-
Notifications
You must be signed in to change notification settings - Fork 7k
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
Re-enable vision MPS builds #8485
Conversation
This reverts commit 4393f7d.
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/vision/8485
Note: Links to docs will display an error until the docs builds have been completed. ❌ 5 New FailuresAs of commit 8c6084c with merge base f1bcbd3 (): NEW FAILURES - The following jobs have failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
@@ -68,7 +68,7 @@ jobs: | |||
export GPU_ARCH_TYPE=cpu | |||
export GPU_ARCH_VERSION='' | |||
|
|||
./.github/scripts/unittest.sh | |||
${CONDA_RUN} ./.github/scripts/unittest.sh |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is not related to the issue at hand but I found it during the investigation. This is the correct way to invoke a script using pytorch/test-infra/.github/workflows/macos_job.yml@main
. Otherwise, the system conda will be used instead of the one setup by macos_job
and the ci
env created by the script won't be clean up correctly
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks @huydhn , does that mean we should use that for the rest of the jobs as well (including linux and windows)?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We don't need this on Linux because the script is run inside Docker, so cleaning things up is not an issue there and you can use your own conda installation if needed.
On the other hand, this might be needed on Windows, but I haven't seen any disk space issue there before, so maybe it's not that important there.
@@ -68,7 +68,7 @@ jobs: | |||
export GPU_ARCH_TYPE=cpu | |||
export GPU_ARCH_VERSION='' | |||
|
|||
./.github/scripts/unittest.sh | |||
${CONDA_RUN} ./.github/scripts/unittest.sh |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks @huydhn , does that mean we should use that for the rest of the jobs as well (including linux and windows)?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I can see the MPS tests being green now. Thanks a ton @huydhn for identifying the original offending PR from pytorch!
Hey @NicolasHug! You merged this PR, but no labels were added. |
Summary: Co-authored-by: Nicolas Hug <contact@nicolas-hug.com> Reviewed By: vmoens Differential Revision: D60596231 fbshipit-source-id: 8493928d6f94a3abeff6a38c9295d343127027b2
Re-run the test after https://hud.pytorch.org/hud/pytorch/pytorch/75b0720a97ac5d82e8a7a1a6ae7c5f7a87d7183d is included in the next PyTorch nightly build.
cc @seemethere