Speedup bf16 gemm fallback on ARM #126592
Closed
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
By dispatching it to multiple threads and using vectorized dot op This bumps stories110M eval from
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/126592
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 358354d with merge base 93844a3 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
pytorch-bot
bot
added
the
module: cpu
malfet
added
release notes: linalg_frontend
malfet
added
ciflow/trunk
|
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
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Outdated suggestions cannot be applied.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.
By dispatching it to multiple threads and using vectorized dot operation (with fp16 to fp32 upcasts via left shift)
This bumps stories110M eval from 22 to 55 tokens/sec using bfloat16
TODO:
cc @jgong5 @mingfeima @XiaobingSuper @sanchitintel @ashokei @jingxu10