-
Notifications
You must be signed in to change notification settings - Fork 26.3k
Port eq kernel to structured kernels.
#60177
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
Conversation
Tracking issue: #55070 [ghstack-poisoned]
💊 CI failures summary and remediationsAs of commit 3c1fa9e (more details on the Dr. CI page and at hud.pytorch.org/pr/60177):
🕵️ 2 new failures recognized by patternsThe following CI failures do not appear to be due to upstream breakages:
|
Tracking issue: #55070 [ghstack-poisoned]
Tracking issue: #55070 [ghstack-poisoned]
Tracking issue: #55070 [ghstack-poisoned]
|
The 3 CI failures don't seem related to this PR. |
|
@ezyang has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
| native::check_convert(other.item(), self.scalar_type()); | ||
| } else if (self.dim() == 0 && other.dim() != 0) { | ||
| native::check_convert(self.item(), other.scalar_type()); | ||
| } |
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.
Note to reviewers: this is ported code that cannot be deleted (and is not otherwise modified); hopefully to be deleted next PR
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.
@ngimel this logic looks potentially inefficient: if I have a CUDA scalar and I pass it to a comparison operation, I'm gonna sync for not very good reason.
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.
why will be be able to remove it in a future PR?
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.
no, as currently specced this is impossible to remove unless you turn it into an async cuda assert
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.
FYI, now people are rightfully complaining about it (and sorry I missed the original ping a year ago, lol) D36012582
Tracking issue: #55070 Differential Revision: [D29509871](https://our.internmc.facebook.com/intern/diff/D29509871) [ghstack-poisoned]
|
@ezyang has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
Tracking issue: #55070 Differential Revision: [D29509871](https://our.internmc.facebook.com/intern/diff/D29509871) [ghstack-poisoned]
Tracking issue: #55070 Differential Revision: [D29509871](https://our.internmc.facebook.com/intern/diff/D29509871) [ghstack-poisoned]
|
@ezyang has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
|
The CI failures don't seem to be related to this PR. |
Tracking issue: #55070 Differential Revision: [D29509871](https://our.internmc.facebook.com/intern/diff/D29509871) [ghstack-poisoned]
|
@ezyang has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
Stack from ghstack:
ne,lt,gt,le,geto structured kernels. #60942 Port other comparison ops:ne,lt,gt,le,geto structured kernels.eqkernel to structured kernels. #60177 Porteqkernel to structured kernels.Tracking issue: #55070
Differential Revision: D29509871