[ET-VK] Changing all conv 2d pw ints from uint16 to int since it slightly improves perf. #7545
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
…htly improves perf. This diff changes all integers in conv 2d pw op shader from uint16 to int in the Vulkan backend of Executorch. The change is made to improve performance since the shader does not appear to be register bound. Differential Revision: [D67906023](https://our.internmc.facebook.com/intern/diff/D67906023/) [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/7545
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 0cfd4a5 with merge base 39e8538 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
trivedivivek
added a commit
that referenced
this pull request
Jan 7, 2025
…htly improves perf. This diff changes all integers in conv 2d pw op shader from uint16 to int in the Vulkan backend of Executorch. The change is made to improve performance since the shader does not appear to be register bound. Differential Revision: [D67906023](https://our.internmc.facebook.com/intern/diff/D67906023/) ghstack-source-id: 260513229 Pull Request resolved: #7545
|
This pull request was exported from Phabricator. Differential Revision: D67906023 |
…nce it slightly improves perf." This diff changes all integers in conv 2d pw op shader from uint16 to int in the Vulkan backend of Executorch. The change is made to improve performance since the shader does not appear to be register bound. Differential Revision: [D67906023](https://our.internmc.facebook.com/intern/diff/D67906023/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D67906023 |
trivedivivek
added a commit
that referenced
this pull request
Jan 8, 2025
…htly improves perf. Pull Request resolved: #7545 This diff changes all integers in conv 2d pw op shader from uint16 to int in the Vulkan backend of Executorch. The change is made to improve performance since the shader does not appear to be register bound. Differential Revision: [D67906023](https://our.internmc.facebook.com/intern/diff/D67906023/) ghstack-source-id: 260637199
jorgep31415
approved these changes
Jan 8, 2025
…nce it slightly improves perf." This diff changes all integers in conv 2d pw op shader from uint16 to int in the Vulkan backend of Executorch. The change is made to improve performance since the shader does not appear to be register bound. Differential Revision: [D67906023](https://our.internmc.facebook.com/intern/diff/D67906023/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D67906023 |
trivedivivek
added a commit
that referenced
this pull request
Jan 9, 2025
…htly improves perf. Pull Request resolved: #7545 This diff changes all integers in conv 2d pw op shader from uint16 to int in the Vulkan backend of Executorch. The change is made to improve performance since the shader does not appear to be register bound. ghstack-source-id: 260707857 Differential Revision: [D67906023](https://our.internmc.facebook.com/intern/diff/D67906023/)
facebook-github-bot
temporarily deployed
to
cherry-pick-bot
GitHub Actions
Inactive
|
This pull request was exported from Phabricator. Differential Revision: D67906023 |
trivedivivek
added a commit
that referenced
this pull request
Jan 9, 2025
…htly improves perf. (#7566) * [ET-VK] Adding a common utility function to calculate 3d output position based on unique index. Pull Request resolved: #7522 This diff adds an indexing utils header file used in Vulkan backend of Executorch. The header file includes functions for converting a global index to u16 indices based on input sizes. ghstack-source-id: 260707858 @exported-using-ghexport Differential Revision: [D67821941](https://our.internmc.facebook.com/intern/diff/D67821941/) * [ET-VK] Adding batch processing in x axis to conv2d dw shader by caching input texel for reuse. Pull Request resolved: #7526 This diff adds batch processing in the x axis to the conv2d dw shader by reusing input texel overlapping between consecutive tiles. The changes include modifying the glsl code for the conv2d dw output tile, adding a new parameter to the yaml file, and modifying the Convolution.cpp file to use the new parameter. ghstack-source-id: 260707856 Differential Revision: [D67868671](https://our.internmc.facebook.com/intern/diff/D67868671/) * [ET-VK] Changing all conv 2d pw ints from uint16 to int since it slightly improves perf. Pull Request resolved: #7545 This diff changes all integers in conv 2d pw op shader from uint16 to int in the Vulkan backend of Executorch. The change is made to improve performance since the shader does not appear to be register bound. ghstack-source-id: 260707857 Differential Revision: [D67906023](https://our.internmc.facebook.com/intern/diff/D67906023/) --------- Co-authored-by: Vivek Trivedi <[email protected]>
Merged
YIWENX14
pushed a commit
that referenced
this pull request
Jan 28, 2025
…htly improves perf. (#7566) * [ET-VK] Adding a common utility function to calculate 3d output position based on unique index. Pull Request resolved: #7522 This diff adds an indexing utils header file used in Vulkan backend of Executorch. The header file includes functions for converting a global index to u16 indices based on input sizes. ghstack-source-id: 260707858 @exported-using-ghexport Differential Revision: [D67821941](https://our.internmc.facebook.com/intern/diff/D67821941/) * [ET-VK] Adding batch processing in x axis to conv2d dw shader by caching input texel for reuse. Pull Request resolved: #7526 This diff adds batch processing in the x axis to the conv2d dw shader by reusing input texel overlapping between consecutive tiles. The changes include modifying the glsl code for the conv2d dw output tile, adding a new parameter to the yaml file, and modifying the Convolution.cpp file to use the new parameter. ghstack-source-id: 260707856 Differential Revision: [D67868671](https://our.internmc.facebook.com/intern/diff/D67868671/) * [ET-VK] Changing all conv 2d pw ints from uint16 to int since it slightly improves perf. Pull Request resolved: #7545 This diff changes all integers in conv 2d pw op shader from uint16 to int in the Vulkan backend of Executorch. The change is made to improve performance since the shader does not appear to be register bound. ghstack-source-id: 260707857 Differential Revision: [D67906023](https://our.internmc.facebook.com/intern/diff/D67906023/) --------- Co-authored-by: Vivek Trivedi <[email protected]>
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):
This diff changes all integers in conv 2d pw op shader from uint16 to int in the Vulkan backend of Executorch. The change is made to improve performance since the shader does not appear to be register bound.
Differential Revision: D67906023