[llvm] r333843 - [X86][AVX512] Cleanup intrinsics tests
Simon Pilgrim via llvm-commits
llvm-commits at lists.llvm.org
Sun Jun 3 07:56:04 PDT 2018
Modified: llvm/trunk/test/CodeGen/X86/avx512vl-intrinsics-fast-isel.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/X86/avx512vl-intrinsics-fast-isel.ll?rev=333843&r1=333842&r2=333843&view=diff
==============================================================================
--- llvm/trunk/test/CodeGen/X86/avx512vl-intrinsics-fast-isel.ll (original)
+++ llvm/trunk/test/CodeGen/X86/avx512vl-intrinsics-fast-isel.ll Sun Jun 3 07:56:04 2018
@@ -1,16 +1,16 @@
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc < %s -fast-isel -mtriple=i386-unknown-unknown -mattr=+avx512f,+avx512vl | FileCheck %s --check-prefix=ALL --check-prefix=X32
-; RUN: llc < %s -fast-isel -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512vl | FileCheck %s --check-prefix=ALL --check-prefix=X64
+; RUN: llc < %s -fast-isel -mtriple=i386-unknown-unknown -mattr=+avx512f,+avx512vl | FileCheck %s --check-prefixes=CHECK,X86
+; RUN: llc < %s -fast-isel -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512vl | FileCheck %s --check-prefixes=CHECK,X64
; NOTE: This should use IR equivalent to what is generated by clang/test/CodeGen/avx512vl-builtins.c
define <4 x float> @test_mm_mask_cvtepi32_ps(<4 x float> %__W, i8 zeroext %__U, <2 x i64> %__A) {
-; X32-LABEL: test_mm_mask_cvtepi32_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtdq2ps %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_cvtepi32_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtdq2ps %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_cvtepi32_ps:
; X64: # %bb.0: # %entry
@@ -27,11 +27,11 @@ entry:
}
define <4 x float> @test_mm_maskz_cvtepi32_ps(i16 zeroext %__U, <2 x i64> %__A) {
-; X32-LABEL: test_mm_maskz_cvtepi32_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: kmovw {{[0-9]+}}(%esp), %k1
-; X32-NEXT: vcvtdq2ps %xmm0, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_cvtepi32_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: kmovw {{[0-9]+}}(%esp), %k1
+; X86-NEXT: vcvtdq2ps %xmm0, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_cvtepi32_ps:
; X64: # %bb.0: # %entry
@@ -50,12 +50,12 @@ entry:
}
define <8 x float> @test_mm256_mask_cvtepi32_ps(<8 x float> %__W, i8 zeroext %__U, <4 x i64> %__A) {
-; X32-LABEL: test_mm256_mask_cvtepi32_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtdq2ps %ymm1, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_cvtepi32_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtdq2ps %ymm1, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_cvtepi32_ps:
; X64: # %bb.0: # %entry
@@ -71,11 +71,11 @@ entry:
}
define <8 x float> @test_mm256_maskz_cvtepi32_ps(i16 zeroext %__U, <4 x i64> %__A) {
-; X32-LABEL: test_mm256_maskz_cvtepi32_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: kmovw {{[0-9]+}}(%esp), %k1
-; X32-NEXT: vcvtdq2ps %ymm0, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_cvtepi32_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: kmovw {{[0-9]+}}(%esp), %k1
+; X86-NEXT: vcvtdq2ps %ymm0, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_cvtepi32_ps:
; X64: # %bb.0: # %entry
@@ -92,12 +92,12 @@ entry:
}
define <2 x i64> @test_mm_mask_cvtpd_epi32(<2 x i64> %__W, i8 zeroext %__U, <2 x double> %__A) {
-; X32-LABEL: test_mm_mask_cvtpd_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtpd2dq %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_cvtpd_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtpd2dq %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_cvtpd_epi32:
; X64: # %bb.0: # %entry
@@ -112,12 +112,12 @@ entry:
}
define <2 x i64> @test_mm_maskz_cvtpd_epi32(i8 zeroext %__U, <2 x double> %__A) {
-; X32-LABEL: test_mm_maskz_cvtpd_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtpd2dq %xmm0, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_cvtpd_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtpd2dq %xmm0, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_cvtpd_epi32:
; X64: # %bb.0: # %entry
@@ -131,13 +131,13 @@ entry:
}
define <2 x i64> @test_mm256_mask_cvtpd_epi32(<2 x i64> %__W, i8 zeroext %__U, <4 x double> %__A) {
-; X32-LABEL: test_mm256_mask_cvtpd_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtpd2dq %ymm1, %xmm0 {%k1}
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_cvtpd_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtpd2dq %ymm1, %xmm0 {%k1}
+; X86-NEXT: vzeroupper
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_cvtpd_epi32:
; X64: # %bb.0: # %entry
@@ -156,13 +156,13 @@ entry:
}
define <2 x i64> @test_mm256_maskz_cvtpd_epi32(i8 zeroext %__U, <4 x double> %__A) {
-; X32-LABEL: test_mm256_maskz_cvtpd_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtpd2dq %ymm0, %xmm0 {%k1} {z}
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_cvtpd_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtpd2dq %ymm0, %xmm0 {%k1} {z}
+; X86-NEXT: vzeroupper
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_cvtpd_epi32:
; X64: # %bb.0: # %entry
@@ -180,12 +180,12 @@ entry:
}
define <4 x float> @test_mm_mask_cvtpd_ps(<4 x float> %__W, i8 zeroext %__U, <2 x double> %__A) {
-; X32-LABEL: test_mm_mask_cvtpd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtpd2ps %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_cvtpd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtpd2ps %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_cvtpd_ps:
; X64: # %bb.0: # %entry
@@ -198,12 +198,12 @@ entry:
}
define <4 x float> @test_mm_maskz_cvtpd_ps(i8 zeroext %__U, <2 x double> %__A) {
-; X32-LABEL: test_mm_maskz_cvtpd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtpd2ps %xmm0, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_cvtpd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtpd2ps %xmm0, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_cvtpd_ps:
; X64: # %bb.0: # %entry
@@ -216,13 +216,13 @@ entry:
}
define <4 x float> @test_mm256_mask_cvtpd_ps(<4 x float> %__W, i8 zeroext %__U, <4 x double> %__A) {
-; X32-LABEL: test_mm256_mask_cvtpd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtpd2ps %ymm1, %xmm0 {%k1}
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_cvtpd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtpd2ps %ymm1, %xmm0 {%k1}
+; X86-NEXT: vzeroupper
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_cvtpd_ps:
; X64: # %bb.0: # %entry
@@ -239,13 +239,13 @@ entry:
}
define <4 x float> @test_mm256_maskz_cvtpd_ps(i8 zeroext %__U, <4 x double> %__A) {
-; X32-LABEL: test_mm256_maskz_cvtpd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtpd2ps %ymm0, %xmm0 {%k1} {z}
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_cvtpd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtpd2ps %ymm0, %xmm0 {%k1} {z}
+; X86-NEXT: vzeroupper
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_cvtpd_ps:
; X64: # %bb.0: # %entry
@@ -262,15 +262,10 @@ entry:
}
define <2 x i64> @test_mm_cvtpd_epu32(<2 x double> %__A) {
-; X32-LABEL: test_mm_cvtpd_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vcvtpd2udq %xmm0, %xmm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_cvtpd_epu32:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vcvtpd2udq %xmm0, %xmm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_cvtpd_epu32:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vcvtpd2udq %xmm0, %xmm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = tail call <4 x i32> @llvm.x86.avx512.mask.cvtpd2udq.128(<2 x double> %__A, <4 x i32> zeroinitializer, i8 -1) #8
%1 = bitcast <4 x i32> %0 to <2 x i64>
@@ -278,12 +273,12 @@ entry:
}
define <2 x i64> @test_mm_mask_cvtpd_epu32(<2 x i64> %__W, i8 zeroext %__U, <2 x double> %__A) {
-; X32-LABEL: test_mm_mask_cvtpd_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtpd2udq %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_cvtpd_epu32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtpd2udq %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_cvtpd_epu32:
; X64: # %bb.0: # %entry
@@ -298,12 +293,12 @@ entry:
}
define <2 x i64> @test_mm_maskz_cvtpd_epu32(i8 zeroext %__U, <2 x double> %__A) {
-; X32-LABEL: test_mm_maskz_cvtpd_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtpd2udq %xmm0, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_cvtpd_epu32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtpd2udq %xmm0, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_cvtpd_epu32:
; X64: # %bb.0: # %entry
@@ -317,17 +312,11 @@ entry:
}
define <2 x i64> @test_mm256_cvtpd_epu32(<4 x double> %__A) {
-; X32-LABEL: test_mm256_cvtpd_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vcvtpd2udq %ymm0, %xmm0
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_cvtpd_epu32:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vcvtpd2udq %ymm0, %xmm0
-; X64-NEXT: vzeroupper
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_cvtpd_epu32:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vcvtpd2udq %ymm0, %xmm0
+; CHECK-NEXT: vzeroupper
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = tail call <4 x i32> @llvm.x86.avx512.mask.cvtpd2udq.256(<4 x double> %__A, <4 x i32> zeroinitializer, i8 -1) #8
%1 = bitcast <4 x i32> %0 to <2 x i64>
@@ -335,13 +324,13 @@ entry:
}
define <2 x i64> @test_mm256_mask_cvtpd_epu32(<2 x i64> %__W, i8 zeroext %__U, <4 x double> %__A) {
-; X32-LABEL: test_mm256_mask_cvtpd_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtpd2udq %ymm1, %xmm0 {%k1}
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_cvtpd_epu32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtpd2udq %ymm1, %xmm0 {%k1}
+; X86-NEXT: vzeroupper
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_cvtpd_epu32:
; X64: # %bb.0: # %entry
@@ -357,13 +346,13 @@ entry:
}
define <2 x i64> @test_mm256_maskz_cvtpd_epu32(i8 zeroext %__U, <4 x double> %__A) {
-; X32-LABEL: test_mm256_maskz_cvtpd_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtpd2udq %ymm0, %xmm0 {%k1} {z}
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_cvtpd_epu32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtpd2udq %ymm0, %xmm0 {%k1} {z}
+; X86-NEXT: vzeroupper
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_cvtpd_epu32:
; X64: # %bb.0: # %entry
@@ -378,12 +367,12 @@ entry:
}
define <2 x i64> @test_mm_mask_cvtps_epi32(<2 x i64> %__W, i8 zeroext %__U, <4 x float> %__A) {
-; X32-LABEL: test_mm_mask_cvtps_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtps2dq %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_cvtps_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtps2dq %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_cvtps_epi32:
; X64: # %bb.0: # %entry
@@ -401,12 +390,12 @@ entry:
}
define <2 x i64> @test_mm_maskz_cvtps_epi32(i8 zeroext %__U, <4 x float> %__A) {
-; X32-LABEL: test_mm_maskz_cvtps_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtps2dq %xmm0, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_cvtps_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtps2dq %xmm0, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_cvtps_epi32:
; X64: # %bb.0: # %entry
@@ -423,12 +412,12 @@ entry:
}
define <4 x i64> @test_mm256_mask_cvtps_epi32(<4 x i64> %__W, i8 zeroext %__U, <8 x float> %__A) {
-; X32-LABEL: test_mm256_mask_cvtps_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtps2dq %ymm1, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_cvtps_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtps2dq %ymm1, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_cvtps_epi32:
; X64: # %bb.0: # %entry
@@ -445,12 +434,12 @@ entry:
}
define <4 x i64> @test_mm256_maskz_cvtps_epi32(i8 zeroext %__U, <8 x float> %__A) {
-; X32-LABEL: test_mm256_maskz_cvtps_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtps2dq %ymm0, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_cvtps_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtps2dq %ymm0, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_cvtps_epi32:
; X64: # %bb.0: # %entry
@@ -466,12 +455,12 @@ entry:
}
define <2 x double> @test_mm_mask_cvtps_pd(<2 x double> %__W, i8 zeroext %__U, <4 x float> %__A) local_unnamed_addr #0 {
-; X32-LABEL: test_mm_mask_cvtps_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtps2pd %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_cvtps_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtps2pd %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_cvtps_pd:
; X64: # %bb.0: # %entry
@@ -488,12 +477,12 @@ entry:
}
define <2 x double> @test_mm_maskz_cvtps_pd(i8 zeroext %__U, <4 x float> %__A) local_unnamed_addr #0 {
-; X32-LABEL: test_mm_maskz_cvtps_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtps2pd %xmm0, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_cvtps_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtps2pd %xmm0, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_cvtps_pd:
; X64: # %bb.0: # %entry
@@ -510,12 +499,12 @@ entry:
}
define <4 x double> @test_mm256_mask_cvtps_pd(<4 x double> %__W, i8 zeroext %__U, <4 x float> %__A) local_unnamed_addr #0 {
-; X32-LABEL: test_mm256_mask_cvtps_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtps2pd %xmm1, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_cvtps_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtps2pd %xmm1, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_cvtps_pd:
; X64: # %bb.0: # %entry
@@ -531,12 +520,12 @@ entry:
}
define <4 x double> @test_mm256_maskz_cvtps_pd(i8 zeroext %__U, <4 x float> %__A) local_unnamed_addr #0 {
-; X32-LABEL: test_mm256_maskz_cvtps_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtps2pd %xmm0, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_cvtps_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtps2pd %xmm0, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_cvtps_pd:
; X64: # %bb.0: # %entry
@@ -552,15 +541,10 @@ entry:
}
define <2 x i64> @test_mm_cvtps_epu32(<4 x float> %__A) {
-; X32-LABEL: test_mm_cvtps_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vcvtps2udq %xmm0, %xmm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_cvtps_epu32:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vcvtps2udq %xmm0, %xmm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_cvtps_epu32:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vcvtps2udq %xmm0, %xmm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = tail call <4 x i32> @llvm.x86.avx512.mask.cvtps2udq.128(<4 x float> %__A, <4 x i32> zeroinitializer, i8 -1) #8
%1 = bitcast <4 x i32> %0 to <2 x i64>
@@ -568,12 +552,12 @@ entry:
}
define <2 x i64> @test_mm_mask_cvtps_epu32(<2 x i64> %__W, i8 zeroext %__U, <4 x float> %__A) {
-; X32-LABEL: test_mm_mask_cvtps_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtps2udq %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_cvtps_epu32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtps2udq %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_cvtps_epu32:
; X64: # %bb.0: # %entry
@@ -588,12 +572,12 @@ entry:
}
define <2 x i64> @test_mm_maskz_cvtps_epu32(i8 zeroext %__U, <4 x float> %__A) {
-; X32-LABEL: test_mm_maskz_cvtps_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtps2udq %xmm0, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_cvtps_epu32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtps2udq %xmm0, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_cvtps_epu32:
; X64: # %bb.0: # %entry
@@ -607,15 +591,10 @@ entry:
}
define <4 x i64> @test_mm256_cvtps_epu32(<8 x float> %__A) {
-; X32-LABEL: test_mm256_cvtps_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vcvtps2udq %ymm0, %ymm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_cvtps_epu32:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vcvtps2udq %ymm0, %ymm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_cvtps_epu32:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vcvtps2udq %ymm0, %ymm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = tail call <8 x i32> @llvm.x86.avx512.mask.cvtps2udq.256(<8 x float> %__A, <8 x i32> zeroinitializer, i8 -1) #8
%1 = bitcast <8 x i32> %0 to <4 x i64>
@@ -623,12 +602,12 @@ entry:
}
define <4 x i64> @test_mm256_mask_cvtps_epu32(<4 x i64> %__W, i8 zeroext %__U, <8 x float> %__A) {
-; X32-LABEL: test_mm256_mask_cvtps_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtps2udq %ymm1, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_cvtps_epu32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtps2udq %ymm1, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_cvtps_epu32:
; X64: # %bb.0: # %entry
@@ -643,12 +622,12 @@ entry:
}
define <4 x i64> @test_mm256_maskz_cvtps_epu32(i8 zeroext %__U, <8 x float> %__A) {
-; X32-LABEL: test_mm256_maskz_cvtps_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtps2udq %ymm0, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_cvtps_epu32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtps2udq %ymm0, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_cvtps_epu32:
; X64: # %bb.0: # %entry
@@ -662,12 +641,12 @@ entry:
}
define <2 x i64> @test_mm_mask_cvttpd_epi32(<2 x i64> %__W, i8 zeroext %__U, <2 x double> %__A) {
-; X32-LABEL: test_mm_mask_cvttpd_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvttpd2dq %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_cvttpd_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvttpd2dq %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_cvttpd_epi32:
; X64: # %bb.0: # %entry
@@ -682,12 +661,12 @@ entry:
}
define <2 x i64> @test_mm_maskz_cvttpd_epi32(i8 zeroext %__U, <2 x double> %__A) {
-; X32-LABEL: test_mm_maskz_cvttpd_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvttpd2dq %xmm0, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_cvttpd_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvttpd2dq %xmm0, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_cvttpd_epi32:
; X64: # %bb.0: # %entry
@@ -701,13 +680,13 @@ entry:
}
define <2 x i64> @test_mm256_mask_cvttpd_epi32(<2 x i64> %__W, i8 zeroext %__U, <4 x double> %__A) {
-; X32-LABEL: test_mm256_mask_cvttpd_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvttpd2dq %ymm1, %xmm0 {%k1}
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_cvttpd_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvttpd2dq %ymm1, %xmm0 {%k1}
+; X86-NEXT: vzeroupper
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_cvttpd_epi32:
; X64: # %bb.0: # %entry
@@ -726,13 +705,13 @@ entry:
}
define <2 x i64> @test_mm256_maskz_cvttpd_epi32(i8 zeroext %__U, <4 x double> %__A) {
-; X32-LABEL: test_mm256_maskz_cvttpd_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvttpd2dq %ymm0, %xmm0 {%k1} {z}
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_cvttpd_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvttpd2dq %ymm0, %xmm0 {%k1} {z}
+; X86-NEXT: vzeroupper
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_cvttpd_epi32:
; X64: # %bb.0: # %entry
@@ -750,15 +729,10 @@ entry:
}
define <2 x i64> @test_mm_cvttpd_epu32(<2 x double> %__A) {
-; X32-LABEL: test_mm_cvttpd_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vcvttpd2udq %xmm0, %xmm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_cvttpd_epu32:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vcvttpd2udq %xmm0, %xmm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_cvttpd_epu32:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vcvttpd2udq %xmm0, %xmm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = tail call <4 x i32> @llvm.x86.avx512.mask.cvttpd2udq.128(<2 x double> %__A, <4 x i32> zeroinitializer, i8 -1) #8
%1 = bitcast <4 x i32> %0 to <2 x i64>
@@ -766,12 +740,12 @@ entry:
}
define <2 x i64> @test_mm_mask_cvttpd_epu32(<2 x i64> %__W, i8 zeroext %__U, <2 x double> %__A) {
-; X32-LABEL: test_mm_mask_cvttpd_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvttpd2udq %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_cvttpd_epu32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvttpd2udq %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_cvttpd_epu32:
; X64: # %bb.0: # %entry
@@ -786,12 +760,12 @@ entry:
}
define <2 x i64> @test_mm_maskz_cvttpd_epu32(i8 zeroext %__U, <2 x double> %__A) {
-; X32-LABEL: test_mm_maskz_cvttpd_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvttpd2udq %xmm0, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_cvttpd_epu32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvttpd2udq %xmm0, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_cvttpd_epu32:
; X64: # %bb.0: # %entry
@@ -805,17 +779,11 @@ entry:
}
define <2 x i64> @test_mm256_cvttpd_epu32(<4 x double> %__A) {
-; X32-LABEL: test_mm256_cvttpd_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vcvttpd2udq %ymm0, %xmm0
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_cvttpd_epu32:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vcvttpd2udq %ymm0, %xmm0
-; X64-NEXT: vzeroupper
