#15824 Workaround LLK issue in max_pool #16849
Open
+1
−2
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
ncvetkovicTT
approved these changes
Jan 17, 2025
There was a problem hiding this comment.
It's a valid workaround until we find out why block_ct_dim=1 doesn't work in llk_pack_untilize. Thanks @pavlejosipovic
pack_untilize_dst is causing issues on GS and WH in case where reduce op is called after max_pool. Workaround is to use default value for block_ct_dim in pack_untilize_dst_init_short in max_pool kernel. This doesn't affect correctness of max_pool op and resolves the issue on subsequent reduce op. Perf looks good on model device perf tests.
pavlejosipovic
force-pushed
the
pjosipovic/workaround_for_16847
branch
from
70c5c8d to
63e574d
Compare
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Outdated suggestions cannot be applied.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.
pack_untilize_dst is causing issues on GS and WH
in case where reduce op is called after max_pool.
Workaround is to use default value for block_ct_dim in pack_untilize_dst_init_short in max_pool kernel.
This doesn't affect correctness of max_pool op
and resolves the issue on subsequent reduce op.
Perf looks good on model device perf tests.
Ticket
Link to Github Issue
This issue should still be properly solved on LLK level, but given it's P0 in metal this should serve as a workaround until it does.
Checklist