HanatoK

> > I would like to have a base-class function for the atoms map, because it's not a NAMD-specific issue. LAMMPS and GROMACS are not using it yet, but simply...

> > > I have tried to solve the issue and revise the NAMD patch that I posted here in https://gitlab.com/tcbgUIUC/namd/-/merge_requests/486. Could you take a look at that NAMD MR...

> > You mean that there currently is no callback to Colvars to state that a new `structure` command has been loaded? If so, we should design a test that...

> > The fix looks good! I wish I had seen the bug... Do you think a variant of the `reinitatoms` test could trigger the issue reliably? Otherwise I'm fine...

Hi @jhenin and @giacomofiorin ! Could you try https://github.com/Colvars/colvars/pull/887/commits/bc27da1c83e3b2bac6e1e8ddb652d7d138ba3293 locally? The ARM64 build currently skip the `003_reinitatoms` test due to possibly the same crash.

> Indeed! This does reveal the crash, and therefore the fix. Now I wish I had an ARM64 test box to investigate - do you? I have one, but the...

> (As an alternative, we can just forget about the `nan` and merge this PR as is, with the needed updated to the description) Maybe you can revert https://github.com/Colvars/colvars/pull/887/commits/0a0e3c2c0625ac9e0b8842029626a9ebddbec2de before...

I have made a NAMD MR to pass the step information to the CUDAGM clients (https://gitlab.com/tcbgUIUC/namd/-/merge_requests/492). The corresponding Colvars PR in https://github.com/Colvars/colvars/pull/899 could use the updated interface to invalidate the...

> Looks like the checks are the same between 18 and 20, but good to have the latter back for the future (thanks @HanatoK). > > I would like to...

Hi @cchapellier ! I am not sure what your issue is. Could you elaborate more about "failing"?