[llvm] 30f1004 - [X86] Add test coverage for #164853 (#165245)
via llvm-commits
llvm-commits at lists.llvm.org
Mon Oct 27 06:28:12 PDT 2025
Author: Simon Pilgrim
Date: 2025-10-27T13:28:07Z
New Revision: 30f1004bec5fb29625b067885118b0e90104f567
URL: https://github.com/llvm/llvm-project/commit/30f1004bec5fb29625b067885118b0e90104f567
DIFF: https://github.com/llvm/llvm-project/commit/30f1004bec5fb29625b067885118b0e90104f567.diff
LOG: [X86] Add test coverage for #164853 (#165245)
Show examples of where truncated loads of non-constant, but known
aligned, shift amounts can be folded into address math to avoid
loads/spills of large/illegal scalar integers
Added:
llvm/test/CodeGen/X86/trunc-srl-load.ll
Modified:
Removed:
################################################################################
diff --git a/llvm/test/CodeGen/X86/trunc-srl-load.ll b/llvm/test/CodeGen/X86/trunc-srl-load.ll
new file mode 100644
index 0000000000000..4dae1433b2196
--- /dev/null
+++ b/llvm/test/CodeGen/X86/trunc-srl-load.ll
@@ -0,0 +1,1672 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=i686-unknown | FileCheck %s --check-prefixes=X86
+; RUN: llc < %s -mtriple=x86_64-unknown -mcpu=x86-64 | FileCheck %s --check-prefixes=X64,SSE
+; RUN: llc < %s -mtriple=x86_64-unknown -mcpu=x86-64-v2 | FileCheck %s --check-prefixes=X64,SSE
+; RUN: llc < %s -mtriple=x86_64-unknown -mcpu=x86-64-v3 | FileCheck %s --check-prefixes=X64,AVX,AVX2
+; RUN: llc < %s -mtriple=x86_64-unknown -mcpu=x86-64-v4 | FileCheck %s --check-prefixes=X64,AVX,AVX512
+
+; Tests showing for the analysis of non-constant shift amounts to improve load address math
+
+; Alignment of shift amounts should allow sub-integer loads.
+
+define i16 @extractSub64_16(ptr %word, i32 %idx) nounwind {
+; X86-LABEL: extractSub64_16:
+; X86: # %bb.0:
+; X86-NEXT: pushl %esi
+; X86-NEXT: movb {{[0-9]+}}(%esp), %ch
+; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
+; X86-NEXT: movl (%eax), %edx
+; X86-NEXT: movl 4(%eax), %esi
+; X86-NEXT: movb %ch, %cl
+; X86-NEXT: andb $16, %cl
+; X86-NEXT: movl %esi, %eax
+; X86-NEXT: shrl %cl, %eax
+; X86-NEXT: shrdl %cl, %esi, %edx
+; X86-NEXT: testb $32, %ch
+; X86-NEXT: jne .LBB0_2
+; X86-NEXT: # %bb.1:
+; X86-NEXT: movl %edx, %eax
+; X86-NEXT: .LBB0_2:
+; X86-NEXT: # kill: def $ax killed $ax killed $eax
+; X86-NEXT: popl %esi
+; X86-NEXT: retl
+;
+; SSE-LABEL: extractSub64_16:
+; SSE: # %bb.0:
+; SSE-NEXT: movl %esi, %ecx
+; SSE-NEXT: movq (%rdi), %rax
+; SSE-NEXT: andb $48, %cl
+; SSE-NEXT: # kill: def $cl killed $cl killed $ecx
+; SSE-NEXT: shrq %cl, %rax
+; SSE-NEXT: # kill: def $ax killed $ax killed $rax
+; SSE-NEXT: retq
+;
+; AVX-LABEL: extractSub64_16:
+; AVX: # %bb.0:
+; AVX-NEXT: # kill: def $esi killed $esi def $rsi
+; AVX-NEXT: andb $48, %sil
+; AVX-NEXT: shrxq %rsi, (%rdi), %rax
+; AVX-NEXT: # kill: def $ax killed $ax killed $rax
+; AVX-NEXT: retq
+ %idx_bounds = and i32 %idx, 63
+ %idx_align = and i32 %idx_bounds, -16
+ %sh = zext nneg i32 %idx_align to i64
+ %ld = load i64, ptr %word, align 8
+ %sub = lshr i64 %ld, %sh
+ %res = trunc i64 %sub to i16
+ ret i16 %res
+}
+
+define i16 @extractSub128_16(ptr %word, i32 %idx) nounwind {
+; X86-LABEL: extractSub128_16:
+; X86: # %bb.0:
+; X86-NEXT: pushl %ebp
+; X86-NEXT: movl %esp, %ebp
+; X86-NEXT: pushl %edi
+; X86-NEXT: pushl %esi
+; X86-NEXT: andl $-16, %esp
+; X86-NEXT: subl $32, %esp
+; X86-NEXT: movzbl 12(%ebp), %eax
+; X86-NEXT: movl 8(%ebp), %ecx
+; X86-NEXT: movl (%ecx), %edx
+; X86-NEXT: movl 4(%ecx), %esi
+; X86-NEXT: movl 8(%ecx), %edi
+; X86-NEXT: movl 12(%ecx), %ecx
+; X86-NEXT: movl %ecx, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %edi, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %esi, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %edx, (%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %eax, %ecx
+; X86-NEXT: andb $16, %cl
+; X86-NEXT: shrb $3, %al
+; X86-NEXT: andb $12, %al
+; X86-NEXT: movzbl %al, %edx
+; X86-NEXT: movl (%esp,%edx), %eax
+; X86-NEXT: movl 4(%esp,%edx), %edx
+; X86-NEXT: shrdl %cl, %edx, %eax
+; X86-NEXT: # kill: def $ax killed $ax killed $eax
+; X86-NEXT: leal -8(%ebp), %esp
+; X86-NEXT: popl %esi
+; X86-NEXT: popl %edi
+; X86-NEXT: popl %ebp
+; X86-NEXT: retl
+;
+; SSE-LABEL: extractSub128_16:
+; SSE: # %bb.0:
+; SSE-NEXT: movq (%rdi), %rax
+; SSE-NEXT: movq 8(%rdi), %rdx
+; SSE-NEXT: movl %esi, %ecx
+; SSE-NEXT: andb $48, %cl
+; SSE-NEXT: movq %rdx, %rdi
+; SSE-NEXT: shrq %cl, %rdi
+; SSE-NEXT: shrdq %cl, %rdx, %rax
+; SSE-NEXT: testb $64, %sil
+; SSE-NEXT: cmovneq %rdi, %rax
+; SSE-NEXT: # kill: def $ax killed $ax killed $rax
+; SSE-NEXT: retq
+;
+; AVX-LABEL: extractSub128_16:
+; AVX: # %bb.0:
+; AVX-NEXT: movq (%rdi), %rdx
+; AVX-NEXT: movq 8(%rdi), %rax
+; AVX-NEXT: movl %esi, %ecx
+; AVX-NEXT: andb $48, %cl
+; AVX-NEXT: shrdq %cl, %rax, %rdx
+; AVX-NEXT: shrxq %rcx, %rax, %rax
+; AVX-NEXT: testb $64, %sil
+; AVX-NEXT: cmoveq %rdx, %rax
+; AVX-NEXT: # kill: def $ax killed $ax killed $rax
+; AVX-NEXT: retq
+ %idx_bounds = and i32 %idx, 127
+ %idx_align = and i32 %idx_bounds, -16
