va_args support for Win64.

[oota-llvm.git] / test / CodeGen / X86 / 2006-05-01-SchedCausingSpills.ll

diff --git a/test/CodeGen/X86/2006-05-01-SchedCausingSpills.ll b/test/CodeGen/X86/2006-05-01-SchedCausingSpills.ll
index f28366699c3d49adff0e29bd65ee025cd90de9e4..b0453299669e2680f90528dfe151789109bfa472 100644 (file)
--- a/test/CodeGen/X86/2006-05-01-SchedCausingSpills.ll
+++ b/test/CodeGen/X86/2006-05-01-SchedCausingSpills.ll
@@ -1,4 +1,4 @@
-; RUN: llvm-as < %s | llc -march=x86 -mcpu=yonah -stats |& \
+; RUN: llc < %s -march=x86 -mcpu=yonah -stats |& \
 ; RUN:   not grep {Number of register spills}
 ; END.
 
@@ -17,8 +17,8 @@ define i32 @foo(<4 x float>* %a, <4 x float>* %b, <4 x float>* %c, <4 x float>*
        %tmp75 = bitcast <4 x float> %tmp74 to <4 x i32>                ; <<4 x i32>> [#uses=1]
        %tmp88 = tail call <4 x float> @llvm.x86.sse.cmp.ps( <4 x float> %tmp44, <4 x float> %tmp61, i8 1 )             ; <<4 x float>> [#uses=1]
        %tmp89 = bitcast <4 x float> %tmp88 to <4 x i32>                ; <<4 x i32>> [#uses=1]
-       %tmp98 = tail call <4 x i32> @llvm.x86.sse2.packssdw.128( <4 x i32> %tmp75, <4 x i32> %tmp89 )          ; <<4 x i32>> [#uses=1]
-       %tmp102 = bitcast <4 x i32> %tmp98 to <8 x i16>         ; <<8 x i16>> [#uses=1]
+       %tmp98 = tail call <8 x i16> @llvm.x86.sse2.packssdw.128( <4 x i32> %tmp75, <4 x i32> %tmp89 )          ; <<4 x i32>> [#uses=1]
+       %tmp102 = bitcast <8 x i16> %tmp98 to <8 x i16>         ; <<8 x i16>> [#uses=1]
        %tmp.upgrd.1 = shufflevector <8 x i16> %tmp102, <8 x i16> undef, <8 x i32> < i32 0, i32 1, i32 2, i32 3, i32 6, i32 5, i32 4, i32 7 >           ; <<8 x i16>> [#uses=1]
        %tmp105 = shufflevector <8 x i16> %tmp.upgrd.1, <8 x i16> undef, <8 x i32> < i32 2, i32 1, i32 0, i32 3, i32 4, i32 5, i32 6, i32 7 >           ; <<8 x i16>> [#uses=1]
        %tmp105.upgrd.2 = bitcast <8 x i16> %tmp105 to <4 x float>              ; <<4 x float>> [#uses=1]
@@ -32,8 +32,8 @@ define i32 @foo(<4 x float>* %a, <4 x float>* %b, <4 x float>* %c, <4 x float>*
        %tmp134 = bitcast <4 x float> %tmp133 to <4 x i32>              ; <<4 x i32>> [#uses=1]
        %tmp147 = tail call <4 x float> @llvm.x86.sse.cmp.ps( <4 x float> %tmp44, <4 x float> %tmp120, i8 1 )           ; <<4 x float>> [#uses=1]
        %tmp148 = bitcast <4 x float> %tmp147 to <4 x i32>              ; <<4 x i32>> [#uses=1]
-       %tmp159 = tail call <4 x i32> @llvm.x86.sse2.packssdw.128( <4 x i32> %tmp134, <4 x i32> %tmp148 )               ; <<4 x i32>> [#uses=1]
-       %tmp163 = bitcast <4 x i32> %tmp159 to <8 x i16>                ; <<8 x i16>> [#uses=1]
+       %tmp159 = tail call <8 x i16> @llvm.x86.sse2.packssdw.128( <4 x i32> %tmp134, <4 x i32> %tmp148 )               ; <<4 x i32>> [#uses=1]
+       %tmp163 = bitcast <8 x i16> %tmp159 to <8 x i16>                ; <<8 x i16>> [#uses=1]
        %tmp164 = shufflevector <8 x i16> %tmp163, <8 x i16> undef, <8 x i32> < i32 0, i32 1, i32 2, i32 3, i32 6, i32 5, i32 4, i32 7 >                ; <<8 x i16>> [#uses=1]
        %tmp166 = shufflevector <8 x i16> %tmp164, <8 x i16> undef, <8 x i32> < i32 2, i32 1, i32 0, i32 3, i32 4, i32 5, i32 6, i32 7 >                ; <<8 x i16>> [#uses=1]
