Skip to content

Disable failing Tensor subscript assignment tests in GPU mode. #24340

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

Merged
merged 1 commit into from
Apr 27, 2019

Conversation

bgogul
Copy link
Contributor

@bgogul bgogul commented Apr 27, 2019

@bgogul
Copy link
Contributor Author

bgogul commented Apr 27, 2019

@swift-ci please test tensorflow

@bgogul bgogul requested review from dan-zheng and rxwei April 27, 2019 05:39
@bgogul
Copy link
Contributor Author

bgogul commented Apr 27, 2019

@eaplatanios, Just FYI.

Copy link
Contributor

@rxwei rxwei left a comment

Choose a reason for hiding this comment

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

I think only the assignment ones are failing, no?

@eaplatanios
Copy link

@bgogul @rxwei Yes, that is correct. Only the assignment ones fail but "EllipsisIndexing" is also using an assignment inside the test so it's ok to either disable that for now or just remove the assignment from inside that test.

I'll try to debug this in the next few days. I think it's an issue with the GPU kernel implementation but haven't had a chance to look at it yet.

Copy link
Contributor

@dan-zheng dan-zheng left a comment

Choose a reason for hiding this comment

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

Thanks @bgogul!

@bgogul
Copy link
Contributor Author

bgogul commented Apr 29, 2019

@eaplatanios FWIW, I investigated this a bit last Friday, but did not make much progress. As far as I ca see, the issue seems to be in the tensorflow implementation and not on our end. @pschuh has seen int32 cause issues with GPU (not GPE) support in tensorflow in the past.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants