diff options
Diffstat (limited to 'llvm/test/CodeGen/X86/sse3.ll')
| -rw-r--r-- | llvm/test/CodeGen/X86/sse3.ll | 202 |
1 files changed, 161 insertions, 41 deletions
diff --git a/llvm/test/CodeGen/X86/sse3.ll b/llvm/test/CodeGen/X86/sse3.ll index 79b949a6ccb..1e7b9da6a32 100644 --- a/llvm/test/CodeGen/X86/sse3.ll +++ b/llvm/test/CodeGen/X86/sse3.ll @@ -1,20 +1,30 @@ ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py -; These are tests for SSE3 codegen. +; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+sse3 | FileCheck %s --check-prefix=X86 +; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse3 | FileCheck %s --check-prefix=X64 -; RUN: llc < %s -mtriple=x86_64-apple-darwin9 --mattr=+sse3 | FileCheck %s --check-prefix=X64 +; These are tests for SSE3 codegen. ; Test for v8xi16 lowering where we extract the first element of the vector and ; placed it in the second element of the result. define void @t0(<8 x i16>* %dest, <8 x i16>* %old) nounwind { +; X86-LABEL: t0: +; X86: # BB#0: # %entry +; X86-NEXT: movl {{[0-9]+}}(%esp), %eax +; X86-NEXT: movl {{[0-9]+}}(%esp), %ecx +; X86-NEXT: movl $1, %edx +; X86-NEXT: movd %edx, %xmm0 +; X86-NEXT: punpcklwd {{.*#+}} xmm0 = xmm0[0],mem[0],xmm0[1],mem[1],xmm0[2],mem[2],xmm0[3],mem[3] +; X86-NEXT: movdqa %xmm0, (%eax) +; X86-NEXT: retl +; ; X64-LABEL: t0: -; X64: ## BB#0: ## %entry +; X64: # BB#0: # %entry ; X64-NEXT: movl $1, %eax ; X64-NEXT: movd %eax, %xmm0 ; X64-NEXT: punpcklwd {{.*#+}} xmm0 = xmm0[0],mem[0],xmm0[1],mem[1],xmm0[2],mem[2],xmm0[3],mem[3] ; X64-NEXT: movdqa %xmm0, (%rdi) ; X64-NEXT: retq -; X64-NEXT: ## -- End function entry: %tmp3 = load <8 x i16>, <8 x i16>* %old %tmp6 = shufflevector <8 x i16> %tmp3, @@ -25,15 +35,25 @@ entry: } define <8 x i16> @t1(<8 x i16>* %A, <8 x i16>* %B) nounwind { +; X86-LABEL: t1: +; X86: # BB#0: +; X86-NEXT: movl {{[0-9]+}}(%esp), %eax +; X86-NEXT: movl {{[0-9]+}}(%esp), %ecx +; X86-NEXT: movaps {{.*#+}} xmm0 = [0,65535,65535,65535,65535,65535,65535,65535] +; X86-NEXT: movaps %xmm0, %xmm1 +; X86-NEXT: andnps (%ecx), %xmm1 +; X86-NEXT: andps (%eax), %xmm0 +; X86-NEXT: orps %xmm1, %xmm0 +; X86-NEXT: retl +; ; X64-LABEL: t1: -; X64: ## BB#0: +; X64: # BB#0: ; X64-NEXT: movaps {{.*#+}} xmm0 = [0,65535,65535,65535,65535,65535,65535,65535] ; X64-NEXT: movaps %xmm0, %xmm1 ; X64-NEXT: andnps (%rsi), %xmm1 ; X64-NEXT: andps (%rdi), %xmm0 ; X64-NEXT: orps %xmm1, %xmm0 ; X64-NEXT: retq -; X64-NEXT: ## -- End function %tmp1 = load <8 x i16>, <8 x i16>* %A %tmp2 = load <8 x i16>, <8 x i16>* %B %tmp3 = shufflevector <8 x i16> %tmp1, <8 x i16> %tmp2, <8 x i32> < i32 8, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7 > @@ -42,86 +62,131 @@ define <8 x i16> @t1(<8 x i16>* %A, <8 x i16>* %B) nounwind { } define <8 x i16> @t2(<8 x i16> %A, <8 x i16> %B) nounwind { +; X86-LABEL: t2: +; X86: # BB#0: +; X86-NEXT: movdqa {{.*#+}} xmm2 = [0,65535,65535,0,65535,65535,65535,65535] +; X86-NEXT: pand %xmm2, %xmm0 +; X86-NEXT: pshuflw {{.*#+}} xmm1 = xmm1[1,1,2,1,4,5,6,7] +; X86-NEXT: pandn %xmm1, %xmm2 +; X86-NEXT: por %xmm2, %xmm0 +; X86-NEXT: retl +; ; X64-LABEL: t2: -; X64: ## BB#0: +; X64: # BB#0: ; X64-NEXT: movdqa {{.*#+}} xmm2 = [0,65535,65535,0,65535,65535,65535,65535] ; X64-NEXT: pand %xmm2, %xmm0 ; X64-NEXT: pshuflw {{.*#+}} xmm1 = xmm1[1,1,2,1,4,5,6,7] ; X64-NEXT: pandn %xmm1, %xmm2 ; X64-NEXT: por %xmm2, %xmm0 ; X64-NEXT: retq -; X64-NEXT: ## -- End function %tmp = shufflevector <8 x i16> %A, <8 x i16> %B, <8 x i32> < i32 9, i32 1, i32 2, i32 9, i32 4, i32 5, i32 6, i32 7 > ret <8 x i16> %tmp } define <8 x i16> @t3(<8 x i16> %A, <8 x i16> %B) nounwind { +; X86-LABEL: t3: +; X86: # BB#0: +; X86-NEXT: pshufd {{.*#+}} xmm0 = xmm0[3,1,2,0] +; X86-NEXT: pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6,5] +; X86-NEXT: pshufd {{.*#+}} xmm0 = xmm0[3,1,2,0] +; X86-NEXT: pshuflw {{.*#+}} xmm0 = xmm0[0,3,2,1,4,5,6,7] +; X86-NEXT: pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,6,5,4] +; X86-NEXT: retl +; ; X64-LABEL: t3: -; X64: ## BB#0: +; X64: # BB#0: ; X64-NEXT: pshufd {{.*#+}} xmm0 = xmm0[3,1,2,0] ; X64-NEXT: pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6,5] ; X64-NEXT: pshufd {{.*#+}} xmm0 = xmm0[3,1,2,0] ; X64-NEXT: pshuflw {{.*#+}} xmm0 = xmm0[0,3,2,1,4,5,6,7] ; X64-NEXT: pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,6,5,4] ; X64-NEXT: retq -; X64-NEXT: ## -- End function %tmp = shufflevector <8 x i16> %A, <8 x i16> %A, <8 x i32> < i32 8, i32 3, i32 2, i32 13, i32 7, i32 6, i32 5, i32 4 > ret <8 x i16> %tmp } define <8 x i16> @t4(<8 x i16> %A, <8 x i16> %B) nounwind { +; X86-LABEL: t4: +; X86: # BB#0: +; X86-NEXT: pshufd {{.*#+}} xmm0 = xmm0[2,1,0,3] +; X86-NEXT: pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,5,4,7] +; X86-NEXT: pshufd {{.*#+}} xmm0 = xmm0[3,1,2,0] +; X86-NEXT: pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,7,4,7] +; X86-NEXT: retl +; ; X64-LABEL: t4: -; X64: ## BB#0: +; X64: # BB#0: ; X64-NEXT: pshufd {{.*#+}} xmm0 = xmm0[2,1,0,3] ; X64-NEXT: pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,5,4,7] ; X64-NEXT: pshufd {{.*#+}} xmm0 = xmm0[3,1,2,0] ; X64-NEXT: pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,7,4,7] ; X64-NEXT: retq -; X64-NEXT: ## -- End function %tmp = shufflevector <8 x i16> %A, <8 x i16> %B, <8 x i32> < i32 0, i32 7, i32 2, i32 3, i32 1, i32 5, i32 6, i32 5 > ret <8 x i16> %tmp } define <8 x i16> @t5(<8 x i16> %A, <8 x i16> %B) nounwind { +; X86-LABEL: t5: +; X86: # BB#0: +; X86-NEXT: punpckldq {{.*#+}} xmm1 = xmm1[0],xmm0[0],xmm1[1],xmm0[1] +; X86-NEXT: movdqa %xmm1, %xmm0 +; X86-NEXT: retl +; ; X64-LABEL: t5: -; X64: ## BB#0: +; X64: # BB#0: ; X64-NEXT: punpckldq {{.*#+}} xmm1 = xmm1[0],xmm0[0],xmm1[1],xmm0[1] ; X64-NEXT: movdqa %xmm1, %xmm0 ; X64-NEXT: retq -; X64-NEXT: ## -- End function %tmp = shufflevector <8 x i16> %A, <8 x i16> %B, <8 x i32> < i32 8, i32 9, i32 0, i32 1, i32 10, i32 11, i32 2, i32 3 > ret <8 x i16> %tmp } define <8 x i16> @t6(<8 x i16> %A, <8 x i16> %B) nounwind { +; X86-LABEL: t6: +; X86: # BB#0: +; X86-NEXT: movss {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3] +; X86-NEXT: retl +; ; X64-LABEL: t6: -; X64: ## BB#0: +; X64: # BB#0: ; X64-NEXT: movss {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3] ; X64-NEXT: retq -; X64-NEXT: ## -- End function %tmp = shufflevector <8 x i16> %A, <8 x i16> %B, <8 x i32> < i32 8, i32 9, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7 > ret <8 x i16> %tmp } define <8 x i16> @t7(<8 x i16> %A, <8 x i16> %B) nounwind { +; X86-LABEL: t7: +; X86: # BB#0: +; X86-NEXT: pshuflw {{.*#+}} xmm0 = xmm0[0,0,3,2,4,5,6,7] +; X86-NEXT: pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,6,4,7] +; X86-NEXT: retl +; ; X64-LABEL: t7: -; X64: ## BB#0: +; X64: # BB#0: ; X64-NEXT: pshuflw {{.*#+}} xmm0 = xmm0[0,0,3,2,4,5,6,7] ; X64-NEXT: pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,6,4,7] ; X64-NEXT: retq -; X64-NEXT: ## -- End function %tmp = shufflevector <8 x i16> %A, <8 x i16> %B, <8 x i32> < i32 0, i32 0, i32 3, i32 2, i32 4, i32 6, i32 4, i32 7 > ret <8 x i16> %tmp } define void @t8(<2 x i64>* %res, <2 x i64>* %A) nounwind { +; X86-LABEL: t8: +; X86: # BB#0: +; X86-NEXT: movl {{[0-9]+}}(%esp), %eax +; X86-NEXT: movl {{[0-9]+}}(%esp), %ecx +; X86-NEXT: pshuflw {{.*#+}} xmm0 = mem[2,1,0,3,4,5,6,7] +; X86-NEXT: pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,5,4,7] +; X86-NEXT: movdqa %xmm0, (%eax) +; X86-NEXT: retl +; ; X64-LABEL: t8: -; X64: ## BB#0: +; X64: # BB#0: ; X64-NEXT: pshuflw {{.