[PATCH] D149393: [AMDGPU][IGLP] Parameterize the SchedGroup processing / linking in Solver

[Jeffrey Byrnes via Phabricator via llvm-commits]

jrbyrnes created this revision.
jrbyrnes added a reviewer: kerbowa.
Herald added subscribers: kosarev, foad, hiraditya, tpr, dstuttard, yaxunl, jvesely, kzhuravl, arsenm.
Herald added a project: All.
jrbyrnes requested review of this revision.
Herald added subscribers: llvm-commits, wdng.
Herald added a project: LLVM.

Currently the PipelineSolver processes SchedGroups in BOTTOM_UP manner. However, there is no compelling reason to require this. Providing the option for TOP_DOWN affords greater experimentation capability, and make usage a bit more intuitive. Importantly, it makes designing rules much easier.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D149393

Files:
  llvm/lib/Target/AMDGPU/AMDGPUIGroupLP.cpp
  llvm/test/CodeGen/AMDGPU/llvm.amdgcn.iglp.opt.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D149393.517750.patch
Type: text/x-patch
Size: 18580 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20230427/414376e6/attachment.bin>