Skip to content

Adding Tiled 2D and 3D Quantizer Linear Base Implementation #5492

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 1 commit into from

Conversation

kushrast
Copy link

Summary:
Adding Tiled Implementation of Weight-Only Quantized Linear operator

This diff adds Texture Implementation, will add Buffer impl next.

Diff Stack

  1. Add Tiled Implementation of Weight-Only Quantized Linear
  2. Add Optimized Quantized Linear Shader and code to invoke shader from Quantized Linear CPP operator
  3. [Will Not Land] Use Optimized Quantized Linear implementation

Differential Revision: D61309097

@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Sep 19, 2024
Copy link

pytorch-bot bot commented Sep 19, 2024

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/5492

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit daa1c0f with merge base ab628cc (image):
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D61309097

@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D61309097

kushrast pushed a commit to kushrast/executorch that referenced this pull request Sep 19, 2024
…5492)

Summary:
Pull Request resolved: pytorch#5492

Adding Tiled Implementation of Weight-Only Quantized Linear operator

This diff adds Texture Implementation, will add Buffer impl next.

# Diff Stack
1. Add Tiled Implementation of Weight-Only Quantized Linear
2. Add Optimized Quantized Linear Shader and code to invoke shader from Quantized Linear CPP operator
3. [Will Not Land] Use Optimized Quantized Linear implementation

Differential Revision: D61309097
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D61309097

kushrast pushed a commit to kushrast/executorch that referenced this pull request Sep 19, 2024
…5492)

Summary:
Pull Request resolved: pytorch#5492

Adding Tiled Implementation of Weight-Only Quantized Linear operator

This diff adds Texture Implementation, will add Buffer impl next.

# Diff Stack
1. Add Tiled Implementation of Weight-Only Quantized Linear
2. Add Optimized Quantized Linear Shader and code to invoke shader from Quantized Linear CPP operator
3. [Will Not Land] Use Optimized Quantized Linear implementation

Differential Revision: D61309097
kushrast pushed a commit to kushrast/executorch that referenced this pull request Oct 15, 2024
…#5492)

Summary:

Removes q_linear.h and moves implementation directly to q_8w_linear.glsl

Reviewed By: nathanaelsee

Differential Revision: D61309097
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D61309097

kushrast pushed a commit to kushrast/executorch that referenced this pull request Oct 17, 2024
…#5492)

Summary:

Removes q_linear.h and moves implementation directly to q_8w_linear.glsl

Reviewed By: nathanaelsee

Differential Revision: D61309097
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D61309097

kushrast pushed a commit to kushrast/executorch that referenced this pull request Oct 17, 2024
…#5492)

Summary:

Removes q_linear.h and moves implementation directly to q_8w_linear.glsl

Reviewed By: nathanaelsee

Differential Revision: D61309097
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D61309097

…#5492)

Summary:

Removes q_linear.h and moves implementation directly to q_8w_linear.glsl

Reviewed By: nathanaelsee

Differential Revision: D61309097
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D61309097

@facebook-github-bot
Copy link
Contributor

This pull request has been merged in 8f6c16e.

nathanaelsee added a commit to nathanaelsee/executorch that referenced this pull request Feb 25, 2025
Summary:
The function defs used to live in a separate header file before D61309097 (pytorch#5492), but were moved to the shader file directly to mirror a similar refactoring for the fp linear shader.

It's easier to reason about the shader without having to jump back and forth from the minimal main() functions to the q_8w_linear functions.

Differential Revision: D70127663
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. fb-exported Merged
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants