projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 96fe919) | patch
AMDGPU: Run SIInsertWaits as pre-emit pass
authorMatt Arsenault <Matthew.Arsenault@amd.com>
Mon, 6 Jul 2015 17:02:20 +0000 (17:02 +0000)
committerMatt Arsenault <Matthew.Arsenault@amd.com>
Mon, 6 Jul 2015 17:02:20 +0000 (17:02 +0000)
commitb560f9ca2fee8c4e060d486367cf19f3b1774921
tree35a25d116305da64833f910f9862c26e48dfc9d1tree | snapshot
parent96fe9196e953b24fa054a9aa4718d3e061c984f3commit | diff
AMDGPU: Run SIInsertWaits as pre-emit pass

Running this after the scheduler enables scheduling
waits later so other ALU instructions can run while
this would be waiting.

When combined with enabling the post-RA scheduler, this
gives about a ~20% improvement on sgemm.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@241473 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/AMDGPU/AMDGPUTargetMachine.cpp diff | blob | history
C/C++ LLVM-based compilers that forbids OOTA behaviors