Rearranging calculations in quantized linear tiled implementaiton to reduce register usage and gain performance. #15111
+31
−33
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/15111
Note: Links to docs will display an error until the docs builds have been completed. ❌ 2 New FailuresAs of commit 8b5a2c4 with merge base 810b8e8 ( NEW FAILURES - The following jobs have failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
bot
added
the
CLA Signed
|
@trivedivivek has exported this pull request. If you are a Meta employee, you can view the originating Diff in D84608770. |
This PR needs a
|
trivedivivek
added a commit
to trivedivivek/executorch
that referenced
this pull request
Oct 15, 2025
…reduce register usage and gain performance. (pytorch#15111) Summary: ### Summary This diff rearranges calculations in the quantized linear tiled implementation to reduce register usage and gain performance. The changes include: - Moving the declaration of `mat1` within the for loop to reduce register usage. - Adding a conditional statement to calculate `in_row_txstride` when `IN_STORAGE` is "buffer". The changes aim to improve performance by reducing register usage in the shader program. Differential Revision: D84608770
trivedivivek
force-pushed
the
export-D84608770
branch
from
9942fe1 to
7b153ca
Compare
SS-JIA
approved these changes
Oct 15, 2025
trivedivivek
added a commit
to trivedivivek/executorch
that referenced
this pull request
Oct 15, 2025
…reduce register usage and gain performance. (pytorch#15111) Summary: ### Summary This diff rearranges calculations in the quantized linear tiled implementation to reduce register usage and gain performance. The changes include: - Moving the declaration of `mat1` within the for loop to reduce register usage. - Adding a conditional statement to calculate `in_row_txstride` when `IN_STORAGE` is "buffer". The changes aim to improve performance by reducing register usage in the shader program. Reviewed By: SS-JIA Differential Revision: D84608770
trivedivivek
force-pushed
the
export-D84608770
branch
from
7b153ca to
bad1ad1
Compare
trivedivivek
added a commit
to trivedivivek/executorch
that referenced
this pull request
Oct 15, 2025
…reduce register usage and gain performance. (pytorch#15111) Summary: ### Summary This diff rearranges calculations in the quantized linear tiled implementation to reduce register usage and gain performance. The changes include: - Moving the declaration of `mat1` within the for loop to reduce register usage. - Adding a conditional statement to calculate `in_row_txstride` when `IN_STORAGE` is "buffer". The changes aim to improve performance by reducing register usage in the shader program. Reviewed By: SS-JIA Differential Revision: D84608770
trivedivivek
force-pushed
the
export-D84608770
branch
from
bad1ad1 to
735ee34
Compare
trivedivivek
added a commit
to trivedivivek/executorch
that referenced
this pull request
Oct 16, 2025
…reduce register usage and gain performance. (pytorch#15111) Summary: ### Summary This diff rearranges calculations in the quantized linear tiled implementation to reduce register usage and gain performance. The changes include: - Moving the declaration of `mat1` within the for loop to reduce register usage. - Adding a conditional statement to calculate `in_row_txstride` when `IN_STORAGE` is "buffer". The changes aim to improve performance by reducing register usage in the shader program. Reviewed By: SS-JIA Differential Revision: D84608770
trivedivivek
added a commit
to trivedivivek/executorch
that referenced
this pull request
Oct 16, 2025
…reduce register usage and gain performance. (pytorch#15111) Summary: ### Summary This diff rearranges calculations in the quantized linear tiled implementation to reduce register usage and gain performance. The changes include: - Moving the declaration of `mat1` within the for loop to reduce register usage. - Adding a conditional statement to calculate `in_row_txstride` when `IN_STORAGE` is "buffer". The changes aim to improve performance by reducing register usage in the shader program. Reviewed By: SS-JIA Differential Revision: D84608770
trivedivivek
added a commit
to trivedivivek/executorch
that referenced
this pull request
Oct 16, 2025
…reduce register usage and gain performance. (pytorch#15111) Summary: ### Summary This diff rearranges calculations in the quantized linear tiled implementation to reduce register usage and gain performance. The changes include: - Moving the declaration of `mat1` within the for loop to reduce register usage. - Adding a conditional statement to calculate `in_row_txstride` when `IN_STORAGE` is "buffer". The changes aim to improve performance by reducing register usage in the shader program. Reviewed By: SS-JIA Differential Revision: D84608770
trivedivivek
added a commit
to trivedivivek/executorch
that referenced
this pull request
Oct 16, 2025
…reduce register usage and gain performance. (pytorch#15111) Summary: ### Summary This diff rearranges calculations in the quantized linear tiled implementation to reduce register usage and gain performance. The changes include: - Moving the declaration of `mat1` within the for loop to reduce register usage. - Adding a conditional statement to calculate `in_row_txstride` when `IN_STORAGE` is "buffer". The changes aim to improve performance by reducing register usage in the shader program. Reviewed By: SS-JIA Differential Revision: D84608770
trivedivivek
added a commit
to trivedivivek/executorch
that referenced
this pull request
Oct 16, 2025
…reduce register usage and gain performance. (pytorch#15111) Summary: ### Summary This diff rearranges calculations in the quantized linear tiled implementation to reduce register usage and gain performance. The changes include: - Moving the declaration of `mat1` within the for loop to reduce register usage. - Adding a conditional statement to calculate `in_row_txstride` when `IN_STORAGE` is "buffer". The changes aim to improve performance by reducing register usage in the shader program. Reviewed By: SS-JIA Differential Revision: D84608770
trivedivivek
force-pushed
the
export-D84608770
branch
from
735ee34 to
700764e
Compare
trivedivivek
added a commit
to trivedivivek/executorch
that referenced
this pull request
Oct 16, 2025
…reduce register usage and gain performance. (pytorch#15111) Summary: ### Summary This diff rearranges calculations in the quantized linear tiled implementation to reduce register usage and gain performance. The changes include: - Moving the declaration of `mat1` within the for loop to reduce register usage. - Adding a conditional statement to calculate `in_row_txstride` when `IN_STORAGE` is "buffer". The changes aim to improve performance by reducing register usage in the shader program. Reviewed By: SS-JIA Differential Revision: D84608770
trivedivivek
added a commit
to trivedivivek/executorch
that referenced
this pull request
Oct 16, 2025
…reduce register usage and gain performance. (pytorch#15111) Summary: ### Summary This diff rearranges calculations in the quantized linear tiled implementation to reduce register usage and gain performance. The changes include: - Moving the declaration of `mat1` within the for loop to reduce register usage. - Adding a conditional statement to calculate `in_row_txstride` when `IN_STORAGE` is "buffer". The changes aim to improve performance by reducing register usage in the shader program. Reviewed By: SS-JIA Differential Revision: D84608770
trivedivivek
added a commit
to trivedivivek/executorch
that referenced
this pull request
Oct 16, 2025
…reduce register usage and gain performance. (pytorch#15111) Summary: ### Summary This diff rearranges calculations in the quantized linear tiled implementation to reduce register usage and gain performance. The changes include: - Moving the declaration of `mat1` within the for loop to reduce register usage. - Adding a conditional statement to calculate `in_row_txstride` when `IN_STORAGE` is "buffer". The changes aim to improve performance by reducing register usage in the shader program. Reviewed By: SS-JIA Differential Revision: D84608770
trivedivivek
added a commit
to trivedivivek/executorch
that referenced
this pull request
Oct 16, 2025
…reduce register usage and gain performance. (pytorch#15111) Summary: ### Summary This diff rearranges calculations in the quantized linear tiled implementation to reduce register usage and gain performance. The changes include: - Moving the declaration of `mat1` within the for loop to reduce register usage. - Adding a conditional statement to calculate `in_row_txstride` when `IN_STORAGE` is "buffer". The changes aim to improve performance by reducing register usage in the shader program. Reviewed By: SS-JIA Differential Revision: D84608770
…reduce register usage and gain performance. (pytorch#15111) Summary: ### Summary This diff rearranges calculations in the quantized linear tiled implementation to reduce register usage and gain performance. The changes include: - Moving the declaration of `mat1` within the for loop to reduce register usage. - Adding a conditional statement to calculate `in_row_txstride` when `IN_STORAGE` is "buffer". The changes aim to improve performance by reducing register usage in the shader program. Reviewed By: SS-JIA Differential Revision: D84608770
trivedivivek
force-pushed
the
export-D84608770
branch
from
700764e to
8b5a2c4
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
Summary:
Summary
This diff rearranges calculations in the quantized linear tiled implementation to reduce register usage and gain performance.
The changes include:
mat1within the for loop to reduce register usage.in_row_txstridewhenIN_STORAGEis "buffer".The changes aim to improve performance by reducing register usage in the shader program.
Differential Revision: D84608770