Skip to content

Fix error message for a wrong fork CUDA #23030

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.

Sign up for GitHub

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

Jump to bottom
Closed
zhangguanheng66 wants to merge 9 commits into from
Closed

Fix error message for a wrong fork CUDA #23030

zhangguanheng66 wants to merge 9 commits into from

Conversation

@zhangguanheng66
Copy link
Contributor

@zhangguanheng66 zhangguanheng66 commented Jul 18, 2019 •
edited
Loading

Fix #17357
Unblock 1.2 release.

@pytorchbot pytorchbot added module: cuda Related to torch.cuda, and CUDA support in general module: internals Related to internal abstractions in c10 and ATen labels Jul 18, 2019
colesbury
colesbury requested changes Jul 18, 2019
View reviewed changes
Copy link
Member

@colesbury colesbury left a comment

Choose a reason for hiding this comment

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

At a minimum, I think THCPModule_getDevice_wrap and THCPModule_getDeviceCount_wrap should also call torch::utils::cuda_lazy_init so that a program that calls torch.cuda.get_device_count() in a bad fork gets a better error message.

colesbury
colesbury reviewed Jul 18, 2019
View reviewed changes
colesbury
colesbury approved these changes Jul 18, 2019
View reviewed changes
Copy link
Member

@colesbury colesbury left a comment

Choose a reason for hiding this comment

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

Looks good other than the inline comment in THCPModule_getDevice_wrap which should be fixed before this is landed

facebook-github-bot
facebook-github-bot reviewed Jul 18, 2019
View reviewed changes
Copy link
Contributor

@facebook-github-bot facebook-github-bot left a comment

Choose a reason for hiding this comment

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

@zhangguanheng66 has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

@ezyang ezyang mentioned this pull request Jul 19, 2019
Closed
@pytorchbot pytorchbot added the module: multiprocessing Related to torch.multiprocessing label Jul 19, 2019
facebook-github-bot
facebook-github-bot reviewed Jul 19, 2019
View reviewed changes
Copy link
Contributor

@facebook-github-bot facebook-github-bot left a comment

Choose a reason for hiding this comment

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

@zhangguanheng66 has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

facebook-github-bot
facebook-github-bot reviewed Jul 22, 2019
View reviewed changes
Copy link
Contributor

@facebook-github-bot facebook-github-bot left a comment

Choose a reason for hiding this comment

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

@zhangguanheng66 has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

facebook-github-bot
facebook-github-bot reviewed Jul 22, 2019
View reviewed changes
Copy link
Contributor

@facebook-github-bot facebook-github-bot left a comment

Choose a reason for hiding this comment

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

@zhangguanheng66 has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

@facebook-github-bot
Copy link
Contributor

facebook-github-bot commented Jul 22, 2019

@zhangguanheng66 merged this pull request in a6e45a6.

@zhangguanheng66 zhangguanheng66 mentioned this pull request Jul 23, 2019
Closed
facebook-github-bot pushed a commit that referenced this pull request Jul 24, 2019
@facebook-github-bot
Re-land "Fix error message for a wrong fork CUDA" (#23209)
aa660b8 
Summary:
Re-land #23030
Pull Request resolved: #23209

Differential Revision: D16440000

Pulled By: zhangguanheng66

fbshipit-source-id: e05683275522835a33d5a7e6d76b7e94774e4d98
@zhangguanheng66 zhangguanheng66 mentioned this pull request Jul 24, 2019
Closed
facebook-github-bot pushed a commit that referenced this pull request Jul 25, 2019
@facebook-github-bot
Fix error message for a wrong fork CUDA (#23322)
b22adeb 
Summary:
Re-land #23030
Pull Request resolved: #23322

Differential Revision: D16469442

Pulled By: zhangguanheng66

fbshipit-source-id: 70b63ab6265efa3f289111ef0ce46bb3c0d353bc
@peterbell10 peterbell10 mentioned this pull request Nov 1, 2019
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@facebook-github-bot facebook-github-bot facebook-github-bot left review comments

@colesbury colesbury colesbury approved these changes

Assignees

No one assigned

Labels

Merged module: cuda Related to torch.cuda, and CUDA support in general module: internals Related to internal abstractions in c10 and ATen module: multiprocessing Related to torch.multiprocessing

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

cuda runtime error (3): we're not detecting bad forks

5 participants

@zhangguanheng66 @facebook-github-bot @colesbury @pytorchbot @mruberry