-
Cuda SpGEMM is very slow, not sure why -> needs investigation
-
I saw that #208 got merged into master, but I am still unable to build PyMFEM against my own parallel MFEM installation. I am using
```
python3 setup.py install --mfem-source=/Users/lyons/src/mfem_…
-
Hi,
I'm using AMGCL for reconstruction in cosmology (large unstructured Poisson systems), and it works extremely well (so congratulations and thank you for this very nice piece of software).
My mat…
-
Which unit tests are relevant and up to date?
We have extensively test on different GPUs, CUDA 8 vs 9.0 and quite a few keep failing under Linux. Besides the missing matrices, which we have filled i…
bnase updated
2 years ago
-
Nightly cuda/11.2.2 builds (no UVM) are failing in the following unit tests with kokkos-kernels@develop:
```
03:05:58 The following tests FAILED:
03:05:58 1784 - MueLu_UnitTestsBlockedTpetra_MPI…
-
**Describe the issue**
On some build (can't say for the moment why some works, other not):
src/tests/dense_lu.cu(114): error: identifier "cudaMallocAsync" is undefined
**Environment informati…
-
Design: #368
1. [ ] Sparse block lowering. (transformation)
1. [x] Sparse/Dense coordinates transformation. (@MasterJH5574 WIP)
1. The same sparse iterator viewed in different sparse…
-
It looks like the cuSparse function to multiply two sparse matrices uses a lot of memory, preventing benchmarking on matrices larger than 2000x2000. Let's try to figure out how much memory exactly and…
-
Can make spGEMM support also 3D tensor (might think as batched):
suppose one sparse matrix has dimensions (B, M, N), the other has dimensions (B, N, F)
and the multiplication result has dimensions (…
-
Hi all, this is not an issue so I'm not following the template. I'm wondering if it makes more sense to set the default storage mode of a `_rocsparse_mat_descr` to unsorted here:
https://github.com…