Convert vperm2f128 and vperm2i128 intrinsics back to using llvm intrinsics. Unfortunately, these instructions have behavior that can't be modeled with shuffle vector.
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@154906 91177308-0d34-0410-b5e6-96231b3b80d8
diff --git a/test/CodeGen/builtins-x86.c b/test/CodeGen/builtins-x86.c
index 30138d6..acb5554 100644
--- a/test/CodeGen/builtins-x86.c
+++ b/test/CodeGen/builtins-x86.c
@@ -414,6 +414,9 @@
tmp_V4i = __builtin_ia32_cvttpd2dq256(tmp_V4d);
tmp_V4i = __builtin_ia32_cvtpd2dq256(tmp_V4d);
tmp_V8i = __builtin_ia32_cvttps2dq256(tmp_V8f);
+ tmp_V4d = __builtin_ia32_vperm2f128_pd256(tmp_V4d, tmp_V4d, 0x7);
+ tmp_V8f = __builtin_ia32_vperm2f128_ps256(tmp_V8f, tmp_V8f, 0x7);
+ tmp_V8i = __builtin_ia32_vperm2f128_si256(tmp_V8i, tmp_V8i, 0x7);
tmp_V4d = __builtin_ia32_vinsertf128_pd256(tmp_V4d, tmp_V2d, 0x7);
tmp_V8f = __builtin_ia32_vinsertf128_ps256(tmp_V8f, tmp_V4f, 0x7);
tmp_V8i = __builtin_ia32_vinsertf128_si256(tmp_V8i, tmp_V4i, 0x7);