projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: ee75cf3) | patch
[AMDGPU] Use fma for lowering frem
authorJay Foad <jay.foad@amd.com>
Fri, 24 Jul 2020 10:41:57 +0000 (11:41 +0100)
committerJay Foad <jay.foad@amd.com>
Tue, 4 Aug 2020 15:18:23 +0000 (16:18 +0100)
commit8ec8ad868d9b970245e827b14306fbd11d11a9b2
treed0141b984cb8d582e163f52d1703ac6f1365193dtree | snapshot
parentee75cf36bb1790a51cd1fd7c022b0ece101eb248commit | diff
[AMDGPU] Use fma for lowering frem

This gives shorter f64 code and perhaps better accuracy.

Differential Revision: https://reviews.llvm.org/D84516
llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp diff | blob | history
llvm/test/CodeGen/AMDGPU/frem.ll diff | blob | history
Domain: System / Toolchain;