Skip to content

[SYCL][CUDA] Partially reverts back the update on guess local work size function. #10055

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.

Sign up for GitHub

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

Jump to bottom
Merged
merged 2 commits into from
Jun 26, 2023
Merged

[SYCL][CUDA] Partially reverts back the update on guess local work size function. #10055

merged 2 commits into from
Jun 26, 2023

Conversation

mmoadeli
Copy link
Contributor

@mmoadeli mmoadeli commented Jun 23, 2023 •
edited
Loading

  • Reverts back the update to calculation of threads per block for 0th dimension when primary ranges are involved. That could cause out of range access.

@mmoadeli
Reverts back the update to calculation of threads per blcok for 0th d…
a1596cb 
…imension as it could cause out of range access.
@mmoadeli mmoadeli requested a review from a team as a code owner June 23, 2023 17:11
@mmoadeli mmoadeli requested a review from steffenlarsen June 23, 2023 17:11
@mmoadeli mmoadeli linked an issue Jun 23, 2023 that may be closed by this pull request
[SYCL][CUDA] Wrong global range when sycl::range is passed a prime #10051
Closed
@mmoadeli mmoadeli changed the title Partially reverts back the update on guess local work size for prime numbers. Partially reverts back the update on guess local work size function. Jun 23, 2023
@mmoadeli mmoadeli temporarily deployed to aws June 23, 2023 17:27 — with GitHub Actions Inactive
@mmoadeli mmoadeli temporarily deployed to aws June 23, 2023 18:05 — with GitHub Actions Inactive
@mmoadeli mmoadeli temporarily deployed to aws June 23, 2023 20:13 — with GitHub Actions Inactive
@mmoadeli mmoadeli temporarily deployed to aws June 23, 2023 20:23 — with GitHub Actions Inactive
@mmoadeli mmoadeli temporarily deployed to aws June 23, 2023 20:50 — with GitHub Actions Inactive
@mmoadeli mmoadeli temporarily deployed to aws June 23, 2023 21:05 — with GitHub Actions Inactive
steffenlarsen
steffenlarsen approved these changes Jun 26, 2023
View reviewed changes
Copy link
Contributor

@steffenlarsen steffenlarsen left a comment

Choose a reason for hiding this comment

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

LGTM.

@steffenlarsen
Copy link
Contributor

ESIMD Failed Tests (1):
SYCL :: ESIMD/api/functional/operators/operator_assignment_glb_mask.cpp
Unrelated as this only touches CUDA. Likely a desync issue.

@steffenlarsen steffenlarsen merged commit 2ab99aa into intel:sycl Jun 26, 2023
@mmoadeli mmoadeli deleted the prime-number-undo branch July 7, 2023 10:38
@mmoadeli mmoadeli changed the title Partially reverts back the update on guess local work size function. [SYCL][CUDA] Partially reverts back the update on guess local work size function. Jul 14, 2023
fabiomestre pushed a commit to fabiomestre/llvm that referenced this pull request Sep 26, 2023
@mmoadeli
Partially reverts back the update on guess local work size function. (i…
7847b96 
…ntel#10055)

* Reverts back the update to calculation of threads per block for 0th
dimension when primary ranges are involved. That could cause out of
range access.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@steffenlarsen steffenlarsen steffenlarsen approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

[SYCL][CUDA] Wrong global range when sycl::range is passed a prime
2 participants
@mmoadeli @steffenlarsen