+ %sh = zext nneg i32 %idx_align to i128
+ %ld = load i128, ptr %word, align 8
+ %sub = lshr i128 %ld, %sh
+ %res = trunc i128 %sub to i16
+ ret i16 %res
+}
+
+define i32 @extractSub128_32(ptr %word, i32 %idx) nounwind {
+; X86-LABEL: extractSub128_32:
+; X86: # %bb.0:
+; X86-NEXT: pushl %ebp
+; X86-NEXT: movl %esp, %ebp
+; X86-NEXT: pushl %edi
+; X86-NEXT: pushl %esi
+; X86-NEXT: andl $-16, %esp
+; X86-NEXT: subl $32, %esp
+; X86-NEXT: movzbl 12(%ebp), %eax
+; X86-NEXT: movl 8(%ebp), %ecx
+; X86-NEXT: movl (%ecx), %edx
+; X86-NEXT: movl 4(%ecx), %esi
+; X86-NEXT: movl 8(%ecx), %edi
+; X86-NEXT: movl 12(%ecx), %ecx
+; X86-NEXT: movl %ecx, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %edi, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %esi, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %edx, (%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: andb $96, %al
+; X86-NEXT: shrb $3, %al
+; X86-NEXT: movzbl %al, %eax
+; X86-NEXT: movl (%esp,%eax), %eax
+; X86-NEXT: leal -8(%ebp), %esp
+; X86-NEXT: popl %esi
+; X86-NEXT: popl %edi
+; X86-NEXT: popl %ebp
+; X86-NEXT: retl
+;
+; SSE-LABEL: extractSub128_32:
+; SSE: # %bb.0:
+; SSE-NEXT: movq (%rdi), %rax
+; SSE-NEXT: movq 8(%rdi), %rdx
+; SSE-NEXT: movl %esi, %ecx
+; SSE-NEXT: andb $32, %cl
+; SSE-NEXT: movq %rdx, %rdi
+; SSE-NEXT: shrq %cl, %rdi
+; SSE-NEXT: shrdq %cl, %rdx, %rax
+; SSE-NEXT: testb $64, %sil
+; SSE-NEXT: cmovneq %rdi, %rax
+; SSE-NEXT: # kill: def $eax killed $eax killed $rax
+; SSE-NEXT: retq
+;
+; AVX-LABEL: extractSub128_32:
+; AVX: # %bb.0:
+; AVX-NEXT: movq (%rdi), %rdx
+; AVX-NEXT: movq 8(%rdi), %rax
+; AVX-NEXT: movl %esi, %ecx
+; AVX-NEXT: andb $32, %cl
+; AVX-NEXT: shrdq %cl, %rax, %rdx
+; AVX-NEXT: shrxq %rcx, %rax, %rax
+; AVX-NEXT: testb $64, %sil
+; AVX-NEXT: cmoveq %rdx, %rax
+; AVX-NEXT: # kill: def $eax killed $eax killed $rax
+; AVX-NEXT: retq
+ %idx_bounds = and i32 %idx, 127
+ %idx_align = and i32 %idx_bounds, -32
+ %sh = zext nneg i32 %idx_align to i128
+ %ld = load i128, ptr %word, align 8
+ %sub = lshr i128 %ld, %sh
+ %res = trunc i128 %sub to i32
+ ret i32 %res
+}
+
+define i64 @extractSub128_64(ptr %word, i32 %idx) nounwind {
+; X86-LABEL: extractSub128_64:
+; X86: # %bb.0:
+; X86-NEXT: pushl %ebp
+; X86-NEXT: movl %esp, %ebp
+; X86-NEXT: pushl %edi
+; X86-NEXT: pushl %esi
+; X86-NEXT: andl $-16, %esp
+; X86-NEXT: subl $32, %esp
+; X86-NEXT: movzbl 12(%ebp), %eax
+; X86-NEXT: movl 8(%ebp), %ecx
+; X86-NEXT: movl (%ecx), %edx
+; X86-NEXT: movl 4(%ecx), %esi
+; X86-NEXT: movl 8(%ecx), %edi
+; X86-NEXT: movl 12(%ecx), %ecx
+; X86-NEXT: movl %ecx, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %edi, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %esi, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %edx, (%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: andb $64, %al
+; X86-NEXT: shrb $3, %al
+; X86-NEXT: movzbl %al, %ecx
+; X86-NEXT: movl (%esp,%ecx), %eax
+; X86-NEXT: movl 4(%esp,%ecx), %edx
+; X86-NEXT: leal -8(%ebp), %esp
+; X86-NEXT: popl %esi
+; X86-NEXT: popl %edi
+; X86-NEXT: popl %ebp
+; X86-NEXT: retl
+;
+; X64-LABEL: extractSub128_64:
+; X64: # %bb.0:
+; X64-NEXT: testb $64, %sil
+; X64-NEXT: je .LBB3_1
+; X64-NEXT: # %bb.2:
+; X64-NEXT: movq 8(%rdi), %rax
+; X64-NEXT: retq
+; X64-NEXT: .LBB3_1:
+; X64-NEXT: movq (%rdi), %rax
+; X64-NEXT: retq
+ %idx_bounds = and i32 %idx, 127
+ %idx_align = and i32 %idx_bounds, -64
+ %sh = zext nneg i32 %idx_align to i128
+ %ld = load i128, ptr %word, align 8
+ %sub = lshr i128 %ld, %sh
+ %res = trunc i128 %sub to i64
+ ret i64 %res
+}
+
+define i8 @extractSub512_8(ptr %word, i32 %idx) nounwind {
+; X86-LABEL: extractSub512_8:
+; X86: # %bb.0:
+; X86-NEXT: pushl %ebp
+; X86-NEXT: movl %esp, %ebp
+; X86-NEXT: pushl %ebx
+; X86-NEXT: pushl %edi
+; X86-NEXT: pushl %esi
+; X86-NEXT: andl $-16, %esp
+; X86-NEXT: subl $192, %esp
+; X86-NEXT: movl 8(%ebp), %eax
+; X86-NEXT: movl (%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 4(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 8(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 12(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 16(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 20(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 24(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 28(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 32(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 36(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 40(%eax), %ebx
+; X86-NEXT: movl 44(%eax), %edi
+; X86-NEXT: movl 48(%eax), %esi
+; X86-NEXT: movl 52(%eax), %edx
+; X86-NEXT: movl 56(%eax), %ecx
+; X86-NEXT: movl 60(%eax), %eax
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %ecx, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %edx, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %esi, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %edi, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %ebx, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl 12(%ebp), %edx
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %edx, %ecx
+; X86-NEXT: andl $24, %ecx
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: shrl $3, %edx
+; X86-NEXT: andl $60, %edx
+; X86-NEXT: movl 48(%esp,%edx), %eax
+; X86-NEXT: movl 52(%esp,%edx), %edx
+; X86-NEXT: # kill: def $cl killed $cl killed $ecx
+; X86-NEXT: shrdl %cl, %edx, %eax
+; X86-NEXT: # kill: def $al killed $al killed $eax
+; X86-NEXT: leal -12(%ebp), %esp
+; X86-NEXT: popl %esi
+; X86-NEXT: popl %edi
+; X86-NEXT: popl %ebx
+; X86-NEXT: popl %ebp
+; X86-NEXT: retl
+;
+; SSE-LABEL: extractSub512_8:
+; SSE: # %bb.0:
+; SSE-NEXT: pushq %rax
+; SSE-NEXT: # kill: def $esi killed $esi def $rsi
+; SSE-NEXT: movups (%rdi), %xmm0
+; SSE-NEXT: movups 16(%rdi), %xmm1
+; SSE-NEXT: movups 32(%rdi), %xmm2
+; SSE-NEXT: movups 48(%rdi), %xmm3
+; SSE-NEXT: xorps %xmm4, %xmm4
+; SSE-NEXT: movaps %xmm4, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm4, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm4, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm4, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm3, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm2, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm1, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movl %esi, %ecx
+; SSE-NEXT: andl $56, %ecx
+; SSE-NEXT: shrl $3, %esi
+; SSE-NEXT: andl $56, %esi
+; SSE-NEXT: movq -128(%rsp,%rsi), %rdx
+; SSE-NEXT: shrq %cl, %rdx
+; SSE-NEXT: movl -120(%rsp,%rsi), %eax
+; SSE-NEXT: addl %eax, %eax
+; SSE-NEXT: notl %ecx
+; SSE-NEXT: # kill: def $cl killed $cl killed $ecx
+; SSE-NEXT: shlq %cl, %rax
+; SSE-NEXT: orl %edx, %eax
+; SSE-NEXT: # kill: def $al killed $al killed $rax
+; SSE-NEXT: popq %rcx
+; SSE-NEXT: retq
+;
+; AVX2-LABEL: extractSub512_8:
+; AVX2: # %bb.0:
+; AVX2-NEXT: pushq %rax
+; AVX2-NEXT: # kill: def $esi killed $esi def $rsi
+; AVX2-NEXT: vmovups (%rdi), %ymm0
+; AVX2-NEXT: vmovups 32(%rdi), %ymm1
+; AVX2-NEXT: vxorps %xmm2, %xmm2, %xmm2
+; AVX2-NEXT: vmovups %ymm2, -{{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm2, -{{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm1, -{{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm0, -{{[0-9]+}}(%rsp)
+; AVX2-NEXT: movl %esi, %ecx
+; AVX2-NEXT: andl $56, %ecx
+; AVX2-NEXT: shrl $3, %esi
+; AVX2-NEXT: andl $56, %esi
+; AVX2-NEXT: shrxq %rcx, -128(%rsp,%rsi), %rax
+; AVX2-NEXT: # kill: def $ecx killed $ecx killed $rcx def $rcx
+; AVX2-NEXT: notl %ecx
+; AVX2-NEXT: movl -120(%rsp,%rsi), %edx
+; AVX2-NEXT: addl %edx, %edx
+; AVX2-NEXT: shlxq %rcx, %rdx, %rcx
+; AVX2-NEXT: orl %ecx, %eax
+; AVX2-NEXT: # kill: def $al killed $al killed $rax
+; AVX2-NEXT: popq %rcx
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
+;
+; AVX512-LABEL: extractSub512_8:
+; AVX512: # %bb.0:
+; AVX512-NEXT: pushq %rax
+; AVX512-NEXT: vmovups (%rdi), %ymm0
+; AVX512-NEXT: vmovups 32(%rdi), %ymm1
+; AVX512-NEXT: vxorps %xmm2, %xmm2, %xmm2
+; AVX512-NEXT: vmovups %ymm2, -{{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm2, -{{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm1, -{{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm0, -{{[0-9]+}}(%rsp)
+; AVX512-NEXT: # kill: def $esi killed $esi def $rsi
+; AVX512-NEXT: movl %esi, %ecx
+; AVX512-NEXT: andl $56, %ecx
+; AVX512-NEXT: shrl $3, %esi
+; AVX512-NEXT: andl $56, %esi
+; AVX512-NEXT: shrxq %rcx, -128(%rsp,%rsi), %rax
+; AVX512-NEXT: # kill: def $ecx killed $ecx killed $rcx def $rcx
+; AVX512-NEXT: notl %ecx
+; AVX512-NEXT: movl -120(%rsp,%rsi), %edx
+; AVX512-NEXT: addl %edx, %edx
+; AVX512-NEXT: shlxq %rcx, %rdx, %rcx
+; AVX512-NEXT: orl %ecx, %eax
+; AVX512-NEXT: # kill: def $al killed $al killed $rax
+; AVX512-NEXT: popq %rcx
+; AVX512-NEXT: vzeroupper
+; AVX512-NEXT: retq
+ %idx_bounds = and i32 %idx, 511
+ %idx_align = and i32 %idx_bounds, -8
+ %ld = load i512, ptr %word, align 8
+ %sh = zext nneg i32 %idx_align to i512
+ %sub = lshr i512 %ld, %sh
+ %res = trunc i512 %sub to i8
+ ret i8 %res
+}
+
+define i64 @extractSub512_64(ptr %word, i32 %idx) nounwind {
+; X86-LABEL: extractSub512_64:
+; X86: # %bb.0:
+; X86-NEXT: pushl %ebp
+; X86-NEXT: movl %esp, %ebp
+; X86-NEXT: pushl %ebx
+; X86-NEXT: pushl %edi
+; X86-NEXT: pushl %esi
+; X86-NEXT: andl $-16, %esp
+; X86-NEXT: subl $192, %esp
+; X86-NEXT: movl 8(%ebp), %eax
+; X86-NEXT: movl (%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 4(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 8(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 12(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 16(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 20(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 24(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 28(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 32(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 36(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 40(%eax), %ebx
+; X86-NEXT: movl 44(%eax), %edi
+; X86-NEXT: movl 48(%eax), %esi
+; X86-NEXT: movl 52(%eax), %edx
+; X86-NEXT: movl 56(%eax), %ecx
+; X86-NEXT: movl 60(%eax), %eax
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %ecx, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %edx, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %esi, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %edi, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %ebx, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl 12(%ebp), %ecx
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: shrl $3, %ecx
+; X86-NEXT: andl $56, %ecx
+; X86-NEXT: movl 48(%esp,%ecx), %eax
+; X86-NEXT: movl 52(%esp,%ecx), %edx
+; X86-NEXT: leal -12(%ebp), %esp
+; X86-NEXT: popl %esi
+; X86-NEXT: popl %edi
+; X86-NEXT: popl %ebx
+; X86-NEXT: popl %ebp
+; X86-NEXT: retl
+;
+; SSE-LABEL: extractSub512_64:
+; SSE: # %bb.0:
+; SSE-NEXT: pushq %rax
+; SSE-NEXT: # kill: def $esi killed $esi def $rsi
+; SSE-NEXT: movups (%rdi), %xmm0
+; SSE-NEXT: movups 16(%rdi), %xmm1
+; SSE-NEXT: movups 32(%rdi), %xmm2
+; SSE-NEXT: movups 48(%rdi), %xmm3
+; SSE-NEXT: xorps %xmm4, %xmm4
+; SSE-NEXT: movaps %xmm4, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm4, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm4, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm4, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm3, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm2, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm1, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: shrl $3, %esi
+; SSE-NEXT: andl $56, %esi
+; SSE-NEXT: movq -128(%rsp,%rsi), %rax
+; SSE-NEXT: popq %rcx
+; SSE-NEXT: retq
+;
+; AVX2-LABEL: extractSub512_64:
+; AVX2: # %bb.0:
+; AVX2-NEXT: pushq %rax
+; AVX2-NEXT: vmovups (%rdi), %ymm0
+; AVX2-NEXT: vmovups 32(%rdi), %ymm1
+; AVX2-NEXT: vxorps %xmm2, %xmm2, %xmm2
+; AVX2-NEXT: vmovups %ymm2, -{{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm2, -{{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm1, -{{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm0, -{{[0-9]+}}(%rsp)
+; AVX2-NEXT: # kill: def $esi killed $esi def $rsi
+; AVX2-NEXT: shrl $3, %esi
+; AVX2-NEXT: andl $56, %esi
+; AVX2-NEXT: movq -128(%rsp,%rsi), %rax
+; AVX2-NEXT: popq %rcx
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
+;
+; AVX512-LABEL: extractSub512_64:
+; AVX512: # %bb.0:
+; AVX512-NEXT: pushq %rax
+; AVX512-NEXT: vmovups (%rdi), %ymm0
+; AVX512-NEXT: vmovups 32(%rdi), %ymm1
+; AVX512-NEXT: vxorps %xmm2, %xmm2, %xmm2
+; AVX512-NEXT: vmovups %ymm2, -{{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm2, -{{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm1, -{{[0-9]+}}(%rsp)
+; AVX512-NEXT: # kill: def $esi killed $esi def $rsi
+; AVX512-NEXT: vmovups %ymm0, -{{[0-9]+}}(%rsp)
+; AVX512-NEXT: shrl $3, %esi
+; AVX512-NEXT: andl $56, %esi
+; AVX512-NEXT: movq -128(%rsp,%rsi), %rax
+; AVX512-NEXT: popq %rcx
+; AVX512-NEXT: vzeroupper
+; AVX512-NEXT: retq
+ %idx_bounds = and i32 %idx, 511
+ %idx_align = and i32 %idx_bounds, -64
+ %sh = zext nneg i32 %idx_align to i512
+ %ld = load i512, ptr %word, align 8
+ %sub = lshr i512 %ld, %sh
+ %res = trunc i512 %sub to i64
+ ret i64 %res
+}
+
+define i128 @extractSub512_128(ptr %word, i32 %idx) nounwind {
+; X86-LABEL: extractSub512_128:
+; X86: # %bb.0:
+; X86-NEXT: pushl %ebp
+; X86-NEXT: movl %esp, %ebp
+; X86-NEXT: pushl %ebx
+; X86-NEXT: pushl %edi
+; X86-NEXT: pushl %esi
+; X86-NEXT: andl $-16, %esp
+; X86-NEXT: subl $192, %esp
+; X86-NEXT: movl 12(%ebp), %eax
+; X86-NEXT: movl (%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 4(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 8(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 12(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 16(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 20(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 24(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 28(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 32(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 36(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 40(%eax), %ebx
+; X86-NEXT: movl 44(%eax), %edi
+; X86-NEXT: movl 48(%eax), %esi
+; X86-NEXT: movl 52(%eax), %edx
+; X86-NEXT: movl 56(%eax), %ecx
+; X86-NEXT: movl 60(%eax), %eax
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %ecx, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %edx, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %esi, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %edi, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %ebx, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl 8(%ebp), %eax
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
+; X86-NEXT: movl %ecx, {{[0-9]+}}(%esp)
+; X86-NEXT: movl 16(%ebp), %edi
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
+; X86-NEXT: movl %ecx, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
+; X86-NEXT: movl %ecx, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
+; X86-NEXT: movl %ecx, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: shrl $3, %edi
+; X86-NEXT: andl $48, %edi
+; X86-NEXT: movl 48(%esp,%edi), %ecx
+; X86-NEXT: movl 52(%esp,%edi), %edx
+; X86-NEXT: movl 56(%esp,%edi), %esi
+; X86-NEXT: movl 60(%esp,%edi), %edi
+; X86-NEXT: movl %edi, 12(%eax)
