pytorch-labs / tritonbench

Tritonbench is a collection of PyTorch custom operators with example inputs to measure their performance.
BSD 3-Clause "New" or "Revised" License
21 stars 3 forks source link

Fix nsys when running multiple ops #75

Closed xuzhao9 closed 3 days ago

xuzhao9 commented 3 days ago

We need to isolate the operator to a single op when spawning subprocess for nsys.

Test plan:

 python run.py --op embedding,rms_norm  --num-inputs 1  --metrics nsys_rep —csv --dump-csv
facebook-github-bot commented 3 days ago

@xuzhao9 has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.

facebook-github-bot commented 3 days ago

@xuzhao9 merged this pull request in pytorch-labs/tritonbench@e40e68faab2e7486c38f3dd96c0c23220c80ef60.