-
Notifications
You must be signed in to change notification settings - Fork 864
WeeklyTelcon_20220712
Geoffrey Paulsen edited this page Jul 12, 2022
·
1 revision
- Dialup Info: (Do not post to public mailing list or public wiki)
- Austen Lauria (IBM)
- Brendan Cunningham (Cornelis Networks)
- David Bernhold (ORNL)
- Edgar Gabriel (UoH)
- Geoffrey Paulsen (IBM)
- Howard Pritchard (LANL)
- Joseph Schuchart
- Josh Fisher (Cornelis Networks)
- Josh Hursey (IBM)
- Todd Kordenbrock (Sandia)
- Tommy Janjusic (nVidia)
- Akshay Venkatesh (NVIDIA)
- Artem Polyakov (nVidia)
- Aurelien Bouteiller (UTK)
- Brandon Yates (Intel)
- Brian Barrett (AWS)
- Charles Shereda (LLNL)
- Christoph Niethammer (HLRS)
- Erik Zeiske
- Geoffroy Vallee (ARM)
- George Bosilca (UTK)
- Harumi Kuno (HPE)
- Hessam Mirsadeghi (UCX/nVidia)
- Jeff Squyres (Cisco)
- Joshua Ladd (nVidia)
- Marisa Roman (Cornelius)
- Mark Allen (IBM)
- Matias Cabral (Intel)
- Matthew Dosanjh (Sandia)
- Michael Heinz (Cornelis Networks)
- Nathan Hjelm (Google)
- Noah Evans (Sandia)
- Raghu Raja (AWS)
- Ralph Castain (Intel)
- Sam Gutierrez (LLNL)10513
- Scott Breyer (Sandia?)
- Shintaro iwasaki
- Thomas Naughton (ORNL)
- William Zhang (AWS)
- Xin Zhao (nVidia)
-
v4.1.5
- Schedule: targeting ~6 mon (Nov 1)
- No driver on schedule yet.
-
Looks like 5 PRs are reviewed and ready to be merged (https://github.com/open-mpi/ompi/pulls?q=is%3Apr+is%3Aopen+base%3Av4.1.x+)
- William Would like merged PR#10513
-
Brendan summarized 3 CVEs in libevent 2.0.x series, that are fixed in newer versions of libevent.
- Two of the CVEs are almost certainly NOT built when Open MPI builds libevent.
- Third CVE is in some search function supporting search (haven't traced if we actually compile it in for Open MPI).
- Fixed in libevent 2.1.8
- libevent fixes will be needed on v4.1.x and v4.0.x
- Brian mentioned on RM Slack channel that upgrading libevent version would be best path
- Brendan will post PR to show how little change is needed...
- Balance between risk and reward.
-
Last week discussion: Would not want to try upgrading libevent to v2.1.8 for v4.1.x
- Fixes are contained to simple fixes (bounds/range checking fix)
- Cherry-picks is much less risky
- Decided Brendan will PR the small cherry-picks needed to v4.1.x and try v4.0.x.
- Schedule:
- Still Looking at end of summer
- PRRTE (longest blocker task) is targeting late summer.
- Still using Critical v5.0.x Issues (https://github.com/open-mpi/ompi/projects/3) yesterday
- Still Looking at end of summer
- Keeping an eye on PRRTE branch (ralph will create new v3 branch soon).
- When this happens we'll swap out submodule pointer.
- OMPI
main
prrte submodule pointer is already pointing to prrtemain
- Sessions PR fix merged to v5.0.x
- Ralph is deleting PRRTE v2.1 branch, and create a new release branch off of PRRTE
master
- https://groups.google.com/g/pmix/c/6bT-mQq4Sqs/m/8Y0NsSbeBgAJ
- Requires OMPIx v4.1.2
- new branch not yet posted to PRRTE, but once it does Austen will PR a OMPI v5.0.x submoule pointer update.
- Joseph is trying to do HAN / Adapt runs.
-
main
branch. - x86 with infiniband.
- Sometimes the startup hangs (really annouing).
- Don't know if it's a prrte run hang?
- Tried to run with GDB/DDT, but of course then it doesn't hang.
- He'll try to see if it's before MPI_Init in PRRTE or something else.
-
- sm_cuda component was moved into framework.
- nVidia has some issues building, and will try again to test
- 10492 and link to 10487
- Geoff STILL needs to test on ppc64le, will get to this, THIS week.
- Joseph will post some additional info thing in the ticket
- we're to review the admin steering committee in July (per our rules):
- we're to review the technical steering committee in July (per our rules):
- We should also review all the OMPI github, slack, and coverity members during the month of July.
- Jeff will kick that off sometime this week or next week.
- In the call we mentioned this, but no real discussion.
- Wiki for face to face: https://github.com/open-mpi/ompi/wiki/Meeting-2022
- Might be better to do a half-day/day-long virtual working session.
- Due to company's travel policies, and convenience.
- Could do administrative tasks here too.
- Might be better to do a half-day/day-long virtual working session.