*#+}} xmm0 = mem[2,1,0,3,4,5,6,7] ; X64-NEXT: pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,5,4,7] ; X64-NEXT: movdqa %xmm0, (%rdi) ; X64-NEXT: retq -; X64-NEXT: ## -- End function %tmp = load <2 x i64>, <2 x i64>* %A %tmp.upgrd.1 = bitcast <2 x i64> %tmp to <8 x i16> %tmp0 = extractelement <8 x i16> %tmp.upgrd.1, i32 0 @@ -146,13 +211,21 @@ define void @t8(<2 x i64>* %res, <2 x i64>* %A) nounwind { } define void @t9(<4 x float>* %r, <2 x i32>* %A) nounwind { +; X86-LABEL: t9: +; X86: # BB#0: +; X86-NEXT: movl {{[0-9]+}}(%esp), %eax +; X86-NEXT: movl {{[0-9]+}}(%esp), %ecx +; X86-NEXT: movapd (%ecx), %xmm0 +; X86-NEXT: movhpd {{.*#+}} xmm0 = xmm0[0],mem[0] +; X86-NEXT: movapd %xmm0, (%ecx) +; X86-NEXT: retl +; ; X64-LABEL: t9: -; X64: ## BB#0: +; X64: # BB#0: ; X64-NEXT: movapd (%rdi), %xmm0 ; X64-NEXT: movhpd {{.*#+}} xmm0 = xmm0[0],mem[0] ; X64-NEXT: movapd %xmm0, (%rdi) ; X64-NEXT: retq -; X64-NEXT: ## -- End function %tmp = load <4 x float>, <4 x float>* %r %tmp.upgrd.3 = bitcast <2 x i32>* %A to double* %tmp.upgrd.4 = load double, double* %tmp.upgrd.3 @@ -180,16 +253,21 @@ define void @t9(<4 x float>* %r, <2 x i32>* %A) nounwind { @g2 = external constant <4 x i16> define void @t10() nounwind { +; X86-LABEL: t10: +; X86: # BB#0: +; X86-NEXT: pshuflw {{.*#+}} xmm0 = mem[0,2,2,3,4,5,6,7] +; X86-NEXT: pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,6,6,7] +; X86-NEXT: pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3] +; X86-NEXT: movq %xmm0, g2 +; X86-NEXT: retl +; ; X64-LABEL: t10: -; X64: ## BB#0: -; X64-NEXT: movq _g1@{{.*}}(%rip), %rax +; X64: # BB#0: ; X64-NEXT: pshuflw {{.*#+}} xmm0 = mem[0,2,2,3,4,5,6,7] ; X64-NEXT: pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,6,6,7] ; X64-NEXT: pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3] -; X64-NEXT: movq _g2@{{.*}}(%rip), %rax -; X64-NEXT: movq %xmm0, (%rax) +; X64-NEXT: movq %xmm0, {{.*}}(%rip) ; X64-NEXT: retq -; X64-NEXT: ## -- End function load <4 x i32>, <4 x i32>* @g1, align 16 bitcast <4 x i32> %1 to <8 x i16> shufflevector <8 x i16> %2, <8 x i16> undef, <8 x i32> < i32 0, i32 2, i32 4, i32 6, i32 undef, i32 undef, i32 undef, i32 undef > @@ -202,12 +280,17 @@ define void @t10() nounwind { ; Pack various elements via shuffles. define <8 x i16> @t11(<8 x i16> %T0, <8 x i16> %T1) nounwind readnone { +; X86-LABEL: t11: +; X86: # BB#0: # %entry +; X86-NEXT: psrld $16, %xmm0 +; 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: t11: -; X64: ## BB#0: ## %entry +; X64: # BB#0: # %entry ; X64-NEXT: psrld $16, %xmm0 ; X64-NEXT: punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3] ; X64-NEXT: retq -; X64-NEXT: ## -- End function entry: %tmp7 = shufflevector <8 x i16> %T0, <8 x i16> %T1, <8 x i32> < i32 1, i32 8, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef , i32 undef > ret <8 x i16> %tmp7 @@ -215,13 +298,19 @@ entry: } define <8 x i16> @t12(<8 x i16> %T0, <8 x i16> %T1) nounwind readnone { +; X86-LABEL: t12: +; X86: # BB#0: # %entry +; X86-NEXT: punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3] +; X86-NEXT: pshuflw {{.