[PATCH] D86253: [AMDGPU] Avoid sorting stalls in regbank-reassign

[Stanislav Mekhanoshin via Phabricator via llvm-commits]

rampitec updated this revision to Diff 286917.
rampitec retitled this revision from "[AMDGPU] Disable stall weight calculation in regbank-reassign" to "[AMDGPU] Avoid sorting stalls in regbank-reassign".
rampitec edited the summary of this revision.
rampitec added a comment.

In fact the slowest part was sorting. I just have changed data structure to avoid sorting. Weight calculation itself turns to be not that much expensive.

This almost halves time the pass takes.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D86253/new/

https://reviews.llvm.org/D86253

Files:
  llvm/lib/Target/AMDGPU/GCNRegBankReassign.cpp
  llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.image.sample.g16.ll
  llvm/test/CodeGen/AMDGPU/GlobalISel/uaddsat.ll
  llvm/test/CodeGen/AMDGPU/atomic_optimizations_local_pointer.ll
  llvm/test/CodeGen/AMDGPU/llvm.amdgcn.image.sample.a16.dim.ll
  llvm/test/CodeGen/AMDGPU/llvm.amdgcn.image.sample.g16.encode.ll
  llvm/test/CodeGen/AMDGPU/llvm.amdgcn.image.sample.g16.ll
  llvm/test/CodeGen/AMDGPU/regbank-reassign.mir

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D86253.286917.patch
Type: text/x-patch
Size: 38053 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20200820/08a5bfa5/attachment-0001.bin>