[PATCH] D72737: [AMDGPU] Bundle loads before pre-RA scheduler

[Austin Kerbow via Phabricator via llvm-commits]

kerbowa added a comment.

Like you say the machine scheduler can choose not to prioritize cluster edges, so this is much more restrictive. We should be //really //sure we always want this clustering. There are probably cases where the cost/benefit is not in favor of bundling, but it's not easy to predict.

By the way I think we actually do try to insert waitcnt into bundles. I wonder if it would be worth it to just update the memory legalizer to work with bundles as well.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D72737/new/

https://reviews.llvm.org/D72737