+; X86-NEXT: movl %esi, 8(%eax)
+; X86-NEXT: movl %edx, 4(%eax)
+; X86-NEXT: movl %ecx, (%eax)
+; X86-NEXT: leal -12(%ebp), %esp
+; X86-NEXT: popl %esi
+; X86-NEXT: popl %edi
+; X86-NEXT: popl %ebx
+; X86-NEXT: popl %ebp
+; X86-NEXT: retl $4
+;
+; SSE-LABEL: extractSub512_128:
+; SSE: # %bb.0:
+; SSE-NEXT: pushq %rax
+; SSE-NEXT: # kill: def $esi killed $esi def $rsi
+; SSE-NEXT: movups (%rdi), %xmm0
+; SSE-NEXT: movups 16(%rdi), %xmm1
+; SSE-NEXT: movups 32(%rdi), %xmm2
+; SSE-NEXT: movups 48(%rdi), %xmm3
+; SSE-NEXT: xorps %xmm4, %xmm4
+; SSE-NEXT: movaps %xmm4, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm4, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm4, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm4, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm3, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm2, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm1, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, -{{[0-9]+}}(%rsp)
+; SSE-NEXT: shrl $3, %esi
+; SSE-NEXT: andl $48, %esi
+; SSE-NEXT: movq -128(%rsp,%rsi), %rax
+; SSE-NEXT: movq -120(%rsp,%rsi), %rdx
+; SSE-NEXT: popq %rcx
+; SSE-NEXT: retq
+;
+; AVX-LABEL: extractSub512_128:
+; AVX: # %bb.0:
+; AVX-NEXT: pushq %rax
+; AVX-NEXT: # kill: def $esi killed $esi def $rsi
+; AVX-NEXT: vmovups (%rdi), %ymm0
+; AVX-NEXT: vmovups 32(%rdi), %ymm1
+; AVX-NEXT: vxorps %xmm2, %xmm2, %xmm2
+; AVX-NEXT: vmovups %ymm2, -{{[0-9]+}}(%rsp)
+; AVX-NEXT: vmovups %ymm2, -{{[0-9]+}}(%rsp)
+; AVX-NEXT: vmovups %ymm1, -{{[0-9]+}}(%rsp)
+; AVX-NEXT: vmovups %ymm0, -{{[0-9]+}}(%rsp)
+; AVX-NEXT: shrl $3, %esi
+; AVX-NEXT: andl $48, %esi
+; AVX-NEXT: movq -128(%rsp,%rsi), %rax
+; AVX-NEXT: movq -120(%rsp,%rsi), %rdx
+; AVX-NEXT: popq %rcx
+; AVX-NEXT: vzeroupper
+; AVX-NEXT: retq
+ %idx_bounds = and i32 %idx, 511
+ %idx_align = and i32 %idx_bounds, -128
+ %sh = zext nneg i32 %idx_align to i512
+ %ld = load i512, ptr %word, align 8
+ %sub = lshr i512 %ld, %sh
+ %res = trunc i512 %sub to i128
+ ret i128 %res
+}
+
+define i64 @extractSub4096_64(ptr %word, i32 %idx) nounwind {
+; X86-LABEL: extractSub4096_64:
+; X86: # %bb.0:
+; X86-NEXT: pushl %ebp
+; X86-NEXT: movl %esp, %ebp
+; X86-NEXT: pushl %ebx
+; X86-NEXT: pushl %edi
+; X86-NEXT: pushl %esi
+; X86-NEXT: andl $-16, %esp
+; X86-NEXT: subl $1536, %esp # imm = 0x600
+; X86-NEXT: movl 8(%ebp), %eax
+; X86-NEXT: movl 4(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 8(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 12(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 16(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 20(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 24(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 28(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 32(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 36(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 40(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 44(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 48(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 52(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 56(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 60(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 64(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 68(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 72(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 76(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 80(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 84(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 88(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 92(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 96(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 100(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 104(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 108(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 112(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 116(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 120(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 124(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 128(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 132(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 136(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 140(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 144(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 148(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 152(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 156(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 160(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 164(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 168(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 172(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 176(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 180(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 184(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 188(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 192(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 196(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 200(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 204(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 208(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 212(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 216(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 220(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 224(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 228(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 232(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 236(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 240(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 244(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 248(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 252(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 256(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 260(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 264(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 268(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 272(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 276(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 280(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 284(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 288(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 292(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 296(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 300(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 304(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 308(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 312(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 316(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 320(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 324(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 328(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 332(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 336(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 340(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 344(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 348(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 352(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 356(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 360(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 364(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 368(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 372(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 376(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 380(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl (%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 384(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 388(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 392(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 396(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 400(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 404(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 408(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 412(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 416(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 420(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 424(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 428(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 432(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 436(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 440(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 444(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 448(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 452(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 456(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 460(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 464(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 468(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 472(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 476(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 480(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 484(%eax), %ecx
+; X86-NEXT: movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT: movl 488(%eax), %ebx
+; X86-NEXT: movl 492(%eax), %edi
+; X86-NEXT: movl 496(%eax), %esi
+; X86-NEXT: movl 500(%eax), %edx
+; X86-NEXT: movl 504(%eax), %ecx
+; X86-NEXT: movl 508(%eax), %eax
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %ecx, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %edx, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %esi, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %edi, {{[0-9]+}}(%esp)
+; X86-NEXT: movl %ebx, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $4032, %ecx # imm = 0xFC0
+; X86-NEXT: andl 12(%ebp), %ecx
+; X86-NEXT: movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT: movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: movl $0, {{[0-9]+}}(%esp)
+; X86-NEXT: shrl $3, %ecx
+; X86-NEXT: movl 496(%esp,%ecx), %eax
+; X86-NEXT: movl 500(%esp,%ecx), %edx
+; X86-NEXT: leal -12(%ebp), %esp
+; X86-NEXT: popl %esi
+; X86-NEXT: popl %edi
+; X86-NEXT: popl %ebx
+; X86-NEXT: popl %ebp
+; X86-NEXT: retl
+;
+; SSE-LABEL: extractSub4096_64:
+; SSE: # %bb.0:
+; SSE-NEXT: subq $1176, %rsp # imm = 0x498
+; SSE-NEXT: # kill: def $esi killed $esi def $rsi
+; SSE-NEXT: movups (%rdi), %xmm0
+; SSE-NEXT: movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
+; SSE-NEXT: movups 16(%rdi), %xmm0
+; SSE-NEXT: movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
+; SSE-NEXT: movups 32(%rdi), %xmm0
+; SSE-NEXT: movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
+; SSE-NEXT: movups 48(%rdi), %xmm0
+; SSE-NEXT: movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
+; SSE-NEXT: movups 64(%rdi), %xmm0
+; SSE-NEXT: movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
+; SSE-NEXT: movups 80(%rdi), %xmm0
+; SSE-NEXT: movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
+; SSE-NEXT: movups 96(%rdi), %xmm0
+; SSE-NEXT: movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
+; SSE-NEXT: movups 112(%rdi), %xmm0
+; SSE-NEXT: movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
+; SSE-NEXT: movups 128(%rdi), %xmm0
+; SSE-NEXT: movaps %xmm0, (%rsp) # 16-byte Spill
+; SSE-NEXT: movups 144(%rdi), %xmm0
+; SSE-NEXT: movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
+; SSE-NEXT: movups 160(%rdi), %xmm0
+; SSE-NEXT: movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
+; SSE-NEXT: movups 176(%rdi), %xmm0
+; SSE-NEXT: movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
+; SSE-NEXT: movups 192(%rdi), %xmm0
+; SSE-NEXT: movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
+; SSE-NEXT: movups 208(%rdi), %xmm0
+; SSE-NEXT: movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
+; SSE-NEXT: movups 224(%rdi), %xmm0
+; SSE-NEXT: movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
+; SSE-NEXT: movups 240(%rdi), %xmm0
+; SSE-NEXT: movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
+; SSE-NEXT: movups 256(%rdi), %xmm0
+; SSE-NEXT: movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
+; SSE-NEXT: movups 272(%rdi), %xmm15
+; SSE-NEXT: movups 288(%rdi), %xmm14
+; SSE-NEXT: movups 304(%rdi), %xmm13
+; SSE-NEXT: movups 320(%rdi), %xmm12
+; SSE-NEXT: movups 336(%rdi), %xmm11
+; SSE-NEXT: movups 352(%rdi), %xmm10
+; SSE-NEXT: movups 368(%rdi), %xmm9
+; SSE-NEXT: movups 384(%rdi), %xmm8
+; SSE-NEXT: movups 400(%rdi), %xmm7
+; SSE-NEXT: movups 416(%rdi), %xmm6
+; SSE-NEXT: movups 432(%rdi), %xmm5
+; SSE-NEXT: movups 448(%rdi), %xmm4
+; SSE-NEXT: movups 464(%rdi), %xmm3
+; SSE-NEXT: movups 480(%rdi), %xmm2
+; SSE-NEXT: movups 496(%rdi), %xmm1
+; SSE-NEXT: xorps %xmm0, %xmm0
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm1, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm2, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm3, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm4, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm5, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm6, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm7, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm8, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm9, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm10, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm11, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm12, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm13, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm14, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps %xmm15, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps (%rsp), %xmm0 # 16-byte Reload
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
+; SSE-NEXT: movaps %xmm0, {{[0-9]+}}(%rsp)
+; SSE-NEXT: andl $4032, %esi # imm = 0xFC0
+; SSE-NEXT: shrl $3, %esi
+; SSE-NEXT: movq 144(%rsp,%rsi), %rax
+; SSE-NEXT: addq $1176, %rsp # imm = 0x498
+; SSE-NEXT: retq
+;
+; AVX2-LABEL: extractSub4096_64:
+; AVX2: # %bb.0:
+; AVX2-NEXT: subq $936, %rsp # imm = 0x3A8
+; AVX2-NEXT: vmovups (%rdi), %ymm0
+; AVX2-NEXT: vmovups %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
+; AVX2-NEXT: vmovups 32(%rdi), %ymm1
+; AVX2-NEXT: vmovups 64(%rdi), %ymm2
+; AVX2-NEXT: vmovups 96(%rdi), %ymm3
+; AVX2-NEXT: vmovups 128(%rdi), %ymm4
+; AVX2-NEXT: vmovups 160(%rdi), %ymm5
+; AVX2-NEXT: vmovups 192(%rdi), %ymm6
+; AVX2-NEXT: vmovups 224(%rdi), %ymm7
+; AVX2-NEXT: vmovups 256(%rdi), %ymm8
+; AVX2-NEXT: vmovups 288(%rdi), %ymm9
+; AVX2-NEXT: vmovups 320(%rdi), %ymm10
+; AVX2-NEXT: vmovups 352(%rdi), %ymm11
+; AVX2-NEXT: vmovups 384(%rdi), %ymm12
+; AVX2-NEXT: vmovups 416(%rdi), %ymm13
+; AVX2-NEXT: vmovups 448(%rdi), %ymm14
+; AVX2-NEXT: vmovups 480(%rdi), %ymm15
+; AVX2-NEXT: vxorps %xmm0, %xmm0, %xmm0
+; AVX2-NEXT: vmovups %ymm0, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm0, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm0, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm0, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm0, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm0, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm0, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm0, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm0, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm0, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm0, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm0, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm0, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm0, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm0, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm0, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm15, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm14, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm13, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm12, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm11, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm10, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm9, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm8, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm7, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm6, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm5, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm4, {{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm3, (%rsp)
+; AVX2-NEXT: vmovups %ymm2, -{{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups %ymm1, -{{[0-9]+}}(%rsp)
+; AVX2-NEXT: vmovups {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
+; AVX2-NEXT: vmovups %ymm0, -{{[0-9]+}}(%rsp)
+; AVX2-NEXT: # kill: def $esi killed $esi def $rsi
+; AVX2-NEXT: andl $4032, %esi # imm = 0xFC0
+; AVX2-NEXT: shrl $3, %esi
+; AVX2-NEXT: movq -96(%rsp,%rsi), %rax
+; AVX2-NEXT: addq $936, %rsp # imm = 0x3A8
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
+;
+; AVX512-LABEL: extractSub4096_64:
+; AVX512: # %bb.0:
+; AVX512-NEXT: subq $904, %rsp # imm = 0x388
+; AVX512-NEXT: # kill: def $esi killed $esi def $rsi
+; AVX512-NEXT: vmovups (%rdi), %ymm0
+; AVX512-NEXT: vmovups 32(%rdi), %ymm1
+; AVX512-NEXT: vmovups 64(%rdi), %ymm2
+; AVX512-NEXT: vmovups 96(%rdi), %ymm3
+; AVX512-NEXT: vmovups 128(%rdi), %ymm4
+; AVX512-NEXT: vmovups 160(%rdi), %ymm5
+; AVX512-NEXT: vmovups 192(%rdi), %ymm6
+; AVX512-NEXT: vmovups 224(%rdi), %ymm7
+; AVX512-NEXT: vmovups 256(%rdi), %ymm8
+; AVX512-NEXT: vmovups 288(%rdi), %ymm9
+; AVX512-NEXT: vmovups 320(%rdi), %ymm10
+; AVX512-NEXT: vmovups 352(%rdi), %ymm11
+; AVX512-NEXT: vmovups 384(%rdi), %ymm12
+; AVX512-NEXT: vmovups 416(%rdi), %ymm13
+; AVX512-NEXT: andl $4032, %esi # imm = 0xFC0
+; AVX512-NEXT: vmovups 448(%rdi), %ymm14
+; AVX512-NEXT: vmovups 480(%rdi), %ymm15
+; AVX512-NEXT: vxorps %xmm16, %xmm16, %xmm16
+; AVX512-NEXT: vmovups %ymm16, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm16, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm16, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm16, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm16, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm16, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm16, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm16, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm16, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm16, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm16, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm16, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm16, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm16, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm16, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm16, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm15, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm14, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm13, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm12, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm11, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm10, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm9, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm8, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm7, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm6, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm5, {{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm4, (%rsp)
+; AVX512-NEXT: vmovups %ymm3, -{{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm2, -{{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm1, -{{[0-9]+}}(%rsp)
+; AVX512-NEXT: vmovups %ymm0, -{{[0-9]+}}(%rsp)
+; AVX512-NEXT: shrl $3, %esi
+; AVX512-NEXT: movq -128(%rsp,%rsi), %rax
+; AVX512-NEXT: addq $904, %rsp # imm = 0x388
+; AVX512-NEXT: vzeroupper
+; AVX512-NEXT: retq
+ %idx_bounds = and i32 %idx, 4095
+ %idx_align = and i32 %idx_bounds, -64
+ %sh = zext nneg i32 %idx_align to i4096
+ %ld = load i4096, ptr %word, align 8
+ %sub = lshr i4096 %ld, %sh
+ %res = trunc i4096 %sub to i64
+ ret i64 %res
+}
More information about the llvm-commits
mailing list