[X86] Add test showing failure to fold freeze(vpermps(x,y)) -> vpermps(freeze(x),freeze(y)) (#160837)
diff --git a/llvm/test/CodeGen/X86/vector-shuffle-combining-avx2.ll b/llvm/test/CodeGen/X86/vector-shuffle-combining-avx2.ll
index 298858a..06468f6 100644
--- a/llvm/test/CodeGen/X86/vector-shuffle-combining-avx2.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-combining-avx2.ll
@@ -933,6 +933,26 @@
ret i32 %tmp6
}
+define <8 x float> @freeze_permps(<8 x float> %a0) {
+; AVX2-LABEL: freeze_permps:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vmovaps {{.*#+}} ymm1 = [7,6,5,4,3,2,1,0]
+; AVX2-NEXT: vpermps %ymm0, %ymm1, %ymm0
+; AVX2-NEXT: vpermps %ymm0, %ymm1, %ymm0
+; AVX2-NEXT: ret{{[l|q]}}
+;
+; AVX512-LABEL: freeze_permps:
+; AVX512: # %bb.0:
+; AVX512-NEXT: vpmovsxbd {{.*#+}} ymm1 = [7,6,5,4,3,2,1,0]
+; AVX512-NEXT: vpermps %ymm0, %ymm1, %ymm0
+; AVX512-NEXT: vpermps %ymm0, %ymm1, %ymm0
+; AVX512-NEXT: ret{{[l|q]}}
+ %s0 = call <8 x float> @llvm.x86.avx2.permps(<8 x float> %a0, <8 x i32> <i32 7, i32 6, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0>)
+ %f0 = freeze <8 x float> %s0
+ %s1 = call <8 x float> @llvm.x86.avx2.permps(<8 x float> %f0, <8 x i32> <i32 7, i32 6, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0>)
+ ret <8 x float> %s1
+}
+
define <32 x i8> @PR27320(<8 x i32> %a0) {
; CHECK-LABEL: PR27320:
; CHECK: # %bb.0: