[ET-VK] Using shared memory to save position in conv2d dw output op.#7923
Conversation
This diff introduces a change to conv2d dw op to save output positions in shared memory, which reduces register usage and improves performance. Differential Revision: [D68400890](https://our.internmc.facebook.com/intern/diff/D68400890/) [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/7923
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 8ffe881 with merge base 7bc06d1 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
This pull request was exported from Phabricator. Differential Revision: D68400890 |
This diff introduces a change to conv2d dw op to save output positions in shared memory, which reduces register usage and improves performance. Differential Revision: [D68400890](https://our.internmc.facebook.com/intern/diff/D68400890/) ghstack-source-id: 262823001 Pull Request resolved: #7923
…output op." This diff introduces a change to conv2d dw op to save output positions in shared memory, which reduces register usage and improves performance. Differential Revision: [D68400890](https://our.internmc.facebook.com/intern/diff/D68400890/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D68400890 |
…output op." This diff introduces a change to conv2d dw op to save output positions in shared memory, which reduces register usage and improves performance. Differential Revision: [D68400890](https://our.internmc.facebook.com/intern/diff/D68400890/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D68400890 |
…output op." This diff introduces a change to conv2d dw op to save output positions in shared memory, which reduces register usage and improves performance. Differential Revision: [D68400890](https://our.internmc.facebook.com/intern/diff/D68400890/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D68400890 |
…output op." This diff introduces a change to conv2d dw op to save output positions in shared memory, which reduces register usage and improves performance. Differential Revision: [D68400890](https://our.internmc.facebook.com/intern/diff/D68400890/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D68400890 |
…output op." This diff introduces a change to conv2d dw op to save output positions in shared memory, which reduces register usage and improves performance. Differential Revision: [D68400890](https://our.internmc.facebook.com/intern/diff/D68400890/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D68400890 |
…output op." This diff introduces a change to conv2d dw op to save output positions in shared memory, which reduces register usage and improves performance. Differential Revision: [D68400890](https://our.internmc.facebook.com/intern/diff/D68400890/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D68400890 |
…output op." This diff introduces a change to conv2d dw op to save output positions in shared memory, which reduces register usage and improves performance. Differential Revision: [D68400890](https://our.internmc.facebook.com/intern/diff/D68400890/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D68400890 |
53d1b0d
into
gh/trivedivivek/51/base
…8007) Pull Request resolved: #7923 This diff introduces a change to conv2d dw op to save output positions in shared memory, which reduces register usage and improves performance. ghstack-source-id: 263440666 @exported-using-ghexport Differential Revision: [D68400890](https://our.internmc.facebook.com/intern/diff/D68400890/) Co-authored-by: Vivek Trivedi <5340687+trivedivivek@users.noreply.github.com>
* [ET-VK] Using shared memory to save position in conv2d dw output op. Pull Request resolved: #7923 This diff introduces a change to conv2d dw op to save output positions in shared memory, which reduces register usage and improves performance. ghstack-source-id: 263440666 @exported-using-ghexport Differential Revision: [D68400890](https://our.internmc.facebook.com/intern/diff/D68400890/) * [ET-VK] Using push constants for conv2d dw. Pull Request resolved: #7928 This diff is related to the use of push constants for convolutional dw (depthwise) in Executorch's Vulkan backend. This optimization improves memory usage. ghstack-source-id: 263440665 @exported-using-ghexport Differential Revision: [D68493849](https://our.internmc.facebook.com/intern/diff/D68493849/) * [ET-VK] Using TmpTensor for width packed versions of q_linear op shader to reduce memory usage. (#8009) Pull Request resolved: #7929 This diff introduces the use of temporary tensors to reduce memory usage in the width packed versions of the q_linear op shader. ghstack-source-id: 263456691 @exported-using-ghexport Differential Revision: [D68561647](https://our.internmc.facebook.com/intern/diff/D68561647/) Co-authored-by: Vivek Trivedi <5340687+trivedivivek@users.noreply.github.com> --------- Co-authored-by: Vivek Trivedi <5340687+trivedivivek@users.noreply.github.com>
Stack from ghstack (oldest at bottom):
This diff introduces a change to conv2d dw op to save output positions in shared memory, which reduces register usage and improves performance.
Differential Revision: D68400890