Merged
Conversation
This is needed as SimpleKernel may be invoked before calling any Runtime APIs.
Member
|
Jenkins CI test (for commit aa7f10e) succeeded without errors! |
Member
Author
|
One consideration is that this feature exposes another route to use Driver API without implicit CUDA context (ref #916). I think it can be ignored though. import cupy
kern = cupy.SimpleKernel(r'''
extern "C" __global__
void test() {
printf("Hello world\n");
}
''', 'test')
cupy.cuda.Device(0).use()
kern((1,), (1,), ())
cupy.cuda.Stream.null.synchronize()
cupy.cuda.Device(1).use()
# cupy.cuda.runtime.memGetInfo() # uncomment to avoid FAIL
kern((1,), (1,), ()) # FAIL
cupy.cuda.Stream.null.synchronize() |
Member
|
Please use |
Member
Author
|
I discussed offline with @okuta -san regarding other design considerations:
|
Member
Author
|
This PR is ready for review. |
Member
Author
|
Jenkins, test this please. |
Member
|
Jenkins CI test (for commit 1138781) failed with status FAILURE. |
Member
Author
|
Jenkins, test this please. |
Member
|
Jenkins CI test (for commit 3a2a898) succeeded without errors! |
Member
|
LGTM! |
This was referenced Jul 17, 2018
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
SimpleKernelRawKernelis intended to expose public API to bring in raw CUDA kernel code to CuPy world.This is one of the frequently requested feature (#1258, #1242, etc.).
Currently it is just a thin wrapper around
cupy.core.compile_with_cachewith automatic per-device memoize. We'd like to discuss the design of SimpleKernel.After that, I should: