diff options
Diffstat (limited to 'llvm/test/CodeGen/X86/vec_ss_load_fold.ll')
| -rw-r--r-- | llvm/test/CodeGen/X86/vec_ss_load_fold.ll | 45 |
1 files changed, 26 insertions, 19 deletions
diff --git a/llvm/test/CodeGen/X86/vec_ss_load_fold.ll b/llvm/test/CodeGen/X86/vec_ss_load_fold.ll index 5b298d8c6eb..076f748009b 100644 --- a/llvm/test/CodeGen/X86/vec_ss_load_fold.ll +++ b/llvm/test/CodeGen/X86/vec_ss_load_fold.ll @@ -13,18 +13,20 @@ define i16 @test1(float %f) nounwind { ; CHECK-NEXT: minss LCPI0_2, %xmm0 ; CHECK-NEXT: maxss %xmm1, %xmm0 ; CHECK-NEXT: cvttss2si %xmm0, %eax +; CHECK-NEXT: ## kill: %AX<def> %AX<kill> %EAX<kill> ; CHECK-NEXT: retl - %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] - %tmp12 = insertelement <4 x float> %tmp11, float 0.000000e+00, i32 3 ; <<4 x float>> [#uses=1] - %tmp28 = tail call <4 x float> @llvm.x86.sse.sub.ss( <4 x float> %tmp12, <4 x float> < float 1.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00 > ) ; <<4 x float>> [#uses=1] - %tmp37 = tail call <4 x float> @llvm.x86.sse.mul.ss( <4 x float> %tmp28, <4 x float> < float 5.000000e-01, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00 > ) ; <<4 x float>> [#uses=1] - %tmp48 = tail call <4 x float> @llvm.x86.sse.min.ss( <4 x float> %tmp37, <4 x float> < float 6.553500e+04, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00 > ) ; <<4 x float>> [#uses=1] - %tmp59 = tail call <4 x float> @llvm.x86.sse.max.ss( <4 x float> %tmp48, <4 x float> zeroinitializer ) ; <<4 x float>> [#uses=1] - %tmp.upgrd.1 = tail call i32 @llvm.x86.sse.cvttss2si( <4 x float> %tmp59 ) ; <i32> [#uses=1] - %tmp69 = trunc i32 %tmp.upgrd.1 to i16 ; <i16> [#uses=1] - ret i16 %tmp69 +; + %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] + %tmp12 = insertelement <4 x float> %tmp11, float 0.000000e+00, i32 3 ; <<4 x float>> [#uses=1] + %tmp28 = tail call <4 x float> @llvm.x86.sse.sub.ss( <4 x float> %tmp12, <4 x float> < float 1.000000e+00, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00 > ) ; <<4 x float>> [#uses=1] + %tmp37 = tail call <4 x float> @llvm.x86.sse.mul.ss( <4 x float> %tmp28, <4 x float> < float 5.000000e-01, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00 > ) ; <<4 x float>> [#uses=1] + %tmp48 = tail call <4 x float> @llvm.x86.sse.min.ss( <4 x float> %tmp37, <4 x float> < float 6.553500e+04, float 0.000000e+00, float 0.000000e+00, float 0.000000e+00 > ) ; <<4 x float>> [#uses=1] + %tmp59 = tail call <4 x float> @llvm.x86.sse.max.ss( <4 x float> %tmp48, <4 x float> zeroinitializer ) ; <<4 x float>> [#uses=1] + %tmp.upgrd.1 = tail call i32 @llvm.x86.sse.cvttss2si( <4 x float> %tmp59 ) ; <i32> [#uses=1] + %tmp69 = trunc i32 %tmp.upgrd.1 to i16 ; <i16> [#uses=1] + ret i16 %tmp69 } define i16 @test2(float %f) nounwind { @@ -37,15 +39,17 @@ define i16 @test2(float %f) nounwind { ; CHECK-NEXT: xorps %xmm1, %xmm1 ; CHECK-NEXT: maxss %xmm1, %xmm0 ; CHECK-NEXT: cvttss2si %xmm0, %eax +; CHECK-NEXT: ## kill: %AX<def> %AX<kill> %EAX<kill> ; CHECK-NEXT: retl - %tmp28 = fsub float %f, 1.000000e+00 ; <float> [#uses=1] - %tmp37 = fmul float %tmp28, 5.000000e-01 ; <float> [#uses=1] - %tmp375 = insertelement <4 x float> undef, float %tmp37, i32 0 ; <<4 x float>> [#uses=1] - %tmp48 = tail call <4 x float> @llvm.x86.sse.min.ss( <4 x float> %tmp375, <4 x float> < float 6.553500e+04, float undef, float undef, float undef > ) ; <<4 x float>> [#uses=1] - %tmp59 = tail call <4 x float> @llvm.x86.sse.max.ss( <4 x float> %tmp48, <4 x float> < float 0.000000e+00, float undef, float undef, float undef > ) ; <<4 x float>> [#uses=1] - %tmp = tail call i32 @llvm.x86.sse.cvttss2si( <4 x float> %tmp59 ) ; <i32> [#uses=1] - %tmp69 = trunc i32 %tmp to i16 ; <i16> [#uses=1] - ret i16 %tmp69 +; + %tmp28 = fsub float %f, 1.000000e+00 ; <float> [#uses=1] + %tmp37 = fmul float %tmp28, 5.000000e-01 ; <float> [#uses=1] + %tmp375 = insertelement <4 x float> undef, float %tmp37, i32 0 ; <<4 x float>> [#uses=1] + %tmp48 = tail call <4 x float> @llvm.x86.sse.min.ss( <4 x float> %tmp375, <4 x float> < float 6.553500e+04, float undef, float undef, float undef > ) ; <<4 x float>> [#uses=1] + %tmp59 = tail call <4 x float> @llvm.x86.sse.max.ss( <4 x float> %tmp48, <4 x float> < float 0.000000e+00, float undef, float undef, float undef > ) ; <<4 x float>> [#uses=1] + %tmp = tail call i32 @llvm.x86.sse.cvttss2si( <4 x float> %tmp59 ) ; <i32> [#uses=1] + %tmp69 = trunc i32 %tmp to i16 ; <i16> [#uses=1] + ret i16 %tmp69 } declare <4 x float> @llvm.x86.sse.sub.ss(<4 x float>, <4 x float>) @@ -68,6 +72,7 @@ define <4 x float> @test3(<4 x float> %A, float *%b, i32 %C) nounwind { ; CHECK-NEXT: movl {{[0-9]+}}(%esp), %eax ; CHECK-NEXT: roundss $4, (%eax), %xmm0 ; CHECK-NEXT: retl +; %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) @@ -86,6 +91,7 @@ define <4 x float> @test4(<4 x float> %A, float *%b, i32 %C) nounwind { ; CHECK-NEXT: roundss $4, %xmm1, %xmm0 ; CHECK-NEXT: addl $28, %esp ; CHECK-NEXT: retl +; %a = load float , float *%b %B = insertelement <4 x float> undef, float %a, i32 0 %q = call <4 x float> @f() @@ -101,6 +107,7 @@ define <2 x double> @test5() nounwind uwtable readnone noinline { ; CHECK-NEXT: movl $128, %eax ; CHECK-NEXT: cvtsi2sdl %eax, %xmm0 ; CHECK-NEXT: retl +; entry: %0 = tail call <2 x double> @llvm.x86.sse2.cvtsi2sd(<2 x double> <double 4.569870e+02, double 1.233210e+02>, i32 128) nounwind readnone |
