Done
> On Jun 10, 2015, at 1:55 PM, Steve Wise wrote:
>
> If you're trying to release 1.8.6, I recommend you revert the change that
> breaks things until we can figure this out.
>
>> -Original Message-
>> From: Jeff Squyres (jsquyres) [mailto:jsquy...@cisco.com]
>> Sent: Wednesday, Jun
If you're trying to release 1.8.6, I recommend you revert the change that
breaks things until we can figure this out.
> -Original Message-
> From: Jeff Squyres (jsquyres) [mailto:jsquy...@cisco.com]
> Sent: Wednesday, June 10, 2015 3:30 PM
> To: Open MPI User's List
> Cc: Nathan Hjelm; St
Nathan / Steve -- you guys are nominally the owners of the openib BTL: can you
please investigate?
> On Jun 10, 2015, at 4:15 PM, Ralph Castain wrote:
>
> Odd - without that setting, the value is essentially undefined, so it’s hard
> to understand how that is any better. Maybe the whole align
Odd - without that setting, the value is essentially undefined, so it’s hard to
understand how that is any better. Maybe the whole alignment thing is busted,
and leaving it undefined (which usually defaults to zero, but not always)
causes it to be turned “off”?
I don’t really care, mind you - b
Ralph --
This change was not correct
(https://github.com/open-mpi/ompi/commit/ce915b5757d428d3e914dcef50bd4b2636561bca).
It is causing memory corruption in the openib BTL.
> On May 25, 2015, at 11:56 AM, Ralph Castain wrote:
>
> I don’t see a problem with it. FWIW: I’m getting ready to rel
Gilles,
I installed the 4.8 series of compilers on the Pi 2 (gcc, g++, and
gfortran).
I did a fresh untar of openmpi-1.8.5. I set the environment variables
CC, CXX, and FC to point to the 4.8 compilers. However, the build failed
again.
Making all in config
make[1]: Entering directory '/work/pi
Just fixed it - thanks!
> On Jun 10, 2015, at 10:31 AM, Nathan Hjelm wrote:
>
>
> Looks like a bug in the CM optimization that went into 1.10 yesterday. I
> will let the author know to fix this one.
>
> -Nathan
>
> On Wed, Jun 10, 2015 at 11:19:04AM +0200, Siegmar Gross wrote:
>> Hi Nathan,
Looks like a bug in the CM optimization that went into 1.10 yesterday. I
will let the author know to fix this one.
-Nathan
On Wed, Jun 10, 2015 at 11:19:04AM +0200, Siegmar Gross wrote:
> Hi Nathan,
>
> > Bah. Another proc->super usage. Thought I got them all. Will fix now.
>
> Today I tried t
On Jun 10, 2015, at 6:36 AM, Cian Davis wrote:
>
> An old version of software I use (FDS 5.5.3 in this case - a CFD solver for
> fire) was compiled against LAM-MPI. It looks for the particular executables
> that came with LAM-MPI and looks for a running lamd.
Ah, bummer.
> I tried a while ago
> On Jun 10, 2015, at 6:27 AM, Jeff Squyres (jsquyres)
> wrote:
>
>> i just found an other issue with this scenario :
>> symlinks in the profile directories (ompi/mpi/c/profile,
>> ompi/mpi/fortran/mpif-h/profile, oshmem/shmem/c/profile) are not recreated
>> and points to the previous source t
it is a bit different for those links since they are created only if they
do not exist.
I guess thede links should be recreated if they are older than Makefile
btw, does make check for the date of the symlink or the linked file ?
On Wednesday, June 10, 2015, Jeff Squyres (jsquyres)
wrote:
> On
An old version of software I use (FDS 5.5.3 in this case - a CFD solver for
fire) was compiled against LAM-MPI. It looks for the particular executables
that came with LAM-MPI and looks for a running lamd. I tried a while ago to
symlink against more current OpenMPI distributions without success (I w
On Jun 10, 2015, at 12:00 AM, Gilles Gouaillardet wrote:
>
> that can happen indeed, in a complex but legitimate environment :
>
> mkdir ~/src
> cd ~/src
> tar xvfj openmpi-1.8.tar.bz2
> mkdir ~/build/openmpi-v1.8
> cd ~/build/openmpi-v1.8
> ~/src/openmpi-1.8/configure
> make
>
> then a few day
On Jun 10, 2015, at 1:35 AM, Siegmar Gross
wrote:
>
>> I don't see any reason why this should be happening to you only
>> sometimes; this code has been unchanged in *forever*. :-(
>
> It only happens since openmpi-v1.8.5-40-g7b9e672 which I tried to
> build June 1th and on all versions followi
Just curious -- is there a reason you can't upgrade?
I ask because:
- LAM/MPI has been dead for nearly a decade.
- Open MPI is source compatible with LAM/MPI.
> On Jun 10, 2015, at 5:07 AM, Cian Davis wrote:
>
>
> Hi All,
> While OpenMPI is the way forward, there is a fair amount of softwar
Hi Nathan,
> Bah. Another proc->super usage. Thought I got them all. Will fix now.
Today I tried to build openmpi-v1.10-dev-70-g5fefddb and I still get an
error.
CC pml_cm.lo
In file included from
../../../../../openmpi-v1.10-dev-70-g5fefddb/ompi/mca/pml/cm/pml_cm.c:25:0:
../../../../..
Hi,
you can find some source rpms for various distros
for example at http://rpms.famillecollet.com/rpmphp/zoom.php?rpm=lam
Cheers,
Gilles
On 6/10/2015 6:07 PM, Cian Davis wrote:
Hi All,
While OpenMPI is the way forward, there is a fair amount of software
out there still compiled against LAM
Hi All,
While OpenMPI is the way forward, there is a fair amount of software out
there still compiled against LAM-MPI. As Indiana University has stopped
hosting (http://blogs.cisco.com/performance/a-farewell-to-lammpi), there is
no easy way to get the sources.
I was wondering if someone know of so
Hi Gilles,
> then i cannot figure out how this can happen ...
>
> i see you remove the install dir (/usr/local/...)
> but i do not see you removing the build directory :
>
> i see
>
> mkdir openmpi-1.8.6-${SYSTEM_ENV}.${MACHINE_ENV}.64_gcc
> cd openmpi-1.8.6-${SYSTEM_ENV}.${MACHINE_ENV}.64_gcc
Siegmar,
then i cannot figure out how this can happen ...
i see you remove the install dir (/usr/local/...)
but i do not see you removing the build directory :
i see
mkdir openmpi-1.8.6-${SYSTEM_ENV}.${MACHINE_ENV}.64_gcc
cd openmpi-1.8.6-${SYSTEM_ENV}.${MACHINE_ENV}.64_gcc
but i do not see
Hi Gilles,
> a simple workaround is you always run configure in an empty directory
> in this example, please recreate
>
> rm -rf /export2/src/openmpi-1.8.6/openmpi-1.8.6-Linux.x86_64.64_gcc
> mkdir /export2/src/openmpi-1.8.6/openmpi-1.8.6-Linux.x86_64.64_gcc
>
> before invoking configure and mak
Hi Jeff,
> I don't see any reason why this should be happening to you only
> sometimes; this code has been unchanged in *forever*. :-(
It only happens since openmpi-v1.8.5-40-g7b9e672 which I tried to
build June 1th and on all versions following-up which I tried
(openmpi-v1.8.5-43-g0dd996b, open
Siegmar,
a simple workaround is you always run configure in an empty directory
in this example, please recreate
rm -rf /export2/src/openmpi-1.8.6/openmpi-1.8.6-Linux.x86_64.64_gcc
mkdir /export2/src/openmpi-1.8.6/openmpi-1.8.6-Linux.x86_64.64_gcc
before invoking configure and make
Jeff,
that
23 matches
Mail list logo
