[ET-VK] Creating specialized version of conv2d pw shader for X and Y stride = 1 and padding = 0.#11137
Conversation
…stride = 1 and padding = 0. This diff creates a specialized version of the conv2d pw shader for X and Y stride equals 1 and padding equals 0. * It adds a new file `conv2d_pw_s1p0.glsl`, which implements the conv2d pw shader for X and Y stride equals 1 and padding equals 0. * It adds a new file `conv2d_pw_s1p0.yaml`, which defines the parameters and shader variants for the specialized conv2d pw shader. * The file `Convolution.cpp` is modified to add a new parameter `stride_1_padding_0` to the `conv2d` function, which enables the use of the specialized shader. Differential Revision: [D75423931](https://our.internmc.facebook.com/intern/diff/D75423931/) [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/11137
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (1 Unrelated Failure)As of commit 9dacfc0 with merge base 380eb5f ( BROKEN TRUNK - The following job 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. |
|
This pull request was exported from Phabricator. Differential Revision: D75423931 |
…or X and Y stride = 1 and padding = 0." This diff creates a specialized version of the conv2d pw shader for X and Y stride equals 1 and padding equals 0. * It adds a new file `conv2d_pw_s1p0.glsl`, which implements the conv2d pw shader for X and Y stride equals 1 and padding equals 0. * It adds a new file `conv2d_pw_s1p0.yaml`, which defines the parameters and shader variants for the specialized conv2d pw shader. * The file `Convolution.cpp` is modified to add a new parameter `stride_1_padding_0` to the `conv2d` function, which enables the use of the specialized shader. Differential Revision: [D75423931](https://our.internmc.facebook.com/intern/diff/D75423931/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D75423931 |
…or X and Y stride = 1 and padding = 0." This diff creates a specialized version of the conv2d pw shader for X and Y stride equals 1 and padding equals 0. * It adds a new file `conv2d_pw_s1p0.glsl`, which implements the conv2d pw shader for X and Y stride equals 1 and padding equals 0. * It adds a new file `conv2d_pw_s1p0.yaml`, which defines the parameters and shader variants for the specialized conv2d pw shader. * The file `Convolution.cpp` is modified to add a new parameter `stride_1_padding_0` to the `conv2d` function, which enables the use of the specialized shader. Differential Revision: [D75423931](https://our.internmc.facebook.com/intern/diff/D75423931/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D75423931 |
…or X and Y stride = 1 and padding = 0." This diff creates a specialized version of the conv2d pw shader for X and Y stride equals 1 and padding equals 0. * It adds a new file `conv2d_pw_s1p0.glsl`, which implements the conv2d pw shader for X and Y stride equals 1 and padding equals 0. * It adds a new file `conv2d_pw_s1p0.yaml`, which defines the parameters and shader variants for the specialized conv2d pw shader. * The file `Convolution.cpp` is modified to add a new parameter `stride_1_padding_0` to the `conv2d` function, which enables the use of the specialized shader. Differential Revision: [D75423931](https://our.internmc.facebook.com/intern/diff/D75423931/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D75423931 |
…or X and Y stride = 1 and padding = 0." This diff creates a specialized version of the conv2d pw shader for X and Y stride equals 1 and padding equals 0. * It adds a new file `conv2d_pw_s1p0.glsl`, which implements the conv2d pw shader for X and Y stride equals 1 and padding equals 0. * It adds a new file `conv2d_pw_s1p0.yaml`, which defines the parameters and shader variants for the specialized conv2d pw shader. * The file `Convolution.cpp` is modified to add a new parameter `stride_1_padding_0` to the `conv2d` function, which enables the use of the specialized shader. Differential Revision: [D75423931](https://our.internmc.facebook.com/intern/diff/D75423931/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D75423931 |
…or X and Y stride = 1 and padding = 0." This diff creates a specialized version of the conv2d pw shader for X and Y stride equals 1 and padding equals 0. * It adds a new file `conv2d_pw_s1p0.glsl`, which implements the conv2d pw shader for X and Y stride equals 1 and padding equals 0. * It adds a new file `conv2d_pw_s1p0.yaml`, which defines the parameters and shader variants for the specialized conv2d pw shader. * The file `Convolution.cpp` is modified to add a new parameter `stride_1_padding_0` to the `conv2d` function, which enables the use of the specialized shader. Differential Revision: [D75423931](https://our.internmc.facebook.com/intern/diff/D75423931/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D75423931 |
73732ac
into
gh/trivedivivek/97/base
…stride = 1 and padding = 0. (#11190) This PR was created by the merge bot to help merge the original PR into the main branch. ghstack PR number: #11137 by @trivedivivek ^ 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/trivedivivek/97/base ghstack PR head: https://github.com/pytorch/executorch/tree/gh/trivedivivek/97/head Merge bot PR base: https://github.com/pytorch/executorch/tree/gh/trivedivivek/96/orig Merge bot PR head: https://github.com/pytorch/executorch/tree/gh/trivedivivek/97/orig @diff-train-skip-merge --------- Co-authored-by: Vivek Trivedi <5340687+trivedivivek@users.noreply.github.com>
Stack from ghstack (oldest at bottom):
This diff creates a specialized version of the conv2d pw shader for X and Y stride equals 1 and padding equals 0.
conv2d_pw_s1p0.glsl, which implements the conv2d pw shader for X and Y stride equals 1 and padding equals 0.conv2d_pw_s1p0.yaml, which defines the parameters and shader variants for the specialized conv2d pw shader.Convolution.cppis modified to add a new parameterstride_1_padding_0to theconv2dfunction, which enables the use of the specialized shader.Differential Revision: D75423931