[PATCH] D20620: [Clang][AVX512][Builtin] Fix palignr intrinsics header
michael zuckerman via cfe-commits
cfe-commits at lists.llvm.org
Wed May 25 06:50:57 PDT 2016
m_zuckerman created this revision.
m_zuckerman added reviewers: AsafBadouh, igorb, delena.
m_zuckerman added a subscriber: cfe-commits.
We don't need to multiply by eight the IMM. The instruction is doing that.
http://reviews.llvm.org/D20620
Files:
lib/Headers/avx512bwintrin.h
Index: lib/Headers/avx512bwintrin.h
===================================================================
--- lib/Headers/avx512bwintrin.h
+++ lib/Headers/avx512bwintrin.h
@@ -2145,19 +2145,19 @@
#define _mm512_alignr_epi8(A, B, N) __extension__ ({\
(__m512i)__builtin_ia32_palignr512_mask((__v64qi)(__m512i)(A), \
- (__v64qi)(__m512i)(B), (int)(N) * 8, \
+ (__v64qi)(__m512i)(B), (int)(N), \
(__v64qi)_mm512_undefined_pd(), \
(__mmask64)-1); })
#define _mm512_mask_alignr_epi8(W, U, A, B, N) __extension__({\
(__m512i)__builtin_ia32_palignr512_mask((__v64qi)(__m512i)(A), \
- (__v64qi)(__m512i)(B), (int)(N) * 8, \
+ (__v64qi)(__m512i)(B), (int)(N), \
(__v64qi)(__m512i)(W), \
(__mmask64)(U)); })
#define _mm512_maskz_alignr_epi8(U, A, B, N) __extension__({\
(__m512i)__builtin_ia32_palignr512_mask((__v64qi)(__m512i)(A), \
- (__v64qi)(__m512i)(B), (int)(N) * 8, \
+ (__v64qi)(__m512i)(B), (int)(N), \
(__v64qi)_mm512_setzero_si512(), \
(__mmask64)(U)); })
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D20620.58406.patch
Type: text/x-patch
Size: 1480 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20160525/608bae5b/attachment.bin>
More information about the cfe-commits
mailing list