[PATCH] D29971: [AMDGPU] Revert failed scheduling

[Stanislav Mekhanoshin via Phabricator via llvm-commits]

rampitec created this revision.
Herald added subscribers: tpr, tony-tye, yaxunl, nhaehnle, wdng, kzhuravl, qcolombet.

This patch reverts region's scheduling to the original untouched state
in case if we have have decreased occupancy.

In addition it switches to use TargetRegisterInfo occupancy callback
for pressure limits instead of gradually increasing limits which were
just passed by. We are going to stay with the best schedule so we do
not need to tolerate worsened scheduling anymore.

Using this reversal mechanism we can also try different scheduling
strategies to pick a best one.


Repository:
  rL LLVM

https://reviews.llvm.org/D29971

Files:
  lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
  lib/Target/AMDGPU/GCNSchedStrategy.cpp
  lib/Target/AMDGPU/GCNSchedStrategy.h
  test/CodeGen/AMDGPU/attr-amdgpu-num-sgpr.ll
  test/CodeGen/AMDGPU/mul.ll
  test/CodeGen/AMDGPU/schedule-regpressure-limit.ll
  test/CodeGen/AMDGPU/vgpr-spill-emergency-stack-slot.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D29971.88461.patch
Type: text/x-patch
Size: 56841 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20170214/af6064d4/attachment-0001.bin>