Skip to content

Update sdp guards for performance #87241

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.

Sign up for GitHub

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

Jump to bottom
Closed
wants to merge 3 commits into from
Closed

Update sdp guards for performance #87241

wants to merge 3 commits into from

Conversation

@drisspg
Copy link
Contributor

@drisspg drisspg commented Oct 18, 2022

Summary

Makes the contiguous check for the nt input more strict/correct as well as makes some performance improvements to the checks

@pytorch-bot
Copy link

pytorch-bot bot commented Oct 18, 2022 •
edited
Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/87241

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

✅ No Failures, 5 Pending

As of commit 013511a:
💚 Looks good so far! There are no failures yet. 💚

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

@drisspg drisspg requested a review from cpuhrsch October 18, 2022 22:06
drisspg
drisspg commented Oct 18, 2022
View reviewed changes
aten/src/ATen/native/nested/cuda/NestedTensorTransformerFunctions.cpp
int64_t offset_constant = (tensor_offsets[1] - tensor_offsets[0]) /
tensor_size_ptr[0] * tensor_stride_ptr[0];

for (int64_t i = 2; i < n_tensors; i++) {
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Could probably roll this into the previous loop but this uses the stride matrix and size matrix to compute numel so is based off the first check for descending ordering

@cpuhrsch cpuhrsch changed the title Update sdp nested for performance Update sdp guards for performance Oct 18, 2022
@pytorch-bot pytorch-bot bot added the ciflow/trunk Trigger trunk jobs on your pull request label Oct 18, 2022
@drisspg
Copy link
Contributor Author

drisspg commented Oct 18, 2022

No Checks:
Screen Shot 2022-10-18 at 3 47 02 PM

With Checks:
Screen Shot 2022-10-18 at 3 48 25 PM

@drisspg
Copy link
Contributor Author

drisspg commented Oct 18, 2022

@pytorchbot merge -l

@pytorchmergebot
Copy link
Collaborator

Merge started

The -l land checks flag is deprecated and no longer needed. Instead we now automatically add the ciflow\trunk label to your PR once it's approved

Your change will be merged once all checks on your PR pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging
Check the merge workflow status
here

@github-actions
Copy link
Contributor

Hey @drisspg.
You've committed this PR, but it does not have both a 'release notes: ...' and 'topics: ...' label. Please add one of each to the PR. The 'release notes: ...' label should represent the part of PyTorch that this PR changes (fx, autograd, distributed, etc) and the 'topics: ...' label should represent the kind of PR it is (not user facing, new feature, bug fix, perf improvement, etc). The list of valid labels can be found here for the 'release notes: ...' and here for the 'topics: ...'.
For changes that are 'topic: not user facing' there is no need for a release notes label.

@cpuhrsch cpuhrsch added the topic: not user facing topic category label Oct 19, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@cpuhrsch cpuhrsch cpuhrsch approved these changes

Assignees

No one assigned

Labels

ciflow/trunk Trigger trunk jobs on your pull request Merged topic: not user facing topic category

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@drisspg @pytorchmergebot @cpuhrsch