Evan Ramos

I believe we should migrate away from the current system of `src/arch/*/conv-mach*.sh` files, moving their functionality to conditionals inside CMakeLists.txt and configure.ac that emit equivalent contents to the generated `conv-mach-opt.sh`...

megacon, Converse pingpong and pingpong_multipairs, `examples/charm++/load_balancing/kNeighbor`, and `benchmarks/charm++/kNeighbor` are tests that take more than a minute on our MPI Linux SMP CI.

I rebased this from the pieglobals-testing branch onto the main branch for review.

Strange segfault in the Spack build. Unfortunately it doesn't show the most important part of the backtrace. ``` 22403 [1:15] bcast_subset_async_delayed 0x5555557717fd CsdScheduleForever 22404 [1:16] bcast_subset_async_delayed 0x555555771a05 CsdScheduler 22405 [1:17]...

The issue was actually a bug in my reimplementation. By default, `std::priority_queue` sorts in descending order, whereas we want to sort the time field in ascending order. I fixed that...

> Perhaps we should separate this PR into mac-arm64 support and merge that, then open a separate issue/PR for mpi-arm64 and ppc64le support Merging Apple Silicon support first is a...

I've been able to look into the mpi-linux-arm8 issue under Asahi Linux on an M1 Mac. A null pointer is dereferenced inside libc during a printf call from the tlsglobals...

Updated this with the macOS parts split out, now merged, and the docs updated. No movement on the failures.

I've debugged the AArch64 failures more and made the following observations: - If I remove all (f)printf statements from the privatization test, all tests pass. This even includes pieglobals with...

#3743 does the cleanup necessary to test CMI_IO_BUFFER_EXPLICIT on netlrts. The tests are able to pass, but they are frequently obstructed by an unrelated error very similar to what I...