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_cvttpd_epu32:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vcvttpd2udq %ymm0, %xmm0
+; CHECK-NEXT: vzeroupper
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = tail call <4 x i32> @llvm.x86.avx512.mask.cvttpd2udq.256(<4 x double> %__A, <4 x i32> zeroinitializer, i8 -1) #8
%1 = bitcast <4 x i32> %0 to <2 x i64>
@@ -823,13 +791,13 @@ entry:
}
define <2 x i64> @test_mm256_mask_cvttpd_epu32(<2 x i64> %__W, i8 zeroext %__U, <4 x double> %__A) {
-; X32-LABEL: test_mm256_mask_cvttpd_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvttpd2udq %ymm1, %xmm0 {%k1}
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_cvttpd_epu32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvttpd2udq %ymm1, %xmm0 {%k1}
+; X86-NEXT: vzeroupper
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_cvttpd_epu32:
; X64: # %bb.0: # %entry
@@ -845,13 +813,13 @@ entry:
}
define <2 x i64> @test_mm256_maskz_cvttpd_epu32(i8 zeroext %__U, <4 x double> %__A) {
-; X32-LABEL: test_mm256_maskz_cvttpd_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvttpd2udq %ymm0, %xmm0 {%k1} {z}
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_cvttpd_epu32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvttpd2udq %ymm0, %xmm0 {%k1} {z}
+; X86-NEXT: vzeroupper
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_cvttpd_epu32:
; X64: # %bb.0: # %entry
@@ -866,12 +834,12 @@ entry:
}
define <2 x i64> @test_mm_mask_cvttps_epi32(<2 x i64> %__W, i8 zeroext %__U, <4 x float> %__A) {
-; X32-LABEL: test_mm_mask_cvttps_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvttps2dq %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_cvttps_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvttps2dq %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_cvttps_epi32:
; X64: # %bb.0: # %entry
@@ -889,12 +857,12 @@ entry:
}
define <2 x i64> @test_mm_maskz_cvttps_epi32(i8 zeroext %__U, <4 x float> %__A) {
-; X32-LABEL: test_mm_maskz_cvttps_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvttps2dq %xmm0, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_cvttps_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvttps2dq %xmm0, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_cvttps_epi32:
; X64: # %bb.0: # %entry
@@ -911,12 +879,12 @@ entry:
}
define <4 x i64> @test_mm256_mask_cvttps_epi32(<4 x i64> %__W, i8 zeroext %__U, <8 x float> %__A) {
-; X32-LABEL: test_mm256_mask_cvttps_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvttps2dq %ymm1, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_cvttps_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvttps2dq %ymm1, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_cvttps_epi32:
; X64: # %bb.0: # %entry
@@ -933,12 +901,12 @@ entry:
}
define <4 x i64> @test_mm256_maskz_cvttps_epi32(i8 zeroext %__U, <8 x float> %__A) {
-; X32-LABEL: test_mm256_maskz_cvttps_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvttps2dq %ymm0, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_cvttps_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvttps2dq %ymm0, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_cvttps_epi32:
; X64: # %bb.0: # %entry
@@ -954,15 +922,10 @@ entry:
}
define <2 x i64> @test_mm_cvttps_epu32(<4 x float> %__A) {
-; X32-LABEL: test_mm_cvttps_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vcvttps2udq %xmm0, %xmm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_cvttps_epu32:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vcvttps2udq %xmm0, %xmm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_cvttps_epu32:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vcvttps2udq %xmm0, %xmm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = tail call <4 x i32> @llvm.x86.avx512.mask.cvttps2udq.128(<4 x float> %__A, <4 x i32> zeroinitializer, i8 -1) #8
%1 = bitcast <4 x i32> %0 to <2 x i64>
@@ -970,12 +933,12 @@ entry:
}
define <2 x i64> @test_mm_mask_cvttps_epu32(<2 x i64> %__W, i8 zeroext %__U, <4 x float> %__A) {
-; X32-LABEL: test_mm_mask_cvttps_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvttps2udq %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_cvttps_epu32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvttps2udq %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_cvttps_epu32:
; X64: # %bb.0: # %entry
@@ -990,12 +953,12 @@ entry:
}
define <2 x i64> @test_mm_maskz_cvttps_epu32(i8 zeroext %__U, <4 x float> %__A) {
-; X32-LABEL: test_mm_maskz_cvttps_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvttps2udq %xmm0, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_cvttps_epu32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvttps2udq %xmm0, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_cvttps_epu32:
; X64: # %bb.0: # %entry
@@ -1009,15 +972,10 @@ entry:
}
define <4 x i64> @test_mm256_cvttps_epu32(<8 x float> %__A) {
-; X32-LABEL: test_mm256_cvttps_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vcvttps2udq %ymm0, %ymm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_cvttps_epu32:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vcvttps2udq %ymm0, %ymm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_cvttps_epu32:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vcvttps2udq %ymm0, %ymm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = tail call <8 x i32> @llvm.x86.avx512.mask.cvttps2udq.256(<8 x float> %__A, <8 x i32> zeroinitializer, i8 -1) #8
%1 = bitcast <8 x i32> %0 to <4 x i64>
@@ -1025,12 +983,12 @@ entry:
}
define <4 x i64> @test_mm256_mask_cvttps_epu32(<4 x i64> %__W, i8 zeroext %__U, <8 x float> %__A) {
-; X32-LABEL: test_mm256_mask_cvttps_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvttps2udq %ymm1, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_cvttps_epu32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvttps2udq %ymm1, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_cvttps_epu32:
; X64: # %bb.0: # %entry
@@ -1045,12 +1003,12 @@ entry:
}
define <4 x i64> @test_mm256_maskz_cvttps_epu32(i8 zeroext %__U, <8 x float> %__A) {
-; X32-LABEL: test_mm256_maskz_cvttps_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvttps2udq %ymm0, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_cvttps_epu32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvttps2udq %ymm0, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_cvttps_epu32:
; X64: # %bb.0: # %entry
@@ -1064,15 +1022,10 @@ entry:
}
define <2 x double> @test_mm_cvtepu32_pd(<2 x i64> %__A) local_unnamed_addr #0 {
-; X32-LABEL: test_mm_cvtepu32_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vcvtudq2pd %xmm0, %xmm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_cvtepu32_pd:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vcvtudq2pd %xmm0, %xmm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_cvtepu32_pd:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vcvtudq2pd %xmm0, %xmm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = bitcast <2 x i64> %__A to <4 x i32>
%shuffle.i = shufflevector <4 x i32> %0, <4 x i32> undef, <2 x i32> <i32 0, i32 1>
@@ -1081,12 +1034,12 @@ entry:
}
define <2 x double> @test_mm_mask_cvtepu32_pd(<2 x double> %__W, i8 zeroext %__U, <2 x i64> %__A) local_unnamed_addr #0 {
-; X32-LABEL: test_mm_mask_cvtepu32_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtudq2pd %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_cvtepu32_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtudq2pd %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_cvtepu32_pd:
; X64: # %bb.0: # %entry
@@ -1104,12 +1057,12 @@ entry:
}
define <2 x double> @test_mm_maskz_cvtepu32_pd(i8 zeroext %__U, <2 x i64> %__A) local_unnamed_addr #0 {
-; X32-LABEL: test_mm_maskz_cvtepu32_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtudq2pd %xmm0, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_cvtepu32_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtudq2pd %xmm0, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_cvtepu32_pd:
; X64: # %bb.0: # %entry
@@ -1127,15 +1080,10 @@ entry:
}
define <4 x double> @test_mm256_cvtepu32_pd(<2 x i64> %__A) local_unnamed_addr #0 {
-; X32-LABEL: test_mm256_cvtepu32_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vcvtudq2pd %xmm0, %ymm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_cvtepu32_pd:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vcvtudq2pd %xmm0, %ymm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_cvtepu32_pd:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vcvtudq2pd %xmm0, %ymm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = bitcast <2 x i64> %__A to <4 x i32>
%conv.i = uitofp <4 x i32> %0 to <4 x double>
@@ -1143,12 +1091,12 @@ entry:
}
define <4 x double> @test_mm256_mask_cvtepu32_pd(<4 x double> %__W, i8 zeroext %__U, <2 x i64> %__A) local_unnamed_addr #0 {
-; X32-LABEL: test_mm256_mask_cvtepu32_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtudq2pd %xmm1, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_cvtepu32_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtudq2pd %xmm1, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_cvtepu32_pd:
; X64: # %bb.0: # %entry
@@ -1165,12 +1113,12 @@ entry:
}
define <4 x double> @test_mm256_maskz_cvtepu32_pd(i8 zeroext %__U, <2 x i64> %__A) local_unnamed_addr #0 {
-; X32-LABEL: test_mm256_maskz_cvtepu32_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtudq2pd %xmm0, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_cvtepu32_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtudq2pd %xmm0, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_cvtepu32_pd:
; X64: # %bb.0: # %entry
@@ -1187,15 +1135,10 @@ entry:
}
define <4 x float> @test_mm_cvtepu32_ps(<2 x i64> %__A) {
-; X32-LABEL: test_mm_cvtepu32_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vcvtudq2ps %xmm0, %xmm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_cvtepu32_ps:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vcvtudq2ps %xmm0, %xmm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_cvtepu32_ps:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vcvtudq2ps %xmm0, %xmm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = bitcast <2 x i64> %__A to <4 x i32>
%conv.i = uitofp <4 x i32> %0 to <4 x float>
@@ -1203,12 +1146,12 @@ entry:
}
define <4 x float> @test_mm_mask_cvtepu32_ps(<4 x float> %__W, i8 zeroext %__U, <2 x i64> %__A) {
-; X32-LABEL: test_mm_mask_cvtepu32_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtudq2ps %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_cvtepu32_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtudq2ps %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_cvtepu32_ps:
; X64: # %bb.0: # %entry
@@ -1225,12 +1168,12 @@ entry:
}
define <4 x float> @test_mm_maskz_cvtepu32_ps(i8 zeroext %__U, <2 x i64> %__A) {
-; X32-LABEL: test_mm_maskz_cvtepu32_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtudq2ps %xmm0, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_cvtepu32_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtudq2ps %xmm0, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_cvtepu32_ps:
; X64: # %bb.0: # %entry
@@ -1247,15 +1190,10 @@ entry:
}
define <8 x float> @test_mm256_cvtepu32_ps(<4 x i64> %__A) {
-; X32-LABEL: test_mm256_cvtepu32_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vcvtudq2ps %ymm0, %ymm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_cvtepu32_ps:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vcvtudq2ps %ymm0, %ymm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_cvtepu32_ps:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vcvtudq2ps %ymm0, %ymm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = bitcast <4 x i64> %__A to <8 x i32>
%conv.i = uitofp <8 x i32> %0 to <8 x float>
@@ -1263,12 +1201,12 @@ entry:
}
define <8 x float> @test_mm256_mask_cvtepu32_ps(<8 x float> %__W, i8 zeroext %__U, <4 x i64> %__A) {
-; X32-LABEL: test_mm256_mask_cvtepu32_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtudq2ps %ymm1, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_cvtepu32_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtudq2ps %ymm1, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_cvtepu32_ps:
; X64: # %bb.0: # %entry
@@ -1284,12 +1222,12 @@ entry:
}
define <8 x float> @test_mm256_maskz_cvtepu32_ps(i8 zeroext %__U, <4 x i64> %__A) {
-; X32-LABEL: test_mm256_maskz_cvtepu32_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vcvtudq2ps %ymm0, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_cvtepu32_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vcvtudq2ps %ymm0, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_cvtepu32_ps:
; X64: # %bb.0: # %entry
@@ -1305,27 +1243,22 @@ entry:
}
define <8 x float> @test_mm256_shuffle_f32x4(<8 x float> %__A, <8 x float> %__B) {
-; X32-LABEL: test_mm256_shuffle_f32x4:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vperm2f128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[2,3]
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_shuffle_f32x4:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vperm2f128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[2,3]
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_shuffle_f32x4:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vperm2f128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[2,3]
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%shuffle = shufflevector <8 x float> %__A, <8 x float> %__B, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 12, i32 13, i32 14, i32 15>
ret <8 x float> %shuffle
}
define <8 x float> @test_mm256_mask_shuffle_f32x4(<8 x float> %__W, i8 zeroext %__U, <8 x float> %__A, <8 x float> %__B) {
-; X32-LABEL: test_mm256_mask_shuffle_f32x4:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vshuff32x4 {{.*#+}} ymm0 {%k1} = ymm1[4,5,6,7],ymm2[4,5,6,7]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_shuffle_f32x4:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vshuff32x4 {{.*#+}} ymm0 {%k1} = ymm1[4,5,6,7],ymm2[4,5,6,7]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_shuffle_f32x4:
; X64: # %bb.0: # %entry
@@ -1340,12 +1273,12 @@ entry:
}
define <8 x float> @test_mm256_maskz_shuffle_f32x4(i8 zeroext %__U, <8 x float> %__A, <8 x float> %__B) {
-; X32-LABEL: test_mm256_maskz_shuffle_f32x4:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vshuff32x4 {{.*#+}} ymm0 {%k1} {z} = ymm0[4,5,6,7],ymm1[4,5,6,7]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_shuffle_f32x4:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vshuff32x4 {{.*#+}} ymm0 {%k1} {z} = ymm0[4,5,6,7],ymm1[4,5,6,7]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_shuffle_f32x4:
; X64: # %bb.0: # %entry
@@ -1360,27 +1293,22 @@ entry:
}
define <4 x double> @test_mm256_shuffle_f64x2(<4 x double> %__A, <4 x double> %__B) {
-; X32-LABEL: test_mm256_shuffle_f64x2:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vperm2f128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[2,3]
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_shuffle_f64x2:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vperm2f128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[2,3]
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_shuffle_f64x2:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vperm2f128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[2,3]
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%shuffle = shufflevector <4 x double> %__A, <4 x double> %__B, <4 x i32> <i32 2, i32 3, i32 6, i32 7>
ret <4 x double> %shuffle
}
define <4 x double> @test_mm256_mask_shuffle_f64x2(<4 x double> %__W, i8 zeroext %__U, <4 x double> %__A, <4 x double> %__B) {
-; X32-LABEL: test_mm256_mask_shuffle_f64x2:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vshuff64x2 {{.*#+}} ymm0 {%k1} = ymm1[2,3],ymm2[2,3]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_shuffle_f64x2:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vshuff64x2 {{.*#+}} ymm0 {%k1} = ymm1[2,3],ymm2[2,3]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_shuffle_f64x2:
; X64: # %bb.0: # %entry
@@ -1396,12 +1324,12 @@ entry:
}
define <4 x double> @test_mm256_maskz_shuffle_f64x2(i8 zeroext %__U, <4 x double> %__A, <4 x double> %__B) {
-; X32-LABEL: test_mm256_maskz_shuffle_f64x2:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vshuff64x2 {{.*#+}} ymm0 {%k1} {z} = ymm0[2,3],ymm1[2,3]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_shuffle_f64x2:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vshuff64x2 {{.*#+}} ymm0 {%k1} {z} = ymm0[2,3],ymm1[2,3]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_shuffle_f64x2:
; X64: # %bb.0: # %entry
@@ -1417,27 +1345,22 @@ entry:
}
define <4 x i64> @test_mm256_shuffle_i32x4(<4 x i64> %__A, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_shuffle_i32x4:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vperm2i128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[2,3]
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_shuffle_i32x4:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vperm2i128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[2,3]
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_shuffle_i32x4:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vperm2i128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[2,3]
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%shuffle = shufflevector <4 x i64> %__A, <4 x i64> %__B, <4 x i32> <i32 2, i32 3, i32 6, i32 7>
ret <4 x i64> %shuffle
}
define <4 x i64> @test_mm256_mask_shuffle_i32x4(<4 x i64> %__W, i8 zeroext %__U, <4 x i64> %__A, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_mask_shuffle_i32x4:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vshufi32x4 {{.*#+}} ymm0 {%k1} = ymm1[4,5,6,7],ymm2[4,5,6,7]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_shuffle_i32x4:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vshufi32x4 {{.*#+}} ymm0 {%k1} = ymm1[4,5,6,7],ymm2[4,5,6,7]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_shuffle_i32x4:
; X64: # %bb.0: # %entry
@@ -1455,12 +1378,12 @@ entry:
}
define <4 x i64> @test_mm256_maskz_shuffle_i32x4(i8 zeroext %__U, <4 x i64> %__A, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_maskz_shuffle_i32x4:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vshufi32x4 {{.*#+}} ymm0 {%k1} {z} = ymm0[4,5,6,7],ymm1[4,5,6,7]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_shuffle_i32x4:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vshufi32x4 {{.*#+}} ymm0 {%k1} {z} = ymm0[4,5,6,7],ymm1[4,5,6,7]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_shuffle_i32x4:
; X64: # %bb.0: # %entry
@@ -1477,27 +1400,22 @@ entry:
}
define <4 x i64> @test_mm256_shuffle_i64x2(<4 x i64> %__A, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_shuffle_i64x2:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vperm2i128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[2,3]
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_shuffle_i64x2:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vperm2i128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[2,3]
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_shuffle_i64x2:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vperm2i128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[2,3]
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%shuffle = shufflevector <4 x i64> %__A, <4 x i64> %__B, <4 x i32> <i32 2, i32 3, i32 6, i32 7>
ret <4 x i64> %shuffle
}
define <4 x i64> @test_mm256_mask_shuffle_i64x2(<4 x i64> %__W, i8 zeroext %__U, <4 x i64> %__A, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_mask_shuffle_i64x2:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vshufi64x2 {{.*#+}} ymm0 {%k1} = ymm1[2,3],ymm2[2,3]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_shuffle_i64x2:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vshufi64x2 {{.*#+}} ymm0 {%k1} = ymm1[2,3],ymm2[2,3]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_shuffle_i64x2:
; X64: # %bb.0: # %entry
@@ -1513,12 +1431,12 @@ entry:
}
define <4 x i64> @test_mm256_maskz_shuffle_i64x2(i8 zeroext %__U, <4 x i64> %__A, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_maskz_shuffle_i64x2:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vshufi64x2 {{.*#+}} ymm0 {%k1} {z} = ymm0[2,3],ymm1[2,3]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_shuffle_i64x2:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vshufi64x2 {{.*#+}} ymm0 {%k1} {z} = ymm0[2,3],ymm1[2,3]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_shuffle_i64x2:
; X64: # %bb.0: # %entry
@@ -1534,19 +1452,12 @@ entry:
}
define zeroext i8 @test_mm_test_epi32_mask(<2 x i64> %__A, <2 x i64> %__B) {
-; X32-LABEL: test_mm_test_epi32_mask:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vptestmd %xmm0, %xmm1, %k0
-; X32-NEXT: kmovw %k0, %eax
-; X32-NEXT: movzbl %al, %eax
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_test_epi32_mask:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vptestmd %xmm0, %xmm1, %k0
-; X64-NEXT: kmovw %k0, %eax
-; X64-NEXT: movzbl %al, %eax
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_test_epi32_mask:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vptestmd %xmm0, %xmm1, %k0
+; CHECK-NEXT: kmovw %k0, %eax
+; CHECK-NEXT: movzbl %al, %eax
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%and.i.i = and <2 x i64> %__B, %__A
%0 = bitcast <2 x i64> %and.i.i to <4 x i32>
@@ -1557,14 +1468,14 @@ entry:
}
define zeroext i8 @test_mm_mask_test_epi32_mask(i8 zeroext %__U, <2 x i64> %__A, <2 x i64> %__B) {
-; X32-LABEL: test_mm_mask_test_epi32_mask:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vptestmd %xmm0, %xmm1, %k0 {%k1}
-; X32-NEXT: kmovw %k0, %eax
-; X32-NEXT: movzbl %al, %eax
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_test_epi32_mask:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vptestmd %xmm0, %xmm1, %k0 {%k1}
+; X86-NEXT: kmovw %k0, %eax
+; X86-NEXT: movzbl %al, %eax
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_test_epi32_mask:
; X64: # %bb.0: # %entry
@@ -1586,21 +1497,13 @@ entry:
}
define zeroext i8 @test_mm256_test_epi32_mask(<4 x i64> %__A, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_test_epi32_mask:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vptestmd %ymm0, %ymm1, %k0
-; X32-NEXT: kmovw %k0, %eax
-; X32-NEXT: movzbl %al, %eax
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_test_epi32_mask:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vptestmd %ymm0, %ymm1, %k0
-; X64-NEXT: kmovw %k0, %eax
-; X64-NEXT: movzbl %al, %eax
-; X64-NEXT: vzeroupper
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_test_epi32_mask:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vptestmd %ymm0, %ymm1, %k0
+; CHECK-NEXT: kmovw %k0, %eax
+; CHECK-NEXT: movzbl %al, %eax
+; CHECK-NEXT: vzeroupper
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%and.i.i = and <4 x i64> %__B, %__A
%0 = bitcast <4 x i64> %and.i.i to <8 x i32>
@@ -1610,15 +1513,15 @@ entry:
}
define zeroext i8 @test_mm256_mask_test_epi32_mask(i8 zeroext %__U, <4 x i64> %__A, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_mask_test_epi32_mask:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vptestmd %ymm0, %ymm1, %k0 {%k1}
-; X32-NEXT: kmovw %k0, %eax
-; X32-NEXT: movzbl %al, %eax
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_test_epi32_mask:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vptestmd %ymm0, %ymm1, %k0 {%k1}
+; X86-NEXT: kmovw %k0, %eax
+; X86-NEXT: movzbl %al, %eax
+; X86-NEXT: vzeroupper
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_test_epi32_mask:
; X64: # %bb.0: # %entry
@@ -1639,19 +1542,12 @@ entry:
}
define zeroext i8 @test_mm_test_epi64_mask(<2 x i64> %__A, <2 x i64> %__B) {
-; X32-LABEL: test_mm_test_epi64_mask:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vptestmq %xmm0, %xmm1, %k0
-; X32-NEXT: kmovw %k0, %eax
-; X32-NEXT: movzbl %al, %eax
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_test_epi64_mask:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vptestmq %xmm0, %xmm1, %k0
-; X64-NEXT: kmovw %k0, %eax
-; X64-NEXT: movzbl %al, %eax
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_test_epi64_mask:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vptestmq %xmm0, %xmm1, %k0
+; CHECK-NEXT: kmovw %k0, %eax
+; CHECK-NEXT: movzbl %al, %eax
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%and.i.i = and <2 x i64> %__B, %__A
%0 = icmp ne <2 x i64> %and.i.i, zeroinitializer
@@ -1661,14 +1557,14 @@ entry:
}
define zeroext i8 @test_mm_mask_test_epi64_mask(i8 zeroext %__U, <2 x i64> %__A, <2 x i64> %__B) {
-; X32-LABEL: test_mm_mask_test_epi64_mask:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vptestmq %xmm0, %xmm1, %k0 {%k1}
-; X32-NEXT: kmovw %k0, %eax
-; X32-NEXT: movzbl %al, %eax
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_test_epi64_mask:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vptestmq %xmm0, %xmm1, %k0 {%k1}
+; X86-NEXT: kmovw %k0, %eax
+; X86-NEXT: movzbl %al, %eax
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_test_epi64_mask:
; X64: # %bb.0: # %entry
@@ -1689,21 +1585,13 @@ entry:
}
define zeroext i8 @test_mm256_test_epi64_mask(<4 x i64> %__A, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_test_epi64_mask:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vptestmq %ymm0, %ymm1, %k0
-; X32-NEXT: kmovw %k0, %eax
-; X32-NEXT: movzbl %al, %eax
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_test_epi64_mask:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vptestmq %ymm0, %ymm1, %k0
-; X64-NEXT: kmovw %k0, %eax
-; X64-NEXT: movzbl %al, %eax
-; X64-NEXT: vzeroupper
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_test_epi64_mask:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vptestmq %ymm0, %ymm1, %k0
+; CHECK-NEXT: kmovw %k0, %eax
+; CHECK-NEXT: movzbl %al, %eax
+; CHECK-NEXT: vzeroupper
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%and.i.i = and <4 x i64> %__B, %__A
%0 = icmp ne <4 x i64> %and.i.i, zeroinitializer
@@ -1713,15 +1601,15 @@ entry:
}
define zeroext i8 @test_mm256_mask_test_epi64_mask(i8 zeroext %__U, <4 x i64> %__A, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_mask_test_epi64_mask:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vptestmq %ymm0, %ymm1, %k0 {%k1}
-; X32-NEXT: kmovw %k0, %eax
-; X32-NEXT: movzbl %al, %eax
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_test_epi64_mask:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vptestmq %ymm0, %ymm1, %k0 {%k1}
+; X86-NEXT: kmovw %k0, %eax
+; X86-NEXT: movzbl %al, %eax
+; X86-NEXT: vzeroupper
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_test_epi64_mask:
; X64: # %bb.0: # %entry
@@ -1743,19 +1631,12 @@ entry:
}
define zeroext i8 @test_mm_testn_epi32_mask(<2 x i64> %__A, <2 x i64> %__B) {
-; X32-LABEL: test_mm_testn_epi32_mask:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vptestnmd %xmm0, %xmm1, %k0
-; X32-NEXT: kmovw %k0, %eax
-; X32-NEXT: movzbl %al, %eax
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_testn_epi32_mask:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vptestnmd %xmm0, %xmm1, %k0
-; X64-NEXT: kmovw %k0, %eax
-; X64-NEXT: movzbl %al, %eax
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_testn_epi32_mask:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vptestnmd %xmm0, %xmm1, %k0
+; CHECK-NEXT: kmovw %k0, %eax
+; CHECK-NEXT: movzbl %al, %eax
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%and.i.i = and <2 x i64> %__B, %__A
%0 = bitcast <2 x i64> %and.i.i to <4 x i32>
@@ -1766,14 +1647,14 @@ entry:
}
define zeroext i8 @test_mm_mask_testn_epi32_mask(i8 zeroext %__U, <2 x i64> %__A, <2 x i64> %__B) {
-; X32-LABEL: test_mm_mask_testn_epi32_mask:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vptestnmd %xmm0, %xmm1, %k0 {%k1}
-; X32-NEXT: kmovw %k0, %eax
-; X32-NEXT: movzbl %al, %eax
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_testn_epi32_mask:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vptestnmd %xmm0, %xmm1, %k0 {%k1}
+; X86-NEXT: kmovw %k0, %eax
+; X86-NEXT: movzbl %al, %eax
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_testn_epi32_mask:
; X64: # %bb.0: # %entry
@@ -1795,21 +1676,13 @@ entry:
}
define zeroext i8 @test_mm256_testn_epi32_mask(<4 x i64> %__A, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_testn_epi32_mask:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vptestnmd %ymm0, %ymm1, %k0
-; X32-NEXT: kmovw %k0, %eax
-; X32-NEXT: movzbl %al, %eax
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_testn_epi32_mask:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vptestnmd %ymm0, %ymm1, %k0
-; X64-NEXT: kmovw %k0, %eax
-; X64-NEXT: movzbl %al, %eax
-; X64-NEXT: vzeroupper
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_testn_epi32_mask:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vptestnmd %ymm0, %ymm1, %k0
+; CHECK-NEXT: kmovw %k0, %eax
+; CHECK-NEXT: movzbl %al, %eax
+; CHECK-NEXT: vzeroupper
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%and.i.i = and <4 x i64> %__B, %__A
%0 = bitcast <4 x i64> %and.i.i to <8 x i32>
@@ -1819,15 +1692,15 @@ entry:
}
define zeroext i8 @test_mm256_mask_testn_epi32_mask(i8 zeroext %__U, <4 x i64> %__A, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_mask_testn_epi32_mask:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vptestnmd %ymm0, %ymm1, %k0 {%k1}
-; X32-NEXT: kmovw %k0, %eax
-; X32-NEXT: movzbl %al, %eax
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_testn_epi32_mask:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vptestnmd %ymm0, %ymm1, %k0 {%k1}
+; X86-NEXT: kmovw %k0, %eax
+; X86-NEXT: movzbl %al, %eax
+; X86-NEXT: vzeroupper
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_testn_epi32_mask:
; X64: # %bb.0: # %entry
@@ -1848,19 +1721,12 @@ entry:
}
define zeroext i8 @test_mm_testn_epi64_mask(<2 x i64> %__A, <2 x i64> %__B) {
-; X32-LABEL: test_mm_testn_epi64_mask:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vptestnmq %xmm0, %xmm1, %k0
-; X32-NEXT: kmovw %k0, %eax
-; X32-NEXT: movzbl %al, %eax
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_testn_epi64_mask:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vptestnmq %xmm0, %xmm1, %k0
-; X64-NEXT: kmovw %k0, %eax
-; X64-NEXT: movzbl %al, %eax
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_testn_epi64_mask:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vptestnmq %xmm0, %xmm1, %k0
+; CHECK-NEXT: kmovw %k0, %eax
+; CHECK-NEXT: movzbl %al, %eax
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%and.i.i = and <2 x i64> %__B, %__A
%0 = icmp eq <2 x i64> %and.i.i, zeroinitializer
@@ -1870,14 +1736,14 @@ entry:
}
define zeroext i8 @test_mm_mask_testn_epi64_mask(i8 zeroext %__U, <2 x i64> %__A, <2 x i64> %__B) {
-; X32-LABEL: test_mm_mask_testn_epi64_mask:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vptestnmq %xmm0, %xmm1, %k0 {%k1}
-; X32-NEXT: kmovw %k0, %eax
-; X32-NEXT: movzbl %al, %eax
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_testn_epi64_mask:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vptestnmq %xmm0, %xmm1, %k0 {%k1}
+; X86-NEXT: kmovw %k0, %eax
+; X86-NEXT: movzbl %al, %eax
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_testn_epi64_mask:
; X64: # %bb.0: # %entry
@@ -1898,21 +1764,13 @@ entry:
}
define zeroext i8 @test_mm256_testn_epi64_mask(<4 x i64> %__A, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_testn_epi64_mask:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vptestnmq %ymm0, %ymm1, %k0
-; X32-NEXT: kmovw %k0, %eax
-; X32-NEXT: movzbl %al, %eax
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_testn_epi64_mask:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vptestnmq %ymm0, %ymm1, %k0
-; X64-NEXT: kmovw %k0, %eax
-; X64-NEXT: movzbl %al, %eax
-; X64-NEXT: vzeroupper
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_testn_epi64_mask:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vptestnmq %ymm0, %ymm1, %k0
+; CHECK-NEXT: kmovw %k0, %eax
+; CHECK-NEXT: movzbl %al, %eax
+; CHECK-NEXT: vzeroupper
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%and.i.i = and <4 x i64> %__B, %__A
%0 = icmp eq <4 x i64> %and.i.i, zeroinitializer
@@ -1922,15 +1780,15 @@ entry:
}
define zeroext i8 @test_mm256_mask_testn_epi64_mask(i8 zeroext %__U, <4 x i64> %__A, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_mask_testn_epi64_mask:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vptestnmq %ymm0, %ymm1, %k0 {%k1}
-; X32-NEXT: kmovw %k0, %eax
-; X32-NEXT: movzbl %al, %eax
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_testn_epi64_mask:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vptestnmq %ymm0, %ymm1, %k0 {%k1}
+; X86-NEXT: kmovw %k0, %eax
+; X86-NEXT: movzbl %al, %eax
+; X86-NEXT: vzeroupper
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_testn_epi64_mask:
; X64: # %bb.0: # %entry
@@ -1952,12 +1810,12 @@ entry:
}
define <2 x i64> @test_mm_mask_set1_epi32(<2 x i64> %__O, i8 zeroext %__M) {
-; X32-LABEL: test_mm_mask_set1_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpbroadcastd {{\.LCPI.*}}, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_set1_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpbroadcastd {{\.LCPI.*}}, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_set1_epi32:
; X64: # %bb.0: # %entry
@@ -1974,12 +1832,12 @@ entry:
}
define <2 x i64> @test_mm_maskz_set1_epi32(i8 zeroext %__M) {
-; X32-LABEL: test_mm_maskz_set1_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpbroadcastd {{\.LCPI.*}}, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_set1_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpbroadcastd {{\.LCPI.*}}, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_set1_epi32:
; X64: # %bb.0: # %entry
@@ -1995,12 +1853,12 @@ entry:
}
define <4 x i64> @test_mm256_mask_set1_epi32(<4 x i64> %__O, i8 zeroext %__M) {
-; X32-LABEL: test_mm256_mask_set1_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpbroadcastd {{\.LCPI.*}}, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_set1_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpbroadcastd {{\.LCPI.*}}, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_set1_epi32:
; X64: # %bb.0: # %entry
@@ -2016,12 +1874,12 @@ entry:
}
define <4 x i64> @test_mm256_maskz_set1_epi32(i8 zeroext %__M) {
-; X32-LABEL: test_mm256_maskz_set1_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpbroadcastd {{\.LCPI.*}}, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_set1_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpbroadcastd {{\.LCPI.*}}, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_set1_epi32:
; X64: # %bb.0: # %entry
@@ -2036,15 +1894,15 @@ entry:
}
define <2 x i64> @test_mm_mask_set1_epi64(<2 x i64> %__O, i8 zeroext %__M, i64 %__A) {
-; X32-LABEL: test_mm_mask_set1_epi64:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
-; X32-NEXT: movb {{[0-9]+}}(%esp), %cl
-; X32-NEXT: vmovd %eax, %xmm1
-; X32-NEXT: vpbroadcastb %xmm1, %xmm1
-; X32-NEXT: kmovw %ecx, %k1
-; X32-NEXT: vmovdqa64 %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_set1_epi64:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
+; X86-NEXT: movb {{[0-9]+}}(%esp), %cl
+; X86-NEXT: vmovd %eax, %xmm1
+; X86-NEXT: vpbroadcastb %xmm1, %xmm1
+; X86-NEXT: kmovw %ecx, %k1
+; X86-NEXT: vmovdqa64 %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_set1_epi64:
; X64: # %bb.0: # %entry
@@ -2065,15 +1923,15 @@ entry:
}
define <2 x i64> @test_mm_maskz_set1_epi64(i8 zeroext %__M, i64 %__A) {
-; X32-LABEL: test_mm_maskz_set1_epi64:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
-; X32-NEXT: movb {{[0-9]+}}(%esp), %cl
-; X32-NEXT: vmovd %eax, %xmm0
-; X32-NEXT: vpbroadcastb %xmm0, %xmm0
-; X32-NEXT: kmovw %ecx, %k1
-; X32-NEXT: vmovdqa64 %xmm0, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_set1_epi64:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
+; X86-NEXT: movb {{[0-9]+}}(%esp), %cl
+; X86-NEXT: vmovd %eax, %xmm0
+; X86-NEXT: vpbroadcastb %xmm0, %xmm0
+; X86-NEXT: kmovw %ecx, %k1
+; X86-NEXT: vmovdqa64 %xmm0, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_set1_epi64:
; X64: # %bb.0: # %entry
@@ -2095,14 +1953,14 @@ entry:
define <4 x i64> @test_mm256_mask_set1_epi64(<4 x i64> %__O, i8 zeroext %__M, i64 %__A) {
-; X32-LABEL: test_mm256_mask_set1_epi64:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: vmovd {{.*#+}} xmm1 = mem[0],zero,zero,zero
-; X32-NEXT: vpinsrd $1, {{[0-9]+}}(%esp), %xmm1, %xmm1
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpbroadcastq %xmm1, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_set1_epi64:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: vmovd {{.*#+}} xmm1 = mem[0],zero,zero,zero
+; X86-NEXT: vpinsrd $1, {{[0-9]+}}(%esp), %xmm1, %xmm1
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpbroadcastq %xmm1, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_set1_epi64:
; X64: # %bb.0: # %entry
@@ -2119,14 +1977,14 @@ entry:
}
define <4 x i64> @test_mm256_maskz_set1_epi64(i8 zeroext %__M, i64 %__A) {
-; X32-LABEL: test_mm256_maskz_set1_epi64:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: vmovd {{.*#+}} xmm0 = mem[0],zero,zero,zero
-; X32-NEXT: vpinsrd $1, {{[0-9]+}}(%esp), %xmm0, %xmm0
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpbroadcastq %xmm0, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_set1_epi64:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: vmovd {{.*#+}} xmm0 = mem[0],zero,zero,zero
+; X86-NEXT: vpinsrd $1, {{[0-9]+}}(%esp), %xmm0, %xmm0
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpbroadcastq %xmm0, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_set1_epi64:
; X64: # %bb.0: # %entry
@@ -2143,15 +2001,10 @@ entry:
}
define <2 x i64> @test_mm_broadcastd_epi32(<2 x i64> %a0) {
-; X32-LABEL: test_mm_broadcastd_epi32:
-; X32: # %bb.0:
-; X32-NEXT: vbroadcastss %xmm0, %xmm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_broadcastd_epi32:
-; X64: # %bb.0:
-; X64-NEXT: vbroadcastss %xmm0, %xmm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_broadcastd_epi32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vbroadcastss %xmm0, %xmm0
+; CHECK-NEXT: ret{{[l|q]}}
%arg0 = bitcast <2 x i64> %a0 to <4 x i32>
%res0 = shufflevector <4 x i32> %arg0, <4 x i32> undef, <4 x i32> zeroinitializer
%res1 = bitcast <4 x i32> %res0 to <2 x i64>
@@ -2159,12 +2012,12 @@ define <2 x i64> @test_mm_broadcastd_epi
}
define <2 x i64> @test_mm_mask_broadcastd_epi32(<2 x i64> %a0, i8 %a1, <2 x i64> %a2) {
-; X32-LABEL: test_mm_mask_broadcastd_epi32:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpbroadcastd %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_broadcastd_epi32:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpbroadcastd %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_broadcastd_epi32:
; X64: # %bb.0:
@@ -2182,12 +2035,12 @@ define <2 x i64> @test_mm_mask_broadcast
}
define <2 x i64> @test_mm_maskz_broadcastd_epi32(i8 %a0, <2 x i64> %a1) {
-; X32-LABEL: test_mm_maskz_broadcastd_epi32:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpbroadcastd %xmm0, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_broadcastd_epi32:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpbroadcastd %xmm0, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_broadcastd_epi32:
; X64: # %bb.0:
@@ -2204,15 +2057,10 @@ define <2 x i64> @test_mm_maskz_broadcas
}
define <4 x i64> @test_mm256_broadcastd_epi32(<2 x i64> %a0) {
-; X32-LABEL: test_mm256_broadcastd_epi32:
-; X32: # %bb.0:
-; X32-NEXT: vbroadcastss %xmm0, %ymm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_broadcastd_epi32:
-; X64: # %bb.0:
-; X64-NEXT: vbroadcastss %xmm0, %ymm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_broadcastd_epi32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vbroadcastss %xmm0, %ymm0
+; CHECK-NEXT: ret{{[l|q]}}
%arg0 = bitcast <2 x i64> %a0 to <4 x i32>
%res0 = shufflevector <4 x i32> %arg0, <4 x i32> undef, <8 x i32> zeroinitializer
%res1 = bitcast <8 x i32> %res0 to <4 x i64>
@@ -2220,12 +2068,12 @@ define <4 x i64> @test_mm256_broadcastd_
}
define <4 x i64> @test_mm256_mask_broadcastd_epi32(<4 x i64> %a0, i8 %a1, <2 x i64> %a2) {
-; X32-LABEL: test_mm256_mask_broadcastd_epi32:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpbroadcastd %xmm1, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_broadcastd_epi32:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpbroadcastd %xmm1, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_broadcastd_epi32:
; X64: # %bb.0:
@@ -2242,12 +2090,12 @@ define <4 x i64> @test_mm256_mask_broadc
}
define <4 x i64> @test_mm256_maskz_broadcastd_epi32(i8 %a0, <2 x i64> %a1) {
-; X32-LABEL: test_mm256_maskz_broadcastd_epi32:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpbroadcastd %xmm0, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_broadcastd_epi32:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpbroadcastd %xmm0, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_broadcastd_epi32:
; X64: # %bb.0:
@@ -2263,26 +2111,21 @@ define <4 x i64> @test_mm256_maskz_broad
}
define <2 x i64> @test_mm_broadcastq_epi64(<2 x i64> %a0) {
-; X32-LABEL: test_mm_broadcastq_epi64:
-; X32: # %bb.0:
-; X32-NEXT: vpbroadcastq %xmm0, %xmm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_broadcastq_epi64:
-; X64: # %bb.0:
-; X64-NEXT: vpbroadcastq %xmm0, %xmm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_broadcastq_epi64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpbroadcastq %xmm0, %xmm0
+; CHECK-NEXT: ret{{[l|q]}}
%res = shufflevector <2 x i64> %a0, <2 x i64> undef, <2 x i32> zeroinitializer
ret <2 x i64> %res
}
define <2 x i64> @test_mm_mask_broadcastq_epi64(<2 x i64> %a0, i8 %a1, <2 x i64> %a2) {
-; X32-LABEL: test_mm_mask_broadcastq_epi64:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpbroadcastq %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_broadcastq_epi64:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpbroadcastq %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_broadcastq_epi64:
; X64: # %bb.0:
@@ -2297,12 +2140,12 @@ define <2 x i64> @test_mm_mask_broadcast
}
define <2 x i64> @test_mm_maskz_broadcastq_epi64(i8 %a0, <2 x i64> %a1) {
-; X32-LABEL: test_mm_maskz_broadcastq_epi64:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpbroadcastq %xmm0, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_broadcastq_epi64:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpbroadcastq %xmm0, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_broadcastq_epi64:
; X64: # %bb.0:
@@ -2317,26 +2160,21 @@ define <2 x i64> @test_mm_maskz_broadcas
}
define <4 x i64> @test_mm256_broadcastq_epi64(<2 x i64> %a0) {
-; X32-LABEL: test_mm256_broadcastq_epi64:
-; X32: # %bb.0:
-; X32-NEXT: vbroadcastsd %xmm0, %ymm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_broadcastq_epi64:
-; X64: # %bb.0:
-; X64-NEXT: vbroadcastsd %xmm0, %ymm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_broadcastq_epi64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vbroadcastsd %xmm0, %ymm0
+; CHECK-NEXT: ret{{[l|q]}}
%res = shufflevector <2 x i64> %a0, <2 x i64> undef, <4 x i32> zeroinitializer
ret <4 x i64> %res
}
define <4 x i64> @test_mm256_mask_broadcastq_epi64(<4 x i64> %a0, i8 %a1, <2 x i64> %a2) {
-; X32-LABEL: test_mm256_mask_broadcastq_epi64:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpbroadcastq %xmm1, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_broadcastq_epi64:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpbroadcastq %xmm1, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_broadcastq_epi64:
; X64: # %bb.0:
@@ -2351,12 +2189,12 @@ define <4 x i64> @test_mm256_mask_broadc
}
define <4 x i64> @test_mm256_maskz_broadcastq_epi64(i8 %a0, <2 x i64> %a1) {
-; X32-LABEL: test_mm256_maskz_broadcastq_epi64:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpbroadcastq %xmm0, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_broadcastq_epi64:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpbroadcastq %xmm0, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_broadcastq_epi64:
; X64: # %bb.0:
@@ -2371,26 +2209,21 @@ define <4 x i64> @test_mm256_maskz_broad
}
define <2 x double> @test_mm_broadcastsd_pd(<2 x double> %a0) {
-; X32-LABEL: test_mm_broadcastsd_pd:
-; X32: # %bb.0:
-; X32-NEXT: vmovddup {{.*#+}} xmm0 = xmm0[0,0]
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_broadcastsd_pd:
-; X64: # %bb.0:
-; X64-NEXT: vmovddup {{.*#+}} xmm0 = xmm0[0,0]
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_broadcastsd_pd:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovddup {{.*#+}} xmm0 = xmm0[0,0]
+; CHECK-NEXT: ret{{[l|q]}}
%res = shufflevector <2 x double> %a0, <2 x double> undef, <2 x i32> zeroinitializer
ret <2 x double> %res
}
define <2 x double> @test_mm_mask_broadcastsd_pd(<2 x double> %a0, i8 %a1, <2 x double> %a2) {
-; X32-LABEL: test_mm_mask_broadcastsd_pd:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vmovddup {{.*#+}} xmm0 {%k1} = xmm1[0,0]
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_broadcastsd_pd:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vmovddup {{.*#+}} xmm0 {%k1} = xmm1[0,0]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_broadcastsd_pd:
; X64: # %bb.0:
@@ -2405,12 +2238,12 @@ define <2 x double> @test_mm_mask_broadc
}
define <2 x double> @test_mm_maskz_broadcastsd_pd(i8 %a0, <2 x double> %a1) {
-; X32-LABEL: test_mm_maskz_broadcastsd_pd:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vmovddup {{.*#+}} xmm0 {%k1} {z} = xmm0[0,0]
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_broadcastsd_pd:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vmovddup {{.*#+}} xmm0 {%k1} {z} = xmm0[0,0]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_broadcastsd_pd:
; X64: # %bb.0:
@@ -2425,26 +2258,21 @@ define <2 x double> @test_mm_maskz_broad
}
define <4 x double> @test_mm256_broadcastsd_pd(<2 x double> %a0) {
-; X32-LABEL: test_mm256_broadcastsd_pd:
-; X32: # %bb.0:
-; X32-NEXT: vbroadcastsd %xmm0, %ymm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_broadcastsd_pd:
-; X64: # %bb.0:
-; X64-NEXT: vbroadcastsd %xmm0, %ymm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_broadcastsd_pd:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vbroadcastsd %xmm0, %ymm0
+; CHECK-NEXT: ret{{[l|q]}}
%res = shufflevector <2 x double> %a0, <2 x double> undef, <4 x i32> zeroinitializer
ret <4 x double> %res
}
define <4 x double> @test_mm256_mask_broadcastsd_pd(<4 x double> %a0, i8 %a1, <2 x double> %a2) {
-; X32-LABEL: test_mm256_mask_broadcastsd_pd:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vbroadcastsd %xmm1, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_broadcastsd_pd:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vbroadcastsd %xmm1, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_broadcastsd_pd:
; X64: # %bb.0:
@@ -2459,12 +2287,12 @@ define <4 x double> @test_mm256_mask_bro
}
define <4 x double> @test_mm256_maskz_broadcastsd_pd(i8 %a0, <2 x double> %a1) {
-; X32-LABEL: test_mm256_maskz_broadcastsd_pd:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vbroadcastsd %xmm0, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_broadcastsd_pd:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vbroadcastsd %xmm0, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_broadcastsd_pd:
; X64: # %bb.0:
@@ -2479,26 +2307,21 @@ define <4 x double> @test_mm256_maskz_br
}
define <4 x float> @test_mm_broadcastss_ps(<4 x float> %a0) {
-; X32-LABEL: test_mm_broadcastss_ps:
-; X32: # %bb.0:
-; X32-NEXT: vbroadcastss %xmm0, %xmm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_broadcastss_ps:
-; X64: # %bb.0:
-; X64-NEXT: vbroadcastss %xmm0, %xmm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_broadcastss_ps:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vbroadcastss %xmm0, %xmm0
+; CHECK-NEXT: ret{{[l|q]}}
%res = shufflevector <4 x float> %a0, <4 x float> undef, <4 x i32> zeroinitializer
ret <4 x float> %res
}
define <4 x float> @test_mm_mask_broadcastss_ps(<4 x float> %a0, i8 %a1, <4 x float> %a2) {
-; X32-LABEL: test_mm_mask_broadcastss_ps:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vbroadcastss %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_broadcastss_ps:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vbroadcastss %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_broadcastss_ps:
; X64: # %bb.0:
@@ -2513,12 +2336,12 @@ define <4 x float> @test_mm_mask_broadca
}
define <4 x float> @test_mm_maskz_broadcastss_ps(i8 %a0, <4 x float> %a1) {
-; X32-LABEL: test_mm_maskz_broadcastss_ps:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vbroadcastss %xmm0, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_broadcastss_ps:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vbroadcastss %xmm0, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_broadcastss_ps:
; X64: # %bb.0:
@@ -2533,26 +2356,21 @@ define <4 x float> @test_mm_maskz_broadc
}
define <8 x float> @test_mm256_broadcastss_ps(<4 x float> %a0) {
-; X32-LABEL: test_mm256_broadcastss_ps:
-; X32: # %bb.0:
-; X32-NEXT: vbroadcastss %xmm0, %ymm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_broadcastss_ps:
-; X64: # %bb.0:
-; X64-NEXT: vbroadcastss %xmm0, %ymm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_broadcastss_ps:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vbroadcastss %xmm0, %ymm0
+; CHECK-NEXT: ret{{[l|q]}}
%res = shufflevector <4 x float> %a0, <4 x float> undef, <8 x i32> zeroinitializer
ret <8 x float> %res
}
define <8 x float> @test_mm256_mask_broadcastss_ps(<8 x float> %a0, i8 %a1, <4 x float> %a2) {
-; X32-LABEL: test_mm256_mask_broadcastss_ps:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vbroadcastss %xmm1, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_broadcastss_ps:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vbroadcastss %xmm1, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_broadcastss_ps:
; X64: # %bb.0:
@@ -2566,12 +2384,12 @@ define <8 x float> @test_mm256_mask_broa
}
define <8 x float> @test_mm256_maskz_broadcastss_ps(i8 %a0, <4 x float> %a1) {
-; X32-LABEL: test_mm256_maskz_broadcastss_ps:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vbroadcastss %xmm0, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_broadcastss_ps:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vbroadcastss %xmm0, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_broadcastss_ps:
; X64: # %bb.0:
@@ -2585,26 +2403,21 @@ define <8 x float> @test_mm256_maskz_bro
}
define <2 x double> @test_mm_movddup_pd(<2 x double> %a0) {
-; X32-LABEL: test_mm_movddup_pd:
-; X32: # %bb.0:
-; X32-NEXT: vmovddup {{.*#+}} xmm0 = xmm0[0,0]
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_movddup_pd:
-; X64: # %bb.0:
-; X64-NEXT: vmovddup {{.*#+}} xmm0 = xmm0[0,0]
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_movddup_pd:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovddup {{.*#+}} xmm0 = xmm0[0,0]
+; CHECK-NEXT: ret{{[l|q]}}
%res = shufflevector <2 x double> %a0, <2 x double> undef, <2 x i32> zeroinitializer
ret <2 x double> %res
}
define <2 x double> @test_mm_mask_movddup_pd(<2 x double> %a0, i8 %a1, <2 x double> %a2) {
-; X32-LABEL: test_mm_mask_movddup_pd:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vmovddup {{.*#+}} xmm0 {%k1} = xmm1[0,0]
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_movddup_pd:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vmovddup {{.*#+}} xmm0 {%k1} = xmm1[0,0]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_movddup_pd:
; X64: # %bb.0:
@@ -2619,12 +2432,12 @@ define <2 x double> @test_mm_mask_movddu
}
define <2 x double> @test_mm_maskz_movddup_pd(i8 %a0, <2 x double> %a1) {
-; X32-LABEL: test_mm_maskz_movddup_pd:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vmovddup {{.*#+}} xmm0 {%k1} {z} = xmm0[0,0]
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_movddup_pd:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vmovddup {{.*#+}} xmm0 {%k1} {z} = xmm0[0,0]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_movddup_pd:
; X64: # %bb.0:
@@ -2639,26 +2452,21 @@ define <2 x double> @test_mm_maskz_movdd
}
define <4 x double> @test_mm256_movddup_pd(<4 x double> %a0) {
-; X32-LABEL: test_mm256_movddup_pd:
-; X32: # %bb.0:
-; X32-NEXT: vmovddup {{.*#+}} ymm0 = ymm0[0,0,2,2]
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_movddup_pd:
-; X64: # %bb.0:
-; X64-NEXT: vmovddup {{.*#+}} ymm0 = ymm0[0,0,2,2]
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_movddup_pd:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovddup {{.*#+}} ymm0 = ymm0[0,0,2,2]
+; CHECK-NEXT: ret{{[l|q]}}
%res = shufflevector <4 x double> %a0, <4 x double> undef, <4 x i32> <i32 0, i32 0, i32 2, i32 2>
ret <4 x double> %res
}
define <4 x double> @test_mm256_mask_movddup_pd(<4 x double> %a0, i8 %a1, <4 x double> %a2) {
-; X32-LABEL: test_mm256_mask_movddup_pd:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vmovddup {{.*#+}} ymm0 {%k1} = ymm1[0,0,2,2]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_movddup_pd:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vmovddup {{.*#+}} ymm0 {%k1} = ymm1[0,0,2,2]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_movddup_pd:
; X64: # %bb.0:
@@ -2673,12 +2481,12 @@ define <4 x double> @test_mm256_mask_mov
}
define <4 x double> @test_mm256_maskz_movddup_pd(i8 %a0, <4 x double> %a1) {
-; X32-LABEL: test_mm256_maskz_movddup_pd:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vmovddup {{.*#+}} ymm0 {%k1} {z} = ymm0[0,0,2,2]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_movddup_pd:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vmovddup {{.*#+}} ymm0 {%k1} {z} = ymm0[0,0,2,2]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_movddup_pd:
; X64: # %bb.0:
@@ -2693,26 +2501,21 @@ define <4 x double> @test_mm256_maskz_mo
}
define <4 x float> @test_mm_movehdup_ps(<4 x float> %a0) {
-; X32-LABEL: test_mm_movehdup_ps:
-; X32: # %bb.0:
-; X32-NEXT: vmovshdup {{.*#+}} xmm0 = xmm0[1,1,3,3]
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_movehdup_ps:
-; X64: # %bb.0:
-; X64-NEXT: vmovshdup {{.*#+}} xmm0 = xmm0[1,1,3,3]
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_movehdup_ps:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovshdup {{.*#+}} xmm0 = xmm0[1,1,3,3]
+; CHECK-NEXT: ret{{[l|q]}}
%res = shufflevector <4 x float> %a0, <4 x float> undef, <4 x i32> <i32 1, i32 1, i32 3, i32 3>
ret <4 x float> %res
}
define <4 x float> @test_mm_mask_movehdup_ps(<4 x float> %a0, i8 %a1, <4 x float> %a2) {
-; X32-LABEL: test_mm_mask_movehdup_ps:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vmovshdup {{.*#+}} xmm0 {%k1} = xmm1[1,1,3,3]
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_movehdup_ps:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vmovshdup {{.*#+}} xmm0 {%k1} = xmm1[1,1,3,3]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_movehdup_ps:
; X64: # %bb.0:
@@ -2727,12 +2530,12 @@ define <4 x float> @test_mm_mask_movehdu
}
define <4 x float> @test_mm_maskz_movehdup_ps(i8 %a0, <4 x float> %a1) {
-; X32-LABEL: test_mm_maskz_movehdup_ps:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vmovshdup {{.*#+}} xmm0 {%k1} {z} = xmm0[1,1,3,3]
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_movehdup_ps:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vmovshdup {{.*#+}} xmm0 {%k1} {z} = xmm0[1,1,3,3]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_movehdup_ps:
; X64: # %bb.0:
@@ -2747,26 +2550,21 @@ define <4 x float> @test_mm_maskz_movehd
}
define <8 x float> @test_mm256_movehdup_ps(<8 x float> %a0) {
-; X32-LABEL: test_mm256_movehdup_ps:
-; X32: # %bb.0:
-; X32-NEXT: vmovshdup {{.*#+}} ymm0 = ymm0[1,1,3,3,5,5,7,7]
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_movehdup_ps:
-; X64: # %bb.0:
-; X64-NEXT: vmovshdup {{.*#+}} ymm0 = ymm0[1,1,3,3,5,5,7,7]
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_movehdup_ps:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovshdup {{.*#+}} ymm0 = ymm0[1,1,3,3,5,5,7,7]
+; CHECK-NEXT: ret{{[l|q]}}
%res = shufflevector <8 x float> %a0, <8 x float> undef, <8 x i32> <i32 1, i32 1, i32 3, i32 3, i32 5, i32 5, i32 7, i32 7>
ret <8 x float> %res
}
define <8 x float> @test_mm256_mask_movehdup_ps(<8 x float> %a0, i8 %a1, <8 x float> %a2) {
-; X32-LABEL: test_mm256_mask_movehdup_ps:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vmovshdup {{.*#+}} ymm0 {%k1} = ymm1[1,1,3,3,5,5,7,7]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_movehdup_ps:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vmovshdup {{.*#+}} ymm0 {%k1} = ymm1[1,1,3,3,5,5,7,7]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_movehdup_ps:
; X64: # %bb.0:
@@ -2780,12 +2578,12 @@ define <8 x float> @test_mm256_mask_move
}
define <8 x float> @test_mm256_maskz_movehdup_ps(i8 %a0, <8 x float> %a1) {
-; X32-LABEL: test_mm256_maskz_movehdup_ps:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vmovshdup {{.*#+}} ymm0 {%k1} {z} = ymm0[1,1,3,3,5,5,7,7]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_movehdup_ps:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vmovshdup {{.*#+}} ymm0 {%k1} {z} = ymm0[1,1,3,3,5,5,7,7]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_movehdup_ps:
; X64: # %bb.0:
@@ -2799,26 +2597,21 @@ define <8 x float> @test_mm256_maskz_mov
}
define <4 x float> @test_mm_moveldup_ps(<4 x float> %a0) {
-; X32-LABEL: test_mm_moveldup_ps:
-; X32: # %bb.0:
-; X32-NEXT: vmovsldup {{.*#+}} xmm0 = xmm0[0,0,2,2]
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_moveldup_ps:
-; X64: # %bb.0:
-; X64-NEXT: vmovsldup {{.*#+}} xmm0 = xmm0[0,0,2,2]
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_moveldup_ps:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovsldup {{.*#+}} xmm0 = xmm0[0,0,2,2]
+; CHECK-NEXT: ret{{[l|q]}}
%res = shufflevector <4 x float> %a0, <4 x float> undef, <4 x i32> <i32 0, i32 0, i32 2, i32 2>
ret <4 x float> %res
}
define <4 x float> @test_mm_mask_moveldup_ps(<4 x float> %a0, i8 %a1, <4 x float> %a2) {
-; X32-LABEL: test_mm_mask_moveldup_ps:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vmovsldup {{.*#+}} xmm0 {%k1} = xmm1[0,0,2,2]
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_moveldup_ps:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vmovsldup {{.*#+}} xmm0 {%k1} = xmm1[0,0,2,2]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_moveldup_ps:
; X64: # %bb.0:
@@ -2833,12 +2626,12 @@ define <4 x float> @test_mm_mask_moveldu
}
define <4 x float> @test_mm_maskz_moveldup_ps(i8 %a0, <4 x float> %a1) {
-; X32-LABEL: test_mm_maskz_moveldup_ps:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vmovsldup {{.*#+}} xmm0 {%k1} {z} = xmm0[0,0,2,2]
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_moveldup_ps:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vmovsldup {{.*#+}} xmm0 {%k1} {z} = xmm0[0,0,2,2]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_moveldup_ps:
; X64: # %bb.0:
@@ -2853,26 +2646,21 @@ define <4 x float> @test_mm_maskz_moveld
}
define <8 x float> @test_mm256_moveldup_ps(<8 x float> %a0) {
-; X32-LABEL: test_mm256_moveldup_ps:
-; X32: # %bb.0:
-; X32-NEXT: vmovsldup {{.*#+}} ymm0 = ymm0[0,0,2,2,4,4,6,6]
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_moveldup_ps:
-; X64: # %bb.0:
-; X64-NEXT: vmovsldup {{.*#+}} ymm0 = ymm0[0,0,2,2,4,4,6,6]
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_moveldup_ps:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vmovsldup {{.*#+}} ymm0 = ymm0[0,0,2,2,4,4,6,6]
+; CHECK-NEXT: ret{{[l|q]}}
%res = shufflevector <8 x float> %a0, <8 x float> undef, <8 x i32> <i32 0, i32 0, i32 2, i32 2, i32 4, i32 4, i32 6, i32 6>
ret <8 x float> %res
}
define <8 x float> @test_mm256_mask_moveldup_ps(<8 x float> %a0, i8 %a1, <8 x float> %a2) {
-; X32-LABEL: test_mm256_mask_moveldup_ps:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vmovsldup {{.*#+}} ymm0 {%k1} = ymm1[0,0,2,2,4,4,6,6]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_moveldup_ps:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vmovsldup {{.*#+}} ymm0 {%k1} = ymm1[0,0,2,2,4,4,6,6]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_moveldup_ps:
; X64: # %bb.0:
@@ -2886,12 +2674,12 @@ define <8 x float> @test_mm256_mask_move
}
define <8 x float> @test_mm256_maskz_moveldup_ps(i8 %a0, <8 x float> %a1) {
-; X32-LABEL: test_mm256_maskz_moveldup_ps:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vmovsldup {{.*#+}} ymm0 {%k1} {z} = ymm0[0,0,2,2,4,4,6,6]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_moveldup_ps:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vmovsldup {{.*#+}} ymm0 {%k1} {z} = ymm0[0,0,2,2,4,4,6,6]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_moveldup_ps:
; X64: # %bb.0:
@@ -2905,26 +2693,21 @@ define <8 x float> @test_mm256_maskz_mov
}
define <4 x i64> @test_mm256_permutex_epi64(<4 x i64> %a0) {
-; X32-LABEL: test_mm256_permutex_epi64:
-; X32: # %bb.0:
-; X32-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[3,0,0,0]
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_permutex_epi64:
-; X64: # %bb.0:
-; X64-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[3,0,0,0]
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_permutex_epi64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[3,0,0,0]
+; CHECK-NEXT: ret{{[l|q]}}
%res = shufflevector <4 x i64> %a0, <4 x i64> undef, <4 x i32> <i32 3, i32 0, i32 0, i32 0>
ret <4 x i64> %res
}
define <4 x i64> @test_mm256_mask_permutex_epi64(<4 x i64> %a0, i8 %a1, <4 x i64> %a2) {
-; X32-LABEL: test_mm256_mask_permutex_epi64:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermq {{.*#+}} ymm0 {%k1} = ymm1[1,0,0,0]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_permutex_epi64:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermq {{.*#+}} ymm0 {%k1} = ymm1[1,0,0,0]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_permutex_epi64:
; X64: # %bb.0:
@@ -2939,12 +2722,12 @@ define <4 x i64> @test_mm256_mask_permut
}
define <4 x i64> @test_mm256_maskz_permutex_epi64(i8 %a0, <4 x i64> %a1) {
-; X32-LABEL: test_mm256_maskz_permutex_epi64:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermq {{.*#+}} ymm0 {%k1} {z} = ymm0[1,0,0,0]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_permutex_epi64:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermq {{.*#+}} ymm0 {%k1} {z} = ymm0[1,0,0,0]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_permutex_epi64:
; X64: # %bb.0:
@@ -2959,26 +2742,21 @@ define <4 x i64> @test_mm256_maskz_permu
}
define <4 x double> @test_mm256_permutex_pd(<4 x double> %a0) {
-; X32-LABEL: test_mm256_permutex_pd:
-; X32: # %bb.0:
-; X32-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[3,0,0,0]
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_permutex_pd:
-; X64: # %bb.0:
-; X64-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[3,0,0,0]
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_permutex_pd:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[3,0,0,0]
+; CHECK-NEXT: ret{{[l|q]}}
%res = shufflevector <4 x double> %a0, <4 x double> undef, <4 x i32> <i32 3, i32 0, i32 0, i32 0>
ret <4 x double> %res
}
define <4 x double> @test_mm256_mask_permutex_pd(<4 x double> %a0, i8 %a1, <4 x double> %a2) {
-; X32-LABEL: test_mm256_mask_permutex_pd:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermpd {{.*#+}} ymm0 {%k1} = ymm1[1,0,0,0]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_permutex_pd:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermpd {{.*#+}} ymm0 {%k1} = ymm1[1,0,0,0]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_permutex_pd:
; X64: # %bb.0:
@@ -2993,12 +2771,12 @@ define <4 x double> @test_mm256_mask_per
}
define <4 x double> @test_mm256_maskz_permutex_pd(i8 %a0, <4 x double> %a1) {
-; X32-LABEL: test_mm256_maskz_permutex_pd:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermpd {{.*#+}} ymm0 {%k1} {z} = ymm0[1,0,0,0]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_permutex_pd:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermpd {{.*#+}} ymm0 {%k1} {z} = ymm0[1,0,0,0]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_permutex_pd:
; X64: # %bb.0:
@@ -3013,26 +2791,21 @@ define <4 x double> @test_mm256_maskz_pe
}
define <2 x double> @test_mm_shuffle_pd(<2 x double> %a0, <2 x double> %a1) {
-; X32-LABEL: test_mm_shuffle_pd:
-; X32: # %bb.0:
-; X32-NEXT: vunpckhpd {{.*#+}} xmm0 = xmm0[1],xmm1[1]
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_shuffle_pd:
-; X64: # %bb.0:
-; X64-NEXT: vunpckhpd {{.*#+}} xmm0 = xmm0[1],xmm1[1]
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_shuffle_pd:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vunpckhpd {{.*#+}} xmm0 = xmm0[1],xmm1[1]
+; CHECK-NEXT: ret{{[l|q]}}
%res = shufflevector <2 x double> %a0, <2 x double> %a1, <2 x i32> <i32 1, i32 3>
ret <2 x double> %res
}
define <2 x double> @test_mm_mask_shuffle_pd(<2 x double> %a0, i8 %a1, <2 x double> %a2, <2 x double> %a3) {
-; X32-LABEL: test_mm_mask_shuffle_pd:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vunpckhpd {{.*#+}} xmm0 {%k1} = xmm1[1],xmm2[1]
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_shuffle_pd:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vunpckhpd {{.*#+}} xmm0 {%k1} = xmm1[1],xmm2[1]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_shuffle_pd:
; X64: # %bb.0:
@@ -3047,12 +2820,12 @@ define <2 x double> @test_mm_mask_shuffl
}
define <2 x double> @test_mm_maskz_shuffle_pd(i8 %a0, <2 x double> %a1, <2 x double> %a2) {
-; X32-LABEL: test_mm_maskz_shuffle_pd:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vunpckhpd {{.*#+}} xmm0 {%k1} {z} = xmm0[1],xmm1[1]
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_shuffle_pd:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vunpckhpd {{.*#+}} xmm0 {%k1} {z} = xmm0[1],xmm1[1]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_shuffle_pd:
; X64: # %bb.0:
@@ -3067,26 +2840,21 @@ define <2 x double> @test_mm_maskz_shuff
}
define <4 x double> @test_mm256_shuffle_pd(<4 x double> %a0, <4 x double> %a1) {
-; X32-LABEL: test_mm256_shuffle_pd:
-; X32: # %bb.0:
-; X32-NEXT: vshufpd {{.*#+}} ymm0 = ymm0[1],ymm1[1],ymm0[2],ymm1[2]
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_shuffle_pd:
-; X64: # %bb.0:
-; X64-NEXT: vshufpd {{.*#+}} ymm0 = ymm0[1],ymm1[1],ymm0[2],ymm1[2]
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_shuffle_pd:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vshufpd {{.*#+}} ymm0 = ymm0[1],ymm1[1],ymm0[2],ymm1[2]
+; CHECK-NEXT: ret{{[l|q]}}
%res = shufflevector <4 x double> %a0, <4 x double> %a1, <4 x i32> <i32 1, i32 5, i32 2, i32 6>
ret <4 x double> %res
}
define <4 x double> @test_mm256_mask_shuffle_pd(<4 x double> %a0, i8 %a1, <4 x double> %a2, <4 x double> %a3) {
-; X32-LABEL: test_mm256_mask_shuffle_pd:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vshufpd {{.*#+}} ymm0 {%k1} = ymm1[1],ymm2[1],ymm1[2],ymm2[2]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_shuffle_pd:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vshufpd {{.*#+}} ymm0 {%k1} = ymm1[1],ymm2[1],ymm1[2],ymm2[2]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_shuffle_pd:
; X64: # %bb.0:
@@ -3101,12 +2869,12 @@ define <4 x double> @test_mm256_mask_shu
}
define <4 x double> @test_mm256_maskz_shuffle_pd(i8 %a0, <4 x double> %a1, <4 x double> %a2) {
-; X32-LABEL: test_mm256_maskz_shuffle_pd:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vshufpd {{.*#+}} ymm0 {%k1} {z} = ymm0[1],ymm1[1],ymm0[2],ymm1[2]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_shuffle_pd:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vshufpd {{.*#+}} ymm0 {%k1} {z} = ymm0[1],ymm1[1],ymm0[2],ymm1[2]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_shuffle_pd:
; X64: # %bb.0:
@@ -3121,26 +2889,21 @@ define <4 x double> @test_mm256_maskz_sh
}
define <4 x float> @test_mm_shuffle_ps(<4 x float> %a0, <4 x float> %a1) {
-; X32-LABEL: test_mm_shuffle_ps:
-; X32: # %bb.0:
-; X32-NEXT: vshufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[0,0]
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_shuffle_ps:
-; X64: # %bb.0:
-; X64-NEXT: vshufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[0,0]
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_shuffle_ps:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vshufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[0,0]
+; CHECK-NEXT: ret{{[l|q]}}
%res = shufflevector <4 x float> %a0, <4 x float> %a1, <4 x i32> <i32 0, i32 1, i32 4, i32 4>
ret <4 x float> %res
}
define <4 x float> @test_mm_mask_shuffle_ps(<4 x float> %a0, i8 %a1, <4 x float> %a2, <4 x float> %a3) {
-; X32-LABEL: test_mm_mask_shuffle_ps:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vshufps {{.*#+}} xmm0 {%k1} = xmm1[0,1],xmm2[0,0]
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_shuffle_ps:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vshufps {{.*#+}} xmm0 {%k1} = xmm1[0,1],xmm2[0,0]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_shuffle_ps:
; X64: # %bb.0:
@@ -3155,12 +2918,12 @@ define <4 x float> @test_mm_mask_shuffle
}
define <4 x float> @test_mm_maskz_shuffle_ps(i8 %a0, <4 x float> %a1, <4 x float> %a2) {
-; X32-LABEL: test_mm_maskz_shuffle_ps:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vshufps {{.*#+}} xmm0 {%k1} {z} = xmm0[0,1],xmm1[0,0]
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_shuffle_ps:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vshufps {{.*#+}} xmm0 {%k1} {z} = xmm0[0,1],xmm1[0,0]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_shuffle_ps:
; X64: # %bb.0:
@@ -3175,26 +2938,21 @@ define <4 x float> @test_mm_maskz_shuffl
}
define <8 x float> @test_mm256_shuffle_ps(<8 x float> %a0, <8 x float> %a1) {
-; X32-LABEL: test_mm256_shuffle_ps:
-; X32: # %bb.0:
-; X32-NEXT: vshufps {{.*#+}} ymm0 = ymm0[0,1],ymm1[0,0],ymm0[4,5],ymm1[4,4]
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_shuffle_ps:
-; X64: # %bb.0:
-; X64-NEXT: vshufps {{.*#+}} ymm0 = ymm0[0,1],ymm1[0,0],ymm0[4,5],ymm1[4,4]
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_shuffle_ps:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vshufps {{.*#+}} ymm0 = ymm0[0,1],ymm1[0,0],ymm0[4,5],ymm1[4,4]
+; CHECK-NEXT: ret{{[l|q]}}
%res = shufflevector <8 x float> %a0, <8 x float> %a1, <8 x i32> <i32 0, i32 1, i32 8, i32 8, i32 4, i32 5, i32 12, i32 12>
ret <8 x float> %res
}
define <8 x float> @test_mm256_mask_shuffle_ps(<8 x float> %a0, i8 %a1, <8 x float> %a2, <8 x float> %a3) {
-; X32-LABEL: test_mm256_mask_shuffle_ps:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vshufps {{.*#+}} ymm0 {%k1} = ymm1[0,1],ymm2[0,0],ymm1[4,5],ymm2[4,4]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_shuffle_ps:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vshufps {{.*#+}} ymm0 {%k1} = ymm1[0,1],ymm2[0,0],ymm1[4,5],ymm2[4,4]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_shuffle_ps:
; X64: # %bb.0:
@@ -3208,12 +2966,12 @@ define <8 x float> @test_mm256_mask_shuf
}
define <8 x float> @test_mm256_maskz_shuffle_ps(i8 %a0, <8 x float> %a1, <8 x float> %a2) {
-; X32-LABEL: test_mm256_maskz_shuffle_ps:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vshufps {{.*#+}} ymm0 {%k1} {z} = ymm0[0,1],ymm1[0,0],ymm0[4,5],ymm1[4,4]
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_shuffle_ps:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vshufps {{.*#+}} ymm0 {%k1} {z} = ymm0[0,1],ymm1[0,0],ymm0[4,5],ymm1[4,4]
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_shuffle_ps:
; X64: # %bb.0:
@@ -3227,12 +2985,12 @@ define <8 x float> @test_mm256_maskz_shu
}
define <4 x i64> @test_mm256_mask_mul_epi32(<4 x i64> %__W, i8 zeroext %__M, <4 x i64> %__X, <4 x i64> %__Y) nounwind {
-; X32-LABEL: test_mm256_mask_mul_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpmuldq %ymm1, %ymm2, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_mul_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpmuldq %ymm1, %ymm2, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_mul_epi32:
; X64: # %bb.0: # %entry
@@ -3252,12 +3010,12 @@ entry:
}
define <4 x i64> @test_mm256_maskz_mul_epi32(i8 zeroext %__M, <4 x i64> %__X, <4 x i64> %__Y) nounwind {
-; X32-LABEL: test_mm256_maskz_mul_epi32:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpmuldq %ymm0, %ymm1, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_mul_epi32:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpmuldq %ymm0, %ymm1, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_mul_epi32:
; X64: # %bb.0:
@@ -3276,12 +3034,12 @@ define <4 x i64> @test_mm256_maskz_mul_e
}
define <2 x i64> @test_mm_mask_mul_epi32(<2 x i64> %__W, i8 zeroext %__M, <2 x i64> %__X, <2 x i64> %__Y) nounwind {
-; X32-LABEL: test_mm_mask_mul_epi32:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpmuldq %xmm1, %xmm2, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_mul_epi32:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpmuldq %xmm1, %xmm2, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_mul_epi32:
; X64: # %bb.0:
@@ -3300,12 +3058,12 @@ define <2 x i64> @test_mm_mask_mul_epi32
}
define <2 x i64> @test_mm_maskz_mul_epi32(i8 zeroext %__M, <2 x i64> %__X, <2 x i64> %__Y) nounwind {
-; X32-LABEL: test_mm_maskz_mul_epi32:
-; X32: # %bb.0:
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpmuldq %xmm0, %xmm1, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_mul_epi32:
+; X86: # %bb.0:
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpmuldq %xmm0, %xmm1, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_mul_epi32:
; X64: # %bb.0:
@@ -3324,12 +3082,12 @@ define <2 x i64> @test_mm_maskz_mul_epi3
}
define <4 x i64> @test_mm256_mask_mul_epu32(<4 x i64> %__W, i8 zeroext %__M, <4 x i64> %__X, <4 x i64> %__Y) nounwind {
-; X32-LABEL: test_mm256_mask_mul_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpmuludq %ymm1, %ymm2, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_mul_epu32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpmuludq %ymm1, %ymm2, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_mul_epu32:
; X64: # %bb.0: # %entry
@@ -3347,12 +3105,12 @@ entry:
}
define <4 x i64> @test_mm256_maskz_mul_epu32(i8 zeroext %__M, <4 x i64> %__X, <4 x i64> %__Y) nounwind {
-; X32-LABEL: test_mm256_maskz_mul_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpmuludq %ymm0, %ymm1, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_mul_epu32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpmuludq %ymm0, %ymm1, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_mul_epu32:
; X64: # %bb.0: # %entry
@@ -3370,12 +3128,12 @@ entry:
}
define <2 x i64> @test_mm_mask_mul_epu32(<2 x i64> %__W, i8 zeroext %__M, <2 x i64> %__X, <2 x i64> %__Y) nounwind {
-; X32-LABEL: test_mm_mask_mul_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpmuludq %xmm1, %xmm2, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_mul_epu32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpmuludq %xmm1, %xmm2, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_mul_epu32:
; X64: # %bb.0: # %entry
@@ -3393,12 +3151,12 @@ entry:
}
define <2 x i64> @test_mm_maskz_mul_epu32(i8 zeroext %__M, <2 x i64> %__X, <2 x i64> %__Y) nounwind {
-; X32-LABEL: test_mm_maskz_mul_epu32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpmuludq %xmm0, %xmm1, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_mul_epu32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpmuludq %xmm0, %xmm1, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_mul_epu32:
; X64: # %bb.0: # %entry
@@ -3416,17 +3174,11 @@ entry:
}
define <2 x i64> @test_mm256_cvtepi32_epi16(<4 x i64> %__A) local_unnamed_addr #0 {
-; X32-LABEL: test_mm256_cvtepi32_epi16:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vpmovdw %ymm0, %xmm0
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_cvtepi32_epi16:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vpmovdw %ymm0, %xmm0
-; X64-NEXT: vzeroupper
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_cvtepi32_epi16:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vpmovdw %ymm0, %xmm0
+; CHECK-NEXT: vzeroupper
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = bitcast <4 x i64> %__A to <8 x i32>
%conv.i = trunc <8 x i32> %0 to <8 x i16>
@@ -3435,13 +3187,13 @@ entry:
}
define <2 x i64> @test_mm256_mask_cvtepi32_epi16(<2 x i64> %__O, i8 zeroext %__M, <4 x i64> %__A) {
-; X32-LABEL: test_mm256_mask_cvtepi32_epi16:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpmovdw %ymm1, %xmm0 {%k1}
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_cvtepi32_epi16:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpmovdw %ymm1, %xmm0 {%k1}
+; X86-NEXT: vzeroupper
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_cvtepi32_epi16:
; X64: # %bb.0: # %entry
@@ -3458,13 +3210,13 @@ entry:
}
define <2 x i64> @test_mm256_maskz_cvtepi32_epi16(i8 zeroext %__M, <4 x i64> %__A) {
-; X32-LABEL: test_mm256_maskz_cvtepi32_epi16:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpmovdw %ymm0, %xmm0 {%k1} {z}
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_cvtepi32_epi16:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpmovdw %ymm0, %xmm0 {%k1} {z}
+; X86-NEXT: vzeroupper
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_cvtepi32_epi16:
; X64: # %bb.0: # %entry
@@ -3480,17 +3232,11 @@ entry:
}
define <2 x i64> @test_mm256_cvtepi64_epi32(<4 x i64> %__A) local_unnamed_addr #0 {
-; X32-LABEL: test_mm256_cvtepi64_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vpmovqd %ymm0, %xmm0
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_cvtepi64_epi32:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vpmovqd %ymm0, %xmm0
-; X64-NEXT: vzeroupper
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_cvtepi64_epi32:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vpmovqd %ymm0, %xmm0
+; CHECK-NEXT: vzeroupper
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%conv.i = trunc <4 x i64> %__A to <4 x i32>
%0 = bitcast <4 x i32> %conv.i to <2 x i64>
@@ -3498,13 +3244,13 @@ entry:
}
define <2 x i64> @test_mm256_mask_cvtepi64_epi32(<2 x i64> %__O, i8 zeroext %__M, <4 x i64> %__A) {
-; X32-LABEL: test_mm256_mask_cvtepi64_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpmovqd %ymm1, %xmm0 {%k1}
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_cvtepi64_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpmovqd %ymm1, %xmm0 {%k1}
+; X86-NEXT: vzeroupper
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_cvtepi64_epi32:
; X64: # %bb.0: # %entry
@@ -3523,13 +3269,13 @@ entry:
}
define <2 x i64> @test_mm256_maskz_cvtepi64_epi32(i8 zeroext %__M, <4 x i64> %__A) {
-; X32-LABEL: test_mm256_maskz_cvtepi64_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpmovqd %ymm0, %xmm0 {%k1} {z}
-; X32-NEXT: vzeroupper
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_cvtepi64_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpmovqd %ymm0, %xmm0 {%k1} {z}
+; X86-NEXT: vzeroupper
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_cvtepi64_epi32:
; X64: # %bb.0: # %entry
@@ -3547,15 +3293,10 @@ entry:
}
define <2 x i64> @test_mm_ternarylogic_epi32(<2 x i64> %__A, <2 x i64> %__B, <2 x i64> %__C) {
-; X32-LABEL: test_mm_ternarylogic_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vpternlogd $4, %xmm2, %xmm1, %xmm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_ternarylogic_epi32:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vpternlogd $4, %xmm2, %xmm1, %xmm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_ternarylogic_epi32:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vpternlogd $4, %xmm2, %xmm1, %xmm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = bitcast <2 x i64> %__A to <4 x i32>
%1 = bitcast <2 x i64> %__B to <4 x i32>
@@ -3568,12 +3309,12 @@ entry:
declare <4 x i32> @llvm.x86.avx512.pternlog.d.128(<4 x i32>, <4 x i32>, <4 x i32>, i32) #2
define <2 x i64> @test_mm_mask_ternarylogic_epi32(<2 x i64> %__A, i8 zeroext %__U, <2 x i64> %__B, <2 x i64> %__C) {
-; X32-LABEL: test_mm_mask_ternarylogic_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpternlogd $4, %xmm2, %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_ternarylogic_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpternlogd $4, %xmm2, %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_ternarylogic_epi32:
; X64: # %bb.0: # %entry
@@ -3593,12 +3334,12 @@ entry:
}
define <2 x i64> @test_mm_maskz_ternarylogic_epi32(i8 zeroext %__U, <2 x i64> %__A, <2 x i64> %__B, <2 x i64> %__C) {
-; X32-LABEL: test_mm_maskz_ternarylogic_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpternlogd $4, %xmm2, %xmm1, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_ternarylogic_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpternlogd $4, %xmm2, %xmm1, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_ternarylogic_epi32:
; X64: # %bb.0: # %entry
@@ -3618,15 +3359,10 @@ entry:
}
define <4 x i64> @test_mm256_ternarylogic_epi32(<4 x i64> %__A, <4 x i64> %__B, <4 x i64> %__C) {
-; X32-LABEL: test_mm256_ternarylogic_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vpternlogd $4, %ymm2, %ymm1, %ymm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_ternarylogic_epi32:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vpternlogd $4, %ymm2, %ymm1, %ymm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_ternarylogic_epi32:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vpternlogd $4, %ymm2, %ymm1, %ymm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = bitcast <4 x i64> %__A to <8 x i32>
%1 = bitcast <4 x i64> %__B to <8 x i32>
@@ -3639,12 +3375,12 @@ entry:
declare <8 x i32> @llvm.x86.avx512.pternlog.d.256(<8 x i32>, <8 x i32>, <8 x i32>, i32) #2
define <4 x i64> @test_mm256_mask_ternarylogic_epi32(<4 x i64> %__A, i8 zeroext %__U, <4 x i64> %__B, <4 x i64> %__C) {
-; X32-LABEL: test_mm256_mask_ternarylogic_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpternlogd $4, %ymm2, %ymm1, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_ternarylogic_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpternlogd $4, %ymm2, %ymm1, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_ternarylogic_epi32:
; X64: # %bb.0: # %entry
@@ -3663,12 +3399,12 @@ entry:
}
define <4 x i64> @test_mm256_maskz_ternarylogic_epi32(i8 zeroext %__U, <4 x i64> %__A, <4 x i64> %__B, <4 x i64> %__C) {
-; X32-LABEL: test_mm256_maskz_ternarylogic_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpternlogd $4, %ymm2, %ymm1, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_ternarylogic_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpternlogd $4, %ymm2, %ymm1, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_ternarylogic_epi32:
; X64: # %bb.0: # %entry
@@ -3687,15 +3423,10 @@ entry:
}
define <2 x i64> @test_mm_ternarylogic_epi64(<2 x i64> %__A, <2 x i64> %__B, <2 x i64> %__C) {
-; X32-LABEL: test_mm_ternarylogic_epi64:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vpternlogq $4, %xmm2, %xmm1, %xmm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_ternarylogic_epi64:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vpternlogq $4, %xmm2, %xmm1, %xmm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_ternarylogic_epi64:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vpternlogq $4, %xmm2, %xmm1, %xmm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = tail call <2 x i64> @llvm.x86.avx512.pternlog.q.128(<2 x i64> %__A, <2 x i64> %__B, <2 x i64> %__C, i32 4)
ret <2 x i64> %0
@@ -3704,12 +3435,12 @@ entry:
declare <2 x i64> @llvm.x86.avx512.pternlog.q.128(<2 x i64>, <2 x i64>, <2 x i64>, i32) #2
define <2 x i64> @test_mm_mask_ternarylogic_epi64(<2 x i64> %__A, i8 zeroext %__U, <2 x i64> %__B, <2 x i64> %__C) {
-; X32-LABEL: test_mm_mask_ternarylogic_epi64:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpternlogq $4, %xmm2, %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_ternarylogic_epi64:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpternlogq $4, %xmm2, %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_ternarylogic_epi64:
; X64: # %bb.0: # %entry
@@ -3725,12 +3456,12 @@ entry:
}
define <2 x i64> @test_mm_maskz_ternarylogic_epi64(i8 zeroext %__U, <2 x i64> %__A, <2 x i64> %__B, <2 x i64> %__C) {
-; X32-LABEL: test_mm_maskz_ternarylogic_epi64:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpternlogq $4, %xmm2, %xmm1, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_ternarylogic_epi64:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpternlogq $4, %xmm2, %xmm1, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_ternarylogic_epi64:
; X64: # %bb.0: # %entry
@@ -3746,15 +3477,10 @@ entry:
}
define <4 x i64> @test_mm256_ternarylogic_epi64(<4 x i64> %__A, <4 x i64> %__B, <4 x i64> %__C) {
-; X32-LABEL: test_mm256_ternarylogic_epi64:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vpternlogq $4, %ymm2, %ymm1, %ymm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_ternarylogic_epi64:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vpternlogq $4, %ymm2, %ymm1, %ymm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_ternarylogic_epi64:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vpternlogq $4, %ymm2, %ymm1, %ymm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = tail call <4 x i64> @llvm.x86.avx512.pternlog.q.256(<4 x i64> %__A, <4 x i64> %__B, <4 x i64> %__C, i32 4)
ret <4 x i64> %0
@@ -3763,12 +3489,12 @@ entry:
declare <4 x i64> @llvm.x86.avx512.pternlog.q.256(<4 x i64>, <4 x i64>, <4 x i64>, i32) #2
define <4 x i64> @test_mm256_mask_ternarylogic_epi64(<4 x i64> %__A, i8 zeroext %__U, <4 x i64> %__B, <4 x i64> %__C) {
-; X32-LABEL: test_mm256_mask_ternarylogic_epi64:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpternlogq $4, %ymm2, %ymm1, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_ternarylogic_epi64:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpternlogq $4, %ymm2, %ymm1, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_ternarylogic_epi64:
; X64: # %bb.0: # %entry
@@ -3784,12 +3510,12 @@ entry:
}
define <4 x i64> @test_mm256_maskz_ternarylogic_epi64(i8 zeroext %__U, <4 x i64> %__A, <4 x i64> %__B, <4 x i64> %__C) {
-; X32-LABEL: test_mm256_maskz_ternarylogic_epi64:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpternlogq $4, %ymm2, %ymm1, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_ternarylogic_epi64:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpternlogq $4, %ymm2, %ymm1, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_ternarylogic_epi64:
; X64: # %bb.0: # %entry
@@ -3805,13 +3531,13 @@ entry:
}
define <2 x i64> @test_mm_mask2_permutex2var_epi32(<2 x i64> %__A, <2 x i64> %__I, i8 zeroext %__U, <2 x i64> %__B) {
-; X32-LABEL: test_mm_mask2_permutex2var_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermi2d %xmm2, %xmm0, %xmm1 {%k1}
-; X32-NEXT: vmovdqa %xmm1, %xmm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask2_permutex2var_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermi2d %xmm2, %xmm0, %xmm1 {%k1}
+; X86-NEXT: vmovdqa %xmm1, %xmm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask2_permutex2var_epi32:
; X64: # %bb.0: # %entry
@@ -3832,13 +3558,13 @@ entry:
}
define <4 x i64> @test_mm256_mask2_permutex2var_epi32(<4 x i64> %__A, <4 x i64> %__I, i8 zeroext %__U, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_mask2_permutex2var_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermi2d %ymm2, %ymm0, %ymm1 {%k1}
-; X32-NEXT: vmovdqa %ymm1, %ymm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask2_permutex2var_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermi2d %ymm2, %ymm0, %ymm1 {%k1}
+; X86-NEXT: vmovdqa %ymm1, %ymm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask2_permutex2var_epi32:
; X64: # %bb.0: # %entry
@@ -3858,13 +3584,13 @@ entry:
}
define <2 x double> @test_mm_mask2_permutex2var_pd(<2 x double> %__A, <2 x i64> %__I, i8 zeroext %__U, <2 x double> %__B) {
-; X32-LABEL: test_mm_mask2_permutex2var_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermi2pd %xmm2, %xmm0, %xmm1 {%k1}
-; X32-NEXT: vmovapd %xmm1, %xmm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask2_permutex2var_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermi2pd %xmm2, %xmm0, %xmm1 {%k1}
+; X86-NEXT: vmovapd %xmm1, %xmm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask2_permutex2var_pd:
; X64: # %bb.0: # %entry
@@ -3882,13 +3608,13 @@ entry:
}
define <4 x double> @test_mm256_mask2_permutex2var_pd(<4 x double> %__A, <4 x i64> %__I, i8 zeroext %__U, <4 x double> %__B) {
-; X32-LABEL: test_mm256_mask2_permutex2var_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermi2pd %ymm2, %ymm0, %ymm1 {%k1}
-; X32-NEXT: vmovapd %ymm1, %ymm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask2_permutex2var_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermi2pd %ymm2, %ymm0, %ymm1 {%k1}
+; X86-NEXT: vmovapd %ymm1, %ymm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask2_permutex2var_pd:
; X64: # %bb.0: # %entry
@@ -3906,13 +3632,13 @@ entry:
}
define <4 x float> @test_mm_mask2_permutex2var_ps(<4 x float> %__A, <2 x i64> %__I, i8 zeroext %__U, <4 x float> %__B) {
-; X32-LABEL: test_mm_mask2_permutex2var_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermi2ps %xmm2, %xmm0, %xmm1 {%k1}
-; X32-NEXT: vmovaps %xmm1, %xmm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask2_permutex2var_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermi2ps %xmm2, %xmm0, %xmm1 {%k1}
+; X86-NEXT: vmovaps %xmm1, %xmm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask2_permutex2var_ps:
; X64: # %bb.0: # %entry
@@ -3931,13 +3657,13 @@ entry:
}
define <8 x float> @test_mm256_mask2_permutex2var_ps(<8 x float> %__A, <4 x i64> %__I, i8 zeroext %__U, <8 x float> %__B) {
-; X32-LABEL: test_mm256_mask2_permutex2var_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermi2ps %ymm2, %ymm0, %ymm1 {%k1}
-; X32-NEXT: vmovaps %ymm1, %ymm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask2_permutex2var_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermi2ps %ymm2, %ymm0, %ymm1 {%k1}
+; X86-NEXT: vmovaps %ymm1, %ymm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask2_permutex2var_ps:
; X64: # %bb.0: # %entry
@@ -3955,13 +3681,13 @@ entry:
}
define <2 x i64> @test_mm_mask2_permutex2var_epi64(<2 x i64> %__A, <2 x i64> %__I, i8 zeroext %__U, <2 x i64> %__B) {
-; X32-LABEL: test_mm_mask2_permutex2var_epi64:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermi2q %xmm2, %xmm0, %xmm1 {%k1}
-; X32-NEXT: vmovdqa %xmm1, %xmm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask2_permutex2var_epi64:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermi2q %xmm2, %xmm0, %xmm1 {%k1}
+; X86-NEXT: vmovdqa %xmm1, %xmm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask2_permutex2var_epi64:
; X64: # %bb.0: # %entry
@@ -3978,13 +3704,13 @@ entry:
}
define <4 x i64> @test_mm256_mask2_permutex2var_epi64(<4 x i64> %__A, <4 x i64> %__I, i8 zeroext %__U, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_mask2_permutex2var_epi64:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermi2q %ymm2, %ymm0, %ymm1 {%k1}
-; X32-NEXT: vmovdqa %ymm1, %ymm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask2_permutex2var_epi64:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermi2q %ymm2, %ymm0, %ymm1 {%k1}
+; X86-NEXT: vmovdqa %ymm1, %ymm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask2_permutex2var_epi64:
; X64: # %bb.0: # %entry
@@ -4001,15 +3727,10 @@ entry:
}
define <2 x i64> @test_mm_permutex2var_epi32(<2 x i64> %__A, <2 x i64> %__I, <2 x i64> %__B) {
-; X32-LABEL: test_mm_permutex2var_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vpermt2d %xmm2, %xmm1, %xmm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_permutex2var_epi32:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vpermt2d %xmm2, %xmm1, %xmm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_permutex2var_epi32:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vpermt2d %xmm2, %xmm1, %xmm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = bitcast <2 x i64> %__A to <4 x i32>
%1 = bitcast <2 x i64> %__I to <4 x i32>
@@ -4020,12 +3741,12 @@ entry:
}
define <2 x i64> @test_mm_mask_permutex2var_epi32(<2 x i64> %__A, i8 zeroext %__U, <2 x i64> %__I, <2 x i64> %__B) {
-; X32-LABEL: test_mm_mask_permutex2var_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermt2d %xmm2, %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_permutex2var_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermt2d %xmm2, %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_permutex2var_epi32:
; X64: # %bb.0: # %entry
@@ -4045,12 +3766,12 @@ entry:
}
define <2 x i64> @test_mm_maskz_permutex2var_epi32(i8 zeroext %__U, <2 x i64> %__A, <2 x i64> %__I, <2 x i64> %__B) {
-; X32-LABEL: test_mm_maskz_permutex2var_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermt2d %xmm2, %xmm1, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_permutex2var_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermt2d %xmm2, %xmm1, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_permutex2var_epi32:
; X64: # %bb.0: # %entry
@@ -4070,15 +3791,10 @@ entry:
}
define <4 x i64> @test_mm256_permutex2var_epi32(<4 x i64> %__A, <4 x i64> %__I, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_permutex2var_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vpermt2d %ymm2, %ymm1, %ymm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_permutex2var_epi32:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vpermt2d %ymm2, %ymm1, %ymm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_permutex2var_epi32:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vpermt2d %ymm2, %ymm1, %ymm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = bitcast <4 x i64> %__A to <8 x i32>
%1 = bitcast <4 x i64> %__I to <8 x i32>
@@ -4089,12 +3805,12 @@ entry:
}
define <4 x i64> @test_mm256_mask_permutex2var_epi32(<4 x i64> %__A, i8 zeroext %__U, <4 x i64> %__I, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_mask_permutex2var_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermt2d %ymm2, %ymm1, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_permutex2var_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermt2d %ymm2, %ymm1, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_permutex2var_epi32:
; X64: # %bb.0: # %entry
@@ -4113,12 +3829,12 @@ entry:
}
define <4 x i64> @test_mm256_maskz_permutex2var_epi32(i8 zeroext %__U, <4 x i64> %__A, <4 x i64> %__I, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_maskz_permutex2var_epi32:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermt2d %ymm2, %ymm1, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_permutex2var_epi32:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermt2d %ymm2, %ymm1, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_permutex2var_epi32:
; X64: # %bb.0: # %entry
@@ -4137,27 +3853,22 @@ entry:
}
define <2 x double> @test_mm_permutex2var_pd(<2 x double> %__A, <2 x i64> %__I, <2 x double> %__B) {
-; X32-LABEL: test_mm_permutex2var_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vpermt2pd %xmm2, %xmm1, %xmm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_permutex2var_pd:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vpermt2pd %xmm2, %xmm1, %xmm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_permutex2var_pd:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vpermt2pd %xmm2, %xmm1, %xmm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = tail call <2 x double> @llvm.x86.avx512.vpermi2var.pd.128(<2 x double> %__A, <2 x i64> %__I, <2 x double> %__B)
ret <2 x double> %0
}
define <2 x double> @test_mm_mask_permutex2var_pd(<2 x double> %__A, i8 zeroext %__U, <2 x i64> %__I, <2 x double> %__B) {
-; X32-LABEL: test_mm_mask_permutex2var_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermt2pd %xmm2, %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_permutex2var_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermt2pd %xmm2, %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_permutex2var_pd:
; X64: # %bb.0: # %entry
@@ -4173,12 +3884,12 @@ entry:
}
define <2 x double> @test_mm_maskz_permutex2var_pd(i8 zeroext %__U, <2 x double> %__A, <2 x i64> %__I, <2 x double> %__B) {
-; X32-LABEL: test_mm_maskz_permutex2var_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermt2pd %xmm2, %xmm1, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_permutex2var_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermt2pd %xmm2, %xmm1, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_permutex2var_pd:
; X64: # %bb.0: # %entry
@@ -4194,27 +3905,22 @@ entry:
}
define <4 x double> @test_mm256_permutex2var_pd(<4 x double> %__A, <4 x i64> %__I, <4 x double> %__B) {
-; X32-LABEL: test_mm256_permutex2var_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vpermt2pd %ymm2, %ymm1, %ymm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_permutex2var_pd:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vpermt2pd %ymm2, %ymm1, %ymm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_permutex2var_pd:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vpermt2pd %ymm2, %ymm1, %ymm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = tail call <4 x double> @llvm.x86.avx512.vpermi2var.pd.256(<4 x double> %__A, <4 x i64> %__I, <4 x double> %__B)
ret <4 x double> %0
}
define <4 x double> @test_mm256_mask_permutex2var_pd(<4 x double> %__A, i8 zeroext %__U, <4 x i64> %__I, <4 x double> %__B) {
-; X32-LABEL: test_mm256_mask_permutex2var_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermt2pd %ymm2, %ymm1, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_permutex2var_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermt2pd %ymm2, %ymm1, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_permutex2var_pd:
; X64: # %bb.0: # %entry
@@ -4230,12 +3936,12 @@ entry:
}
define <4 x double> @test_mm256_maskz_permutex2var_pd(i8 zeroext %__U, <4 x double> %__A, <4 x i64> %__I, <4 x double> %__B) {
-; X32-LABEL: test_mm256_maskz_permutex2var_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermt2pd %ymm2, %ymm1, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_permutex2var_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermt2pd %ymm2, %ymm1, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_permutex2var_pd:
; X64: # %bb.0: # %entry
@@ -4251,15 +3957,10 @@ entry:
}
define <4 x float> @test_mm_permutex2var_ps(<4 x float> %__A, <2 x i64> %__I, <4 x float> %__B) {
-; X32-LABEL: test_mm_permutex2var_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vpermt2ps %xmm2, %xmm1, %xmm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_permutex2var_ps:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vpermt2ps %xmm2, %xmm1, %xmm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_permutex2var_ps:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vpermt2ps %xmm2, %xmm1, %xmm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = bitcast <2 x i64> %__I to <4 x i32>
%1 = tail call <4 x float> @llvm.x86.avx512.vpermi2var.ps.128(<4 x float> %__A, <4 x i32> %0, <4 x float> %__B)
@@ -4267,12 +3968,12 @@ entry:
}
define <4 x float> @test_mm_mask_permutex2var_ps(<4 x float> %__A, i8 zeroext %__U, <2 x i64> %__I, <4 x float> %__B) {
-; X32-LABEL: test_mm_mask_permutex2var_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermt2ps %xmm2, %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_permutex2var_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermt2ps %xmm2, %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_permutex2var_ps:
; X64: # %bb.0: # %entry
@@ -4289,12 +3990,12 @@ entry:
}
define <4 x float> @test_mm_maskz_permutex2var_ps(i8 zeroext %__U, <4 x float> %__A, <2 x i64> %__I, <4 x float> %__B) {
-; X32-LABEL: test_mm_maskz_permutex2var_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermt2ps %xmm2, %xmm1, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_permutex2var_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermt2ps %xmm2, %xmm1, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_permutex2var_ps:
; X64: # %bb.0: # %entry
@@ -4311,15 +4012,10 @@ entry:
}
define <8 x float> @test_mm256_permutex2var_ps(<8 x float> %__A, <4 x i64> %__I, <8 x float> %__B) {
-; X32-LABEL: test_mm256_permutex2var_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vpermt2ps %ymm2, %ymm1, %ymm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_permutex2var_ps:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vpermt2ps %ymm2, %ymm1, %ymm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_permutex2var_ps:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vpermt2ps %ymm2, %ymm1, %ymm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = bitcast <4 x i64> %__I to <8 x i32>
%1 = tail call <8 x float> @llvm.x86.avx512.vpermi2var.ps.256(<8 x float> %__A, <8 x i32> %0, <8 x float> %__B)
@@ -4327,12 +4023,12 @@ entry:
}
define <8 x float> @test_mm256_mask_permutex2var_ps(<8 x float> %__A, i8 zeroext %__U, <4 x i64> %__I, <8 x float> %__B) {
-; X32-LABEL: test_mm256_mask_permutex2var_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermt2ps %ymm2, %ymm1, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_permutex2var_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermt2ps %ymm2, %ymm1, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_permutex2var_ps:
; X64: # %bb.0: # %entry
@@ -4348,12 +4044,12 @@ entry:
}
define <8 x float> @test_mm256_maskz_permutex2var_ps(i8 zeroext %__U, <8 x float> %__A, <4 x i64> %__I, <8 x float> %__B) {
-; X32-LABEL: test_mm256_maskz_permutex2var_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermt2ps %ymm2, %ymm1, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_permutex2var_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermt2ps %ymm2, %ymm1, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_permutex2var_ps:
; X64: # %bb.0: # %entry
@@ -4369,27 +4065,22 @@ entry:
}
define <2 x i64> @test_mm_permutex2var_epi64(<2 x i64> %__A, <2 x i64> %__I, <2 x i64> %__B) {
-; X32-LABEL: test_mm_permutex2var_epi64:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vpermt2q %xmm2, %xmm1, %xmm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm_permutex2var_epi64:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vpermt2q %xmm2, %xmm1, %xmm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm_permutex2var_epi64:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vpermt2q %xmm2, %xmm1, %xmm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = tail call <2 x i64> @llvm.x86.avx512.vpermi2var.q.128(<2 x i64> %__A, <2 x i64> %__I, <2 x i64> %__B)
ret <2 x i64> %0
}
define <2 x i64> @test_mm_mask_permutex2var_epi64(<2 x i64> %__A, i8 zeroext %__U, <2 x i64> %__I, <2 x i64> %__B) {
-; X32-LABEL: test_mm_mask_permutex2var_epi64:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermt2q %xmm2, %xmm1, %xmm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_permutex2var_epi64:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermt2q %xmm2, %xmm1, %xmm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_permutex2var_epi64:
; X64: # %bb.0: # %entry
@@ -4405,12 +4096,12 @@ entry:
}
define <2 x i64> @test_mm_maskz_permutex2var_epi64(i8 zeroext %__U, <2 x i64> %__A, <2 x i64> %__I, <2 x i64> %__B) {
-; X32-LABEL: test_mm_maskz_permutex2var_epi64:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermt2q %xmm2, %xmm1, %xmm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_permutex2var_epi64:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermt2q %xmm2, %xmm1, %xmm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_permutex2var_epi64:
; X64: # %bb.0: # %entry
@@ -4426,27 +4117,22 @@ entry:
}
define <4 x i64> @test_mm256_permutex2var_epi64(<4 x i64> %__A, <4 x i64> %__I, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_permutex2var_epi64:
-; X32: # %bb.0: # %entry
-; X32-NEXT: vpermt2q %ymm2, %ymm1, %ymm0
-; X32-NEXT: retl
-;
-; X64-LABEL: test_mm256_permutex2var_epi64:
-; X64: # %bb.0: # %entry
-; X64-NEXT: vpermt2q %ymm2, %ymm1, %ymm0
-; X64-NEXT: retq
+; CHECK-LABEL: test_mm256_permutex2var_epi64:
+; CHECK: # %bb.0: # %entry
+; CHECK-NEXT: vpermt2q %ymm2, %ymm1, %ymm0
+; CHECK-NEXT: ret{{[l|q]}}
entry:
%0 = tail call <4 x i64> @llvm.x86.avx512.vpermi2var.q.256(<4 x i64> %__A, <4 x i64> %__I, <4 x i64> %__B)
ret <4 x i64> %0
}
define <4 x i64> @test_mm256_mask_permutex2var_epi64(<4 x i64> %__A, i8 zeroext %__U, <4 x i64> %__I, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_mask_permutex2var_epi64:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermt2q %ymm2, %ymm1, %ymm0 {%k1}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_permutex2var_epi64:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermt2q %ymm2, %ymm1, %ymm0 {%k1}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_permutex2var_epi64:
