-
**Environment:**
1. **Framework**: TensorFlow, PyTorch
2. **Framework version**:
- TensorFlow: 2.18.0
- PyTorch: 2.4.1
3. **Horovod version**: Attempting to install latest via pip
4. **MPI…
-
Installing a CUDA-aware MPI in the defaults, e.g. with
```bash
spack install mpi+cuda
```
does not properly select `openmpi+cuda` but just the plain `openmpi~cuda` default.
Since "CUDA-Awaren…
-
## Background information
Using openmpi direct GPU communication with CUDA, a memory growth is observed throughout the duration of the run.
After tracing the issue using the ompi logs `--mca mpi_c…
-
The `--cuda-mpi` flag is utilized by users who wish to use CUDA aware MPI. This inherently requires GPU access. Currently the `--gpu` flag is also required, but the functionality added there should be…
-
### Required prerequisites
- [x] Consult the [security policy](https://github.com/NVIDIA/cuda-quantum/security/policy). If reporting a security vulnerability, do not report the bug using this form. U…
-
Hello,
I am building [thirdPartyLibraries](https://github.com/GEOS-DEV/thirdPartyLibs) for [GEOS](https://github.com/GEOS-DEV/GEOS) using
- GCC 11.4.1
- HPC_X (v2.20), Open MPI (v4.1.7a1)
- …
-
## Background information
### What version of Open MPI are you using? (e.g., v3.0.5, v4.0.2, git branch name and hash, etc.)
4.0.5
### Describe how Open MPI was installed (e.g., from a source…
-
### Description
Hi,
`jax.jit` on a function seems to fail when running in an OpenMPI environment. An MWE is shown below:
```python
# error.py
# Run as: mpirun -n 8 python error.py
import…
-
Hi,
I'm seeing various errors in muelu unit tests on nvidia h100 gpu using cuda 12.4 w/ kokkos uvm flag enabled. I'm not sure which approach is preferred for reporting here but I've tested using tw…
-
Intel 24 is too new. Oh dear.
```
cmake
-D CMAKE_INSTALL_PREFIX=/work/00434/eijkhout/mfem/installation-mfem-4.7-milan-intel24-impi21.12
-D CMAKE_COLOR_DIAGNOSTICS=OFF
-D CMAKE_VERBO…