Skip to content

[SYCL] Propagate attributes from redeclarations to SYCL kernel #2063

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 8 commits into from
Sep 8, 2020

Conversation

schittir
Copy link
Contributor

@schittir schittir commented Jul 7, 2020

  • Fix bug that propagates attributes from only canoncial declaration
  • Handle conflicting attributes
  • Add a test

@schittir schittir marked this pull request as draft August 14, 2020 01:54
- Fix bug that propagates attributes from only canoncial declaration
- Add a test
- throw error when same attributes with different parameters are
applied
- add test case
@schittir schittir force-pushed the propagate_redeclaration_attributes branch from d61e65e to 84e458e Compare September 3, 2020 01:16
@schittir schittir marked this pull request as ready for review September 3, 2020 18:11
- address max_work_group_size attribute applied with different
parameters
@schittir schittir requested a review from Fznamznon September 4, 2020 07:16
@elizabethandrews
Copy link
Contributor

I don't think this lit failure is related to your patch. l've seen it before. @bader is this a known issue?

@bader
Copy link
Contributor

bader commented Sep 4, 2020

I haven't seen this failure before.

@elizabethandrews
Copy link
Contributor

Hmm....ok. I'm pretty sure I have. The error message "error in a[0]=7!=7" looks very familiar.

@schittir if you can't reproduce the issue locally, you can try re-building this test and then file a bug report if it is flaky.

@schittir
Copy link
Contributor Author

schittir commented Sep 4, 2020

Hmm....ok. I'm pretty sure I have. The error message "error in a[0]=7!=7" looks very familiar.

@schittir if you can't reproduce the issue locally, you can try re-building this test and then file a bug report if it is flaky.

@elizabethandrews @bader
The error was unrelated to my patch. It disappeared with a rebuild.
Also all my local lit tests passed prior to commit.

@schittir
Copy link
Contributor Author

schittir commented Sep 4, 2020

Hi @premanandrao could you please review this?
Just pinging in case it didn't catch your attention. Thank you!

and change pattern matching for KernelName in lit test
@schittir schittir requested a review from Fznamznon September 8, 2020 07:35
The previous commit had the parameters of the attributes of func3
changed unintentionally. This is inconsequential except for code-review
consistency purpose.
Pass functions as lambda parameters instead of calling them from kernel
body
@bader bader merged commit 7ec85b0 into intel:sycl Sep 8, 2020
Chenyang-L pushed a commit that referenced this pull request Feb 18, 2025
[L0] Refactor to remove default constructor inits
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.

5 participants