From: Simon Pilgrim Date: Wed, 24 Aug 2016 15:07:11 +0000 (+0000) Subject: [X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets X-Git-Tag: android-x86-7.1-r4~28104 X-Git-Url: http://git.osdn.net/view?a=commitdiff_plain;h=e2aca6b0d951fdd3534d493678aa1238a2b682c0;p=android-x86%2Fexternal-llvm.git [X86][SSE] Regenerate scalar math load folding tests for 32 and 64 bit targets git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@279630 91177308-0d34-0410-b5e6-96231b3b80d8 --- diff --git a/test/CodeGen/X86/vec_ss_load_fold.ll b/test/CodeGen/X86/vec_ss_load_fold.ll index 076f748009b..bdc490f461c 100644 --- a/test/CodeGen/X86/vec_ss_load_fold.ll +++ b/test/CodeGen/X86/vec_ss_load_fold.ll @@ -1,21 +1,31 @@ -; NOTE: Assertions have been autogenerated by utils/update_test_checks.py -; RUN: llc < %s -mtriple=i686-apple-darwin9 -mattr=+sse,+sse2,+sse4.1 | FileCheck %s - -target datalayout = "e-p:32:32" +; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py +; RUN: llc < %s -mtriple=i686-apple-darwin9 -mattr=+sse,+sse2,+sse4.1 | FileCheck %s --check-prefix=CHECK --check-prefix=X32 +; RUN: llc < %s -mtriple=x86_64-apple-darwin9 -mattr=+sse,+sse2,+sse4.1 | FileCheck %s --check-prefix=CHECK --check-prefix=X64 define i16 @test1(float %f) nounwind { -; CHECK-LABEL: test1: -; CHECK: ## BB#0: -; CHECK-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero -; CHECK-NEXT: xorps %xmm1, %xmm1 -; CHECK-NEXT: subss LCPI0_0, %xmm0 -; CHECK-NEXT: mulss LCPI0_1, %xmm0 -; CHECK-NEXT: minss LCPI0_2, %xmm0 -; CHECK-NEXT: maxss %xmm1, %xmm0 -; CHECK-NEXT: cvttss2si %xmm0, %eax -; CHECK-NEXT: ## kill: %AX %AX %EAX -; CHECK-NEXT: retl +; X32-LABEL: test1: +; X32: ## BB#0: +; X32-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero +; X32-NEXT: xorps %xmm1, %xmm1 +; X32-NEXT: subss LCPI0_0, %xmm0 +; X32-NEXT: mulss LCPI0_1, %xmm0 +; X32-NEXT: minss LCPI0_2, %xmm0 +; X32-NEXT: maxss %xmm1, %xmm0 +; X32-NEXT: cvttss2si %xmm0, %eax +; X32-NEXT: ## kill: %AX %AX %EAX +; X32-NEXT: retl ; +; X64-LABEL: test1: +; X64: ## BB#0: +; X64-NEXT: xorps %xmm1, %xmm1 +; X64-NEXT: blendps {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3] +; X64-NEXT: subss {{.*}}(%rip), %xmm0 +; X64-NEXT: mulss {{.*}}(%rip), %xmm0 +; X64-NEXT: minss {{.*}}(%rip), %xmm0 +; X64-NEXT: maxss %xmm1, %xmm0 +; X64-NEXT: cvttss2si %xmm0, %eax +; X64-NEXT: ## kill: %AX %AX %EAX +; X64-NEXT: retq %tmp = insertelement <4 x float> undef, float %f, i32 0 ; <<4 x float>> [#uses=1] %tmp10 = insertelement <4 x float> %tmp, float 0.000000e+00, i32 1 ; <<4 x float>> [#uses=1] %tmp11 = insertelement <4 x float> %tmp10, float 0.000000e+00, i32 2 ; <<4 x float>> [#uses=1] @@ -30,18 +40,28 @@ define i16 @test1(float %f) nounwind { } define i16 @test2(float %f) nounwind { -; CHECK-LABEL: test2: -; CHECK: ## BB#0: -; CHECK-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero -; CHECK-NEXT: addss LCPI1_0, %xmm0 -; CHECK-NEXT: mulss LCPI1_1, %xmm0 -; CHECK-NEXT: minss LCPI1_2, %xmm0 -; CHECK-NEXT: xorps %xmm1, %xmm1 -; CHECK-NEXT: maxss %xmm1, %xmm0 -; CHECK-NEXT: cvttss2si %xmm0, %eax -; CHECK-NEXT: ## kill: %AX %AX %EAX -; CHECK-NEXT: retl +; X32-LABEL: test2: +; X32: ## BB#0: +; X32-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero +; X32-NEXT: addss LCPI1_0, %xmm0 +; X32-NEXT: mulss LCPI1_1, %xmm0 +; X32-NEXT: minss LCPI1_2, %xmm0 +; X32-NEXT: xorps %xmm1, %xmm1 +; X32-NEXT: maxss %xmm1, %xmm0 +; X32-NEXT: cvttss2si %xmm0, %eax +; X32-NEXT: ## kill: %AX %AX %EAX +; X32-NEXT: retl ; +; X64-LABEL: test2: +; X64: ## BB#0: +; X64-NEXT: addss {{.