Pratik Nayak
Pratik Nayak
Just adding more performance results on a 7pt stencil and some chosen matrices from Suitesparse, to show that this type of coarsening can help in certain cases. ### Apply speedup...
I think that is because of one for PR to master and one for push, but I think that was by design ? But we can probably remove the action...
Thanks @lahwaacz for the report. I think installing documentation might be a good idea. In line with the usual linux packages, it probably should go into `doc/`. Benchmarks are generally...
Can we extend this approach to enable builds for HPC systems ? Frontier, Horeka etc.
We should either remove bump the SYCL version support or create a special case for it, as this update [currently breaks for SYCL 2022-1](https://gitlab.com/ginkgo-project/ginkgo-public-ci/-/jobs/6271510855). I would also vote for dropping...
Maybe I am misunderstanding something, but in the non-distributed case, this is basically diagonal relaxation with weights equal to row sums, right ? So, I think it definitely makes sense...
LGTM too. Feel free to approve and merge when you are ready.