[PATCH] D105839: [AMDGPU] Improve register computation for indirect calls

[Sebastian Neubauer via Phabricator via llvm-commits]

sebastian-ne created this revision.
sebastian-ne added reviewers: arsenm, foad, madhur13490.
Herald added subscribers: kerbowa, hiraditya, t-tye, tpr, dstuttard, yaxunl, mgorny, nhaehnle, jvesely, kzhuravl.
sebastian-ne requested review of this revision.
Herald added subscribers: llvm-commits, wdng.
Herald added a project: LLVM.

First, collect the register usage in each function, then apply the
maximum register usage of all functions to functions with indirect
calls.

This is more accurate than guessing the maximum register usage without
looking at the actual usage.

As before, assume that indirect calls will hit a function in the
current module.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D105839

Files:
  llvm/lib/Target/AMDGPU/AMDGPU.h
  llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp
  llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.h
  llvm/lib/Target/AMDGPU/AMDGPUResourceUsageAnalysis.cpp
  llvm/lib/Target/AMDGPU/AMDGPUResourceUsageAnalysis.h
  llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
  llvm/lib/Target/AMDGPU/CMakeLists.txt
  llvm/test/CodeGen/AMDGPU/agpr-register-count.ll
  llvm/test/CodeGen/AMDGPU/amdpal-callable.ll
  llvm/test/CodeGen/AMDGPU/call-graph-register-usage.ll
  llvm/test/CodeGen/AMDGPU/indirect-call.ll
  llvm/test/CodeGen/AMDGPU/llc-pipeline.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D105839.358044.patch
Type: text/x-patch
Size: 56946 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20210712/83864ada/attachment.bin>