[PATCH] D101430: [AMDGPU] Refactor hazard recognition IsHazardFn and IsExpiredFn

Wed Apr 28 01:21:39 PDT 2021

foad added a comment.

Copying some of the prior discussion over from D56923 <https://reviews.llvm.org/D56923>:

@foad wrote:

> I think this early return [the one being removed in this patch] is broken, because there might be another predecessor that has a smaller value of MinWaitStates which would not satisfy the IsExpired test. Do you agree? (Sorry for reopening such an old review.)

@critson wrote:

> I do not think it is broken per-say but definitely confusing.
> This allows the IsExpired function to trigger an early exit given a specific waitstate count.
> The early-exit probe is detectable as it occurs with the MachineInstr* set to null.
> And hence avoidable by return false if !MI.
>
> Looking at all the IsExpired implementations, none of them use this functionality (or at least not correctly).
> So I'll prepare a patch to remove it and tidy up.

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D101430/new/

https://reviews.llvm.org/D101430