-
Notifications
You must be signed in to change notification settings - Fork 75.2k
[PluggableDevice] Add pluggable device load mechanism #43610
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
[PluggableDevice] Add pluggable device load mechanism #43610
Conversation
tensorflow/c/experimental/stream_executor/stream_executor_internal.h
Outdated
Show resolved
Hide resolved
|
@gbaned , sorry to late response, we are in a long public holiday last week, will address all comments soon. Thanks! |
I don't have a usecase in mind either right now. We can always expose it later if we find such case. |
|
Tagging |
manivaradarajan
left a comment
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.
For @tensorflow/api-owners: API changes LGTM. Please address other comments before merging.
Thanks @penpornk and @fchollet, yes we are actively working on adding the Metal device plug-in using the modular design's pluggable interface. |
manivaradarajan
left a comment
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.
Please address comment wording suggestions before merging.
support for pluggable device plugin loading mechanism, this PR is following Pluggable device for TensorFlow, which provides the capability of loading a pluggable device plugin in the TensorFlow Proper.
Added by @penpornk on 11/18/20:
This PR is part of an effort to let new third-party devices connect to TensorFlow modularly. There are two RFCs: