Skip to content

[SYCL][Doc] Add parallel_for simplification extension #1580

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 1 commit into from
May 7, 2020

Conversation

rarutyun
Copy link
Contributor

Signed-off-by: Ruslan Arutyunyan ruslan.arutyunyan@intel.com

Contains the following features:

  • Allow parallel_for call with number or braced-init-list as the first argument, i.e.
  • Allow C++14 Generic lambda expression as the kernel for parallel_for, i.e [](auto){}
  • Allow C++14 Generic lambda expression as the kernel for parallel_for_work_group
  • Allow integral type as the kernel argument for parallel_for called with range<1>, e.g. [](int i){}
  • Allow item<1> to size_t conversion
  • Resolve ambiguity for accessor::operator[] when the argument for it is an item
  • (optional) parallel_for kernel shall always take item as an argument (not item or id)

Signed-off-by: Ruslan Arutyunyan <ruslan.arutyunyan@intel.com>
@rarutyun
Copy link
Contributor Author

Probably we should not add the feature marked as optional in Pull Request description to Khronos merge request. Let me know your thoughts

@mkinsner mkinsner changed the title [SYCL] Add parallel_for simplification extension [SYCL][Doc] Add parallel_for simplification extension Apr 24, 2020
@keryell
Copy link
Contributor

keryell commented Apr 24, 2020

That looks pretty triSYCL style... I did not know it was an Intel extension 5 years ago... :-)

@bader bader added the spec extension All issues/PRs related to extensions specifications label May 1, 2020
@bader bader merged commit 856a777 into sycl May 7, 2020
@bader bader deleted the private/rarutyun/parallel_for_simplification_extension branch May 7, 2020 16:55
bb-sycl pushed a commit that referenced this pull request Sep 15, 2022
This extension allows TypeVector to have a Physical Pointer Type
Component Type and introduces gather/scatter instructions.
It will be useful for explicitly vectorized kernels.

Spec: #6613

Signed-off-by: Sidorov, Dmitry <dmitry.sidorov@intel.com

Original commit:
KhronosGroup/SPIRV-LLVM-Translator@49b08e8
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
spec extension All issues/PRs related to extensions specifications
Projects
None yet
Development

Successfully merging this pull request may close these issues.

7 participants