[SYCL][CUDA] Fix flaky interop task test #17161
Closed
Conversation
This file contains 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
In host tasks, `Q.wait()` doesn't wait on interop asynchronous work, it simply waits on the lambba being executed. So for correct execution the CUDA stream must be waited on inside of the lambda. This made the test flaky because in some cases the copies had been completed before being checked, and in some cases they hadn't, this extra wait fixes that.
npmiller
had a problem deploying
to
WindowsCILock
GitHub Actions
Failure
npmiller
had a problem deploying
to
WindowsCILock
GitHub Actions
Failure
npmiller
temporarily deployed
to
WindowsCILock
GitHub Actions
Inactive
npmiller
temporarily deployed
to
WindowsCILock
GitHub Actions
Inactive
againull
approved these changes
Feb 26, 2025
|
Actually looking at this again all the CUDA calls are synchronous so it shouldn't need an extra sync, it might still help but it's likely not the root cause of the issue |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
In host tasks,
Q.wait()doesn't wait on interop asynchronous work, it simply waits on the lambba being executed.So for correct execution the CUDA stream must be waited on inside of the lambda.
This made the test flaky because in some cases the copies had been completed before being checked, and in some cases they hadn't, this extra wait fixes that.
Fixes #17026