Rocco Meli
Rocco Meli
@IAlibay are you suggesting to merge despite the failures?
> * The performance gain with the tuned A100 kernels is minor compared to using the P100 kernels like the tuned P100 kernels work reasonably well for V100. > >...
It could be useful, so why not. However, `relprop` doesn't say anything to me; is this a standard name? If not, can we thing of something more explicit?
Depends on https://github.com/MDAnalysis/mdanalysis/pulls and needs to be added to `.git-blame-ignore-revs`
> I asssume you'll also create/update .git-blame-ignore-revs ? Yes. As discussed in https://github.com/MDAnalysis/mdanalysis/pull/4725#issuecomment-2412050407, I am tagging formatting PRs with `[fmt]` and they will be squash-merged. They will be added to...
Temporarily converted to draft to prevent an accidental merge before the release.
The first point has been merged in https://github.com/eth-cscs/COSMA/pull/144, sorry for the duplication.
Thanks for the reports. Can you please give more details about the exact spec you are using with Spack (and possibly the dependencies)? I built successfully `[email protected]` both locally and...
The DBCSR version was my first suspect too, so I'm glad you found this is the case. I'll add this constraint on the package, I have an open PR to...
I think the CMake check is currently broken: ``` -- Checking for modules 'libxcf03;libxc>=6' -- Found libxcf03, version 7.0.0 -- Found libxc, version 7.0.0 -- Found LibXC: 1 (Required is...