Hello Martin, Sorry to answer your message so late. There are a couple of things I'd like to address for 4.1 1- all the clean-up stuff w.r.t (now) chapter 7, as I presented during my reading in the last ratification meeting. This include (see attached document): * Add process topology functions that create communicators to the various lists throughout the chapter -> Issue #497, PR yet to create * Remove/improve text for callback-copy functions -> Issue #496, PR yet to create * Improve semantic index -> Issue #237, PR yet to create * Library examples to reformat -> Issue #229 (continued), PR yet to create * Process → MPI Process where relevant -> Issue #??? → PR#590 * Fix incorrect usage of ranks/tasks/etc -> Issue #474 → PR#566 2- I'd like to resurrect issue #154 (A new function to query hardware resource names, https://github.com/mpi-forum/mpi-issues/issues/154 ) with PR#143. If you recall, people were OK with this in Albuquerque on the substance but not on the interface. Using an MPI-defined new datastruct as a parameter was suggested. That's was I should have done a while ago and that is to be proposed. 4.1 seems a good target but I might be wrong on this. I'll do my best to attend today's meeting but I will have to leave for a while because of another engagement that I can't postpone nor defer. Thanks, Guillaume On 7/14/21 3:42 PM, Martin Schulz via mpi-forum wrote:
Hi all,
As mentioned last week, we'll pick up our Virtual MPI Forum Meetings again next week on July 21st.
As for the 21st: I would like to use the meeting to set up a (at least preliminary) goal-line for MPI 4.1 - we had several open items from the 4.0 process (and before), which will be the starting point. If there are any particular items you think we should address as part of a "clean up" or "minor functionality" please either send them to me and I will collect them or add them to the system as a blank ticket with the MPI 4.1 tag. Next week we'll consolidate the list and decide how we approach this.
Thanks!
Martin