        %tmp166.upgrd.4 = bitcast <8 x i16> %tmp166 to <4 x float>              ; <<4 x float>> [#uses=1]
@@ -47,8 +47,8 @@ define i32 @foo(<4 x float>* %a, <4 x float>* %b, <4 x float>* %c, <4 x float>*
        %tmp195 = bitcast <4 x float> %tmp194 to <4 x i32>              ; <<4 x i32>> [#uses=1]
        %tmp208 = tail call <4 x float> @llvm.x86.sse.cmp.ps( <4 x float> %tmp44, <4 x float> %tmp181, i8 1 )           ; <<4 x float>> [#uses=1]
        %tmp209 = bitcast <4 x float> %tmp208 to <4 x i32>              ; <<4 x i32>> [#uses=1]
-       %tmp220 = tail call <4 x i32> @llvm.x86.sse2.packssdw.128( <4 x i32> %tmp195, <4 x i32> %tmp209 )               ; <<4 x i32>> [#uses=1]
-       %tmp224 = bitcast <4 x i32> %tmp220 to <8 x i16>                ; <<8 x i16>> [#uses=1]
+       %tmp220 = tail call <8 x i16> @llvm.x86.sse2.packssdw.128( <4 x i32> %tmp195, <4 x i32> %tmp209 )               ; <<4 x i32>> [#uses=1]
+       %tmp224 = bitcast <8 x i16> %tmp220 to <8 x i16>                ; <<8 x i16>> [#uses=1]
        %tmp225 = shufflevector <8 x i16> %tmp224, <8 x i16> undef, <8 x i32> < i32 0, i32 1, i32 2, i32 3, i32 6, i32 5, i32 4, i32 7 >                ; <<8 x i16>> [#uses=1]
        %tmp227 = shufflevector <8 x i16> %tmp225, <8 x i16> undef, <8 x i32> < i32 2, i32 1, i32 0, i32 3, i32 4, i32 5, i32 6, i32 7 >                ; <<8 x i16>> [#uses=1]
        %tmp227.upgrd.6 = bitcast <8 x i16> %tmp227 to <4 x float>              ; <<4 x float>> [#uses=1]
@@ -62,8 +62,8 @@ define i32 @foo(<4 x float>* %a, <4 x float>* %b, <4 x float>* %c, <4 x float>*
        %tmp256 = bitcast <4 x float> %tmp255 to <4 x i32>              ; <<4 x i32>> [#uses=1]
        %tmp269 = tail call <4 x float> @llvm.x86.sse.cmp.ps( <4 x float> %tmp44, <4 x float> %tmp242, i8 1 )           ; <<4 x float>> [#uses=1]
        %tmp270 = bitcast <4 x float> %tmp269 to <4 x i32>              ; <<4 x i32>> [#uses=1]
-       %tmp281 = tail call <4 x i32> @llvm.x86.sse2.packssdw.128( <4 x i32> %tmp256, <4 x i32> %tmp270 )               ; <<4 x i32>> [#uses=1]
-       %tmp285 = bitcast <4 x i32> %tmp281 to <8 x i16>                ; <<8 x i16>> [#uses=1]
+       %tmp281 = tail call <8 x i16> @llvm.x86.sse2.packssdw.128( <4 x i32> %tmp256, <4 x i32> %tmp270 )               ; <<4 x i32>> [#uses=1]
+       %tmp285 = bitcast <8 x i16> %tmp281 to <8 x i16>                ; <<8 x i16>> [#uses=1]
        %tmp286 = shufflevector <8 x i16> %tmp285, <8 x i16> undef, <8 x i32> < i32 0, i32 1, i32 2, i32 3, i32 6, i32 5, i32 4, i32 7 >                ; <<8 x i16>> [#uses=1]
        %tmp288 = shufflevector <8 x i16> %tmp286, <8 x i16> undef, <8 x i32> < i32 2, i32 1, i32 0, i32 3, i32 4, i32 5, i32 6, i32 7 >                ; <<8 x i16>> [#uses=1]
        %tmp288.upgrd.8 = bitcast <8 x i16> %tmp288 to <4 x float>              ; <<4 x float>> [#uses=1]
@@ -73,4 +73,4 @@ define i32 @foo(<4 x float>* %a, <4 x float>* %b, <4 x float>* %c, <4 x float>*
 
 declare <4 x float> @llvm.x86.sse.cmp.ps(<4 x float>, <4 x float>, i8)
 
-declare <4 x i32> @llvm.x86.sse2.packssdw.128(<4 x i32>, <4 x i32>)
+declare <8 x i16> @llvm.x86.sse2.packssdw.128(<4 x i32>, <4 x i32>)