[PATCH] D159452: [AMDGPU] Machine scheduler should take care of the 'amdgpu-num-vgp' function attribute

[Alexander via Phabricator via llvm-commits]

alex-t created this revision.
alex-t added reviewers: jrbyrnes, kerbowa.
Herald added subscribers: foad, hiraditya, tpr, dstuttard, yaxunl, jvesely, kzhuravl, arsenm.
Herald added a project: All.
alex-t requested review of this revision.
Herald added a subscriber: wdng.
Herald added a project: LLVM.

We currently have the 'amdgpu-num-vgpr' attribute parsed during the TargetRegisterInfo initialization.

  GCNSchedStage class method computing occupancy does not read the attribute and consider all the available
   registers for the given target even if the number of VGPRs was explicitly restricted.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D159452

Files:
  llvm/lib/Target/AMDGPU/AMDGPUSubtarget.cpp
  llvm/lib/Target/AMDGPU/GCNSchedStrategy.cpp
  llvm/lib/Target/AMDGPU/GCNSubtarget.h
  llvm/test/CodeGen/AMDGPU/agpr-remat.ll
  llvm/test/CodeGen/AMDGPU/preserve-wwm-copy-dst-reg.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D159452.555886.patch
Type: text/x-patch
Size: 7657 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20230905/d7cd3762/attachment.bin>