Dom Heinzeller

Fixes https://github.com/spack/spack/issues/32732

@skosukhin @haampie Can we move this PR forward? Note that @kgerheiser is currently going through a job transition, I will try to help get this across the finish line if...

@WardF @haampie @skosukhin Sorry to bug you, but can we review and merge this PR some time soon? Thanks.

Hmm, this PR now has conflicts that must be resolved, but I do not have write access to @kgerheiser's fork/branch.

@skosukhin Can you help resolving the conflicts, please, and pushing this PR across the finish line?

> @kgerheiser is there a way to transfer ownership of this PR? What I'll do is check out Kyle's branch, add the zstd change on top of it, then create...

> > @kgerheiser is there a way to transfer ownership of this PR? > > What I'll do is check out Kyle's branch, add the zstd change on top of...

As of now, I am not aware of any.

> @climbfuji I want to confirm with you that the failed gnu tests listed in this issue were able to run when a new gnu compiler (v12) is used, it...

Note that this will increase the memory footprint of the model. Right now, all variables that are required only between `GFS_surface_composites_pre` and `GFS_surface_composites_post` and that do not need to survive...