*#+}} xmm0 = xmm0[0,2,2,3,4,5,6,7] +; X86-NEXT: pshufd {{.*#+}} xmm0 = xmm0[0,1,3,3] +; X86-NEXT: retl +; ; X64-LABEL: t12: -; X64: ## BB#0: ## %entry +; X64: # BB#0: # %entry ; X64-NEXT: punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3] ; X64-NEXT: pshuflw {{.*#+}} xmm0 = xmm0[0,2,2,3,4,5,6,7] ; X64-NEXT: pshufd {{.*#+}} xmm0 = xmm0[0,1,3,3] ; X64-NEXT: retq -; X64-NEXT: ## -- End function entry: %tmp9 = shufflevector <8 x i16> %T0, <8 x i16> %T1, <8 x i32> < i32 0, i32 1, i32 undef, i32 undef, i32 3, i32 11, i32 undef , i32 undef > ret <8 x i16> %tmp9 @@ -229,26 +318,38 @@ entry: } define <8 x i16> @t13(<8 x i16> %T0, <8 x i16> %T1) nounwind readnone { +; X86-LABEL: t13: +; X86: # BB#0: # %entry +; X86-NEXT: punpcklwd {{.*#+}} xmm1 = xmm1[0],xmm0[0],xmm1[1],xmm0[1],xmm1[2],xmm0[2],xmm1[3],xmm0[3] +; X86-NEXT: pshuflw {{.*#+}} xmm0 = xmm1[0,2,2,3,4,5,6,7] +; X86-NEXT: pshufd {{.*#+}} xmm0 = xmm0[0,1,3,3] +; X86-NEXT: retl +; ; X64-LABEL: t13: -; X64: ## BB#0: ## %entry +; X64: # BB#0: # %entry ; X64-NEXT: punpcklwd {{.*#+}} xmm1 = xmm1[0],xmm0[0],xmm1[1],xmm0[1],xmm1[2],xmm0[2],xmm1[3],xmm0[3] ; X64-NEXT: pshuflw {{.*#+}} xmm0 = xmm1[0,2,2,3,4,5,6,7] ; X64-NEXT: pshufd {{.*#+}} xmm0 = xmm0[0,1,3,3] ; X64-NEXT: retq -; X64-NEXT: ## -- End function entry: %tmp9 = shufflevector <8 x i16> %T0, <8 x i16> %T1, <8 x i32> < i32 8, i32 9, i32 undef, i32 undef, i32 11, i32 3, i32 undef , i32 undef > ret <8 x i16> %tmp9 } define <8 x i16> @t14(<8 x i16> %T0, <8 x i16> %T1) nounwind readnone { +; X86-LABEL: t14: +; X86: # BB#0: # %entry +; X86-NEXT: psrlq $16, %xmm0 +; X86-NEXT: punpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm0[0] +; X86-NEXT: movdqa %xmm1, %xmm0 +; X86-NEXT: retl +; ; X64-LABEL: t14: -; X64: ## BB#0: ## %entry +; X64: # BB#0: # %entry ; X64-NEXT: psrlq $16, %xmm0 ; X64-NEXT: punpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm0[0] ; X64-NEXT: movdqa %xmm1, %xmm0 ; X64-NEXT: retq -; X64-NEXT: ## -- End function entry: %tmp9 = shufflevector <8 x i16> %T0, <8 x i16> %T1, <8 x i32> < i32 8, i32 9, i32 undef, i32 undef, i32 undef, i32 2, i32 undef , i32 undef > ret <8 x i16> %tmp9 @@ -256,13 +357,19 @@ entry: ; FIXME: t15 is worse off from disabling of scheduler 2-address hack. define <8 x i16> @t15(<8 x i16> %T0, <8 x i16> %T1) nounwind readnone { +; X86-LABEL: t15: +; X86: # BB#0: # %entry +; X86-NEXT: pshufd {{.*#+}} xmm0 = xmm0[3,1,2,3] +; X86-NEXT: pshuflw {{.*#+}} xmm0 = xmm0[0,1,1,2,4,5,6,7] +; X86-NEXT: punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0] +; X86-NEXT: retl +; ; X64-LABEL: t15: -; X64: ## BB#0: ## %entry +; X64: # BB#0: # %entry ; X64-NEXT: pshufd {{.*#+}} xmm0 = xmm0[3,1,2,3] ; X64-NEXT: pshuflw {{.*#+}} xmm0 = xmm0[0,1,1,2,4,5,6,7] ; X64-NEXT: punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0] ; X64-NEXT: retq -; X64-NEXT: ## -- End function entry: %tmp8 = shufflevector <8 x i16> %T0, <8 x i16> %T1, <8 x i32> < i32 undef, i32 undef, i32 7, i32 2, i32 8, i32 undef, i32 undef , i32 undef > ret <8 x i16> %tmp8 @@ -270,13 +377,19 @@ entry: ; Test yonah where we convert a shuffle to pextrw and pinrsw define <16 x i8> @t16(<16 x i8> %T0) nounwind readnone { +; X86-LABEL: t16: +; X86: # BB#0: # %entry +; X86-NEXT: movdqa {{.*#+}} xmm1 = [0,0,0,0,1,1,1,1,0,0,0,0,0,0,0,0] +; X86-NEXT: punpcklwd {{.*#+}} xmm1 = xmm1[0],xmm0[0],xmm1[1],xmm0[1],xmm1[2],xmm0[2],xmm1[3],xmm0[3] +; X86-NEXT: movdqa %xmm1, %xmm0 +; X86-NEXT: retl +; ; X64-LABEL: t16: -; X64: ## BB#0: ## %entry +; X64: # BB#0: # %entry ; X64-NEXT: movdqa {{.*#+}} xmm1 = [0,0,0,0,1,1,1,1,0,0,0,0,0,0,0,0] ; X64-NEXT: punpcklwd {{.*#+}} xmm1 = xmm1[0],xmm0[0],xmm1[1],xmm0[1],xmm1[2],xmm0[2],xmm1[3],xmm0[3] ; X64-NEXT: movdqa %xmm1, %xmm0 ; X64-NEXT: retq -; X64-NEXT: ## -- End function entry: %tmp8 = shufflevector <16 x i8> <i8 0, i8 0, i8 0, i8 0, i8 1, i8 1, i8 1, i8 1, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0>, <16 x i8> %T0, <16 x i32> < i32 0, i32 1, i32 16, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef , i32 undef > %tmp9 = shufflevector <16 x i8> %tmp8, <16 x i8> %T0, <16 x i32> < i32 0, i32 1, i32 2, i32 17, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef , i32 undef > @@ -285,14 +398,21 @@ entry: ; rdar://8520311 define <4 x i32> @t17() nounwind { +; X86-LABEL: t17: +; X86: # BB#0: # %entry +; X86-NEXT: movaps (%eax), %xmm0 +; X86-NEXT: unpcklps {{.*#+}} xmm0 = xmm0[0,0,1,1] +; X86-NEXT: pxor %xmm1, %xmm1 +; X86-NEXT: punpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1] +; X86-NEXT: retl +; ; X64-LABEL: t17: -; X64: ## BB#0: ## %entry +; X64: # BB#0: # %entry ; X64-NEXT: movaps (%rax), %xmm0 ; X64-NEXT: unpcklps {{.*#+}} xmm0 = xmm0[0,0,1,1] ; X64-NEXT: pxor %xmm1, %xmm1 ; X64-NEXT: punpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1] ; X64-NEXT: retq -; X64-NEXT: ## -- End function entry: %tmp1 = load <4 x float>, <4 x float>* undef, align 16 %tmp2 = shufflevector <4 x float> %tmp1, <4 x float> undef, <4 x i32> <i32 4, i32 1, i32 2, i32 3> |
