[PATCH] D115747: [AMDGPU] Hoist waitcnt out of loops when they unecessarily wait for stores

[Jay Foad via Phabricator via llvm-commits]

foad added a comment.

Can't we always set the state at the start of the loop to the merged state from loop backedges *only*, and then have a final pass that inserts extra waits in preheaders to reconcile the state outside the loop with the state that was assumed for the start of the loop? (Sorry I know that is a bit simplistic and hand-wavy.)


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D115747/new/

https://reviews.llvm.org/D115747