[all-commits] [llvm/llvm-project] 28733d: [amdgpu] Change the RA to basic

Luo, Yuanke via All-commits all-commits at lists.llvm.org
Wed Aug 17 17:18:21 PDT 2022

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 28733d86cf7bf3e4e9667654ad6785aa8e21e04f
  Author: Luo, Yuanke <yuanke.luo at intel.com>
  Date:   2022-08-18 (Thu, 18 Aug 2022)

  Changed paths:
    M llvm/test/CodeGen/AMDGPU/attr-amdgpu-flat-work-group-size-vgpr-limit.ll

  Log Message:
  [amdgpu] Change the RA to basic

Specifying `-regalloc=fast` is not reliable. With fast register allocation,
`LIS = getAnalysisIfAvailable<LiveIntervals>();` get nullptr
in "si-lower-sgpr-spills" pass, so the slot index is not created in the
pass for new inserted instructions. When verifying the machine
instructions, it fails on checking slot index. While greedy-ra is time
consuming basic-ra can be used to reduce compiling time for this test case.

Differential Revision: https://reviews.llvm.org/D131931

More information about the All-commits mailing list