1 files changed, 65 insertions, 0 deletions
diff --git a/test/Transforms/ScalarRepl/vector_promote.ll b/test/Transforms/ScalarRepl/vector_promote.ll
new file mode 100644
index 0000000..4f875b0
--- /dev/null
+++ b/test/Transforms/ScalarRepl/vector_promote.ll
@@ -0,0 +1,65 @@
+; RUN: opt < %s -scalarrepl -S | not grep alloca
+; RUN: opt < %s -scalarrepl -S | grep {load <4 x float>}
+target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64"
+
+define void @test(<4 x float>* %F, float %f) {
+entry:
+	%G = alloca <4 x float>, align 16		; <<4 x float>*> [#uses=3]
+	%tmp = load <4 x float>* %F		; <<4 x float>> [#uses=2]
+	%tmp3 = fadd <4 x float> %tmp, %tmp		; <<4 x float>> [#uses=1]
+	store <4 x float> %tmp3, <4 x float>* %G
+	%G.upgrd.1 = getelementptr <4 x float>* %G, i32 0, i32 0		; <float*> [#uses=1]
+	store float %f, float* %G.upgrd.1
+	%tmp4 = load <4 x float>* %G		; <<4 x float>> [#uses=2]
+	%tmp6 = fadd <4 x float> %tmp4, %tmp4		; <<4 x float>> [#uses=1]
+	store <4 x float> %tmp6, <4 x float>* %F
+	ret void
+}
+
+define void @test2(<4 x float>* %F, float %f) {
+entry:
+	%G = alloca <4 x float>, align 16		; <<4 x float>*> [#uses=3]
+	%tmp = load <4 x float>* %F		; <<4 x float>> [#uses=2]
+	%tmp3 = fadd <4 x float> %tmp, %tmp		; <<4 x float>> [#uses=1]
+	store <4 x float> %tmp3, <4 x float>* %G
+	%tmp.upgrd.2 = getelementptr <4 x float>* %G, i32 0, i32 2		; <float*> [#uses=1]
+	store float %f, float* %tmp.upgrd.2
+	%tmp4 = load <4 x float>* %G		; <<4 x float>> [#uses=2]
+	%tmp6 = fadd <4 x float> %tmp4, %tmp4		; <<4 x float>> [#uses=1]
+	store <4 x float> %tmp6, <4 x float>* %F
+	ret void
+}
+
+define void @test3(<4 x float>* %F, float* %f) {
+entry:
+	%G = alloca <4 x float>, align 16		; <<4 x float>*> [#uses=2]
+	%tmp = load <4 x float>* %F		; <<4 x float>> [#uses=2]
+	%tmp3 = fadd <4 x float> %tmp, %tmp		; <<4 x float>> [#uses=1]
+	store <4 x float> %tmp3, <4 x float>* %G
+	%tmp.upgrd.3 = getelementptr <4 x float>* %G, i32 0, i32 2		; <float*> [#uses=1]
+	%tmp.upgrd.4 = load float* %tmp.upgrd.3		; <float> [#uses=1]
+	store float %tmp.upgrd.4, float* %f
+	ret void
+}
+
+define void @test4(<4 x float>* %F, float* %f) {
+entry:
+	%G = alloca <4 x float>, align 16		; <<4 x float>*> [#uses=2]
+	%tmp = load <4 x float>* %F		; <<4 x float>> [#uses=2]
+	%tmp3 = fadd <4 x float> %tmp, %tmp		; <<4 x float>> [#uses=1]
+	store <4 x float> %tmp3, <4 x float>* %G
+	%G.upgrd.5 = getelementptr <4 x float>* %G, i32 0, i32 0		; <float*> [#uses=1]
+	%tmp.upgrd.6 = load float* %G.upgrd.5		; <float> [#uses=1]
+	store float %tmp.upgrd.6, float* %f
+	ret void
+}
+
+define i32 @test5(float %X) {  ;; should turn into bitcast.
+	%X_addr = alloca [4 x float]
+        %X1 = getelementptr [4 x float]* %X_addr, i32 0, i32 2
+	store float %X, float* %X1
+	%a = bitcast float* %X1 to i32*
+	%tmp = load i32* %a
+	ret i32 %tmp
+}
+