-
Notifications
You must be signed in to change notification settings - Fork 787
[SYCL][USM] Initial commit of flattening for kernel submission on queue #911
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
Signed-off-by: James Brodman <[email protected]>
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 looks good to me, but I do have a comment.
Currently the events vector is the last argument before the kernel functor in all cases. We should think about where it would go once we introduce reductions -- should the events vector still be the last argument before the functor, or should it be the last argument before the reductions?
More generally we might want to spend some time coming up with rules for the ordering of parameters to parallel_for
, to ensure we're being consistent and that code remains easy to read.
Signed-off-by: James Brodman <[email protected]>
Signed-off-by: James Brodman <[email protected]>
Signed-off-by: James Brodman <[email protected]>
Are we good to merge? @bader |
GitHub does not allow merging - "At least 1 approving review is required". @Pennycook, please, review the latest version. |
@bader Ping. |
Please, resolve @romanovvlad comment. |
Signed-off-by: James Brodman <[email protected]>
No, there is no dependency. |
Add utility methods to flatten kernel submission to 1 lambda when using USM.
Events for depends_on are passed as an extra function arg.
Signed-off-by: James Brodman [email protected]