; X64: # %bb.0: # %entry
@@ -4462,12 +4148,12 @@ entry:
}
define <4 x i64> @test_mm256_maskz_permutex2var_epi64(i8 zeroext %__U, <4 x i64> %__A, <4 x i64> %__I, <4 x i64> %__B) {
-; X32-LABEL: test_mm256_maskz_permutex2var_epi64:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vpermt2q %ymm2, %ymm1, %ymm0 {%k1} {z}
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_permutex2var_epi64:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vpermt2q %ymm2, %ymm1, %ymm0 {%k1} {z}
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_permutex2var_epi64:
; X64: # %bb.0: # %entry
@@ -4484,12 +4170,12 @@ entry:
define <2 x double> @test_mm_mask_fmadd_pd(<2 x double> %__A, i8 zeroext %__U, <2 x double> %__B, <2 x double> %__C) {
-; X32-LABEL: test_mm_mask_fmadd_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmadd132pd {{.*#+}} xmm0 = (xmm0 * xmm1) + xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_fmadd_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmadd132pd {{.*#+}} xmm0 = (xmm0 * xmm1) + xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_fmadd_pd:
; X64: # %bb.0: # %entry
@@ -4505,12 +4191,12 @@ entry:
}
define <2 x double> @test_mm_mask_fmsub_pd(<2 x double> %__A, i8 zeroext %__U, <2 x double> %__B, <2 x double> %__C) {
-; X32-LABEL: test_mm_mask_fmsub_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsub132pd {{.*#+}} xmm0 = (xmm0 * xmm1) - xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_fmsub_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsub132pd {{.*#+}} xmm0 = (xmm0 * xmm1) - xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_fmsub_pd:
; X64: # %bb.0: # %entry
@@ -4527,13 +4213,13 @@ entry:
}
define <2 x double> @test_mm_mask3_fmadd_pd(<2 x double> %__A, <2 x double> %__B, <2 x double> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm_mask3_fmadd_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmadd231pd {{.*#+}} xmm2 = (xmm0 * xmm1) + xmm2
-; X32-NEXT: vmovapd %xmm2, %xmm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask3_fmadd_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmadd231pd {{.*#+}} xmm2 = (xmm0 * xmm1) + xmm2
+; X86-NEXT: vmovapd %xmm2, %xmm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask3_fmadd_pd:
; X64: # %bb.0: # %entry
@@ -4550,13 +4236,13 @@ entry:
}
define <2 x double> @test_mm_mask3_fnmadd_pd(<2 x double> %__A, <2 x double> %__B, <2 x double> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm_mask3_fnmadd_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmadd231pd {{.*#+}} xmm2 = -(xmm0 * xmm1) + xmm2
-; X32-NEXT: vmovapd %xmm2, %xmm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask3_fnmadd_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmadd231pd {{.*#+}} xmm2 = -(xmm0 * xmm1) + xmm2
+; X86-NEXT: vmovapd %xmm2, %xmm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask3_fnmadd_pd:
; X64: # %bb.0: # %entry
@@ -4574,12 +4260,12 @@ entry:
}
define <2 x double> @test_mm_maskz_fmadd_pd(i8 zeroext %__U, <2 x double> %__A, <2 x double> %__B, <2 x double> %__C) {
-; X32-LABEL: test_mm_maskz_fmadd_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmadd213pd {{.*#+}} xmm0 = (xmm1 * xmm0) + xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_fmadd_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmadd213pd {{.*#+}} xmm0 = (xmm1 * xmm0) + xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_fmadd_pd:
; X64: # %bb.0: # %entry
@@ -4595,12 +4281,12 @@ entry:
}
define <2 x double> @test_mm_maskz_fmsub_pd(i8 zeroext %__U, <2 x double> %__A, <2 x double> %__B, <2 x double> %__C) {
-; X32-LABEL: test_mm_maskz_fmsub_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsub213pd {{.*#+}} xmm0 = (xmm1 * xmm0) - xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_fmsub_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsub213pd {{.*#+}} xmm0 = (xmm1 * xmm0) - xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_fmsub_pd:
; X64: # %bb.0: # %entry
@@ -4617,12 +4303,12 @@ entry:
}
define <2 x double> @test_mm_maskz_fnmadd_pd(i8 zeroext %__U, <2 x double> %__A, <2 x double> %__B, <2 x double> %__C) {
-; X32-LABEL: test_mm_maskz_fnmadd_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmadd213pd {{.*#+}} xmm0 = -(xmm1 * xmm0) + xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_fnmadd_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmadd213pd {{.*#+}} xmm0 = -(xmm1 * xmm0) + xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_fnmadd_pd:
; X64: # %bb.0: # %entry
@@ -4639,12 +4325,12 @@ entry:
}
define <2 x double> @test_mm_maskz_fnmsub_pd(i8 zeroext %__U, <2 x double> %__A, <2 x double> %__B, <2 x double> %__C) {
-; X32-LABEL: test_mm_maskz_fnmsub_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmsub213pd {{.*#+}} xmm0 = -(xmm1 * xmm0) - xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_fnmsub_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmsub213pd {{.*#+}} xmm0 = -(xmm1 * xmm0) - xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_fnmsub_pd:
; X64: # %bb.0: # %entry
@@ -4662,12 +4348,12 @@ entry:
}
define <4 x double> @test_mm256_mask_fmadd_pd(<4 x double> %__A, i8 zeroext %__U, <4 x double> %__B, <4 x double> %__C) {
-; X32-LABEL: test_mm256_mask_fmadd_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmadd132pd {{.*#+}} ymm0 = (ymm0 * ymm1) + ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_fmadd_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmadd132pd {{.*#+}} ymm0 = (ymm0 * ymm1) + ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_fmadd_pd:
; X64: # %bb.0: # %entry
@@ -4683,12 +4369,12 @@ entry:
}
define <4 x double> @test_mm256_mask_fmsub_pd(<4 x double> %__A, i8 zeroext %__U, <4 x double> %__B, <4 x double> %__C) {
-; X32-LABEL: test_mm256_mask_fmsub_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsub132pd {{.*#+}} ymm0 = (ymm0 * ymm1) - ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_fmsub_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsub132pd {{.*#+}} ymm0 = (ymm0 * ymm1) - ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_fmsub_pd:
; X64: # %bb.0: # %entry
@@ -4705,13 +4391,13 @@ entry:
}
define <4 x double> @test_mm256_mask3_fmadd_pd(<4 x double> %__A, <4 x double> %__B, <4 x double> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm256_mask3_fmadd_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmadd231pd {{.*#+}} ymm2 = (ymm0 * ymm1) + ymm2
-; X32-NEXT: vmovapd %ymm2, %ymm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask3_fmadd_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmadd231pd {{.*#+}} ymm2 = (ymm0 * ymm1) + ymm2
+; X86-NEXT: vmovapd %ymm2, %ymm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask3_fmadd_pd:
; X64: # %bb.0: # %entry
@@ -4728,13 +4414,13 @@ entry:
}
define <4 x double> @test_mm256_mask3_fnmadd_pd(<4 x double> %__A, <4 x double> %__B, <4 x double> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm256_mask3_fnmadd_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmadd231pd {{.*#+}} ymm2 = -(ymm0 * ymm1) + ymm2
-; X32-NEXT: vmovapd %ymm2, %ymm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask3_fnmadd_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmadd231pd {{.*#+}} ymm2 = -(ymm0 * ymm1) + ymm2
+; X86-NEXT: vmovapd %ymm2, %ymm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask3_fnmadd_pd:
; X64: # %bb.0: # %entry
@@ -4752,12 +4438,12 @@ entry:
}
define <4 x double> @test_mm256_maskz_fmadd_pd(i8 zeroext %__U, <4 x double> %__A, <4 x double> %__B, <4 x double> %__C) {
-; X32-LABEL: test_mm256_maskz_fmadd_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmadd213pd {{.*#+}} ymm0 = (ymm1 * ymm0) + ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_fmadd_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmadd213pd {{.*#+}} ymm0 = (ymm1 * ymm0) + ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_fmadd_pd:
; X64: # %bb.0: # %entry
@@ -4773,12 +4459,12 @@ entry:
}
define <4 x double> @test_mm256_maskz_fmsub_pd(i8 zeroext %__U, <4 x double> %__A, <4 x double> %__B, <4 x double> %__C) {
-; X32-LABEL: test_mm256_maskz_fmsub_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsub213pd {{.*#+}} ymm0 = (ymm1 * ymm0) - ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_fmsub_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsub213pd {{.*#+}} ymm0 = (ymm1 * ymm0) - ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_fmsub_pd:
; X64: # %bb.0: # %entry
@@ -4795,12 +4481,12 @@ entry:
}
define <4 x double> @test_mm256_maskz_fnmadd_pd(i8 zeroext %__U, <4 x double> %__A, <4 x double> %__B, <4 x double> %__C) {
-; X32-LABEL: test_mm256_maskz_fnmadd_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmadd213pd {{.*#+}} ymm0 = -(ymm1 * ymm0) + ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_fnmadd_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmadd213pd {{.*#+}} ymm0 = -(ymm1 * ymm0) + ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_fnmadd_pd:
; X64: # %bb.0: # %entry
@@ -4817,12 +4503,12 @@ entry:
}
define <4 x double> @test_mm256_maskz_fnmsub_pd(i8 zeroext %__U, <4 x double> %__A, <4 x double> %__B, <4 x double> %__C) {
-; X32-LABEL: test_mm256_maskz_fnmsub_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmsub213pd {{.*#+}} ymm0 = -(ymm1 * ymm0) - ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_fnmsub_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmsub213pd {{.*#+}} ymm0 = -(ymm1 * ymm0) - ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_fnmsub_pd:
; X64: # %bb.0: # %entry
@@ -4840,12 +4526,12 @@ entry:
}
define <4 x float> @test_mm_mask_fmadd_ps(<4 x float> %__A, i8 zeroext %__U, <4 x float> %__B, <4 x float> %__C) {
-; X32-LABEL: test_mm_mask_fmadd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmadd132ps {{.*#+}} xmm0 = (xmm0 * xmm1) + xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_fmadd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmadd132ps {{.*#+}} xmm0 = (xmm0 * xmm1) + xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_fmadd_ps:
; X64: # %bb.0: # %entry
@@ -4861,12 +4547,12 @@ entry:
}
define <4 x float> @test_mm_mask_fmsub_ps(<4 x float> %__A, i8 zeroext %__U, <4 x float> %__B, <4 x float> %__C) {
-; X32-LABEL: test_mm_mask_fmsub_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsub132ps {{.*#+}} xmm0 = (xmm0 * xmm1) - xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_fmsub_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsub132ps {{.*#+}} xmm0 = (xmm0 * xmm1) - xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_fmsub_ps:
; X64: # %bb.0: # %entry
@@ -4883,13 +4569,13 @@ entry:
}
define <4 x float> @test_mm_mask3_fmadd_ps(<4 x float> %__A, <4 x float> %__B, <4 x float> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm_mask3_fmadd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmadd231ps {{.*#+}} xmm2 = (xmm0 * xmm1) + xmm2
-; X32-NEXT: vmovaps %xmm2, %xmm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask3_fmadd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmadd231ps {{.*#+}} xmm2 = (xmm0 * xmm1) + xmm2
+; X86-NEXT: vmovaps %xmm2, %xmm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask3_fmadd_ps:
; X64: # %bb.0: # %entry
@@ -4906,13 +4592,13 @@ entry:
}
define <4 x float> @test_mm_mask3_fnmadd_ps(<4 x float> %__A, <4 x float> %__B, <4 x float> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm_mask3_fnmadd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmadd231ps {{.*#+}} xmm2 = -(xmm0 * xmm1) + xmm2
-; X32-NEXT: vmovaps %xmm2, %xmm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask3_fnmadd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmadd231ps {{.*#+}} xmm2 = -(xmm0 * xmm1) + xmm2
+; X86-NEXT: vmovaps %xmm2, %xmm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask3_fnmadd_ps:
; X64: # %bb.0: # %entry
@@ -4930,12 +4616,12 @@ entry:
}
define <4 x float> @test_mm_maskz_fmadd_ps(i8 zeroext %__U, <4 x float> %__A, <4 x float> %__B, <4 x float> %__C) {
-; X32-LABEL: test_mm_maskz_fmadd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmadd213ps {{.*#+}} xmm0 = (xmm1 * xmm0) + xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_fmadd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmadd213ps {{.*#+}} xmm0 = (xmm1 * xmm0) + xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_fmadd_ps:
; X64: # %bb.0: # %entry
@@ -4951,12 +4637,12 @@ entry:
}
define <4 x float> @test_mm_maskz_fmsub_ps(i8 zeroext %__U, <4 x float> %__A, <4 x float> %__B, <4 x float> %__C) {
-; X32-LABEL: test_mm_maskz_fmsub_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsub213ps {{.*#+}} xmm0 = (xmm1 * xmm0) - xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_fmsub_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsub213ps {{.*#+}} xmm0 = (xmm1 * xmm0) - xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_fmsub_ps:
; X64: # %bb.0: # %entry
@@ -4973,12 +4659,12 @@ entry:
}
define <4 x float> @test_mm_maskz_fnmadd_ps(i8 zeroext %__U, <4 x float> %__A, <4 x float> %__B, <4 x float> %__C) {
-; X32-LABEL: test_mm_maskz_fnmadd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmadd213ps {{.*#+}} xmm0 = -(xmm1 * xmm0) + xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_fnmadd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmadd213ps {{.*#+}} xmm0 = -(xmm1 * xmm0) + xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_fnmadd_ps:
; X64: # %bb.0: # %entry
@@ -4995,12 +4681,12 @@ entry:
}
define <4 x float> @test_mm_maskz_fnmsub_ps(i8 zeroext %__U, <4 x float> %__A, <4 x float> %__B, <4 x float> %__C) {
-; X32-LABEL: test_mm_maskz_fnmsub_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmsub213ps {{.*#+}} xmm0 = -(xmm1 * xmm0) - xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_fnmsub_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmsub213ps {{.*#+}} xmm0 = -(xmm1 * xmm0) - xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_fnmsub_ps:
; X64: # %bb.0: # %entry
@@ -5018,12 +4704,12 @@ entry:
}
define <8 x float> @test_mm256_mask_fmadd_ps(<8 x float> %__A, i8 zeroext %__U, <8 x float> %__B, <8 x float> %__C) {
-; X32-LABEL: test_mm256_mask_fmadd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmadd132ps {{.*#+}} ymm0 = (ymm0 * ymm1) + ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_fmadd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmadd132ps {{.*#+}} ymm0 = (ymm0 * ymm1) + ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_fmadd_ps:
; X64: # %bb.0: # %entry
@@ -5038,12 +4724,12 @@ entry:
}
define <8 x float> @test_mm256_mask_fmsub_ps(<8 x float> %__A, i8 zeroext %__U, <8 x float> %__B, <8 x float> %__C) {
-; X32-LABEL: test_mm256_mask_fmsub_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsub132ps {{.*#+}} ymm0 = (ymm0 * ymm1) - ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_fmsub_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsub132ps {{.*#+}} ymm0 = (ymm0 * ymm1) - ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_fmsub_ps:
; X64: # %bb.0: # %entry
@@ -5059,13 +4745,13 @@ entry:
}
define <8 x float> @test_mm256_mask3_fmadd_ps(<8 x float> %__A, <8 x float> %__B, <8 x float> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm256_mask3_fmadd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmadd231ps {{.*#+}} ymm2 = (ymm0 * ymm1) + ymm2
-; X32-NEXT: vmovaps %ymm2, %ymm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask3_fmadd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmadd231ps {{.*#+}} ymm2 = (ymm0 * ymm1) + ymm2
+; X86-NEXT: vmovaps %ymm2, %ymm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask3_fmadd_ps:
; X64: # %bb.0: # %entry
@@ -5081,13 +4767,13 @@ entry:
}
define <8 x float> @test_mm256_mask3_fnmadd_ps(<8 x float> %__A, <8 x float> %__B, <8 x float> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm256_mask3_fnmadd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmadd231ps {{.*#+}} ymm2 = -(ymm0 * ymm1) + ymm2
-; X32-NEXT: vmovaps %ymm2, %ymm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask3_fnmadd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmadd231ps {{.*#+}} ymm2 = -(ymm0 * ymm1) + ymm2
+; X86-NEXT: vmovaps %ymm2, %ymm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask3_fnmadd_ps:
; X64: # %bb.0: # %entry
@@ -5104,12 +4790,12 @@ entry:
}
define <8 x float> @test_mm256_maskz_fmadd_ps(i8 zeroext %__U, <8 x float> %__A, <8 x float> %__B, <8 x float> %__C) {
-; X32-LABEL: test_mm256_maskz_fmadd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmadd213ps {{.*#+}} ymm0 = (ymm1 * ymm0) + ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_fmadd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmadd213ps {{.*#+}} ymm0 = (ymm1 * ymm0) + ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_fmadd_ps:
; X64: # %bb.0: # %entry
@@ -5124,12 +4810,12 @@ entry:
}
define <8 x float> @test_mm256_maskz_fmsub_ps(i8 zeroext %__U, <8 x float> %__A, <8 x float> %__B, <8 x float> %__C) {
-; X32-LABEL: test_mm256_maskz_fmsub_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsub213ps {{.*#+}} ymm0 = (ymm1 * ymm0) - ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_fmsub_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsub213ps {{.*#+}} ymm0 = (ymm1 * ymm0) - ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_fmsub_ps:
; X64: # %bb.0: # %entry
@@ -5145,12 +4831,12 @@ entry:
}
define <8 x float> @test_mm256_maskz_fnmadd_ps(i8 zeroext %__U, <8 x float> %__A, <8 x float> %__B, <8 x float> %__C) {
-; X32-LABEL: test_mm256_maskz_fnmadd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmadd213ps {{.*#+}} ymm0 = -(ymm1 * ymm0) + ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_fnmadd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmadd213ps {{.*#+}} ymm0 = -(ymm1 * ymm0) + ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_fnmadd_ps:
; X64: # %bb.0: # %entry
@@ -5166,12 +4852,12 @@ entry:
}
define <8 x float> @test_mm256_maskz_fnmsub_ps(i8 zeroext %__U, <8 x float> %__A, <8 x float> %__B, <8 x float> %__C) {
-; X32-LABEL: test_mm256_maskz_fnmsub_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmsub213ps {{.*#+}} ymm0 = -(ymm1 * ymm0) - ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_fnmsub_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmsub213ps {{.*#+}} ymm0 = -(ymm1 * ymm0) - ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_fnmsub_ps:
; X64: # %bb.0: # %entry
@@ -5188,12 +4874,12 @@ entry:
}
define <2 x double> @test_mm_mask_fmaddsub_pd(<2 x double> %__A, i8 zeroext %__U, <2 x double> %__B, <2 x double> %__C) {
-; X32-LABEL: test_mm_mask_fmaddsub_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmaddsub132pd {{.*#+}} xmm0 = (xmm0 * xmm1) +/- xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_fmaddsub_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmaddsub132pd {{.*#+}} xmm0 = (xmm0 * xmm1) +/- xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_fmaddsub_pd:
; X64: # %bb.0: # %entry
@@ -5212,12 +4898,12 @@ entry:
}
define <2 x double> @test_mm_mask_fmsubadd_pd(<2 x double> %__A, i8 zeroext %__U, <2 x double> %__B, <2 x double> %__C) {
-; X32-LABEL: test_mm_mask_fmsubadd_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsubadd132pd {{.*#+}} xmm0 = (xmm0 * xmm1) -/+ xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_fmsubadd_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsubadd132pd {{.*#+}} xmm0 = (xmm0 * xmm1) -/+ xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_fmsubadd_pd:
; X64: # %bb.0: # %entry
@@ -5236,13 +4922,13 @@ entry:
}
define <2 x double> @test_mm_mask3_fmaddsub_pd(<2 x double> %__A, <2 x double> %__B, <2 x double> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm_mask3_fmaddsub_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmaddsub231pd {{.*#+}} xmm2 = (xmm0 * xmm1) +/- xmm2
-; X32-NEXT: vmovapd %xmm2, %xmm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask3_fmaddsub_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmaddsub231pd {{.*#+}} xmm2 = (xmm0 * xmm1) +/- xmm2
+; X86-NEXT: vmovapd %xmm2, %xmm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask3_fmaddsub_pd:
; X64: # %bb.0: # %entry
@@ -5262,12 +4948,12 @@ entry:
}
define <2 x double> @test_mm_maskz_fmaddsub_pd(i8 zeroext %__U, <2 x double> %__A, <2 x double> %__B, <2 x double> %__C) {
-; X32-LABEL: test_mm_maskz_fmaddsub_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmaddsub213pd {{.*#+}} xmm0 = (xmm1 * xmm0) +/- xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_fmaddsub_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmaddsub213pd {{.*#+}} xmm0 = (xmm1 * xmm0) +/- xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_fmaddsub_pd:
; X64: # %bb.0: # %entry
@@ -5286,12 +4972,12 @@ entry:
}
define <2 x double> @test_mm_maskz_fmsubadd_pd(i8 zeroext %__U, <2 x double> %__A, <2 x double> %__B, <2 x double> %__C) {
