coll/libnbc: fix NBC_Unpack() #5680
Conversation
|
bot:ompi:retest |
|
@ggouaillardet As you pointed out in #5643 (comment), it might be worth putting in the commit message that this is a real bug, but it's in an algorithm that was implemented but not used (i.e., it's not a NEWS-worthy comment, nor anything for users to worry about -- and not even worth PR'ing to any of the current release branches). |
|
OS X CI keeps failing: |
|
Confirmed: OS X CI "piper" is having networking problems at the moment. Ok to merge this PR since it was the only failure. |
|
@jsquyres what? How do you know this doesn't break OS X? seems much better idea to retest. |
|
Ok, sorry -- do we have an ETA when OS X CI will be back? |
|
@bwbarrett I understand your concern, but you can't ask everything to hit a dead stop because the OSX tester is hung/down with no prognosis for recovery. Some of us are quite capable of building on OSX and can see that nothing was broken - no sense in holding everything up in such cases. |
always initialize 'size'. Only the a2a_sched_diss() alltoall algorithm is impacted, and this algo is currently unused, so there is no need to backport nor update the NEWS file for now. Signed-off-by: Gilles Gouaillardet <[email protected]>
ggouaillardet
force-pushed
the
topic/nbc_unpack
branch
from
ee5e037 to
ff48e92
Compare
|
@jsquyres per your comment, I updated the commit message. |
|
bot:ompi:retest |
always initialize 'size'
Signed-off-by: Gilles Gouaillardet [email protected]