Jeff, I must admit that I don't completely understand how your fix work.
Can you explan me why this veriant was failing:
CPPFLAGS="-I$srcdir/opal/libltdl/"
AC_EGREP_HEADER([lt_dladvise_init], [$srcdir/opal/libltdl/ltdl.h]
while the new one:
CPPFLAGS="-I$srcdir -I$srcdir/opal/libltdl/"
Hi Ralph,
No problem. My comment about the pull requests was that even if we know
something is currently
broken in master - resulting in jenkins failures for pull requests - I'd
prefer not to get into the mode
of just going on and merging the request because we know somethings broken
in master.
As for the checks before merge - I suspect this was done exactly that way, if I
am right as to the cause. The problem is that it only reproduces for certain
conditions and environments. When working in a system with such a wide range of
supported setups, you learn to live with the bad as well
> On Dec 3, 2014, at 10:10 AM, Jeff Squyres (jsquyres)
> wrote:
>
> Yes, I am able to replicate. There's some kind of dangling function pointer
> in the ORTE event loop during shutdown. It seems to be some kind of race --
> it only happens in some runs, not all (but
Hello Artem,
No, but I was also told by schedmd that the slurm we have on our systems is
ancient.
So I'm no longer considering this problem very important. We have a
workaround of always configuring
with --disable-dlopen.
Thanks,
Howard
2014-12-02 20:59 GMT-07:00 Artem Polyakov
Yes, I am able to replicate. There's some kind of dangling function pointer in
the ORTE event loop during shutdown. It seems to be some kind of race -- it
only happens in some runs, not all (but still pretty frequent).
Ralph is looking into it.
On Dec 3, 2014, at 1:03 PM, Howard Pritchard
We talked during the telecon about the user-reported issue where they asked for
knem support, it wasn’t available on the system, but we ran anyway at a reduced
performance level. The agreement we had was that OMPI should instead fail at
that point since the user had requested something we could
Thanks!
On Dec 3, 2014, at 7:03 AM, Artem Polyakov wrote:
>
>
> среда, 3 декабря 2014 г. пользователь Jeff Squyres (jsquyres) написал:
> They were equivalent until yesterday. :-)
> I see. Got that!
>
> I was going to file a PR to bring the changes over to v1.8, but not
среда, 3 декабря 2014 г. пользователь Jeff Squyres (jsquyres) написал:
> They were equivalent until yesterday. :-)
I see. Got that!
>
> I was going to file a PR to bring the changes over to v1.8, but not until
> they had shaken out on master.
>
> Would you mind filing a PR?
Sure, will do that
They were equivalent until yesterday. :-)
I was going to file a PR to bring the changes over to v1.8, but not until they
had shaken out on master.
Would you mind filing a PR?
On Dec 3, 2014, at 5:56 AM, Artem Polyakov wrote:
> I finally found the clear reason of this
I finally found the clear reason of this strange situation!
In ompi opal_setup_libltdl.m4 has the following content:
CPPFLAGS="-I$srcdir -I$srcdir/opal/libltdl"
AC_EGREP_HEADER([lt_dladvise_init], [opal/libltdl/ltdl.h],
[OPAL_HAVE_LTDL_ADVISE=1])
And in ompi-release
11 matches
Mail list logo