Results 262 comments of Cameron Smith

I think avoiding empty parts is better than relying on a partitioner to fix them. IIRC, some zoltan balancing methods can create empty parts.

From offline discussions the following was points were raised: - Do you have `libfabric/1.22.0` in your module list? - yes, see the module list in the OP - When you...

> Would you please retry a run on Frontier setting the environment variable FABRIC_IFACE="cxi0"? We've identified an issue and this may be a workaround until we get it fully resolved....

MPI_init_thread results were added to the comment [here](https://github.com/ornladios/ADIOS2/issues/4630#issuecomment-3285790186): search for "MPI_Init_thread".

For what it is worth, I tried forcing use of the MPI dataplane via `DataTransport=MPI` for SST, and initializing mpi with threading enabled (as discussed above), but hit the following...

How about building with address sanitizer?

Hi @bartgol . Thanks for digging into the failing tests and the bisect. Sorry for missing these, I was only running `ctest -R omega` instead of `ctest -R [oO]mega` to...

I'm not sure. If we handled ownership correctly in the omegah dofmanager etc. then it should be fine. I'm digging now and will ping you on slack if I find...

Yeah, I like the idea of having the macro. Unfortunately, I don't see the 'deprecated' attribute (old or new C++14) listed in the oneAPI Intel compiler reference (linked above).

We require C++11 but optionally enable C++14. Our local systems don't currently have access to the new Intel oneAPI compilers, but it looks like we may be able to download...