-
Implement using
- [ ] OpenACC
- [ ] OpenCL
- [ ] CUDA
tkphd updated
6 years ago
-
Hi All,
With recent developments like `OpenMP offload` support or `openacc pragma`, Can we port ANUGA to use GPU and accelerate simulations?
Cheers,
Samir
-
https://www.longlong.asia/2020/02/03/openacc.html
-
The gocean/eg2 example compiles and runs an OpenACC example. To return the final values from the accelerator an OpenACC directive is manually placed in the algorithm file. However, this directive is r…
-
Inspired by @Lancern work on providing some ASM guideline goodness, I created some guidelines for `auto` and `camelCase` usage, see it here: https://llvm.github.io/clangir/GettingStarted/coding-guidel…
-
This extends #512. Reductions and 'omp parallel do' are not needed for my ROMS work, but it would complete the omp directive support in PSyIR, so I opened a separate issue for this.
hiker updated
2 weeks ago
-
Click to expand!
### Issue Type
Bug
### Have you reproduced the bug with TF nightly?
No
### Source
source
### Tensorflow Version
2.12.0
### Custom Code
Yes
### OS Platform and Distributi…
-
At present some of the SAXS-sample components:
- SAXSCylinders.comp
- SAXSNanodiscs.comp
- SAXSNanodiscsWithTags.comp
- SAXSPDB.comp
...have calls to bessel functions j1 and j0 inside TRACE (i.…
-
Here's a list of open PRs -- along with notes for each PR as needed -- that are required to complete work on the [atmosphere/develop-openacc](https://github.com/MPAS-Dev/MPAS-Model/tree/atmosphere/dev…
-
Here is the file the is giving the ICE:
flang -fopenmp -fopenmp-targets=nvptx64-nvidia-cuda -c ../rte/kernels-openacc/mo_optical_props_kernels.F90
And the error message is:
/scratch/naromero/tmp…