projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 8781c42) | patch
[X86][AVX2] Passing the appropriate memory operand class to VPMADDWD instruction.
authorAyman Musa <ayman.musa@intel.com>
Thu, 22 Dec 2016 08:42:46 +0000 (08:42 +0000)
committerAyman Musa <ayman.musa@intel.com>
Thu, 22 Dec 2016 08:42:46 +0000 (08:42 +0000)
commit9ff608cdc64815ae337e06785e5c69781a355221
tree1c5e9d896adf84dc2bd20d8beca41424a9b09818tree | snapshot
parent8781c425b379e0da70de6b5ae5bf9abcdc822dafcommit | diff
[X86][AVX2] Passing the appropriate memory operand class to VPMADDWD instruction.

Replacing the memory operand in the ymm version of VPMADDWD from i128mem to i256mem.

Differential Revision: https://reviews.llvm.org/D28024

llvm-svn: 290333
llvm/lib/Target/X86/X86InstrSSE.td diff | blob | history
llvm/test/MC/X86/intel-syntax-x86-64-avx.s diff | blob | history
llvm/test/MC/X86/x86_64-avx-encoding.s diff | blob | history
Domain: System / Toolchain;