kvoronin

Hi @sandwichmaker and everyone! As a part of this story, will you be interested in integrating [cuDSS](https://developer.nvidia.com/cudss) a GPU-accelerated sparse direct solver? It's an early-access but already has some promising...

@sandwichmaker thanks for the reply! Yes, we're willing to work on a patch. Requirements you listed above are very reasonable. (The only thing to note is that cuDSS might not...

Hi @S-o-T and @sandwichmaker! We have some PoC for an integration patch but currently it stays internal. The intention to bring it to the public is as strong as it...

Yes, @sandwichmaker, I will compare our local patch with what @S-o-T is suggesting and see if there is anything to change/add to the patch. Hopefully we will have best of...

Hi! What OS and compiler did you use? I suspect, that you need to pass -lstdc++ additionally but I am curious to learn your setup. Current CMakeLists.txt expects a compiler...

Hi @Xusj0w0! Yes, it is a problem we are aware of. Basically, system-wide installation of cudss does not work as intended and due to the update-alternatives cannot find the shared...

Same error, similar setup (but while trying to solve it, I've mode the vscode cache folder to a custom location outside my tiny home directory which has a limitation of...

> Has your application received a response yet, and have the restrictions been lifted? I applied more than two weeks ago, but I noticed that the 100MB limit is still...

Hi @cmaureir! I totally agree that options you mention are reasonable to consider. - cusparselt does not have any dependencies which can be split off the main wheels and the...

Update: I've checked that adding `--mca btl_smcuda_use_cuda_ipc 0` fixes the issue. As I understand, this is a workaround rather than a solution. So I think it would be helpful if...