William R Tobin

... I suppose having CI that doesn't have issues with timing out is probably a prerequisite of this.

Yeah that is more or less what this is about. It will requires discussions and planning here at Livermore though, and will require developers have a slightly different understanding of...

Yeah the first 2 ticks can be resolved, especially since for critical profiling I usually retain the full caliper database now instead of just the stdout report. Don't think our...

Yeah anything that can be done to reduce the mesh size but still replicate the error would be beneficial. Debugging this sort of issue at scale and without the files...

Please provide any input files if possible. Documenting that this may be related to #841 and if/when that is resolved this should be checked.

@corbett5 Randy and I discussed that particular point and IIRC we decided it wasn't a necessary restriction. My concern was primarily preventing users from erroneously de- and re-allocating on device...

We're getting a memory failure on ~~lassen~~ frontier in that routine so this branch is/was just to test if bypassing that usage happened to resolve the issue. So far seems...

Its really more me not trusting the hip compiler to handle syntax / usage that isn't super common (as I have experienced numerous times during the HIP port).

Sorry no, just frontier/hip, I was typing that while in a meeting where we were discussing other things on lassen.

The mem-aware packing branch was functionally complete and ready-to-go, the only reason it hasn't been merged has been related to failures of integrated tests (in develop) on our primary GPU...