projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 5a3cf5e) | patch
[X86][FP16] Do not combine fminnum/fmaxnum for FP16 emulation
authorPhoebe Wang <phoebe.wang@intel.com>
Thu, 1 Dec 2022 14:31:51 +0000 (22:31 +0800)
committerPhoebe Wang <phoebe.wang@intel.com>
Thu, 1 Dec 2022 15:24:40 +0000 (23:24 +0800)
commit54ebf1c4a19c7e6556bf193c42250d65120847de
treee5a1600fb493986e8999fc1f51ce3c9d67c353d8tree | snapshot
parent5a3cf5e1a458e0256f57aa1d529df627874ffebacommit | diff
[X86][FP16] Do not combine fminnum/fmaxnum for FP16 emulation

Under the emulation situation, we lack native fmin/fmax instruction support.

Fixes #59258

Reviewed By: skan, spatel

Differential Revision: https://reviews.llvm.org/D139078
llvm/lib/Target/X86/X86ISelLowering.cpp diff | blob | history
llvm/test/CodeGen/X86/pr59258.ll [new file with mode: 0644] blob
Domain: System / Toolchain;