-
Notifications
You must be signed in to change notification settings - Fork 25.6k
Add BUILD_LAZY_CUDA_LINALG option #72306
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
When enable, it will generate `torch_cuda_linalg` library, which would depend on cusolve and magma and registers dynamic bindings to it from LinearAlgebraStubs [ghstack-poisoned]
CI Flow Status⚛️ CI FlowRuleset - Version:
|
🔗 Helpful links
💊 CI failures summary and remediationsAs of commit 5e30c44 (more details on the Dr. CI page): 💚 💚 Looks good so far! There are no failures yet. 💚 💚 This comment was automatically generated by Dr. CI (expand for details).Please report bugs/suggestions to the (internal) Dr. CI Users group. |
@malfet has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
Windows build errors are real |
So what happens now when |
When enable, it will generate `torch_cuda_linalg` library, which would depend on cusolve and magma and registers dynamic bindings to it from LinearAlgebraStubs Differential Revision: [D33992795](https://our.internmc.facebook.com/intern/diff/D33992795) [ghstack-poisoned]
@malfet has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
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.
As discussed with @IvanYashchuk , we both agree that it would be good to have a small write-up (perhaps at the top of LinearAlgebraStubs.cpp
detailing what are the steps that we need to take to add a new function, and possible problems we may find while doing so.
The comment below shows that, at least by the looks of it, one may find some highly non-trivial cases, and it's not clear to us when may this happen.
When enable, it will generate `torch_cuda_linalg` library, which would depend on cusolve and magma and registers dynamic bindings to it from LinearAlgebraStubs Differential Revision: [D33992795](https://our.internmc.facebook.com/intern/diff/D33992795) [ghstack-poisoned]
When enable, it will generate `torch_cuda_linalg` library, which would depend on cusolve and magma and registers dynamic bindings to it from LinearAlgebraStubs Differential Revision: [D33992795](https://our.internmc.facebook.com/intern/diff/D33992795) [ghstack-poisoned]
When enable, it will generate `torch_cuda_linalg` library, which would depend on cusolve and magma and registers dynamic bindings to it from LinearAlgebraStubs Differential Revision: [D33992795](https://our.internmc.facebook.com/intern/diff/D33992795) [ghstack-poisoned]
When enable, it will generate `torch_cuda_linalg` library, which would depend on cusolve and magma and registers dynamic bindings to it from LinearAlgebraStubs Differential Revision: [D33992795](https://our.internmc.facebook.com/intern/diff/D33992795) [ghstack-poisoned]
@malfet has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
When enable, it will generate `torch_cuda_linalg` library, which would depend on cusolve and magma and registers dynamic bindings to it from LinearAlgebraStubs Differential Revision: [D33992795](https://our.internmc.facebook.com/intern/diff/D33992795) [ghstack-poisoned]
@malfet has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
When enable, it will generate `torch_cuda_linalg` library, which would depend on cusolve and magma and registers dynamic bindings to it from LinearAlgebraStubs Differential Revision: [D33992795](https://our.internmc.facebook.com/intern/diff/D33992795) [ghstack-poisoned]
When enable, it will generate `torch_cuda_linalg` library, which would depend on cusolve and magma and registers dynamic bindings to it from LinearAlgebraStubs Differential Revision: [D33992795](https://our.internmc.facebook.com/intern/diff/D33992795) [ghstack-poisoned]
@malfet has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
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 great!
|
||
#if defined(BUILD_LAZY_CUDA_LINALG) | ||
namespace cuda { namespace detail { | ||
struct LinalgDispatch { |
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.
nit: can you add a comment here that these are old-style dispatches and people shouldn't add to this list, and instead prefer implementing with stubs right away?
Summary: Pull Request resolved: #72306 When enable, it will generate `torch_cuda_linalg` library, which would depend on cusolve and magma and registers dynamic bindings to it from LinearAlgebraStubs Test Plan: Imported from OSS Reviewed By: ngimel Differential Revision: D33992795 Pulled By: malfet fbshipit-source-id: d1fa351a320659b29754997c20d754e69bfe36c0
Summary: Pull Request resolved: #72306 When enable, it will generate `torch_cuda_linalg` library, which would depend on cusolve and magma and registers dynamic bindings to it from LinearAlgebraStubs Test Plan: Imported from OSS Reviewed By: ngimel Differential Revision: D33992795 Pulled By: malfet fbshipit-source-id: d1fa351a320659b29754997c20d754e69bfe36c0 (cherry picked from commit d5d6c69)
Sorry, reverting as this is the most likely culprit of causing timeouts in our OSS CI slow tests. See https://hud2.pytorch.org/hud/pytorch/pytorch/master/2 |
This pull request has been reverted by 3127128. To re-land this change, please open another pull request, assignthe same reviewers, fix the CI failures that caused the revert and make sure that the failing CI runs on the PR by applying the proper ciflow label (e.g., ciflow/trunk). |
Summary: Pull Request resolved: pytorch#72306 When enable, it will generate `torch_cuda_linalg` library, which would depend on cusolve and magma and registers dynamic bindings to it from LinearAlgebraStubs Test Plan: Imported from OSS Reviewed By: ngimel Differential Revision: D33992795 Pulled By: malfet fbshipit-source-id: d1fa351a320659b29754997c20d754e69bfe36c0 (cherry picked from commit d5d6c69)
Summary: Pull Request resolved: pytorch#72306 When enable, it will generate `torch_cuda_linalg` library, which would depend on cusolve and magma and registers dynamic bindings to it from LinearAlgebraStubs Test Plan: Imported from OSS Reviewed By: ngimel Differential Revision: D33992795 Pulled By: malfet fbshipit-source-id: d1fa351a320659b29754997c20d754e69bfe36c0 (cherry picked from commit d5d6c69)
Summary: Pull Request resolved: pytorch#72306 When enable, it will generate `torch_cuda_linalg` library, which would depend on cusolve and magma and registers dynamic bindings to it from LinearAlgebraStubs Test Plan: Imported from OSS Reviewed By: ngimel Differential Revision: D33992795 Pulled By: malfet fbshipit-source-id: d1fa351a320659b29754997c20d754e69bfe36c0 (cherry picked from commit d5d6c69)
This pull request has been reverted by 3127128. To re-land this change, please open another pull request, assignthe same reviewers, fix the CI failures that caused the revert and make sure that the failing CI runs on the PR by applying the proper ciflow label (e.g., ciflow/trunk). |
Summary: Pull Request resolved: pytorch/pytorch#72306 When enable, it will generate `torch_cuda_linalg` library, which would depend on cusolve and magma and registers dynamic bindings to it from LinearAlgebraStubs Test Plan: Imported from OSS Reviewed By: ngimel Differential Revision: D33992795 Pulled By: malfet fbshipit-source-id: d1fa351a320659b29754997c20d754e69bfe36c0 (cherry picked from commit d5d6c69a988b9454538ecd28674206da2541de17)
Summary: Pull Request resolved: pytorch/pytorch#72306 When enable, it will generate `torch_cuda_linalg` library, which would depend on cusolve and magma and registers dynamic bindings to it from LinearAlgebraStubs Test Plan: Imported from OSS Reviewed By: ngimel Differential Revision: D33992795 Pulled By: malfet fbshipit-source-id: d1fa351a320659b29754997c20d754e69bfe36c0 (cherry picked from commit d5d6c69a988b9454538ecd28674206da2541de17)
Stack from ghstack:
When enable, it will generate
torch_cuda_linalg
library, which would depend on cusolve and magma and registers dynamic bindings to it from LinearAlgebraStubsDifferential Revision: D33992795