %struct.__va_list_tag* noundef %list)
-// X86_64-CXX: define {{.*}} @_Z1fiP13__va_list_tag(i32 noundef %n, %struct.__va_list_tag* noundef %list)
-// PPC64-C: define {{.*}} @f(i32 noundef signext %n, i8* noundef %list)
-// PPC64-CXX: define {{.*}} @_Z1fiPc(i32 noundef signext %n, i8* noundef %list)
+// AARCH64-C: define {{.*}} @f(i32 noundef %n, ptr noundef %list)
+// AARCH64-CXX: define {{.*}} @_Z1fiSt9__va_list(i32 noundef %n, ptr noundef %list)
+// X86_64-C: define {{.*}} @f(i32 noundef %n, ptr noundef %list)
+// X86_64-CXX: define {{.*}} @_Z1fiP13__va_list_tag(i32 noundef %n, ptr noundef %list)
+// PPC64-C: define {{.*}} @f(i32 noundef signext %n, ptr noundef %list)
+// PPC64-CXX: define {{.*}} @_Z1fiPc(i32 noundef signext %n, ptr noundef %list)
 // AAPCS-C: define {{.*}} @f(i32 noundef %n, [1 x i32] %list.coerce)
 // AAPCS-CXX: define {{.*}} @_Z1fiSt9__va_list(i32 noundef %n, [1 x i32] %list.coerce)
-// SYSTEMZ-C: define {{.*}} @f(i32 noundef signext %n, %struct.__va_list_tag* noundef %list)
-// SYSTEMZ-CXX: define {{.*}} @_Z1fiP13__va_list_tag(i32 noundef signext %n, %struct.__va_list_tag* noundef %list)
-// CHARPTR-C: define {{.*}} @f(i32 noundef %n, i8* noundef %list)
-// CHARPTR-CXX: define {{.*}} @_Z1fiPc(i32 noundef %n, i8* noundef %list)
-// VOIDPTR-C: define {{.*}} @f(i32 noundef %n, i8* noundef %list)
-// VOIDPTR-CXX: define {{.*}} @_Z1fiPv(i32 noundef %n, i8* noundef %list)
+// SYSTEMZ-C: define {{.*}} @f(i32 noundef signext %n, ptr noundef %list)
+// SYSTEMZ-CXX: define {{.*}} @_Z1fiP13__va_list_tag(i32 noundef signext %n, ptr noundef %list)
+// CHARPTR-C: define {{.*}} @f(i32 noundef %n, ptr noundef %list)
+// CHARPTR-CXX: define {{.*}} @_Z1fiPc(i32 noundef %n, ptr noundef %list)
+// VOIDPTR-C: define {{.*}} @f(i32 noundef %n, ptr noundef %list)
+// VOIDPTR-CXX: define {{.*}} @_Z1fiPv(i32 noundef %n, ptr noundef %list)
 void f(int n, va_list list) {}

diff  --git a/clang/test/Headers/wasm.c b/clang/test/Headers/wasm.c
index 53acbf4de4c96..8913a273aaeb1 100644
--- a/clang/test/Headers/wasm.c
+++ b/clang/test/Headers/wasm.c
@@ -3,14 +3,13 @@
 
 // FIXME: This should not be using -O2 and implicitly testing the entire IR opt pipeline.
 
-// RUN: %clang -Xclang -no-opaque-pointers %s -O2 -emit-llvm -S -o - -target wasm32-unknown-unknown -msimd128 -Wall -Weverything -Wno-missing-prototypes -fno-lax-vector-conversions -Werror | FileCheck %s
+// RUN: %clang %s -O2 -emit-llvm -S -o - -target wasm32-unknown-unknown -msimd128 -Wall -Weverything -Wno-missing-prototypes -fno-lax-vector-conversions -Werror | FileCheck %s
 
 #include <wasm_simd128.h>
 
 // CHECK-LABEL: @test_v128_load(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[__V_I:%.*]] = bitcast i8* [[MEM:%.*]] to <4 x i32>*
-// CHECK-NEXT:    [[TMP0:%.*]] = load <4 x i32>, <4 x i32>* [[__V_I]], align 1, !tbaa [[TBAA2:![0-9]+]]
+// CHECK-NEXT:    [[TMP0:%.*]] = load <4 x i32>, ptr [[MEM:%.*]], align 1, !tbaa [[TBAA2:![0-9]+]]
 // CHECK-NEXT:    ret <4 x i32> [[TMP0]]
 //
 v128_t test_v128_load(const void *mem) {
@@ -19,7 +18,7 @@ v128_t test_v128_load(const void *mem) {
 
 // CHECK-LABEL: @test_v128_load8_splat(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i8, i8* [[MEM:%.*]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    [[TMP0:%.*]] = load i8, ptr [[MEM:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    [[VECINIT_I:%.*]] = insertelement <16 x i8> undef, i8 [[TMP0]], i64 0
 // CHECK-NEXT:    [[VECINIT16_I:%.*]] = shufflevector <16 x i8> [[VECINIT_I]], <16 x i8> poison, <16 x i32> zeroinitializer
 // CHECK-NEXT:    [[TMP1:%.*]] = bitcast <16 x i8> [[VECINIT16_I]] to <4 x i32>
@@ -31,8 +30,7 @@ v128_t test_v128_load8_splat(const void *mem) {
 
 // CHECK-LABEL: @test_v128_load16_splat(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[__V1_I:%.*]] = bitcast i8* [[MEM:%.*]] to i16*
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* [[__V1_I]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr [[MEM:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    [[VECINIT_I:%.*]] = insertelement <8 x i16> undef, i16 [[TMP0]], i64 0
 // CHECK-NEXT:    [[VECINIT8_I:%.*]] = shufflevector <8 x i16> [[VECINIT_I]], <8 x i16> poison, <8 x i32> zeroinitializer
 // CHECK-NEXT:    [[TMP1:%.*]] = bitcast <8 x i16> [[VECINIT8_I]] to <4 x i32>
@@ -44,8 +42,7 @@ v128_t test_v128_load16_splat(const void *mem) {
 
 // CHECK-LABEL: @test_v128_load32_splat(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[__V1_I:%.*]] = bitcast i8* [[MEM:%.*]] to i32*
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* [[__V1_I]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr [[MEM:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    [[VECINIT_I:%.*]] = insertelement <4 x i32> undef, i32 [[TMP0]], i64 0
 // CHECK-NEXT:    [[VECINIT4_I:%.*]] = shufflevector <4 x i32> [[VECINIT_I]], <4 x i32> poison, <4 x i32> zeroinitializer
 // CHECK-NEXT:    ret <4 x i32> [[VECINIT4_I]]
@@ -56,8 +53,7 @@ v128_t test_v128_load32_splat(const void *mem) {
 
 // CHECK-LABEL: @test_v128_load64_splat(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[__V1_I:%.*]] = bitcast i8* [[MEM:%.*]] to i64*
-// CHECK-NEXT:    [[TMP0:%.*]] = load i64, i64* [[__V1_I]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    [[TMP0:%.*]] = load i64, ptr [[MEM:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    [[VECINIT_I:%.*]] = insertelement <2 x i64> undef, i64 [[TMP0]], i64 0
 // CHECK-NEXT:    [[VECINIT2_I:%.*]] = shufflevector <2 x i64> [[VECINIT_I]], <2 x i64> poison, <2 x i32> zeroinitializer
 // CHECK-NEXT:    [[TMP1:%.*]] = bitcast <2 x i64> [[VECINIT2_I]] to <4 x i32>
@@ -69,8 +65,7 @@ v128_t test_v128_load64_splat(const void *mem) {
 
 // CHECK-LABEL: @test_i16x8_load8x8(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[__V1_I:%.*]] = bitcast i8* [[MEM:%.*]] to <8 x i8>*
-// CHECK-NEXT:    [[TMP0:%.*]] = load <8 x i8>, <8 x i8>* [[__V1_I]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    [[TMP0:%.*]] = load <8 x i8>, ptr [[MEM:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    [[CONV_I:%.*]] = sext <8 x i8> [[TMP0]] to <8 x i16>
 // CHECK-NEXT:    [[TMP1:%.*]] = bitcast <8 x i16> [[CONV_I]] to <4 x i32>
 // CHECK-NEXT:    ret <4 x i32> [[TMP1]]
@@ -81,8 +76,7 @@ v128_t test_i16x8_load8x8(const void *mem) {
 
 // CHECK-LABEL: @test_u16x8_load8x8(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[__V1_I:%.*]] = bitcast i8* [[MEM:%.*]] to <8 x i8>*
-// CHECK-NEXT:    [[TMP0:%.*]] = load <8 x i8>, <8 x i8>* [[__V1_I]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    [[TMP0:%.*]] = load <8 x i8>, ptr [[MEM:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    [[CONV_I:%.*]] = zext <8 x i8> [[TMP0]] to <8 x i16>
 // CHECK-NEXT:    [[TMP1:%.*]] = bitcast <8 x i16> [[CONV_I]] to <4 x i32>
 // CHECK-NEXT:    ret <4 x i32> [[TMP1]]
@@ -93,8 +87,7 @@ v128_t test_u16x8_load8x8(const void *mem) {
 
 // CHECK-LABEL: @test_i32x4_load16x4(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[__V1_I:%.*]] = bitcast i8* [[MEM:%.*]] to <4 x i16>*
-// CHECK-NEXT:    [[TMP0:%.*]] = load <4 x i16>, <4 x i16>* [[__V1_I]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    [[TMP0:%.*]] = load <4 x i16>, ptr [[MEM:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    [[CONV_I:%.*]] = sext <4 x i16> [[TMP0]] to <4 x i32>
 // CHECK-NEXT:    ret <4 x i32> [[CONV_I]]
 //
@@ -104,8 +97,7 @@ v128_t test_i32x4_load16x4(const void *mem) {
 
 // CHECK-LABEL: @test_u32x4_load16x4(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[__V1_I:%.*]] = bitcast i8* [[MEM:%.*]] to <4 x i16>*
-// CHECK-NEXT:    [[TMP0:%.*]] = load <4 x i16>, <4 x i16>* [[__V1_I]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    [[TMP0:%.*]] = load <4 x i16>, ptr [[MEM:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    [[CONV_I:%.*]] = zext <4 x i16> [[TMP0]] to <4 x i32>
 // CHECK-NEXT:    ret <4 x i32> [[CONV_I]]
 //
@@ -115,8 +107,7 @@ v128_t test_u32x4_load16x4(const void *mem) {
 
 // CHECK-LABEL: @test_i64x2_load32x2(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[__V1_I:%.*]] = bitcast i8* [[MEM:%.*]] to <2 x i32>*
-// CHECK-NEXT:    [[TMP0:%.*]] = load <2 x i32>, <2 x i32>* [[__V1_I]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    [[TMP0:%.*]] = load <2 x i32>, ptr [[MEM:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    [[CONV_I:%.*]] = sext <2 x i32> [[TMP0]] to <2 x i64>
 // CHECK-NEXT:    [[TMP1:%.*]] = bitcast <2 x i64> [[CONV_I]] to <4 x i32>
 // CHECK-NEXT:    ret <4 x i32> [[TMP1]]
@@ -127,8 +118,7 @@ v128_t test_i64x2_load32x2(const void *mem) {
 
 // CHECK-LABEL: @test_u64x2_load32x2(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[__V1_I:%.*]] = bitcast i8* [[MEM:%.*]] to <2 x i32>*
-// CHECK-NEXT:    [[TMP0:%.*]] = load <2 x i32>, <2 x i32>* [[__V1_I]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    [[TMP0:%.*]] = load <2 x i32>, ptr [[MEM:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    [[CONV_I:%.*]] = zext <2 x i32> [[TMP0]] to <2 x i64>
 // CHECK-NEXT:    [[TMP1:%.*]] = bitcast <2 x i64> [[CONV_I]] to <4 x i32>
 // CHECK-NEXT:    ret <4 x i32> [[TMP1]]
@@ -139,8 +129,7 @@ v128_t test_u64x2_load32x2(const void *mem) {
 
 // CHECK-LABEL: @test_v128_load32_zero(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[__V1_I:%.*]] = bitcast i8* [[MEM:%.*]] to i32*
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* [[__V1_I]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr [[MEM:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    [[VECINIT4_I:%.*]] = insertelement <4 x i32> <i32 poison, i32 0, i32 0, i32 0>, i32 [[TMP0]], i64 0
 // CHECK-NEXT:    ret <4 x i32> [[VECINIT4_I]]
 //
@@ -150,8 +139,7 @@ v128_t test_v128_load32_zero(const void *mem) {
 
 // CHECK-LABEL: @test_v128_load64_zero(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[__V1_I:%.*]] = bitcast i8* [[MEM:%.*]] to i64*
-// CHECK-NEXT:    [[TMP0:%.*]] = load i64, i64* [[__V1_I]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    [[TMP0:%.*]] = load i64, ptr [[MEM:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    [[VECINIT2_I:%.*]] = insertelement <2 x i64> <i64 poison, i64 0>, i64 [[TMP0]], i64 0
 // CHECK-NEXT:    [[TMP1:%.*]] = bitcast <2 x i64> [[VECINIT2_I]] to <4 x i32>
 // CHECK-NEXT:    ret <4 x i32> [[TMP1]]
@@ -162,7 +150,7 @@ v128_t test_v128_load64_zero(const void *mem) {
 
 // CHECK-LABEL: @test_v128_load8_lane(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i8, i8* [[PTR:%.*]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    [[TMP0:%.*]] = load i8, ptr [[PTR:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    [[TMP1:%.*]] = bitcast <4 x i32> [[VEC:%.*]] to <16 x i8>
 // CHECK-NEXT:    [[VECINS_I:%.*]] = insertelement <16 x i8> [[TMP1]], i8 [[TMP0]], i64 15
 // CHECK-NEXT:    [[TMP2:%.*]] = bitcast <16 x i8> [[VECINS_I]] to <4 x i32>
@@ -174,7 +162,7 @@ v128_t test_v128_load8_lane(const uint8_t *ptr, v128_t vec) {
 
 // CHECK-LABEL: @test_v128_load16_lane(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i16, i16* [[PTR:%.*]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr [[PTR:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    [[TMP1:%.*]] = bitcast <4 x i32> [[VEC:%.*]] to <8 x i16>
 // CHECK-NEXT:    [[VECINS_I:%.*]] = insertelement <8 x i16> [[TMP1]], i16 [[TMP0]], i64 7
 // CHECK-NEXT:    [[TMP2:%.*]] = bitcast <8 x i16> [[VECINS_I]] to <4 x i32>
@@ -186,7 +174,7 @@ v128_t test_v128_load16_lane(const uint16_t *ptr, v128_t vec) {
 
 // CHECK-LABEL: @test_v128_load32_lane(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* [[PTR:%.*]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr [[PTR:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    [[VECINS_I:%.*]] = insertelement <4 x i32> [[VEC:%.*]], i32 [[TMP0]], i64 3
 // CHECK-NEXT:    ret <4 x i32> [[VECINS_I]]
 //
@@ -196,7 +184,7 @@ v128_t test_v128_load32_lane(const uint32_t *ptr, v128_t vec) {
 
 // CHECK-LABEL: @test_v128_load64_lane(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[TMP0:%.*]] = load i64, i64* [[PTR:%.*]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    [[TMP0:%.*]] = load i64, ptr [[PTR:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    [[TMP1:%.*]] = bitcast <4 x i32> [[VEC:%.*]] to <2 x i64>
 // CHECK-NEXT:    [[VECINS_I:%.*]] = insertelement <2 x i64> [[TMP1]], i64 [[TMP0]], i64 1
 // CHECK-NEXT:    [[TMP2:%.*]] = bitcast <2 x i64> [[VECINS_I]] to <4 x i32>
@@ -208,8 +196,7 @@ v128_t test_v128_load64_lane(const uint64_t *ptr, v128_t vec) {
 
 // CHECK-LABEL: @test_v128_store(
 // CHECK-NEXT:  entry:
-// CHECK-NEXT:    [[__V_I:%.*]] = bitcast i8* [[MEM:%.*]] to <4 x i32>*
-// CHECK-NEXT:    store <4 x i32> [[A:%.*]], <4 x i32>* [[__V_I]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    store <4 x i32> [[A:%.*]], ptr [[MEM:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    ret void
 //
 void test_v128_store(void *mem, v128_t a) {
@@ -220,7 +207,7 @@ void test_v128_store(void *mem, v128_t a) {
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = bitcast <4 x i32> [[VEC:%.*]] to <16 x i8>
 // CHECK-NEXT:    [[VECEXT_I:%.*]] = extractelement <16 x i8> [[TMP0]], i64 15
-// CHECK-NEXT:    store i8 [[VECEXT_I]], i8* [[PTR:%.*]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    store i8 [[VECEXT_I]], ptr [[PTR:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    ret void
 //
 void test_v128_store8_lane(uint8_t *ptr, v128_t vec) {
@@ -231,7 +218,7 @@ void test_v128_store8_lane(uint8_t *ptr, v128_t vec) {
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = bitcast <4 x i32> [[VEC:%.*]] to <8 x i16>
 // CHECK-NEXT:    [[VECEXT_I:%.*]] = extractelement <8 x i16> [[TMP0]], i64 7
-// CHECK-NEXT:    store i16 [[VECEXT_I]], i16* [[PTR:%.*]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    store i16 [[VECEXT_I]], ptr [[PTR:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    ret void
 //
 void test_v128_store16_lane(uint16_t *ptr, v128_t vec) {
@@ -241,7 +228,7 @@ void test_v128_store16_lane(uint16_t *ptr, v128_t vec) {
 // CHECK-LABEL: @test_v128_store32_lane(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[VECEXT_I:%.*]] = extractelement <4 x i32> [[VEC:%.*]], i64 3
-// CHECK-NEXT:    store i32 [[VECEXT_I]], i32* [[PTR:%.*]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    store i32 [[VECEXT_I]], ptr [[PTR:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    ret void
 //
 void test_v128_store32_lane(uint32_t *ptr, v128_t vec) {
@@ -252,7 +239,7 @@ void test_v128_store32_lane(uint32_t *ptr, v128_t vec) {
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = bitcast <4 x i32> [[VEC:%.*]] to <2 x i64>
 // CHECK-NEXT:    [[VECEXT_I:%.*]] = extractelement <2 x i64> [[TMP0]], i64 1
-// CHECK-NEXT:    store i64 [[VECEXT_I]], i64* [[PTR:%.*]], align 1, !tbaa [[TBAA2]]
+// CHECK-NEXT:    store i64 [[VECEXT_I]], ptr [[PTR:%.*]], align 1, !tbaa [[TBAA2]]
 // CHECK-NEXT:    ret void
 //
 void test_v128_store64_lane(uint64_t *ptr, v128_t vec) {

diff  --git a/clang/test/Modules/direct-module-import.m b/clang/test/Modules/direct-module-import.m
index 15dde0f496893..3c75a8c135f0c 100644
--- a/clang/test/Modules/direct-module-import.m
+++ b/clang/test/Modules/direct-module-import.m
@@ -1,8 +1,8 @@
 // UNSUPPORTED: -zos, target={{.*}}-aix{{.*}}
 // RUN: rm -rf %t
-// RUN: %clang_cc1 -no-opaque-pointers -fmodules-cache-path=%t -fmodules -fimplicit-module-maps -F %S/Inputs -include Module/Module.h %s -emit-llvm -o - | FileCheck %s
+// RUN: %clang_cc1 -fmodules-cache-path=%t -fmodules -fimplicit-module-maps -F %S/Inputs -include Module/Module.h %s -emit-llvm -o - | FileCheck %s
 
-// CHECK: call {{.*}}i8* @getModuleVersion
+// CHECK: call {{.*}}ptr @getModuleVersion
 const char* getVer(void) {
   return getModuleVersion();
 }

diff  --git a/clang/test/Modules/templates.mm b/clang/test/Modules/templates.mm
index 10ffee59a3dcd..0e66c9843cc09 100644
--- a/clang/test/Modules/templates.mm
+++ b/clang/test/Modules/templates.mm
@@ -1,6 +1,6 @@
 // RUN: rm -rf %t
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-linux-gnu -std=c++11 -x objective-c++ -fmodules -fimplicit-module-maps -fmodules-cache-path=%t -I %S/Inputs -verify %s -Wno-objc-root-class
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-linux-gnu -std=c++11 -x objective-c++ -fmodules -fimplicit-module-maps -fmodules-cache-path=%t -I %S/Inputs -emit-llvm %s -o - -Wno-objc-root-class | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-linux-gnu -std=c++11 -x objective-c++ -fmodules -fimplicit-module-maps -fmodules-cache-path=%t -I %S/Inputs -verify %s -Wno-objc-root-class
+// RUN: %clang_cc1 -triple x86_64-linux-gnu -std=c++11 -x objective-c++ -fmodules -fimplicit-module-maps -fmodules-cache-path=%t -I %S/Inputs -emit-llvm %s -o - -Wno-objc-root-class | FileCheck %s
 // expected-no-diagnostics
 // REQUIRES: x86-registered-target
 @import templates_left;
@@ -12,10 +12,10 @@ void testInlineRedeclEarly() {
 
 @import templates_right;
 
-// CHECK-DAG: @list_left ={{.*}} global %[[LIST:.*]] { %[[LISTNODE:.*]]* null, i32 8 }, align 8
-// CHECK-DAG: @list_right ={{.*}} global %[[LIST]] { %[[LISTNODE]]* null, i32 12 }, align 8
-// CHECK-DAG: @__const._Z15testMixedStructv.l = {{.*}} constant %[[LIST]] { %{{.*}}* null, i32 1 }, align 8
-// CHECK-DAG: @__const._Z15testMixedStructv.r = {{.*}} constant %[[LIST]] { %{{.*}}* null, i32 2 }, align 8
+// CHECK-DAG: @list_left ={{.*}} global %[[LIST:.*]] { ptr null, i32 8 }, align 8
+// CHECK-DAG: @list_right ={{.*}} global %[[LIST]] { ptr null, i32 12 }, align 8
+// CHECK-DAG: @__const._Z15testMixedStructv.l = {{.*}} constant %[[LIST]] { ptr null, i32 1 }, align 8
+// CHECK-DAG: @__const._Z15testMixedStructv.r = {{.*}} constant %[[LIST]] { ptr null, i32 2 }, align 8
 // CHECK-DAG: @_ZN29WithUndefinedStaticDataMemberIA_iE9undefinedE = external global
 
 void testTemplateClasses() {
@@ -77,19 +77,19 @@ unsigned testMixedStruct() {
   // CHECK: %[[l:.*]] = alloca %[[ListInt:[^ ]*]], align 8
   // CHECK: %[[r:.*]] = alloca %[[ListInt]], align 8
 
-  // CHECK: call {{.*}}memcpy{{.*}}(i8* align {{[0-9]+}} %{{.*}}, i8* align {{[0-9]+}} bitcast ({{.*}}* @__const._Z15testMixedStructv.l to i8*), i64 16,
+  // CHECK: call {{.*}}memcpy{{.*}}(ptr align {{[0-9]+}} %{{.*}}, ptr align {{[0-9]+}} @__const._Z15testMixedStructv.l, i64 16,
   ListInt_left l{0, 1};
 
-  // CHECK: call {{.*}}memcpy{{.*}}(i8* align {{[0-9]+}} %{{.*}}, i8* align {{[0-9]+}} bitcast ({{.*}}* @__const._Z15testMixedStructv.r to i8*), i64 16,
+  // CHECK: call {{.*}}memcpy{{.*}}(ptr align {{[0-9]+}} %{{.*}}, ptr align {{[0-9]+}} @__const._Z15testMixedStructv.r, i64 16,
   ListInt_right r{0, 2};
 
-  // CHECK: call void @_Z10useListIntR4ListIiE(%[[ListInt]]* noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) %[[l]])
+  // CHECK: call void @_Z10useListIntR4ListIiE(ptr noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) %[[l]])
   useListInt(l);
-  // CHECK: call void @_Z10useListIntR4ListIiE(%[[ListInt]]* noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) %[[r]])
+  // CHECK: call void @_Z10useListIntR4ListIiE(ptr noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) %[[r]])
   useListInt(r);
 
-  // CHECK: load i32, i32* bitcast (i8* getelementptr inbounds (i8, i8* bitcast ({{.*}}* @list_left to i8*), i64 8) to i32*)
-  // CHECK: load i32, i32* bitcast (i8* getelementptr inbounds (i8, i8* bitcast ({{.*}}* @list_right to i8*), i64 8) to i32*)
+  // CHECK: load i32, ptr getelementptr inbounds (i8, ptr @list_left, i64 8)
+  // CHECK: load i32, ptr getelementptr inbounds (i8, ptr @list_right, i64 8)
   return list_left.*size_right + list_right.*size_left;
 }
 
@@ -107,12 +107,12 @@ unsigned testMixedStruct() {
 void testStaticDataMember() {
   WithUndefinedStaticDataMember<int[]> load_it;
 
-  // CHECK-LABEL: define linkonce_odr noundef i32* @_Z23getStaticDataMemberLeftv(
-  // CHECK: ret i32* getelementptr inbounds ([0 x i32], [0 x i32]* @_ZN29WithUndefinedStaticDataMemberIA_iE9undefinedE, i64 0, i64 0)
+  // CHECK-LABEL: define linkonce_odr noundef ptr @_Z23getStaticDataMemberLeftv(
+  // CHECK: ret ptr @_ZN29WithUndefinedStaticDataMemberIA_iE9undefinedE
   (void) getStaticDataMemberLeft();
 
-  // CHECK-LABEL: define linkonce_odr noundef i32* @_Z24getStaticDataMemberRightv(
-  // CHECK: ret i32* getelementptr inbounds ([0 x i32], [0 x i32]* @_ZN29WithUndefinedStaticDataMemberIA_iE9undefinedE, i64 0, i64 0)
+  // CHECK-LABEL: define linkonce_odr noundef ptr @_Z24getStaticDataMemberRightv(
+  // CHECK: ret ptr @_ZN29WithUndefinedStaticDataMemberIA_iE9undefinedE
   (void) getStaticDataMemberRight();
 }
 
@@ -125,7 +125,7 @@ void testWithAttributes() {
 
 // Check that returnNonTrivial doesn't return Class0<S0> directly in registers.
 
-// CHECK: declare void @_Z16returnNonTrivialv(%struct.Class0* sret(%struct.Class0) align 8)
+// CHECK: declare void @_Z16returnNonTrivialv(ptr sret(%struct.Class0) align 8)
 
 @import template_nontrivial0;
 @import template_nontrivial1;

diff  --git a/clang/test/PCH/non-trivial-c-compound-literal.m b/clang/test/PCH/non-trivial-c-compound-literal.m
index 57a108e93a033..7e8f69f71d4cf 100644
--- a/clang/test/PCH/non-trivial-c-compound-literal.m
+++ b/clang/test/PCH/non-trivial-c-compound-literal.m
@@ -1,5 +1,5 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -x objective-c -fobjc-arc -emit-pch -o %t %s
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -x objective-c -fobjc-arc -include-pch %t -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -x objective-c -fobjc-arc -emit-pch -o %t %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -x objective-c -fobjc-arc -include-pch %t -emit-llvm -o - %s | FileCheck %s
 
 #ifndef HEADER
 #define HEADER
@@ -15,12 +15,11 @@ static inline id getObj(id a) {
 
 #else
 
-// CHECK: %[[STRUCT_S:.*]] = type { i8* }
+// CHECK: %[[STRUCT_S:.*]] = type { ptr }
 
-// CHECK: define internal i8* @getObj(
+// CHECK: define internal ptr @getObj(
 // CHECK: %[[_COMPOUNDLITERAL:.*]] = alloca %[[STRUCT_S]],
-// CHECK: %[[V5:.*]] = bitcast %[[STRUCT_S]]* %[[_COMPOUNDLITERAL]] to i8**
-// CHECK: call void @__destructor_8_s0(i8** %[[V5]])
+// CHECK: call void @__destructor_8_s0(ptr %[[_COMPOUNDLITERAL]])
 
 id test(id a) {
   return getObj(a);

diff  --git a/clang/test/PCH/objcxx-ivar-class.mm b/clang/test/PCH/objcxx-ivar-class.mm
index 3d714fce8a05a..ef8b6a73f11f2 100644
--- a/clang/test/PCH/objcxx-ivar-class.mm
+++ b/clang/test/PCH/objcxx-ivar-class.mm
@@ -1,16 +1,16 @@
 // UNSUPPORTED: -zos, target={{.*}}-aix{{.*}}
 // Test this without pch.
-// RUN: %clang_cc1 -no-opaque-pointers -include %S/objcxx-ivar-class.h -triple %itanium_abi_triple %s -emit-llvm -o - | FileCheck %s
+// RUN: %clang_cc1 -include %S/objcxx-ivar-class.h -triple %itanium_abi_triple %s -emit-llvm -o - | FileCheck %s
 
 // Test with pch.
-// RUN: %clang_cc1 -no-opaque-pointers -x objective-c++-header -triple %itanium_abi_triple -emit-pch -o %t %S/objcxx-ivar-class.h
-// RUN: %clang_cc1 -no-opaque-pointers -include-pch %t -triple %itanium_abi_triple %s -emit-llvm -o - | FileCheck %s
+// RUN: %clang_cc1 -x objective-c++-header -triple %itanium_abi_triple -emit-pch -o %t %S/objcxx-ivar-class.h
+// RUN: %clang_cc1 -include-pch %t -triple %itanium_abi_triple %s -emit-llvm -o - | FileCheck %s
 
 // CHECK: [C position]
 // CHECK: call {{.*}} @_ZN1SC1ERKS_
 
 // CHECK: [C setPosition:]
-// CHECK: = call {{.*}}%struct.S* @_ZN1SaSERKS_
+// CHECK: = call {{.*}}ptr @_ZN1SaSERKS_
 
 // CHECK: [C .cxx_destruct]
 // CHECK: [C .cxx_construct]

diff  --git a/clang/test/Sema/ms_class_layout.cpp b/clang/test/Sema/ms_class_layout.cpp
index ce5ff7563f4dc..cd0f04abd1398 100644
--- a/clang/test/Sema/ms_class_layout.cpp
+++ b/clang/test/Sema/ms_class_layout.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -std=c++14 -no-opaque-pointers -emit-llvm-only -triple i686-pc-win32 -fdump-record-layouts %s 2>/dev/null \
+// RUN: %clang_cc1 -std=c++14 -emit-llvm-only -triple i686-pc-win32 -fdump-record-layouts %s 2>/dev/null \
 // RUN:            | FileCheck %s --strict-whitespace
 
 #pragma pack(push, 8)
@@ -206,14 +206,14 @@ int main() {
 // CHECK-NEXT: sizeof=80, align=8
 // CHECK-NEXT: nvsize=64, nvalign=8
 
-// CHECK: %class.D = type { i32 (...)**, double }
+// CHECK: %class.D = type { ptr, double }
 
-// CHECK: %class.B = type { i32 (...)**, i32 }
+// CHECK: %class.B = type { ptr, i32 }
 
 // CHECK: %class.A = type { %class.B, i32, i8 }
 
-// CHECK: %class.C = type { %class.D, %class.B, i32*, double, i32, double, i32, [4 x i8], %class.A }
-// CHECK: %class.C.base = type { %class.D, %class.B, i32*, double, i32, double, i32 }
+// CHECK: %class.C = type { %class.D, %class.B, ptr, double, i32, double, i32, [4 x i8], %class.A }
+// CHECK: %class.C.base = type { %class.D, %class.B, ptr, double, i32, double, i32 }
 
 // CHECK-LABEL: 0 | struct BaseStruct{{$}}
 // CHECK-NEXT:  0 |   double v0
@@ -284,7 +284,7 @@ int main() {
 // CHECK-NEXT: sizeof=24, align=8
 // CHECK-NEXT: nvsize=8, nvalign=8
 
-// CHECK: %struct.H = type { %struct.G, i32*, %class.D }
+// CHECK: %struct.H = type { %struct.G, ptr, %class.D }
 
 // CHECK-LABEL: 0 | struct I{{$}}
 // CHECK-NEXT:  0 |   (I vftable pointer)
@@ -296,8 +296,8 @@ int main() {
 // CHECK-NEXT: sizeof=40, align=8
 // CHECK-NEXT: nvsize=24, nvalign=8
 
-// CHECK: %struct.I = type { i32 (...)**, [4 x i8], i32*, double, %class.D }
-// CHECK: %struct.I.base = type { i32 (...)**, [4 x i8], i32*, double }
+// CHECK: %struct.I = type { ptr, [4 x i8], ptr, double, %class.D }
+// CHECK: %struct.I.base = type { ptr, [4 x i8], ptr, double }
 
 // CHECK-LABEL: 0 | struct L{{$}}
 // CHECK-NEXT:  0 |   int l
@@ -316,8 +316,8 @@ int main() {
 // CHECK-NEXT:  8 |     int k
 // CHECK-NEXT: sizeof=12, align=4
 
-//CHECK: %struct.M = type { i32*, i32, %struct.K }
-//CHECK: %struct.M.base = type { i32*, i32 }
+//CHECK: %struct.M = type { ptr, i32, %struct.K }
+//CHECK: %struct.M.base = type { ptr, i32 }
 
 // CHECK-LABEL: 0 | struct N{{$}}
 // CHECK-NEXT:  0 |   (N vftable pointer)
@@ -331,7 +331,7 @@ int main() {
 // CHECK-NEXT: sizeof=20, align=4
 // CHECK-NEXT: nvsize=16, nvalign=4
 
-//CHECK: %struct.N = type { i32 (...)**, %struct.L, %struct.M.base, %struct.K }
+//CHECK: %struct.N = type { ptr, %struct.L, %struct.M.base, %struct.K }
 
 // CHECK-LABEL: 0 | struct O{{$}}
 // CHECK-NEXT:  0 |   (O vftable pointer)
@@ -347,8 +347,8 @@ int main() {
 // CHECK-NEXT:    | [sizeof=40, align=8
 // CHECK-NEXT:    |  nvsize=24, nvalign=8]
 
-// CHECK: struct.O = type { i32 (...)**, [4 x i8], %struct.H.base, %struct.G, %class.D }
-// CHECK: struct.O.base = type { i32 (...)**, [4 x i8], %struct.H.base, %struct.G, [4 x i8] }
+// CHECK: struct.O = type { ptr, [4 x i8], %struct.H.base, %struct.G, %class.D }
+// CHECK: struct.O.base = type { ptr, [4 x i8], %struct.H.base, %struct.G, [4 x i8] }
 
 // CHECK-LABEL: 0 | struct P{{$}}
 // CHECK-NEXT:  0 |   struct M (base)
@@ -419,11 +419,11 @@ int main() {
 // CHECK-NEXT: sizeof=48, align=4
 // CHECK-NEXT: nvsize=12, nvalign=4
 
-// CHECK: %struct.f = type { i32 (...)** }
-// CHECK: %struct.s = type { i32 (...)**, i32*, i32, i32, %struct.f }
-// CHECK: %class.IA = type { i32 (...)** }
-// CHECK: %class.ICh = type { i32 (...)**, i32*, i32, %class.IA }
-// CHECK: %struct.sd = type { i32*, i32, i8, i32, %struct.f, %struct.s.base, i32, %class.IA, %class.ICh.base }
+// CHECK: %struct.f = type { ptr }
+// CHECK: %struct.s = type { ptr, ptr, i32, i32, %struct.f }
+// CHECK: %class.IA = type { ptr }
+// CHECK: %class.ICh = type { ptr, ptr, i32, %class.IA }
+// CHECK: %struct.sd = type { ptr, i32, i8, i32, %struct.f, %struct.s.base, i32, %class.IA, %class.ICh.base }
 
 // CHECK-LABEL: 0 | struct AV{{$}}
 // CHECK-NEXT:  0 |   (AV vftable pointer)
@@ -445,10 +445,10 @@ int main() {
 // CHECK-NEXT: sizeof=12, align=4
 // CHECK-NEXT: nvsize=4, nvalign=4
 
-// CHECK: %struct.AV = type { i32 (...)** }
+// CHECK: %struct.AV = type { ptr }
 // CHECK: %struct.BV = type { %struct.AV }
-// CHECK: %struct.CV = type { i32*, i32, %struct.BV }
-// CHECK: %struct.CV.base = type { i32* }
+// CHECK: %struct.CV = type { ptr, i32, %struct.BV }
+// CHECK: %struct.CV.base = type { ptr }
 
 // CHECK-LABEL: 0 | struct DV{{$}}
 // CHECK-NEXT:  0 |   struct BV (primary base)

diff  --git a/clang/test/SemaCXX/constexpr-vectors.cpp b/clang/test/SemaCXX/constexpr-vectors.cpp
index f5ded584f8c2c..99b045f888d87 100644
--- a/clang/test/SemaCXX/constexpr-vectors.cpp
+++ b/clang/test/SemaCXX/constexpr-vectors.cpp
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -no-opaque-pointers -std=c++14 -Wno-unused-value %s -disable-llvm-passes -triple x86_64-linux-gnu -emit-llvm -o - | FileCheck %s
+// RUN: %clang_cc1 -std=c++14 -Wno-unused-value %s -disable-llvm-passes -triple x86_64-linux-gnu -emit-llvm -o - | FileCheck %s
 
 // FIXME: Unfortunately there is no good way to validate that our values are
 // correct since Vector types don't have operator [] implemented for constexpr.
@@ -681,33 +681,33 @@ using FourBoolsExtVec __attribute__((ext_vector_type(4))) = bool;
 void BoolVecUsage() {
   constexpr auto a = FourBoolsExtVec{true, false, true, false} <
                      FourBoolsExtVec{false, false, true, true};
-  // CHECK: store i8 bitcast (<8 x i1> <i1 false, i1 false, i1 false, i1 true, i1 undef, i1 undef, i1 undef, i1 undef> to i8), i8* %a, align 1
+  // CHECK: store i8 bitcast (<8 x i1> <i1 false, i1 false, i1 false, i1 true, i1 undef, i1 undef, i1 undef, i1 undef> to i8), ptr %a, align 1
   constexpr auto b = FourBoolsExtVec{true, false, true, false} <=
                      FourBoolsExtVec{false, false, true, true};
-  // CHECK: store i8 bitcast (<8 x i1> <i1 false, i1 true, i1 true, i1 true, i1 undef, i1 undef, i1 undef, i1 undef> to i8), i8* %b, align 1
+  // CHECK: store i8 bitcast (<8 x i1> <i1 false, i1 true, i1 true, i1 true, i1 undef, i1 undef, i1 undef, i1 undef> to i8), ptr %b, align 1
   constexpr auto c = FourBoolsExtVec{true, false, true, false} ==
                      FourBoolsExtVec{false, false, true, true};
-  // CHECK: store i8 bitcast (<8 x i1> <i1 false, i1 true, i1 true, i1 false, i1 undef, i1 undef, i1 undef, i1 undef> to i8), i8* %c, align 1
+  // CHECK: store i8 bitcast (<8 x i1> <i1 false, i1 true, i1 true, i1 false, i1 undef, i1 undef, i1 undef, i1 undef> to i8), ptr %c, align 1
   constexpr auto d = FourBoolsExtVec{true, false, true, false} !=
                      FourBoolsExtVec{false, false, true, true};
-  // CHECK: store i8 bitcast (<8 x i1> <i1 true, i1 false, i1 false, i1 true, i1 undef, i1 undef, i1 undef, i1 undef> to i8), i8* %d, align 1
+  // CHECK: store i8 bitcast (<8 x i1> <i1 true, i1 false, i1 false, i1 true, i1 undef, i1 undef, i1 undef, i1 undef> to i8), ptr %d, align 1
   constexpr auto e = FourBoolsExtVec{true, false, true, false} >=
                      FourBoolsExtVec{false, false, true, true};
-  // CHECK: store i8 bitcast (<8 x i1> <i1 true, i1 true, i1 true, i1 false, i1 undef, i1 undef, i1 undef, i1 undef> to i8), i8* %e, align 1
+  // CHECK: store i8 bitcast (<8 x i1> <i1 true, i1 true, i1 true, i1 false, i1 undef, i1 undef, i1 undef, i1 undef> to i8), ptr %e, align 1
   constexpr auto f = FourBoolsExtVec{true, false, true, false} >
                      FourBoolsExtVec{false, false, true, true};
-  // CHECK: store i8 bitcast (<8 x i1> <i1 true, i1 false, i1 false, i1 false, i1 undef, i1 undef, i1 undef, i1 undef> to i8), i8* %f, align 1
+  // CHECK: store i8 bitcast (<8 x i1> <i1 true, i1 false, i1 false, i1 false, i1 undef, i1 undef, i1 undef, i1 undef> to i8), ptr %f, align 1
   constexpr auto g = FourBoolsExtVec{true, false, true, false} &
                      FourBoolsExtVec{false, false, true, true};
-  // CHECK: store i8 bitcast (<8 x i1> <i1 false, i1 false, i1 true, i1 false, i1 undef, i1 undef, i1 undef, i1 undef> to i8), i8* %g, align 1
+  // CHECK: store i8 bitcast (<8 x i1> <i1 false, i1 false, i1 true, i1 false, i1 undef, i1 undef, i1 undef, i1 undef> to i8), ptr %g, align 1
   constexpr auto h = FourBoolsExtVec{true, false, true, false} |
                      FourBoolsExtVec{false, false, true, true};
-  // CHECK: store i8 bitcast (<8 x i1> <i1 true, i1 false, i1 true, i1 true, i1 undef, i1 undef, i1 undef, i1 undef> to i8), i8* %h, align 1
+  // CHECK: store i8 bitcast (<8 x i1> <i1 true, i1 false, i1 true, i1 true, i1 undef, i1 undef, i1 undef, i1 undef> to i8), ptr %h, align 1
   constexpr auto i = FourBoolsExtVec{true, false, true, false} ^
                      FourBoolsExtVec { false, false, true, true };
-  // CHECK: store i8 bitcast (<8 x i1> <i1 true, i1 false, i1 false, i1 true, i1 undef, i1 undef, i1 undef, i1 undef> to i8), i8* %i, align 1
+  // CHECK: store i8 bitcast (<8 x i1> <i1 true, i1 false, i1 false, i1 true, i1 undef, i1 undef, i1 undef, i1 undef> to i8), ptr %i, align 1
   constexpr auto j = !FourBoolsExtVec{true, false, true, false};
-  // CHECK: store i8 bitcast (<8 x i1> <i1 false, i1 true, i1 false, i1 true, i1 undef, i1 undef, i1 undef, i1 undef> to i8), i8* %j, align 1
+  // CHECK: store i8 bitcast (<8 x i1> <i1 false, i1 true, i1 false, i1 true, i1 undef, i1 undef, i1 undef, i1 undef> to i8), ptr %j, align 1
   constexpr auto k = ~FourBoolsExtVec{true, false, true, false};
-  // CHECK: store i8 bitcast (<8 x i1> <i1 false, i1 true, i1 false, i1 true, i1 undef, i1 undef, i1 undef, i1 undef> to i8), i8* %k, align 1
+  // CHECK: store i8 bitcast (<8 x i1> <i1 false, i1 true, i1 false, i1 true, i1 undef, i1 undef, i1 undef, i1 undef> to i8), ptr %k, align 1
 }

