Skip to content

Fixes enabling hierarchical parallelism API. #239

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

Merged
merged 2 commits into from
Jun 25, 2019
Merged

Conversation

kbobrovs
Copy link
Contributor

#221 will be split into 2 parts - enabling fixes and the hierarchical parallelism part itself. This is the enabling part. All (hopefully) the review comments from the original PR related to these fixes were addressed.

Device code in the SYCL runtime library may want to reuse SPIRV-based
initialization of current group and item id, size and offset - factor
them out into a separate header.

Signed-off-by: Konstantin S Bobrovsky <[email protected]>
agozillon
agozillon previously approved these changes Jun 24, 2019
Copy link
Contributor

@agozillon agozillon left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM, thanks for listening to my feedback as annoying as it probably is!

Remove "template overloads" of single_task and parallel_for invocation APIs
leading to template instantiation ambiguity in some cases and causing
compilation errors. The APIs with KernelName and KernelType template
parameters are changed to infer correct KernelName in both lambda and functor
flavors of API invocation.

Signed-off-by: Konstantin S Bobrovsky <[email protected]>
@kbobrovs
Copy link
Contributor Author

LGTM, thanks for listening to my feedback as annoying as it probably is!

The feedback was useful and is always welcome. Thanks!

@bader bader merged commit a480528 into intel:sycl Jun 25, 2019
@kbobrovs kbobrovs deleted the minor_fixes branch July 30, 2020 12:31
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants