-
Notifications
You must be signed in to change notification settings - Fork 1.1k
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
Add run time compilation #11225
base: main
Are you sure you want to change the base?
Add run time compilation #11225
Conversation
✅ Deploy Preview for meta-velox canceled.
|
@oerling has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
@oerling has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
@oerling has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
Summary: - Adds a CompiledModule abstraction on top of Cuda run time compilation. - Adds a cache of run time compiled kernels. The cache returns a kernel immediately and leaves the kernel compiling in the background. The kernel's methods wait for the compilation to be ready. - tests that runtime API and driver API streams are interchangeable when running a dynamically generated kernel. - Add proper use of contexts, one per device. The contexts are needed because of using the driver API to handle run time compilation. - Add device properties to the Device* struct. Differential Revision: D64205005 Pulled By: oerling
This pull request was exported from Phabricator. Differential Revision: D64205005 |
Summary: - Adds a CompiledModule abstraction on top of Cuda run time compilation. - Adds a cache of run time compiled kernels. The cache returns a kernel immediately and leaves the kernel compiling in the background. The kernel's methods wait for the compilation to be ready. - tests that runtime API and driver API streams are interchangeable when running a dynamically generated kernel. - Add proper use of contexts, one per device. The contexts are needed because of using the driver API to handle run time compilation. - Add device properties to the Device* struct. Differential Revision: D64205005 Pulled By: oerling
This pull request was exported from Phabricator. Differential Revision: D64205005 |
Adds a CompiledModule abstraction on top of Cuda run time compilation.
Adds a cache of run time compiled kernels. The cache returns a kernel immediately and leaves the kernel compiling in the background. The kernel's methods wait for the compilation to be ready.
tests that runtime API and driver API streams are interchangeable when running a dynamically generated kernel.
Add proper use of contexts, one per device. The contexts are needed because of using the driver API to handle run time compilation.
Add device properties to the Device* struct.