x64: matmul: Fix buffer B chunk size / Buffer B per thread size #3161
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Outdated suggestions cannot be applied.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.
Description
It partially fixes MFDNN-13311.
buffer_b_chunk_sz
should be aligned withchar *get_buf_B_ptr(int ithr, int k_blk_idx, int n_blk_idx, int gb)
logic implemented here:https://github.com/uxlfoundation/oneDNN/blob/main/src/cpu/x64/matmul/brgemm_matmul.cpp#L1688
The issue accidentally observed on the big number of threads (>193).
The PR only fixes --wtag=abcd cases. --wtag=abdc which is 31 cases of original 72 still fail