1 ; RUN: llvm-upgrade < %s | llvm-as | llc -march=x86 -mattr=+sse,+sse2 | grep minss | grep CPI | wc -l | grep 2 &&
2 ; RUN: llvm-upgrade < %s | llvm-as | llc -march=x86 -mattr=+sse,+sse2 | grep CPI | not grep movss
5 target pointersize = 32
6 target triple = "i686-apple-darwin8.7.2"
8 implementation ; Functions:
10 ushort %test1(float %f) {
11 %tmp = insertelement <4 x float> undef, float %f, uint 0 ; <<4 x float>> [#uses=1]
12 %tmp10 = insertelement <4 x float> %tmp, float 0.000000e+00, uint 1 ; <<4 x float>> [#uses=1]
13 %tmp11 = insertelement <4 x float> %tmp10, float 0.000000e+00, uint 2 ; <<4 x float>> [#uses=1]
14 %tmp12 = insertelement <4 x float> %tmp11, float 0.000000e+00, uint 3 ; <<4 x float>> [#uses=1]
15 %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]
16 %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]
17 %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]
18 %tmp59 = tail call <4 x float> %llvm.x86.sse.max.ss( <4 x float> %tmp48, <4 x float> zeroinitializer ) ; <<4 x float>> [#uses=1]
19 %tmp = tail call int %llvm.x86.sse.cvttss2si( <4 x float> %tmp59 ) ; <int> [#uses=1]
20 %tmp69 = cast int %tmp to ushort ; <ushort> [#uses=1]
24 ushort %test2(float %f) {
25 %tmp28 = sub float %f, 1.000000e+00 ; <float> [#uses=1]
26 %tmp37 = mul float %tmp28, 5.000000e-01 ; <float> [#uses=1]
27 %tmp375 = insertelement <4 x float> undef, float %tmp37, uint 0 ; <<4 x float>> [#uses=1]
28 %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]
29 %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]
30 %tmp = tail call int %llvm.x86.sse.cvttss2si( <4 x float> %tmp59 ) ; <int> [#uses=1]
31 %tmp69 = cast int %tmp to ushort ; <ushort> [#uses=1]
36 declare <4 x float> %llvm.x86.sse.sub.ss(<4 x float>, <4 x float>)
38 declare <4 x float> %llvm.x86.sse.mul.ss(<4 x float>, <4 x float>)
40 declare <4 x float> %llvm.x86.sse.min.ss(<4 x float>, <4 x float>)
42 declare <4 x float> %llvm.x86.sse.max.ss(<4 x float>, <4 x float>)
44 declare int %llvm.x86.sse.cvttss2si(<4 x float>)