[mpiwg-rma] RMA WG Meeting today

Joseph Schuchart schuchart at icl.utk.edu
Thu Jul 14 12:53:08 CDT 2022


All,

We went over the open issues in the MPI Forum github project for the 4.1 
cycle today. It seems that while there is a general consensus that some 
of these issues should be fixed rather sooner than later there also 
seems to be a lack of bandwidth to drive them. In particular, I have 
moved the issues surrounding atomics to 5.0 because the people driving 
them have moved on and the time left for 4.1 is short [2, 3]. If anyone 
wants to drive any of these issues for 4.1 please let me know, I'll be 
happy to assist :)

As mentioned in my earlier mail, I'd like to move to a weekly schedule. 
Not everyone was on the call today so we punted on it. Please let me 
know if a weekly meeting in that same slot won't work for you in general 
(there will always be conflicts from time to time, of course).

See you at the Forum meeting next week,
Joseph

[1] 
https://github.com/orgs/mpi-forum/projects/1/views/1?filterQuery=label%3Awg-rma%2Clabel%3Achap-rma
[2] https://github.com/mpi-forum/mpi-issues/issues/114
[3] https://github.com/mpi-forum/mpi-issues/issues/36

On 7/14/22 08:15, Joseph Schuchart wrote:
> The RMA WG will meet today at 10am Central / 11am Eastern / 5pm 
> Central European time. I'd like to see if we can move to a weekly 
> schedule so we have enough time to discuss all the open topics. Please 
> check if your schedule allows for meeting weekly in the same slot. I'd 
> like to go over the open items in the MPI 4.1 Github project to 
> prepare for next week's meeting:
>
> https://github.com/orgs/mpi-forum/projects/1/views/1?filterQuery=label%3Awg-rma%2Clabel%3Achap-rma 
>
>
> I created a PR from Jeff's draft on allowing MPI_Win_shared_query for 
> windows other than shared memory windows:
>
> https://github.com/mpiwg-rma/mpi-standard/pull/7
>
> As usual, the meeting info can be found at 
> https://github.com/mpiwg-rma/mpi-standard/wiki (ping me if you get a 
> 404 and I will add you to the github group).
>
> See you then
> Joseph



More information about the mpiwg-rma mailing list