diff options
Diffstat (limited to 'llvm/test/CodeGen/X86/vselect.ll')
| -rw-r--r-- | llvm/test/CodeGen/X86/vselect.ll | 75 |
1 files changed, 37 insertions, 38 deletions
diff --git a/llvm/test/CodeGen/X86/vselect.ll b/llvm/test/CodeGen/X86/vselect.ll index d3e2b5477ac..d89b1352aee 100644 --- a/llvm/test/CodeGen/X86/vselect.ll +++ b/llvm/test/CodeGen/X86/vselect.ll @@ -563,43 +563,36 @@ define <2 x i64> @shrunkblend_nonvselectuse(<2 x i1> %cond, <2 x i64> %a, <2 x i define <2 x i32> @simplify_select(i32 %x, <2 x i1> %z) { ; SSE2-LABEL: simplify_select: ; SSE2: # %bb.0: -; SSE2-NEXT: pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3] -; SSE2-NEXT: pslld $31, %xmm0 +; SSE2-NEXT: psllq $63, %xmm0 ; SSE2-NEXT: psrad $31, %xmm0 +; SSE2-NEXT: pshufd {{.*#+}} xmm0 = xmm0[1,1,3,3] ; SSE2-NEXT: movd %edi, %xmm1 -; SSE2-NEXT: pshufd {{.*#+}} xmm2 = xmm1[1,0,1,1] -; SSE2-NEXT: por %xmm1, %xmm2 -; SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[0,0],xmm2[1,1] -; SSE2-NEXT: shufps {{.*#+}} xmm1 = xmm1[2,0],xmm2[2,3] +; SSE2-NEXT: pshufd {{.*#+}} xmm2 = xmm1[0,1,0,1] +; SSE2-NEXT: movdqa %xmm2, %xmm3 +; SSE2-NEXT: punpcklqdq {{.*#+}} xmm3 = xmm3[0],xmm1[0] ; SSE2-NEXT: pand %xmm0, %xmm2 -; SSE2-NEXT: pandn %xmm1, %xmm0 +; SSE2-NEXT: pandn %xmm3, %xmm0 ; SSE2-NEXT: por %xmm2, %xmm0 ; SSE2-NEXT: retq ; ; SSE41-LABEL: simplify_select: ; SSE41: # %bb.0: -; SSE41-NEXT: pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3] -; SSE41-NEXT: pslld $31, %xmm0 -; SSE41-NEXT: movd %edi, %xmm1 -; SSE41-NEXT: pshufd {{.*#+}} xmm2 = xmm1[1,0,1,1] -; SSE41-NEXT: por %xmm1, %xmm2 -; SSE41-NEXT: pshufd {{.*#+}} xmm1 = xmm2[1,1,2,3] -; SSE41-NEXT: pinsrd $1, %edi, %xmm1 -; SSE41-NEXT: blendvps %xmm0, %xmm2, %xmm1 -; SSE41-NEXT: movaps %xmm1, %xmm0 +; SSE41-NEXT: movd %edi, %xmm0 +; SSE41-NEXT: pshufd {{.*#+}} xmm0 = xmm0[0,1,0,1] ; SSE41-NEXT: retq ; -; AVX-LABEL: simplify_select: -; AVX: # %bb.0: -; AVX-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3] -; AVX-NEXT: vpslld $31, %xmm0, %xmm0 -; AVX-NEXT: vmovd %edi, %xmm1 -; AVX-NEXT: vpshufd {{.*#+}} xmm2 = xmm1[1,0,1,1] -; AVX-NEXT: vpor %xmm1, %xmm2, %xmm1 -; AVX-NEXT: vpshufd {{.*#+}} xmm2 = xmm1[1,1,2,3] -; AVX-NEXT: vpinsrd $1, %edi, %xmm2, %xmm2 -; AVX-NEXT: vblendvps %xmm0, %xmm1, %xmm2, %xmm0 -; AVX-NEXT: retq +; AVX1-LABEL: simplify_select: +; AVX1: # %bb.0: +; AVX1-NEXT: vmovd %edi, %xmm0 +; AVX1-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,1,0,1] +; AVX1-NEXT: retq +; +; AVX2-LABEL: simplify_select: +; AVX2: # %bb.0: +; AVX2-NEXT: # kill: def $edi killed $edi def $rdi +; AVX2-NEXT: vmovq %rdi, %xmm0 +; AVX2-NEXT: vpbroadcastq %xmm0, %xmm0 +; AVX2-NEXT: retq %a = insertelement <2 x i32> <i32 0, i32 undef>, i32 %x, i32 1 %b = insertelement <2 x i32> <i32 undef, i32 0>, i32 %x, i32 0 %y = or <2 x i32> %a, %b @@ -650,28 +643,34 @@ define i64 @vselect_any_extend_vector_inreg_crash(<8 x i8>* %x) { ; SSE2-LABEL: vselect_any_extend_vector_inreg_crash: ; SSE2: # %bb.0: ; SSE2-NEXT: movq {{.*#+}} xmm0 = mem[0],zero -; SSE2-NEXT: pcmpeqb {{.*}}(%rip), %xmm0 -; SSE2-NEXT: punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7] -; SSE2-NEXT: punpcklwd {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3] +; SSE2-NEXT: pxor %xmm1, %xmm1 +; SSE2-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] +; SSE2-NEXT: pcmpeqw {{.*}}(%rip), %xmm0 +; SSE2-NEXT: pshufd {{.*#+}} xmm0 = xmm0[0,1,0,3] +; SSE2-NEXT: pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,5,6,7] +; SSE2-NEXT: psllq $56, %xmm0 ; SSE2-NEXT: psrad $24, %xmm0 +; SSE2-NEXT: pshufd {{.*#+}} xmm0 = xmm0[1,1,3,3] ; SSE2-NEXT: movq %xmm0, %rax ; SSE2-NEXT: andl $32768, %eax # imm = 0x8000 ; SSE2-NEXT: retq ; ; SSE41-LABEL: vselect_any_extend_vector_inreg_crash: ; SSE41: # %bb.0: -; SSE41-NEXT: movq {{.*#+}} xmm0 = mem[0],zero -; SSE41-NEXT: pcmpeqb {{.*}}(%rip), %xmm0 -; SSE41-NEXT: pmovsxbq %xmm0, %xmm0 -; SSE41-NEXT: movq %xmm0, %rax -; SSE41-NEXT: andl $32768, %eax # imm = 0x8000 +; SSE41-NEXT: pmovzxbw {{.*#+}} xmm0 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero +; SSE41-NEXT: pcmpeqw {{.*}}(%rip), %xmm0 +; SSE41-NEXT: psllq $56, %xmm0 +; SSE41-NEXT: movl $32768, %eax # imm = 0x8000 +; SSE41-NEXT: movq %rax, %xmm1 +; SSE41-NEXT: xorpd %xmm2, %xmm2 +; SSE41-NEXT: blendvpd %xmm0, %xmm1, %xmm2 +; SSE41-NEXT: movq %xmm2, %rax ; SSE41-NEXT: retq ; ; AVX-LABEL: vselect_any_extend_vector_inreg_crash: ; AVX: # %bb.0: -; AVX-NEXT: vmovq {{.*#+}} xmm0 = mem[0],zero -; AVX-NEXT: vpcmpeqb {{.*}}(%rip), %xmm0, %xmm0 -; AVX-NEXT: vpmovsxbq %xmm0, %xmm0 +; AVX-NEXT: vpmovzxbw {{.*#+}} xmm0 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero +; AVX-NEXT: vpcmpeqw {{.*}}(%rip), %xmm0, %xmm0 ; AVX-NEXT: vmovq %xmm0, %rax ; AVX-NEXT: andl $32768, %eax # imm = 0x8000 ; AVX-NEXT: retq |