diff  --git a/clang/test/SemaCXX/linkage.cpp b/clang/test/SemaCXX/linkage.cpp
index cb82e52b2d8e8..871d64b4c2ebb 100644
--- a/clang/test/SemaCXX/linkage.cpp
+++ b/clang/test/SemaCXX/linkage.cpp
@@ -3,7 +3,7 @@
 // compared against the earlier cached value.  If we had a way of
 // testing linkage directly in Sema, that would be better.
 
-// RUN: %clang_cc1 -no-opaque-pointers -Werror -Wno-non-c-typedef-for-linkage -triple x86_64-apple-darwin10 -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -Werror -Wno-non-c-typedef-for-linkage -triple x86_64-apple-darwin10 -emit-llvm %s -o - | FileCheck %s
 
 // CHECK: @_ZZN5test61A3fooEvE3bar = linkonce_odr global i32 0, align 4
 
@@ -13,7 +13,7 @@ namespace test0 {
     void *foo() { return 0; }
   } A;
 
-  // CHECK: define linkonce_odr noundef i8* @_ZN5test01A3fooEv(
+  // CHECK: define linkonce_odr noundef ptr @_ZN5test01A3fooEv(
 
   void test(A *a) {
     a->foo();
@@ -83,7 +83,7 @@ extern "C" {
         // Test both for mangling in the code generation and warnings from use
         // of internal, undefined names via -Werror.
         // CHECK: call i32 @g(
-        // CHECK: load i32, i32* @a,
+        // CHECK: load i32, ptr @a,
         return g() + a;
       }
     };
@@ -94,8 +94,8 @@ extern "C" {
   }
 }
 
-// CHECK: define linkonce_odr noundef i8* @_ZN5test11A3fooILj0EEEPvv(
-// CHECK: define linkonce_odr noundef i8* @_ZN5test21A1BILj0EE3fooEv(
+// CHECK: define linkonce_odr noundef ptr @_ZN5test11A3fooILj0EEEPvv(
+// CHECK: define linkonce_odr noundef ptr @_ZN5test21A1BILj0EE3fooEv(
 
 namespace test5 {
   struct foo {

diff  --git a/clang/test/SemaObjC/debugger-support.m b/clang/test/SemaObjC/debugger-support.m
index b16e618af207d..c005147d394b9 100644
--- a/clang/test/SemaObjC/debugger-support.m
+++ b/clang/test/SemaObjC/debugger-support.m
@@ -1,14 +1,14 @@
-// RUN: %clang_cc1 -no-opaque-pointers -triple x86_64-apple-darwin10 -fdebugger-support %s -emit-llvm -o - | FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-apple-darwin10 -fdebugger-support %s -emit-llvm -o - | FileCheck %s
 
 // rdar://problem/9416370
 void test0(id x) {
   struct A { int w, x, y, z; };
   struct A result = (struct A) [x makeStruct];
   // CHECK:     define{{.*}} void @test0(
-  // CHECK:      [[X:%.*]] = alloca i8*, align 8
+  // CHECK:      [[X:%.*]] = alloca ptr, align 8
   // CHECK-NEXT: [[RESULT:%.*]] = alloca [[A:%.*]], align 4
-  // CHECK-NEXT: store i8* {{%.*}}, i8** [[X]],
-  // CHECK-NEXT: [[T0:%.*]] = load i8*, i8** [[X]],
-  // CHECK-NEXT: [[T1:%.*]] = load i8*, i8** @OBJC_SELECTOR_REFERENCES_
-  // CHECK-NEXT: [[T2:%.*]] = call { i64, i64 } bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to { i64, i64 } (i8*, i8*)*)(i8* noundef [[T0]], i8* noundef [[T1]])
+  // CHECK-NEXT: store ptr {{%.*}}, ptr [[X]],
+  // CHECK-NEXT: [[T0:%.*]] = load ptr, ptr [[X]],
+  // CHECK-NEXT: [[T1:%.*]] = load ptr, ptr @OBJC_SELECTOR_REFERENCES_
+  // CHECK-NEXT: [[T2:%.*]] = call { i64, i64 } @objc_msgSend(ptr noundef [[T0]], ptr noundef [[T1]])
 }


        


More information about the cfe-commits mailing list