[all-commits] [llvm/llvm-project] c74ed8: [AMDGPU][Scheduler] Support for rematerializing SG...

[Lucas Ramirez via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: c74ed8a0d3cadf4dfe84d6c9b282823439c8ca00
      https://github.com/llvm/llvm-project/commit/c74ed8a0d3cadf4dfe84d6c9b282823439c8ca00
  Author: Lucas Ramirez <11032120+lucas-rami at users.noreply.github.com>
  Date:   2025-06-24 (Tue, 24 Jun 2025)

  Changed paths:
    M llvm/lib/Target/AMDGPU/GCNSchedStrategy.cpp
    M llvm/lib/Target/AMDGPU/GCNSchedStrategy.h
    M llvm/test/CodeGen/AMDGPU/machine-scheduler-sink-trivial-remats-attr.mir
    M llvm/test/CodeGen/AMDGPU/machine-scheduler-sink-trivial-remats.mir

  Log Message:
  -----------
  [AMDGPU][Scheduler] Support for rematerializing SGPRs and AGPRs (#140036)

This adds the ability to rematerialize SGPRs and AGPRs to the
scheduler's `PreRARematStage`, which can currently only rematerialize
ArchVGPRs. This also fixes a small potential issue in the stage where,
in case of spilling, the target occupancy could be set to a lower than
expected value when the function had either one of the "amdgpu-num-sgpr"
or "amdgpu-num-vgpr" attributes set.



To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications