From 8bb12d0a77160aa38716f77f15ba6f968765686e Mon Sep 17 00:00:00 2001 From: Andrea Di Biagio Date: Tue, 20 Oct 2015 11:19:54 +0000 Subject: [x86] Fix maskload/store intrinsic definitions in avxintrin.h According to the Intel documentation, the mask operand of a maskload and maskstore intrinsics is always a vector of packed integer/long integer values. This patch introduces the following two changes: 1. It fixes the avx maskload/store intrinsic definitions in avxintrin.h. 2. It changes BuiltinsX86.def to match the correct gcc definitions for avx maskload/store (see D13861 for more details). Differential Revision: http://reviews.llvm.org/D13861 llvm-svn: 250816 --- clang/test/CodeGen/builtins-x86.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) (limited to 'clang/test/CodeGen/builtins-x86.c') diff --git a/clang/test/CodeGen/builtins-x86.c b/clang/test/CodeGen/builtins-x86.c index 0721ccb9ff3..83b11a023a2 100644 --- a/clang/test/CodeGen/builtins-x86.c +++ b/clang/test/CodeGen/builtins-x86.c @@ -465,14 +465,14 @@ void f0() { __builtin_ia32_movntdq256(tmp_V4LLip, tmp_V4LLi); __builtin_ia32_movntpd256(tmp_dp, tmp_V4d); __builtin_ia32_movntps256(tmp_fp, tmp_V8f); - tmp_V2d = __builtin_ia32_maskloadpd(tmp_V2dCp, tmp_V2d); - tmp_V4f = __builtin_ia32_maskloadps(tmp_V4fCp, tmp_V4f); - tmp_V4d = __builtin_ia32_maskloadpd256(tmp_V4dCp, tmp_V4d); - tmp_V8f = __builtin_ia32_maskloadps256(tmp_V8fCp, tmp_V8f); - __builtin_ia32_maskstorepd(tmp_V2dp, tmp_V2d, tmp_V2d); - __builtin_ia32_maskstoreps(tmp_V4fp, tmp_V4f, tmp_V4f); - __builtin_ia32_maskstorepd256(tmp_V4dp, tmp_V4d, tmp_V4d); - __builtin_ia32_maskstoreps256(tmp_V8fp, tmp_V8f, tmp_V8f); + tmp_V2d = __builtin_ia32_maskloadpd(tmp_V2dCp, tmp_V2LLi); + tmp_V4f = __builtin_ia32_maskloadps(tmp_V4fCp, tmp_V4i); + tmp_V4d = __builtin_ia32_maskloadpd256(tmp_V4dCp, tmp_V4LLi); + tmp_V8f = __builtin_ia32_maskloadps256(tmp_V8fCp, tmp_V8i); + __builtin_ia32_maskstorepd(tmp_V2dp, tmp_V2LLi, tmp_V2d); + __builtin_ia32_maskstoreps(tmp_V4fp, tmp_V4i, tmp_V4f); + __builtin_ia32_maskstorepd256(tmp_V4dp, tmp_V4LLi, tmp_V4d); + __builtin_ia32_maskstoreps256(tmp_V8fp, tmp_V8i, tmp_V8f); #ifdef USE_3DNOW tmp_V8c = __builtin_ia32_pavgusb(tmp_V8c, tmp_V8c); -- cgit v1.2.3