Add SBGEMM for arm neoversev1 #5108
Merged
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
Signed-off-by: Ye Tao <ye.tao@arm.com>
|
@martin-frbg hi, martin, seems my patch fails on a irrelevant mips64 ci pipeline and after printing out TEST 109/111 kernel_regress:skx_avx [OK] due to exceeding maximum execution time. Could you help me on understanding whether this is expected or my changes accidentally cause performance issues? it seems some previous pr also fails after this kernel_regress:skx_avx test. |
|
Thanks - no problem with your PR, just me lacking the time and energy to confirm and merge it |
|
@martin-frbg Do we have any release cadence to upgrade the version to 0.30.0? |
|
@aditew01 0.3.30 is planned for end of the month (please see Milestones) |
nSircombe
added a commit
to nSircombe/Tool-Solutions
that referenced
this pull request
Feb 11, 2025
- Updates changelog - Removes pytorch/pytorch#139387 - Add prepacking for linear weights. Performance gains better realised by ideep reorder caching. - Updates OpenBLAS build to use recent commit from develop which includes: OpenMathLib/OpenBLAS#5108
|
@martin-frbg that sounds great! Thanks for the clarification. Missed the milestone update, but I'd keep a look out. :) |
nSircombe
added a commit
to nSircombe/Tool-Solutions
that referenced
this pull request
Feb 12, 2025
- Updates changelog - Removes pytorch/pytorch#139387 - Add prepacking for linear weights. Performance gains better realised by ideep reorder caching. - Updates OpenBLAS build to use recent commit from develop which includes: OpenMathLib/OpenBLAS#5108
# for free
to join this conversation on GitHub.
Already have an account?
# 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Outdated suggestions cannot be applied.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.
This PR is to add an optimised bf16 gemm kernel for arm neoversev1 machine (sve-256 bit).