--- /dev/null
+; RUN: llvm-as< %s | llc -march=x86 -mcpu=yonah &&
+; RUN: llvm-as< %s | llc -march=x86 -mcpu=yonah | not grep sub.*esp
+
+; This checks that various insert/extract idiom work without going to the
+; stack.
+; XFAIL: *
+
+void %test(<4 x float>* %F, float %f) {
+entry:
+ %tmp = load <4 x float>* %F ; <<4 x float>> [#uses=2]
+ %tmp3 = add <4 x float> %tmp, %tmp ; <<4 x float>> [#uses=1]
+ %tmp10 = insertelement <4 x float> %tmp3, float %f, uint 0 ; <<4 x float>> [#uses=2]
+ %tmp6 = add <4 x float> %tmp10, %tmp10 ; <<4 x float>> [#uses=1]
+ store <4 x float> %tmp6, <4 x float>* %F
+ ret void
+}
+
+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 = add <4 x float> %tmp, %tmp ; <<4 x float>> [#uses=1]
+ store <4 x float> %tmp3, <4 x float>* %G
+ %tmp = getelementptr <4 x float>* %G, int 0, int 2 ; <float*> [#uses=1]
+ store float %f, float* %tmp
+ %tmp4 = load <4 x float>* %G ; <<4 x float>> [#uses=2]
+ %tmp6 = add <4 x float> %tmp4, %tmp4 ; <<4 x float>> [#uses=1]
+ store <4 x float> %tmp6, <4 x float>* %F
+ ret void
+}
+
+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 = add <4 x float> %tmp, %tmp ; <<4 x float>> [#uses=1]
+ store <4 x float> %tmp3, <4 x float>* %G
+ %tmp = getelementptr <4 x float>* %G, int 0, int 2 ; <float*> [#uses=1]
+ %tmp = load float* %tmp ; <float> [#uses=1]
+ store float %tmp, float* %f
+ ret void
+}
+
+void %test4(<4 x float>* %F, float* %f) {
+entry:
+ %tmp = load <4 x float>* %F ; <<4 x float>> [#uses=2]
+ %tmp5.lhs = extractelement <4 x float> %tmp, uint 0 ; <float> [#uses=1]
+ %tmp5.rhs = extractelement <4 x float> %tmp, uint 0 ; <float> [#uses=1]
+ %tmp5 = add float %tmp5.lhs, %tmp5.rhs ; <float> [#uses=1]
+ store float %tmp5, float* %f
+ ret void
+}