projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 0441272) | patch
[X86] Don't fold select for vXi1 on X86 target.
authorLuo, Yuanke <yuanke.luo@intel.com>
Mon, 16 Jan 2023 14:39:06 +0000 (22:39 +0800)
committerLuo, Yuanke <yuanke.luo@intel.com>
Tue, 17 Jan 2023 00:00:41 +0000 (08:00 +0800)
commit159cac50cfc12f8951e80d640909ff08364e65de
tree5e7a644827b2c46180704bc7b081ae3ed9f5a9b4tree | snapshot
parent0441272c4501a72b86cd5c0b8ab96802d1abb8d9commit | diff
[X86] Don't fold select for vXi1 on X86 target.

Since there is no mask instruction for vXi1 with avx512f in X86 target.
Folding select for vXi1 doesn't help to reduce instructions.

Differential Revision: https://reviews.llvm.org/D141782
llvm/lib/Target/X86/X86ISelLowering.cpp diff | blob | history
llvm/test/CodeGen/X86/fold-select.ll diff | blob | history
Domain: System / Toolchain;