diff options
Diffstat (limited to 'llvm/test/CodeGen/X86/extract-insert.ll')
| -rw-r--r-- | llvm/test/CodeGen/X86/extract-insert.ll | 11 |
1 files changed, 1 insertions, 10 deletions
diff --git a/llvm/test/CodeGen/X86/extract-insert.ll b/llvm/test/CodeGen/X86/extract-insert.ll index be5f9ed24fb..97ca884a54b 100644 --- a/llvm/test/CodeGen/X86/extract-insert.ll +++ b/llvm/test/CodeGen/X86/extract-insert.ll @@ -28,30 +28,21 @@ define i8 @extractelt_bitcast(i32 %x) nounwind { ret i8 %ext } -; TODO: This should have folded to avoid vector ops, but the transform -; is guarded by 'hasOneUse'. That limitation apparently makes some AMDGPU -; codegen better. - define i8 @extractelt_bitcast_extra_use(i32 %x, <4 x i8>* %p) nounwind { ; X86-LABEL: extractelt_bitcast_extra_use: ; X86: # %bb.0: ; X86-NEXT: pushl %eax ; X86-NEXT: movl {{[0-9]+}}(%esp), %eax -; X86-NEXT: movd %eax, %xmm0 -; X86-NEXT: punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7] ; X86-NEXT: movl {{[0-9]+}}(%esp), %ecx ; X86-NEXT: movl %eax, (%ecx) -; X86-NEXT: movd %xmm0, %eax ; X86-NEXT: # kill: def $al killed $al killed $eax ; X86-NEXT: popl %ecx ; X86-NEXT: retl ; ; X64-LABEL: extractelt_bitcast_extra_use: ; X64: # %bb.0: -; X64-NEXT: movd %edi, %xmm0 -; X64-NEXT: punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7] +; X64-NEXT: movl %edi, %eax ; X64-NEXT: movl %edi, (%rsi) -; X64-NEXT: movd %xmm0, %eax ; X64-NEXT: # kill: def $al killed $al killed $eax ; X64-NEXT: retq %bc = bitcast i32 %x to <4 x i8> |
