From e62fbca6b7dbb40ed453cc6fb002ef6be7f5c3e6 Mon Sep 17 00:00:00 2001 From: Chandler Carruth Date: Mon, 16 Feb 2015 12:23:22 +0000 Subject: [PATCH] [x86] Switch this test to use checks generated by my update script. NFC git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@229377 91177308-0d34-0410-b5e6-96231b3b80d8 --- test/CodeGen/X86/pmul.ll | 118 ++++++++++++++++++++++++++++----------- 1 file changed, 85 insertions(+), 33 deletions(-) diff --git a/test/CodeGen/X86/pmul.ll b/test/CodeGen/X86/pmul.ll index 8937d6afa0a..ec4077a8a5b 100644 --- a/test/CodeGen/X86/pmul.ll +++ b/test/CodeGen/X86/pmul.ll @@ -3,16 +3,18 @@ define <4 x i32> @a(<4 x i32> %i) nounwind { ; SSE2-LABEL: a: -; SSE2: movdqa {{.*}}, %[[X1:xmm[0-9]+]] -; SSE2-NEXT: pshufd {{.*}} # [[X2:xmm[0-9]+]] = xmm0[1,1,3,3] -; SSE2-NEXT: pmuludq %[[X1]], %xmm0 -; SSE2-NEXT: pmuludq %[[X1]], %[[X2]] -; SSE2-NEXT: shufps {{.*}} # xmm0 = xmm0[0,2],[[X2]][0,2] -; SSE2-NEXT: shufps {{.*}} # xmm0 = xmm0[0,2,1,3] +; SSE2: # BB#0: # %entry +; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [117,117,117,117] +; SSE2-NEXT: pshufd {{.*#+}} xmm2 = xmm0[1,1,3,3] +; SSE2-NEXT: pmuludq %xmm1, %xmm0 +; SSE2-NEXT: pmuludq %xmm1, %xmm2 +; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,2],xmm2[0,2] +; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,2,1,3] ; SSE2-NEXT: retq ; ; SSE41-LABEL: a: -; SSE41: pmulld +; SSE41: # BB#0: # %entry +; SSE41-NEXT: pmulld {{.*}}(%rip), %xmm0 ; SSE41-NEXT: retq entry: %A = mul <4 x i32> %i, < i32 117, i32 117, i32 117, i32 117 > @@ -21,9 +23,19 @@ entry: define <2 x i64> @b(<2 x i64> %i) nounwind { ; ALL-LABEL: b: -; ALL: pmuludq -; ALL: pmuludq -; ALL: pmuludq +; ALL: # BB#0: # %entry +; ALL-NEXT: movdqa {{.*#+}} xmm1 = [117,117] +; ALL-NEXT: movdqa %xmm0, %xmm2 +; ALL-NEXT: pmuludq %xmm1, %xmm2 +; ALL-NEXT: pxor %xmm3, %xmm3 +; ALL-NEXT: pmuludq %xmm0, %xmm3 +; ALL-NEXT: psllq $32, %xmm3 +; ALL-NEXT: paddq %xmm3, %xmm2 +; ALL-NEXT: psrlq $32, %xmm0 +; ALL-NEXT: pmuludq %xmm1, %xmm0 +; ALL-NEXT: psllq $32, %xmm0 +; ALL-NEXT: paddq %xmm2, %xmm0 +; ALL-NEXT: retq entry: %A = mul <2 x i64> %i, < i64 117, i64 117 > ret <2 x i64> %A @@ -31,16 +43,18 @@ entry: define <4 x i32> @c(<4 x i32> %i, <4 x i32> %j) nounwind { ; SSE2-LABEL: c: -; SSE2: pshufd {{.*}} # [[X2:xmm[0-9]+]] = xmm0[1,1,3,3] +; SSE2: # BB#0: # %entry +; SSE2-NEXT: pshufd {{.*#+}} xmm2 = xmm0[1,1,3,3] ; SSE2-NEXT: pmuludq %xmm1, %xmm0 -; SSE2-NEXT: pshufd {{.*}} # xmm1 = xmm1[1,1,3,3] -; SSE2-NEXT: pmuludq %[[X2]], %xmm1 -; SSE2-NEXT: shufps {{.*}} # xmm0 = xmm0[0,2],xmm1[0,2] -; SSE2-NEXT: shufps {{.*}} # xmm0 = xmm0[0,2,1,3] +; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm1[1,1,3,3] +; SSE2-NEXT: pmuludq %xmm2, %xmm1 +; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,2],xmm1[0,2] +; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,2,1,3] ; SSE2-NEXT: retq ; ; SSE41-LABEL: c: -; SSE41: pmulld +; SSE41: # BB#0: # %entry +; SSE41-NEXT: pmulld %xmm1, %xmm0 ; SSE41-NEXT: retq entry: %A = mul <4 x i32> %i, %j @@ -49,9 +63,19 @@ entry: define <2 x i64> @d(<2 x i64> %i, <2 x i64> %j) nounwind { ; ALL-LABEL: d: -; ALL: pmuludq -; ALL: pmuludq -; ALL: pmuludq +; ALL: # BB#0: # %entry +; ALL-NEXT: movdqa %xmm0, %xmm2 +; ALL-NEXT: pmuludq %xmm1, %xmm2 +; ALL-NEXT: movdqa %xmm1, %xmm3 +; ALL-NEXT: psrlq $32, %xmm3 +; ALL-NEXT: pmuludq %xmm0, %xmm3 +; ALL-NEXT: psllq $32, %xmm3 +; ALL-NEXT: paddq %xmm3, %xmm2 +; ALL-NEXT: psrlq $32, %xmm0 +; ALL-NEXT: pmuludq %xmm1, %xmm0 +; ALL-NEXT: psllq $32, %xmm0 +; ALL-NEXT: paddq %xmm2, %xmm0 +; ALL-NEXT: retq entry: %A = mul <2 x i64> %i, %j ret <2 x i64> %A @@ -61,20 +85,31 @@ declare void @foo() define <4 x i32> @e(<4 x i32> %i, <4 x i32> %j) nounwind { ; SSE2-LABEL: e: -; SSE2: movdqa {{[0-9]*}}(%rsp), %xmm0 -; SSE2-NEXT: pshufd {{.*}} # [[X1:xmm[0-9]+]] = xmm0[1,1,3,3] -; SSE2-NEXT: movdqa {{[0-9]*}}(%rsp), %[[X2:xmm[0-9]+]] -; SSE2-NEXT: pmuludq %[[X2]], %xmm0 -; SSE2-NEXT: pshufd {{.*}} # [[X2]] = [[X2]][1,1,3,3] -; SSE2-NEXT: pmuludq %[[X1]], %[[X2]] -; SSE2-NEXT: shufps {{.*}} # xmm0 = xmm0[0,2],[[X2]][0,2] -; SSE2-NEXT: shufps {{.*}} # xmm0 = xmm0[0,2,1,3] -; SSE2-NEXT: addq ${{[0-9]+}}, %rsp +; SSE2: # BB#0: # %entry +; SSE2-NEXT: subq $40, %rsp +; SSE2-NEXT: movaps %xmm1, {{[0-9]+}}(%rsp) # 16-byte Spill +; SSE2-NEXT: movaps %xmm0, (%rsp) # 16-byte Spill +; SSE2-NEXT: callq foo +; SSE2-NEXT: movdqa (%rsp), %xmm0 # 16-byte Reload +; SSE2-NEXT: pshufd {{.*#+}} xmm2 = xmm0[1,1,3,3] +; SSE2-NEXT: movdqa {{[0-9]+}}(%rsp), %xmm1 # 16-byte Reload +; SSE2-NEXT: pmuludq %xmm1, %xmm0 +; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm1[1,1,3,3] +; SSE2-NEXT: pmuludq %xmm2, %xmm1 +; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,2],xmm1[0,2] +; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[0,2,1,3] +; SSE2-NEXT: addq $40, %rsp ; SSE2-NEXT: retq ; ; SSE41-LABEL: e: -; SSE41: pmulld {{[0-9]+}}(%rsp), %xmm -; SSE41-NEXT: addq ${{[0-9]+}}, %rsp +; SSE41: # BB#0: # %entry +; SSE41-NEXT: subq $40, %rsp +; SSE41-NEXT: movaps %xmm1, {{[0-9]+}}(%rsp) # 16-byte Spill +; SSE41-NEXT: movaps %xmm0, (%rsp) # 16-byte Spill +; SSE41-NEXT: callq foo +; SSE41-NEXT: movdqa (%rsp), %xmm0 # 16-byte Reload +; SSE41-NEXT: pmulld {{[0-9]+}}(%rsp), %xmm0 # 16-byte Folded Reload +; SSE41-NEXT: addq $40, %rsp ; SSE41-NEXT: retq entry: ; Use a call to force spills. @@ -85,9 +120,26 @@ entry: define <2 x i64> @f(<2 x i64> %i, <2 x i64> %j) nounwind { ; ALL-LABEL: f: -; ALL: pmuludq -; ALL: pmuludq -; ALL: pmuludq +; ALL: # BB#0: # %entry +; ALL-NEXT: subq $40, %rsp +; ALL-NEXT: movaps %xmm1, {{[0-9]+}}(%rsp) # 16-byte Spill +; ALL-NEXT: movaps %xmm0, (%rsp) # 16-byte Spill +; ALL-NEXT: callq foo +; ALL-NEXT: movdqa (%rsp), %xmm0 # 16-byte Reload +; ALL-NEXT: movdqa %xmm0, %xmm2 +; ALL-NEXT: movdqa {{[0-9]+}}(%rsp), %xmm3 # 16-byte Reload +; ALL-NEXT: pmuludq %xmm3, %xmm2 +; ALL-NEXT: movdqa %xmm3, %xmm1 +; ALL-NEXT: psrlq $32, %xmm1 +; ALL-NEXT: pmuludq %xmm0, %xmm1 +; ALL-NEXT: psllq $32, %xmm1 +; ALL-NEXT: paddq %xmm1, %xmm2 +; ALL-NEXT: psrlq $32, %xmm0 +; ALL-NEXT: pmuludq %xmm3, %xmm0 +; ALL-NEXT: psllq $32, %xmm0 +; ALL-NEXT: paddq %xmm2, %xmm0 +; ALL-NEXT: addq $40, %rsp +; ALL-NEXT: retq entry: ; Use a call to force spills. call void @foo() -- 2.34.1