Aurelien Bouteiller

problem initially manifest in OpenMPI https://github.com/open-mpi/ompi/pull/6784

@awlauria, @azat the issue is still present, but it manifests in Open MPI, a chunky piece of code. I will need some time to produce a simpler reproducer. The general...

As the large banner on your output indicates, using ULFM with RMA windows is very experimental at this point. We have had success in the past running some code using...

The 'node-died' issue appears to be related to `prted` not finding its own files `share/prte/help-errmgr-base.txt` Are you using an internal prte? (it is indicated in the final lines of 'configure'...

The root cause for the missing help message is here https://github.com/openpmix/prrte/issues/1360

I have the same problem on RHEL 7

No these were bona fide typos. The reason the PR is dependent upon the other two is that it documents functions that are introduced in them.

In my own testings I get an error that looks related to coll_han. ``` [b00:45954:0:45954] Caught signal 11 (Segmentation fault: address not mapped to object at address 0x150) /home/bouteill/ompi/master.debug/ompi/mca/coll/han/../../../../../master/ompi/communicator/communicator.h: [...

I am running in non-singleton case for that test. The han_comm_create_new procedure has no error checking, which means that when the SPLIT_TYPE fails line 121 (there are dead procs) it...

Both old and new APIs will remain available within Open MPI itself for the time being. No need to emulate them on your end. The old API will not be...