projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 9339f68) | patch
[AMDGPU] should expand ROTL i16 to shifts.
authoralex-t <alexander.timofeev@amd.com>
Wed, 16 Sep 2020 16:54:29 +0000 (19:54 +0300)
committeralex-t <alexander.timofeev@amd.com>
Thu, 17 Sep 2020 14:34:33 +0000 (17:34 +0300)
commit0efbb70b719e990fe153373eda5a604344ae36bb
treeaaafb22339903a2dff3f51ed4dd41951bb7a2921tree | snapshot
parent9339f68f21facc34fb0901045d571c818e1fa84acommit | diff
[AMDGPU] should expand ROTL i16 to shifts.

Instruction combining pass turns library rotl implementation to llvm.fshl.i16.
In the selection dag the intrinsic is turned to ISD::ROTL node that cannot be selected.
Need to expand it to shifts again.

Reviewed By: rampitec, arsenm

Differential Revision: https://reviews.llvm.org/D87618
llvm/lib/Target/AMDGPU/SIISelLowering.cpp diff | blob | history
llvm/test/CodeGen/AMDGPU/rotl.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/rotr.ll diff | blob | history
Domain: System / Toolchain;