Split op oneDNN AVX2 fix#33944
Merged
juncaipeng merged 3 commits intoPaddlePaddle:developfrom Jul 7, 2021
Merged
Conversation
|
Thanks for your contribution! |
arlesniak
reviewed
Jul 5, 2021
lidanqing-vv
reviewed
Jul 5, 2021
Contributor
Contributor
|
#33944 will solve this error on AVX2. But testing this model on AVX2 environment may not speedup as good as on AVX512 (6148 itself suppots AVX512). Because oneDNN improve very well on AVX512. But in AVX2, some oneDNN op will call reference version (slower). So to achieve expected performance, please use avx512. But this fix will solve error on AVX2. @juncaipeng Please approve and merge this PR |
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
PR types
Bug fixes
PR changes
OPs
Describe
Fixed bug that was occuring when AVX2 instruction set was used. In that scenario blocking format was chosen and split op doesn't support that. It is a fix for #33918