*}}(%rip), %xmm0 +; X64-NEXT: mulss {{.*}}(%rip), %xmm0 +; X64-NEXT: minss {{.*}}(%rip), %xmm0 +; X64-NEXT: xorps %xmm1, %xmm1 +; X64-NEXT: maxss %xmm1, %xmm0 +; X64-NEXT: cvttss2si %xmm0, %eax +; X64-NEXT: ## kill: %AX %AX %EAX +; X64-NEXT: retq %tmp28 = fsub float %f, 1.000000e+00 ; [#uses=1] %tmp37 = fmul float %tmp28, 5.000000e-01 ; [#uses=1] %tmp375 = insertelement <4 x float> undef, float %tmp37, i32 0 ; <<4 x float>> [#uses=1] @@ -67,12 +87,16 @@ declare <4 x float> @llvm.x86.sse41.round.ss(<4 x float>, <4 x float>, i32) declare <4 x float> @f() define <4 x float> @test3(<4 x float> %A, float *%b, i32 %C) nounwind { -; CHECK-LABEL: test3: -; CHECK: ## BB#0: -; CHECK-NEXT: movl {{[0-9]+}}(%esp), %eax -; CHECK-NEXT: roundss $4, (%eax), %xmm0 -; CHECK-NEXT: retl +; X32-LABEL: test3: +; X32: ## BB#0: +; X32-NEXT: movl {{[0-9]+}}(%esp), %eax +; X32-NEXT: roundss $4, (%eax), %xmm0 +; X32-NEXT: retl ; +; X64-LABEL: test3: +; X64: ## BB#0: +; X64-NEXT: roundss $4, (%rdi), %xmm0 +; X64-NEXT: retq %a = load float , float *%b %B = insertelement <4 x float> undef, float %a, i32 0 %X = call <4 x float> @llvm.x86.sse41.round.ss(<4 x float> %A, <4 x float> %B, i32 4) @@ -80,18 +104,28 @@ define <4 x float> @test3(<4 x float> %A, float *%b, i32 %C) nounwind { } define <4 x float> @test4(<4 x float> %A, float *%b, i32 %C) nounwind { -; CHECK-LABEL: test4: -; CHECK: ## BB#0: -; CHECK-NEXT: subl $28, %esp -; CHECK-NEXT: movl {{[0-9]+}}(%esp), %eax -; CHECK-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero -; CHECK-NEXT: movaps %xmm0, (%esp) ## 16-byte Spill -; CHECK-NEXT: calll _f -; CHECK-NEXT: movaps (%esp), %xmm1 ## 16-byte Reload -; CHECK-NEXT: roundss $4, %xmm1, %xmm0 -; CHECK-NEXT: addl $28, %esp -; CHECK-NEXT: retl +; X32-LABEL: test4: +; X32: ## BB#0: +; X32-NEXT: subl $28, %esp +; X32-NEXT: movl {{[0-9]+}}(%esp), %eax +; X32-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero +; X32-NEXT: movaps %xmm0, (%esp) ## 16-byte Spill +; X32-NEXT: calll _f +; X32-NEXT: movaps (%esp), %xmm1 ## 16-byte Reload +; X32-NEXT: roundss $4, %xmm1, %xmm0 +; X32-NEXT: addl $28, %esp +; X32-NEXT: retl ; +; X64-LABEL: test4: +; X64: ## BB#0: +; X64-NEXT: subq $24, %rsp +; X64-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero +; X64-NEXT: movaps %xmm0, (%rsp) ## 16-byte Spill +; X64-NEXT: callq _f +; X64-NEXT: movaps (%rsp), %xmm1 ## 16-byte Reload +; X64-NEXT: roundss $4, %xmm1, %xmm0 +; X64-NEXT: addq $24, %rsp +; X64-NEXT: retq %a = load float , float *%b %B = insertelement <4 x float> undef, float %a, i32 0 %q = call <4 x float> @f() @@ -101,16 +135,21 @@ define <4 x float> @test4(<4 x float> %A, float *%b, i32 %C) nounwind { ; PR13576 define <2 x double> @test5() nounwind uwtable readnone noinline { -; CHECK-LABEL: test5: -; CHECK: ## BB#0: ## %entry -; CHECK-NEXT: movaps {{.*#+}} xmm0 = [4.569870e+02,1.233210e+02] -; CHECK-NEXT: movl $128, %eax -; CHECK-NEXT: cvtsi2sdl %eax, %xmm0 -; CHECK-NEXT: retl +; X32-LABEL: test5: +; X32: ## BB#0: ## %entry +; X32-NEXT: movaps {{.*#+}} xmm0 = [4.569870e+02,1.233210e+02] +; X32-NEXT: movl $128, %eax +; X32-NEXT: cvtsi2sdl %eax, %xmm0 +; X32-NEXT: retl ; +; X64-LABEL: test5: +; X64: ## BB#0: ## %entry +; X64-NEXT: movaps {{.*#+}} xmm0 = [4.569870e+02,1.233210e+02] +; X64-NEXT: movl $128, %eax +; X64-NEXT: cvtsi2sdl %eax, %xmm0 +; X64-NEXT: retq entry: - %0 = tail call <2 x double> @llvm.x86.sse2.cvtsi2sd(<2 x double> , i32 128) nounwind readnone + %0 = tail call <2 x double> @llvm.x86.sse2.cvtsi2sd(<2 x double> , i32 128) nounwind readnone ret <2 x double> %0 }