-; X32-LABEL: test_mm_maskz_fmsubadd_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsubadd213pd {{.*#+}} xmm0 = (xmm1 * xmm0) -/+ xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_fmsubadd_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsubadd213pd {{.*#+}} xmm0 = (xmm1 * xmm0) -/+ xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_fmsubadd_pd:
; X64: # %bb.0: # %entry
@@ -5310,12 +4996,12 @@ entry:
}
define <4 x double> @test_mm256_mask_fmaddsub_pd(<4 x double> %__A, i8 zeroext %__U, <4 x double> %__B, <4 x double> %__C) {
-; X32-LABEL: test_mm256_mask_fmaddsub_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmaddsub132pd {{.*#+}} ymm0 = (ymm0 * ymm1) +/- ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_fmaddsub_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmaddsub132pd {{.*#+}} ymm0 = (ymm0 * ymm1) +/- ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_fmaddsub_pd:
; X64: # %bb.0: # %entry
@@ -5334,12 +5020,12 @@ entry:
}
define <4 x double> @test_mm256_mask_fmsubadd_pd(<4 x double> %__A, i8 zeroext %__U, <4 x double> %__B, <4 x double> %__C) {
-; X32-LABEL: test_mm256_mask_fmsubadd_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsubadd132pd {{.*#+}} ymm0 = (ymm0 * ymm1) -/+ ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_fmsubadd_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsubadd132pd {{.*#+}} ymm0 = (ymm0 * ymm1) -/+ ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_fmsubadd_pd:
; X64: # %bb.0: # %entry
@@ -5358,13 +5044,13 @@ entry:
}
define <4 x double> @test_mm256_mask3_fmaddsub_pd(<4 x double> %__A, <4 x double> %__B, <4 x double> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm256_mask3_fmaddsub_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmaddsub231pd {{.*#+}} ymm2 = (ymm0 * ymm1) +/- ymm2
-; X32-NEXT: vmovapd %ymm2, %ymm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask3_fmaddsub_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmaddsub231pd {{.*#+}} ymm2 = (ymm0 * ymm1) +/- ymm2
+; X86-NEXT: vmovapd %ymm2, %ymm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask3_fmaddsub_pd:
; X64: # %bb.0: # %entry
@@ -5384,12 +5070,12 @@ entry:
}
define <4 x double> @test_mm256_maskz_fmaddsub_pd(i8 zeroext %__U, <4 x double> %__A, <4 x double> %__B, <4 x double> %__C) {
-; X32-LABEL: test_mm256_maskz_fmaddsub_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmaddsub213pd {{.*#+}} ymm0 = (ymm1 * ymm0) +/- ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_fmaddsub_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmaddsub213pd {{.*#+}} ymm0 = (ymm1 * ymm0) +/- ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_fmaddsub_pd:
; X64: # %bb.0: # %entry
@@ -5408,12 +5094,12 @@ entry:
}
define <4 x double> @test_mm256_maskz_fmsubadd_pd(i8 zeroext %__U, <4 x double> %__A, <4 x double> %__B, <4 x double> %__C) {
-; X32-LABEL: test_mm256_maskz_fmsubadd_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsubadd213pd {{.*#+}} ymm0 = (ymm1 * ymm0) -/+ ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_fmsubadd_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsubadd213pd {{.*#+}} ymm0 = (ymm1 * ymm0) -/+ ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_fmsubadd_pd:
; X64: # %bb.0: # %entry
@@ -5432,12 +5118,12 @@ entry:
}
define <4 x float> @test_mm_mask_fmaddsub_ps(<4 x float> %__A, i8 zeroext %__U, <4 x float> %__B, <4 x float> %__C) {
-; X32-LABEL: test_mm_mask_fmaddsub_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmaddsub132ps {{.*#+}} xmm0 = (xmm0 * xmm1) +/- xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_fmaddsub_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmaddsub132ps {{.*#+}} xmm0 = (xmm0 * xmm1) +/- xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_fmaddsub_ps:
; X64: # %bb.0: # %entry
@@ -5456,12 +5142,12 @@ entry:
}
define <4 x float> @test_mm_mask_fmsubadd_ps(<4 x float> %__A, i8 zeroext %__U, <4 x float> %__B, <4 x float> %__C) {
-; X32-LABEL: test_mm_mask_fmsubadd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsubadd132ps {{.*#+}} xmm0 = (xmm0 * xmm1) -/+ xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_fmsubadd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsubadd132ps {{.*#+}} xmm0 = (xmm0 * xmm1) -/+ xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_fmsubadd_ps:
; X64: # %bb.0: # %entry
@@ -5480,13 +5166,13 @@ entry:
}
define <4 x float> @test_mm_mask3_fmaddsub_ps(<4 x float> %__A, <4 x float> %__B, <4 x float> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm_mask3_fmaddsub_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmaddsub231ps {{.*#+}} xmm2 = (xmm0 * xmm1) +/- xmm2
-; X32-NEXT: vmovaps %xmm2, %xmm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask3_fmaddsub_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmaddsub231ps {{.*#+}} xmm2 = (xmm0 * xmm1) +/- xmm2
+; X86-NEXT: vmovaps %xmm2, %xmm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask3_fmaddsub_ps:
; X64: # %bb.0: # %entry
@@ -5506,12 +5192,12 @@ entry:
}
define <4 x float> @test_mm_maskz_fmaddsub_ps(i8 zeroext %__U, <4 x float> %__A, <4 x float> %__B, <4 x float> %__C) {
-; X32-LABEL: test_mm_maskz_fmaddsub_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmaddsub213ps {{.*#+}} xmm0 = (xmm1 * xmm0) +/- xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_fmaddsub_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmaddsub213ps {{.*#+}} xmm0 = (xmm1 * xmm0) +/- xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_fmaddsub_ps:
; X64: # %bb.0: # %entry
@@ -5530,12 +5216,12 @@ entry:
}
define <4 x float> @test_mm_maskz_fmsubadd_ps(i8 zeroext %__U, <4 x float> %__A, <4 x float> %__B, <4 x float> %__C) {
-; X32-LABEL: test_mm_maskz_fmsubadd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsubadd213ps {{.*#+}} xmm0 = (xmm1 * xmm0) -/+ xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_maskz_fmsubadd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsubadd213ps {{.*#+}} xmm0 = (xmm1 * xmm0) -/+ xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_maskz_fmsubadd_ps:
; X64: # %bb.0: # %entry
@@ -5554,12 +5240,12 @@ entry:
}
define <8 x float> @test_mm256_mask_fmaddsub_ps(<8 x float> %__A, i8 zeroext %__U, <8 x float> %__B, <8 x float> %__C) {
-; X32-LABEL: test_mm256_mask_fmaddsub_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmaddsub132ps {{.*#+}} ymm0 = (ymm0 * ymm1) +/- ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_fmaddsub_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmaddsub132ps {{.*#+}} ymm0 = (ymm0 * ymm1) +/- ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_fmaddsub_ps:
; X64: # %bb.0: # %entry
@@ -5577,12 +5263,12 @@ entry:
}
define <8 x float> @test_mm256_mask_fmsubadd_ps(<8 x float> %__A, i8 zeroext %__U, <8 x float> %__B, <8 x float> %__C) {
-; X32-LABEL: test_mm256_mask_fmsubadd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsubadd132ps {{.*#+}} ymm0 = (ymm0 * ymm1) -/+ ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_fmsubadd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsubadd132ps {{.*#+}} ymm0 = (ymm0 * ymm1) -/+ ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_fmsubadd_ps:
; X64: # %bb.0: # %entry
@@ -5600,13 +5286,13 @@ entry:
}
define <8 x float> @test_mm256_mask3_fmaddsub_ps(<8 x float> %__A, <8 x float> %__B, <8 x float> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm256_mask3_fmaddsub_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmaddsub231ps {{.*#+}} ymm2 = (ymm0 * ymm1) +/- ymm2
-; X32-NEXT: vmovaps %ymm2, %ymm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask3_fmaddsub_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmaddsub231ps {{.*#+}} ymm2 = (ymm0 * ymm1) +/- ymm2
+; X86-NEXT: vmovaps %ymm2, %ymm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask3_fmaddsub_ps:
; X64: # %bb.0: # %entry
@@ -5625,12 +5311,12 @@ entry:
}
define <8 x float> @test_mm256_maskz_fmaddsub_ps(i8 zeroext %__U, <8 x float> %__A, <8 x float> %__B, <8 x float> %__C) {
-; X32-LABEL: test_mm256_maskz_fmaddsub_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmaddsub213ps {{.*#+}} ymm0 = (ymm1 * ymm0) +/- ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_fmaddsub_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmaddsub213ps {{.*#+}} ymm0 = (ymm1 * ymm0) +/- ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_fmaddsub_ps:
; X64: # %bb.0: # %entry
@@ -5648,12 +5334,12 @@ entry:
}
define <8 x float> @test_mm256_maskz_fmsubadd_ps(i8 zeroext %__U, <8 x float> %__A, <8 x float> %__B, <8 x float> %__C) {
-; X32-LABEL: test_mm256_maskz_fmsubadd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsubadd213ps {{.*#+}} ymm0 = (ymm1 * ymm0) -/+ ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_maskz_fmsubadd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsubadd213ps {{.*#+}} ymm0 = (ymm1 * ymm0) -/+ ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_maskz_fmsubadd_ps:
; X64: # %bb.0: # %entry
@@ -5671,13 +5357,13 @@ entry:
}
define <2 x double> @test_mm_mask3_fmsub_pd(<2 x double> %__A, <2 x double> %__B, <2 x double> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm_mask3_fmsub_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsub231pd {{.*#+}} xmm2 = (xmm0 * xmm1) - xmm2
-; X32-NEXT: vmovapd %xmm2, %xmm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask3_fmsub_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsub231pd {{.*#+}} xmm2 = (xmm0 * xmm1) - xmm2
+; X86-NEXT: vmovapd %xmm2, %xmm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask3_fmsub_pd:
; X64: # %bb.0: # %entry
@@ -5695,13 +5381,13 @@ entry:
}
define <4 x double> @test_mm256_mask3_fmsub_pd(<4 x double> %__A, <4 x double> %__B, <4 x double> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm256_mask3_fmsub_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsub231pd {{.*#+}} ymm2 = (ymm0 * ymm1) - ymm2
-; X32-NEXT: vmovapd %ymm2, %ymm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask3_fmsub_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsub231pd {{.*#+}} ymm2 = (ymm0 * ymm1) - ymm2
+; X86-NEXT: vmovapd %ymm2, %ymm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask3_fmsub_pd:
; X64: # %bb.0: # %entry
@@ -5719,13 +5405,13 @@ entry:
}
define <4 x float> @test_mm_mask3_fmsub_ps(<4 x float> %__A, <4 x float> %__B, <4 x float> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm_mask3_fmsub_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsub231ps {{.*#+}} xmm2 = (xmm0 * xmm1) - xmm2
-; X32-NEXT: vmovaps %xmm2, %xmm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask3_fmsub_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsub231ps {{.*#+}} xmm2 = (xmm0 * xmm1) - xmm2
+; X86-NEXT: vmovaps %xmm2, %xmm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask3_fmsub_ps:
; X64: # %bb.0: # %entry
@@ -5743,13 +5429,13 @@ entry:
}
define <8 x float> @test_mm256_mask3_fmsub_ps(<8 x float> %__A, <8 x float> %__B, <8 x float> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm256_mask3_fmsub_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsub231ps {{.*#+}} ymm2 = (ymm0 * ymm1) - ymm2
-; X32-NEXT: vmovaps %ymm2, %ymm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask3_fmsub_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsub231ps {{.*#+}} ymm2 = (ymm0 * ymm1) - ymm2
+; X86-NEXT: vmovaps %ymm2, %ymm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask3_fmsub_ps:
; X64: # %bb.0: # %entry
@@ -5766,13 +5452,13 @@ entry:
}
define <2 x double> @test_mm_mask3_fmsubadd_pd(<2 x double> %__A, <2 x double> %__B, <2 x double> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm_mask3_fmsubadd_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsubadd231pd {{.*#+}} xmm2 = (xmm0 * xmm1) -/+ xmm2
-; X32-NEXT: vmovapd %xmm2, %xmm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask3_fmsubadd_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsubadd231pd {{.*#+}} xmm2 = (xmm0 * xmm1) -/+ xmm2
+; X86-NEXT: vmovapd %xmm2, %xmm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask3_fmsubadd_pd:
; X64: # %bb.0: # %entry
@@ -5792,13 +5478,13 @@ entry:
}
define <4 x double> @test_mm256_mask3_fmsubadd_pd(<4 x double> %__A, <4 x double> %__B, <4 x double> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm256_mask3_fmsubadd_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsubadd231pd {{.*#+}} ymm2 = (ymm0 * ymm1) -/+ ymm2
-; X32-NEXT: vmovapd %ymm2, %ymm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask3_fmsubadd_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsubadd231pd {{.*#+}} ymm2 = (ymm0 * ymm1) -/+ ymm2
+; X86-NEXT: vmovapd %ymm2, %ymm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask3_fmsubadd_pd:
; X64: # %bb.0: # %entry
@@ -5818,13 +5504,13 @@ entry:
}
define <4 x float> @test_mm_mask3_fmsubadd_ps(<4 x float> %__A, <4 x float> %__B, <4 x float> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm_mask3_fmsubadd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsubadd231ps {{.*#+}} xmm2 = (xmm0 * xmm1) -/+ xmm2
-; X32-NEXT: vmovaps %xmm2, %xmm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask3_fmsubadd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsubadd231ps {{.*#+}} xmm2 = (xmm0 * xmm1) -/+ xmm2
+; X86-NEXT: vmovaps %xmm2, %xmm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask3_fmsubadd_ps:
; X64: # %bb.0: # %entry
@@ -5844,13 +5530,13 @@ entry:
}
define <8 x float> @test_mm256_mask3_fmsubadd_ps(<8 x float> %__A, <8 x float> %__B, <8 x float> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm256_mask3_fmsubadd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfmsubadd231ps {{.*#+}} ymm2 = (ymm0 * ymm1) -/+ ymm2
-; X32-NEXT: vmovaps %ymm2, %ymm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask3_fmsubadd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfmsubadd231ps {{.*#+}} ymm2 = (ymm0 * ymm1) -/+ ymm2
+; X86-NEXT: vmovaps %ymm2, %ymm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask3_fmsubadd_ps:
; X64: # %bb.0: # %entry
@@ -5869,12 +5555,12 @@ entry:
}
define <2 x double> @test_mm_mask_fnmadd_pd(<2 x double> %__A, i8 zeroext %__U, <2 x double> %__B, <2 x double> %__C) {
-; X32-LABEL: test_mm_mask_fnmadd_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmadd132pd {{.*#+}} xmm0 = -(xmm0 * xmm1) + xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_fnmadd_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmadd132pd {{.*#+}} xmm0 = -(xmm0 * xmm1) + xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_fnmadd_pd:
; X64: # %bb.0: # %entry
@@ -5891,12 +5577,12 @@ entry:
}
define <4 x double> @test_mm256_mask_fnmadd_pd(<4 x double> %__A, i8 zeroext %__U, <4 x double> %__B, <4 x double> %__C) {
-; X32-LABEL: test_mm256_mask_fnmadd_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmadd132pd {{.*#+}} ymm0 = -(ymm0 * ymm1) + ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_fnmadd_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmadd132pd {{.*#+}} ymm0 = -(ymm0 * ymm1) + ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_fnmadd_pd:
; X64: # %bb.0: # %entry
@@ -5913,12 +5599,12 @@ entry:
}
define <4 x float> @test_mm_mask_fnmadd_ps(<4 x float> %__A, i8 zeroext %__U, <4 x float> %__B, <4 x float> %__C) {
-; X32-LABEL: test_mm_mask_fnmadd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmadd132ps {{.*#+}} xmm0 = -(xmm0 * xmm1) + xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_fnmadd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmadd132ps {{.*#+}} xmm0 = -(xmm0 * xmm1) + xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_fnmadd_ps:
; X64: # %bb.0: # %entry
@@ -5935,12 +5621,12 @@ entry:
}
define <8 x float> @test_mm256_mask_fnmadd_ps(<8 x float> %__A, i8 zeroext %__U, <8 x float> %__B, <8 x float> %__C) {
-; X32-LABEL: test_mm256_mask_fnmadd_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmadd132ps {{.*#+}} ymm0 = -(ymm0 * ymm1) + ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_fnmadd_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmadd132ps {{.*#+}} ymm0 = -(ymm0 * ymm1) + ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_fnmadd_ps:
; X64: # %bb.0: # %entry
@@ -5956,12 +5642,12 @@ entry:
}
define <2 x double> @test_mm_mask_fnmsub_pd(<2 x double> %__A, i8 zeroext %__U, <2 x double> %__B, <2 x double> %__C) {
-; X32-LABEL: test_mm_mask_fnmsub_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmsub132pd {{.*#+}} xmm0 = -(xmm0 * xmm1) - xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_fnmsub_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmsub132pd {{.*#+}} xmm0 = -(xmm0 * xmm1) - xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_fnmsub_pd:
; X64: # %bb.0: # %entry
@@ -5979,13 +5665,13 @@ entry:
}
define <2 x double> @test_mm_mask3_fnmsub_pd(<2 x double> %__A, <2 x double> %__B, <2 x double> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm_mask3_fnmsub_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmsub231pd {{.*#+}} xmm2 = -(xmm0 * xmm1) - xmm2
-; X32-NEXT: vmovapd %xmm2, %xmm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask3_fnmsub_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmsub231pd {{.*#+}} xmm2 = -(xmm0 * xmm1) - xmm2
+; X86-NEXT: vmovapd %xmm2, %xmm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask3_fnmsub_pd:
; X64: # %bb.0: # %entry
@@ -6004,12 +5690,12 @@ entry:
}
define <4 x double> @test_mm256_mask_fnmsub_pd(<4 x double> %__A, i8 zeroext %__U, <4 x double> %__B, <4 x double> %__C) {
-; X32-LABEL: test_mm256_mask_fnmsub_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmsub132pd {{.*#+}} ymm0 = -(ymm0 * ymm1) - ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_fnmsub_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmsub132pd {{.*#+}} ymm0 = -(ymm0 * ymm1) - ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_fnmsub_pd:
; X64: # %bb.0: # %entry
@@ -6027,13 +5713,13 @@ entry:
}
define <4 x double> @test_mm256_mask3_fnmsub_pd(<4 x double> %__A, <4 x double> %__B, <4 x double> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm256_mask3_fnmsub_pd:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmsub231pd {{.*#+}} ymm2 = -(ymm0 * ymm1) - ymm2
-; X32-NEXT: vmovapd %ymm2, %ymm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask3_fnmsub_pd:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmsub231pd {{.*#+}} ymm2 = -(ymm0 * ymm1) - ymm2
+; X86-NEXT: vmovapd %ymm2, %ymm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask3_fnmsub_pd:
; X64: # %bb.0: # %entry
@@ -6052,12 +5738,12 @@ entry:
}
define <4 x float> @test_mm_mask_fnmsub_ps(<4 x float> %__A, i8 zeroext %__U, <4 x float> %__B, <4 x float> %__C) {
-; X32-LABEL: test_mm_mask_fnmsub_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmsub132ps {{.*#+}} xmm0 = -(xmm0 * xmm1) - xmm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask_fnmsub_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmsub132ps {{.*#+}} xmm0 = -(xmm0 * xmm1) - xmm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask_fnmsub_ps:
; X64: # %bb.0: # %entry
@@ -6075,13 +5761,13 @@ entry:
}
define <4 x float> @test_mm_mask3_fnmsub_ps(<4 x float> %__A, <4 x float> %__B, <4 x float> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm_mask3_fnmsub_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmsub231ps {{.*#+}} xmm2 = -(xmm0 * xmm1) - xmm2
-; X32-NEXT: vmovaps %xmm2, %xmm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm_mask3_fnmsub_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmsub231ps {{.*#+}} xmm2 = -(xmm0 * xmm1) - xmm2
+; X86-NEXT: vmovaps %xmm2, %xmm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm_mask3_fnmsub_ps:
; X64: # %bb.0: # %entry
@@ -6100,12 +5786,12 @@ entry:
}
define <8 x float> @test_mm256_mask_fnmsub_ps(<8 x float> %__A, i8 zeroext %__U, <8 x float> %__B, <8 x float> %__C) {
-; X32-LABEL: test_mm256_mask_fnmsub_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmsub132ps {{.*#+}} ymm0 = -(ymm0 * ymm1) - ymm2
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask_fnmsub_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmsub132ps {{.*#+}} ymm0 = -(ymm0 * ymm1) - ymm2
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask_fnmsub_ps:
; X64: # %bb.0: # %entry
@@ -6122,13 +5808,13 @@ entry:
}
define <8 x float> @test_mm256_mask3_fnmsub_ps(<8 x float> %__A, <8 x float> %__B, <8 x float> %__C, i8 zeroext %__U) {
-; X32-LABEL: test_mm256_mask3_fnmsub_ps:
-; X32: # %bb.0: # %entry
-; X32-NEXT: movb {{[0-9]+}}(%esp), %al
-; X32-NEXT: kmovw %eax, %k1
-; X32-NEXT: vfnmsub231ps {{.*#+}} ymm2 = -(ymm0 * ymm1) - ymm2
-; X32-NEXT: vmovaps %ymm2, %ymm0
-; X32-NEXT: retl
+; X86-LABEL: test_mm256_mask3_fnmsub_ps:
+; X86: # %bb.0: # %entry
+; X86-NEXT: movb {{[0-9]+}}(%esp), %al
+; X86-NEXT: kmovw %eax, %k1
+; X86-NEXT: vfnmsub231ps {{.*#+}} ymm2 = -(ymm0 * ymm1) - ymm2
+; X86-NEXT: vmovaps %ymm2, %ymm0
+; X86-NEXT: retl
;
; X64-LABEL: test_mm256_mask3_fnmsub_ps:
; X64: # %bb.0: # %entry
More information about the llvm-commits
mailing list