Skip to content
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

SYCL/Intel GPU issue tracker #619

Closed
BenWibking opened this issue Apr 23, 2024 · 2 comments
Closed

SYCL/Intel GPU issue tracker #619

BenWibking opened this issue Apr 23, 2024 · 2 comments
Labels
enhancement New feature or request SYCL

Comments

@BenWibking
Copy link
Collaborator

BenWibking commented Apr 23, 2024 •
edited
Loading

Describe the proposal

This issue is to track changes that need to be made to get Quokka to compile with the SYCL GPU backend:

  • Can't use function references (used in ODEIntegrate.hpp for rhs function) 
    /home/runner/work/quokka/quokka/src/ODEIntegrate.hpp:136:2: error: SYCL kernel cannot call through a function pointer
136 |         rhs(t0, y0, ydot0, user_data);
    |         ^
  • Can't use global variables on device (used everywhere in Microphysics...) 
    In file included from /home/runner/work/quokka/quokka/src/EOS.hpp:20:
/home/runner/work/quokka/quokka/extern/Microphysics/interfaces/eos.H:86:35: error: SYCL kernel cannot use a non-const global variable
 86 |   state.rho = amrex::min(EOSData::maxdens, amrex::max(EOSData::mindens, state.rho));
    |
  • Need to add fp-model=precise to compiler flags: 
    export CMAKE_CXX_FLAGS="-ffp-model=precise"

Additional context
SYCL 2020 specification lists many restrictions that CUDA and HIP do not have.
https://registry.khronos.org/SYCL/specs/sycl-2020/html/sycl-2020.html#sec:language.restrictions.kernels
@BenWibking BenWibking added enhancement New feature or request SYCL labels Apr 23, 2024
@BenWibking BenWibking changed the title issues to fix to compile on Intel GPUs SYCL/Intel GPU issue tracker Apr 23, 2024
@BenWibking BenWibking mentioned this issue Jun 22, 2024
Merged
7 tasks
github-merge-queue bot pushed a commit that referenced this issue Jun 24, 2024
@BenWibking
update docs re Intel GPUs (#654)
349ef20 
### Description
Updates the documentation to note that Quokka cannot currently be
compiled for Intel GPUs due to design limitations in Intel's GPU
programming model.

### Related issues
#619

### Checklist
_Before this pull request can be reviewed, all of these tasks should be
completed. Denote completed tasks with an `x` inside the square brackets
`[ ]` in the Markdown source below:_
- [x] I have added a description (see above).
- [x] I have added a link to any related issues see (see above).
- [x] I have read the [Contributing
Guide](https://github.com/quokka-astro/quokka/blob/development/CONTRIBUTING.md).
- [ ] I have added tests for any new physics that this PR adds to the
code.
- [ ] I have tested this PR on my local computer and all tests pass.
- [ ] I have manually triggered the GPU tests with the magic comment
`/azp run`.
- [x] I have requested a reviewer for this PR.
@BenWibking
Copy link
Collaborator Author

Device global variable workaround: AMReX-Codes/amrex#4056

@BenWibking
Copy link
Collaborator Author

Intel no longer sells datacenter GPUs, so there's no reason anymore to support SYCL.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
enhancement New feature or request SYCL
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@BenWibking