diff options
Diffstat (limited to 'llvm/test/CodeGen/X86/widen_conversions.ll')
| -rw-r--r-- | llvm/test/CodeGen/X86/widen_conversions.ll | 31 |
1 files changed, 19 insertions, 12 deletions
diff --git a/llvm/test/CodeGen/X86/widen_conversions.ll b/llvm/test/CodeGen/X86/widen_conversions.ll index dd75097a251..9945e26c550 100644 --- a/llvm/test/CodeGen/X86/widen_conversions.ll +++ b/llvm/test/CodeGen/X86/widen_conversions.ll @@ -1,17 +1,24 @@ -; RUN: llc < %s -mcpu=x86-64 -x86-experimental-vector-widening-legalization | FileCheck %s - -target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128" -target triple = "x86_64-unknown-unknown" +; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py +; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+sse2 -x86-experimental-vector-widening-legalization | FileCheck %s --check-prefix=X86 +; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse2 -x86-experimental-vector-widening-legalization | FileCheck %s --check-prefix=X64 define <4 x i32> @zext_v4i8_to_v4i32(<4 x i8>* %ptr) { -; CHECK-LABEL: zext_v4i8_to_v4i32: -; -; CHECK: movd (%{{.*}}), %[[X:xmm[0-9]+]] -; CHECK-NEXT: pxor %[[Z:xmm[0-9]+]], %[[Z]] -; CHECK-NEXT: punpcklbw %[[Z]], %[[X]] -; CHECK-NEXT: punpcklwd %[[Z]], %[[X]] -; CHECK-NEXT: ret - +; X86-LABEL: zext_v4i8_to_v4i32: +; X86: # BB#0: +; X86-NEXT: movl {{[0-9]+}}(%esp), %eax +; X86-NEXT: movd {{.*#+}} xmm0 = mem[0],zero,zero,zero +; X86-NEXT: pxor %xmm1, %xmm1 +; X86-NEXT: punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7] +; X86-NEXT: punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3] +; X86-NEXT: retl +; +; X64-LABEL: zext_v4i8_to_v4i32: +; X64: # BB#0: +; X64-NEXT: movd {{.*#+}} xmm0 = mem[0],zero,zero,zero +; X64-NEXT: pxor %xmm1, %xmm1 +; X64-NEXT: punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7] +; X64-NEXT: punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3] +; X64-NEXT: retq %val = load <4 x i8>, <4 x i8>* %ptr %ext = zext <4 x i8> %val to <4 x i32> ret <4 x i32> %ext |
