I periodically get these on the trunk:
--
alloc-mem: connect/btl_openib_connect_udcm.c:1850: udcm_cq_event_dispatch:
Assertion `((void *)0) != m && ((void *)0) != m->cm_channel' failed.
alloc-mem: connect/btl_openib_connect_udcm.c:1850: udcm_cq_event_dispatch:
Assertion `((void *)0) != m &&
Must be happening at teardown? The assertion was there for debugging purposes.
I will change it to just return if either the context or channel are NULL.
-Nathan
On Friday, July 13, 2012 8:37 AM, devel-boun...@open-mpi.org
[devel-boun...@open-mpi.org] on behalf of Jeff Squyres [jsquy...@cisco.c
Don't forget to CMR to v1.7.
On Jul 13, 2012, at 11:14 AM, wrote:
> Author: hjelmn (Nathan Hjelm)
> Date: 2012-07-13 11:14:48 EDT (Fri, 13 Jul 2012)
> New Revision: 26790
> URL: https://svn.open-mpi.org/trac/ompi/changeset/26790
>
> Log:
> remove assertion in udcm
>
> Text files modified:
>
I don't see one. Probably should have some entry in the "building" area
that describes their use.
On Jul 12, 2012, at 12:30 PM, Nathan Hjelm wrote:
> I wouln't consider sourced variables being overritten by the sourcing
> platform file a problem. I can update the platform file documentation t
Is elan toast? I see on the wiki that it is "removed from trunk" but I see it
in both the trunk and the v1.7 branch.
-Nathan
On Jul 11, 2012, at 12:47 PM, Alex Margolin wrote:
> I'm not sure if anyone remembers, but I was working on Open MPI support for
> MOSIX in the form of several MCA modules (turned out to be BTL, ODLS, and
> RAS). It's pretty much finished now, thanks to your help (I got many useful
> tips and c
On Jul 12, 2012, at 12:04 PM, Paul Kapinos wrote:
> (cross-post to 'users' and 'devel' mailing lists)
Sorry for the delay in replying here; I got slammed with some deadlines this
week...
The short version is that the issue has been confirmed. One root cause is
Mellanox significantly decreasin
On Jul 12, 2012, at 12:04 PM, Paul Kapinos wrote:
> a long time ago, I reported about an error in Open MPI:
> http://www.open-mpi.org/community/lists/users/2012/02/18565.php
>
> Well, in the 1.6 the behaviour has changed: the test case don't hang forever
> and block an InfiniBand interface, but
