[LLVMGPU] Use flat workgroup sizes in vector distribution #18947
+26
−20
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Groverkss
requested review from
MaheshRavishankar,
qedawkins and
kuhar
as code owners
qedawkins
approved these changes
Oct 30, 2024
MaheshRavishankar
approved these changes
Oct 30, 2024
|
Just adding info from offline chat with Kunwar, we can get away with not explicitly delinearized workgroup since vector distribution already does some delinearization. |
raikonenfnu
approved these changes
Oct 30, 2024
kuhar
reviewed
Oct 30, 2024
There was a problem hiding this comment.
Regardless of whether this should be the default heuristic or not, we should either keep some tests making sure non-flat workgroup sizes continue working or add a verifier to disallow it.
@Groverkss could you please submit a follow up PR for this?
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There is no reason for vector distribution to use this weird workgroup split. Use flat workgroup sizes in KernelConfig instead.