Support cosine operator on XNNPACK #15431
Merged
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/15431
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New Failure, 1 Unrelated FailureAs of commit 585fd19 with merge base 82e37df ( NEW FAILURE - The following job has failed:
UNSTABLE - The following job is marked as unstable, possibly due to flakiness on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
bot
added
the
CLA Signed
|
@GregoryComer has exported this pull request. If you are a Meta employee, you can view the originating Diff in D83623619. |
GregoryComer
added
the
release notes: xnnpack
GregoryComer
force-pushed
the
export-D83623619
branch
from
ecfa831 to
97fe98d
Compare
GregoryComer
added a commit
to GregoryComer/executorch
that referenced
this pull request
Dec 15, 2025
Summary: Wire up the unary cosine operator in xnnpack for fp32 and fp16. Differential Revision: D83623619
GregoryComer
force-pushed
the
export-D83623619
branch
from
97fe98d to
b1a698a
Compare
GregoryComer
added a commit
to GregoryComer/executorch
that referenced
this pull request
Dec 15, 2025
Summary: Wire up the unary cosine operator in xnnpack for fp32 and fp16. Differential Revision: D83623619
GregoryComer
force-pushed
the
export-D83623619
branch
from
b1a698a to
b10ba7c
Compare
|
@GregoryComer has imported this pull request. If you are a Meta employee, you can view this in D83623619. |
digantdesai
reviewed
Dec 15, 2025
| input_id = vals_to_ids[get_input_node(node, 0)] | ||
|
|
||
| # output | ||
| output_id = vals_to_ids[node] |
Member
Author
There was a problem hiding this comment.
Will take as a follow-up, if that's okay. This pattern is shared between many ops, so I might do a larger refactor.
digantdesai
reviewed
Dec 15, 2025
| target_name = "cos.default" | ||
|
|
||
| def supported_precision_types(self) -> List[ConfigPrecisionType]: | ||
| return [ConfigPrecisionType.FP32] |
Contributor
There was a problem hiding this comment.
we need to add FP16 and later BF16 here.
Member
Author
There was a problem hiding this comment.
Yeah. The way the partitioner is currently written, FP32 implies FP16. I'll likely refactor this a little bit when we add BF16 support.
digantdesai
approved these changes
Dec 15, 2025
GregoryComer
added a commit
to GregoryComer/executorch
that referenced
this pull request
Dec 16, 2025
Summary: Wire up the unary cosine operator in xnnpack for fp32 and fp16. Reviewed By: digantdesai Differential Revision: D83623619 Pulled By: GregoryComer
GregoryComer
force-pushed
the
export-D83623619
branch
from
b10ba7c to
7249141
Compare
Summary: Wire up the unary cosine operator in xnnpack for fp32 and fp16. Reviewed By: digantdesai Differential Revision: D83623619 Pulled By: GregoryComer
GregoryComer
force-pushed
the
export-D83623619
branch
from
7249141 to
585fd19
Compare
xingguo01
pushed a commit
to xingguo01/executorch
that referenced
this pull request
Dec 18, 2025
Summary: Wire up the unary cosine operator in xnnpack for fp32 and fp16. Differential Revision: D83623619
jirioc
pushed a commit
to nxp-upstream/executorch
that referenced
this pull request
Dec 19, 2025
Summary: Wire up the unary cosine operator in xnnpack for fp32 and fp16. Differential Revision: D83623619
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: Wire up the unary cosine operator in xnnpack for fp32 and fp16.
Differential Revision: D83623619