[ET-VK] Improve q8 matmul by increasing TILE_N4 #14597
Merged
Conversation
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
Title says it all! I found that the latency of executing int8 matmul can be improved by increases the output tile's N4 dimension to 2. The improvement is about 20-25% on Samsung Galaxy S25. Differential Revision: [D83253129](https://our.internmc.facebook.com/intern/diff/D83253129/) [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/14597
Note: Links to docs will display an error until the docs builds have been completed. ❌ 3 New Failures, 1 Cancelled Job, 4 Unrelated FailuresAs of commit 48fa7b0 with merge base c18abc8 ( NEW FAILURES - The following jobs have failed:
BROKEN TRUNK - The following jobs failed but were present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
bot
added
the
CLA Signed
digantdesai
approved these changes
Sep 25, 2025
This PR needs a
|
facebook-github-bot
merged commit 1a28dbb
into
gh/SS-JIA/331/base
126 of 143 checks passed
facebook-github-bot
temporarily deployed
to
cherry-pick-bot
GitHub Actions
Inactive
SS-JIA
pushed a commit
that referenced
this pull request
Sep 25, 2025
Title says it all! I found that the latency of executing int8 matmul can be improved by increases the output tile's N4 dimension to 2. The improvement is about 20-25% on Samsung Galaxy S25. Differential Revision: [D83253129](https://our.internmc.facebook.com/intern/diff/D83253129/) ghstack-source-id: 312106549 Pull Request resolved: #14597
SS-JIA
pushed a commit
that referenced
this pull request
Sep 25, 2025
This PR was created by the merge bot to help merge the original PR into the main branch. ghstack PR number: #14597 by @SS-JIA ^ Please use this as the source of truth for the PR details, comments, and reviews ghstack PR base: https://github.com/pytorch/executorch/tree/gh/SS-JIA/331/base ghstack PR head: https://github.com/pytorch/executorch/tree/gh/SS-JIA/331/head Merge bot PR base: https://github.com/pytorch/executorch/tree/gh/SS-JIA/329/orig Merge bot PR head: https://github.com/pytorch/executorch/tree/gh/SS-JIA/331/orig Differential Revision: [D83253129](https://our.internmc.facebook.com/intern/diff/D83253129/) @diff-train-skip-merge Co-authored-by: ssjia <ssjia@devvm1479.ncg0.facebook.com>
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.
Stack from ghstack (oldest at bottom):
kInt8x4dtype andGPUMemoryLayouts for packed quantized tensors #14329Title says it all! I found that the latency of executing int8 matmul can be improved by increases the output tile's N4 dimension to 2. The improvement is about 20-25% on Samsung Galaxy S25.
Differential Revision: D83253129