[all-commits] [llvm/llvm-project] acb785: [MachineSink] Extend loop sinking capability (#117...

[Jeffrey Byrnes via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: acb7859f075f91b1105c04c37c6aa85db27a898a
      https://github.com/llvm/llvm-project/commit/acb7859f075f91b1105c04c37c6aa85db27a898a
  Author: Jeffrey Byrnes <jeffrey.byrnes at amd.com>
  Date:   2025-01-23 (Thu, 23 Jan 2025)

  Changed paths:
    M llvm/lib/CodeGen/MachineSink.cpp
    A llvm/test/CodeGen/AMDGPU/machine-sink-cycle.mir
    M llvm/test/CodeGen/AMDGPU/machine-sink-ignorable-exec-use.mir
    M llvm/test/CodeGen/AMDGPU/machine-sink-lane-mask.mir
    M llvm/test/CodeGen/SystemZ/machinelicm-sunk-kill-flags.mir

  Log Message:
  -----------
  [MachineSink] Extend loop sinking capability (#117247)

The current MIR cycle sinking capabilities are rather limited. It only
support sinking copies into a single successor block while obeying
limits.

This opt-in feature adds a more aggressive option, that is not limited
to the above concerns. The feature will try to "sink" by duplicating any
top-level preheader instruction (that we are sure is safe to sink) into
any user block, then does some dead code cleanup. In particular, this is
useful for high RP situations when loop bodies have control flow.



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