-
Notifications
You must be signed in to change notification settings - Fork 21.4k
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
Fixed CUDA randint generation for large ranges. #126066
base: main
Are you sure you want to change the base?
Fixed CUDA randint generation for large ranges. #126066
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/126066
Note: Links to docs will display an error until the docs builds have been completed. ❌ 3 New Failures, 8 Pending, 1 Unrelated FailureAs of commit 1b281aa with merge base d578039 ( NEW FAILURES - The following jobs have failed:
FLAKY - The following job failed but was likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
@r-barnes Thanks for reviewing, I added some type annotations and changed the C++ parameters to |
3ea6988
to
849bf9e
Compare
@pytorchbot rebase |
@pytorchbot started a rebase job onto refs/remotes/origin/viable/strict. Check the current status here |
Successfully rebased |
b09c3f1
to
cb7925c
Compare
303b76e
to
0a7226b
Compare
CC @drisspg who might know more about the SDPA tests |
Thanks @eqy. Those tests in |
ada1975
to
993afca
Compare
…ts with torch.rand.
…relied on overlapping random states, which should now be fixed.
cafe610
to
e080abb
Compare
Fixes #125224
For large ranges, calls to CUDA
randint
use a differentunroll_factor
to generate random ints. Thisunroll_factor
was not considered correctly in the calculation of the Philox offsets. Thus, some of the random states were reused, resulting in lower entropy (see #125224).