-sbuild)root@hartree:/home/mba/build# mpirun /bin/true
Segmentation fault
The patch is attached.
Cheers,
Michael
--
Pavan Balaji
http://www.mcs.anl.gov/~balaji
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Conta
better to
use --with-device=ch3:sock which doesn't use shared memory at all.
-- Pavan
--
Pavan Balaji
http://www.mcs.anl.gov/~balaji
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
On 05/06/2012 08:28 PM, Samuel Thibault wrote:
Pavan Balaji, le Sun 06 May 2012 12:14:27 -0500, a écrit :
MPICH2 currently does not support native atomic primitives for gcc/sparc.
It needs to be built with the emulated atomics,
Mmm, really?
I just verified it with our OpenPA developer. We
primitives for
gcc/sparc. It needs to be built with the emulated atomics, which means
that the application needs to be linked with libopa as well. In
general, it is good to use pkg-config to see what all libraries are
required and link with all of them.
--
Pavan Balaji
http://www.mcs.anl.gov
332905598
Can you also report this bug to MPICH2 upstream, with the complete
configure log and other information requested in the README file?
https://trac.mcs.anl.gov/projects/mpich2/newticket
-- Pavan
--
Pavan Balaji
http://www.mcs.anl.gov/~balaji
--
To UNSUBSCRIBE, email
-help should give you the usage instructions. Or you
can find them here.
http://wiki.mcs.anl.gov/mpich2/index.php/Using_the_Hydra_Process_Manager
-- Pavan
--
Pavan Balaji
http://www.mcs.anl.gov/~balaji
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of
--
Pavan Balaji
http://www.mcs.anl.gov/~balaji
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
On 01/12/2012 10:27 AM, Pavan Balaji wrote:
For the inter-library dependencies, I'd not like to have a Debian-only
patch. But you are right, libtool should be able to disable
dependencies on platforms it doesn't support. Let me look into this
some more.
Hmm.. In 1.5.x, it loo
some more.
Btw, the destdir patch should not be required in 1.5.x.
-- Pavan
--
Pavan Balaji
http://www.mcs.anl.gov/~balaji
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
/libtool/manual/html_node/Inter_002dlibrary-dependencies.html
Note that this is not portable, so there'll need to be a configure
option to enable it for platforms that libtool supports it on.
Or did you mean something else?
Thanks,
-- Pavan
--
Pavan Balaji
http://www.mcs.anl.gov/~balaji
g than standard binutils.
Feel free to ignore this, but at some point this will become a
release-critical bug.
-Adam
On Thu, 2011-12-29 at 23:44 -0600, Pavan Balaji wrote:
The best method is to use pkg-config to find what libraries need to be
linked in. Currently, this list is libmpich, libmpl
to
see them all).
It's interesting that this works fine on other platforms, that it either
ignores or manages to find libmpl... But in general, it's best to link
to all of the libraries whose symbols you use, I'm pretty sure it's also
a requirement of binutils-gold.
Thanks,
A
evice to ch3:sock (using the configure option
--with-device=ch3:sock), which doesn't use shared memory at all and
hence doesn't need OPA atomics.
3. You can contribute m68k processor atomics to OPA, so it is natively
supported and OPA doesn't have to fall back to locks.
-- Pa
compiler (g++, I assume; which version), and the exact configure
flags used?
https://trac.mcs.anl.gov/projects/mpich2/newticket
-- Pavan
--
Pavan Balaji
http://www.mcs.anl.gov/~balaji
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubs
On 10/24/2010 11:55 AM, Pavan Balaji wrote:
On 10/24/2010 11:42 AM, Lucas Nussbaum wrote:
On 23/10/10 at 12:51 -0500, Pavan Balaji wrote:
On 10/23/2010 12:40 PM, Lucas Nussbaum wrote:
On 11/11/09 at 10:15 -0600, Pavan Balaji wrote:
We had a short discussion on this within the MPICH group
On 10/24/2010 11:42 AM, Lucas Nussbaum wrote:
On 23/10/10 at 12:51 -0500, Pavan Balaji wrote:
On 10/23/2010 12:40 PM, Lucas Nussbaum wrote:
On 11/11/09 at 10:15 -0600, Pavan Balaji wrote:
We had a short discussion on this within the MPICH group. We believe
that we will be able to modify the
On 10/23/2010 12:40 PM, Lucas Nussbaum wrote:
On 11/11/09 at 10:15 -0600, Pavan Balaji wrote:
We had a short discussion on this within the MPICH group. We believe
that we will be able to modify the code such that enabling it at
configure time should not add any additional performance overhead
available for
people who want to configure with mpd using --with-pm=mpd). The timeline
for releasing 1.3 is late summer unfortunately (1.3a1 is already out).
-- Pavan
--
Pavan Balaji
http://www.mcs.anl.gov/~balaji
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with
Manuel Prinz wrote:
Am Donnerstag, den 08.04.2010, 18:22 -0500 schrieb Pavan Balaji:
We were having a discussion about a common ABI in the context of MPI-3.
But that working group fizzled out. I'll check on what exactly happened
with it in the next meeting (in May). Even if the MPI sta
t
specify ABI compatibility, I'll talk to the Open MPI folks to see if
MPICH2 and Open MPI do whatever the ABI working group was proposing
outside of the standard, which might be sufficient for you guys.
-- Pavan
--
Pavan Balaji
http://www.mcs.anl.gov/~balaji
--
To UNSUBSCRIBE, email
Lucas Nussbaum wrote:
On 06/04/10 at 21:01 -0500, Pavan Balaji wrote:
By default, mpi.h should installed in $prefix/include, which seems
to be what you need (prefix = /usr/lib/mpich2). So, unless you give
an --includedir option to configure, this should work the way you
expect it to. Maybe I&#
, this should work the way you expect it
to. Maybe I'm not understanding the problem correctly here?
-- Pavan
--
Pavan Balaji
http://www.mcs.anl.gov/~balaji
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble?
let me know what you
think about that.
Best regards,
Manuel
--
Pavan Balaji
http://www.mcs.anl.gov/~balaji
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
On 02/24/2010 08:01 AM, Pavan Balaji wrote:
> Hi Lucas,
>
> On 02/24/2010 05:10 AM, Lucas Nussbaum wrote:
>>> % mpicc.mpich2 -show
>>> cc -g -O2 -g -Wall -O2 -I/usr/include/mpich2 -L/usr/lib -L/usr/lib
>>> -Wl,-rpath,/
itself, unless the user sets it as LDFLAGS.
What's the exact configure you are using?
-- Pavan
--
Pavan Balaji
http://www.mcs.anl.gov/~balaji
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
We had a short discussion on this within the MPICH group. We believe
that we will be able to modify the code such that enabling it at
configure time should not add any additional performance overhead for
applications that don't need this support. So starting mpich2-1.3, you
should be able
ode by a
runtime option, as opposed to a compile-time option. This will still
hurt performance a little, but might be a lot more manageable. Let me
check with the developer who is adding this support and see what he thinks.
-- Pavan
--
Pavan Balaji
http://www.mcs.anl.gov/~balaji
--
To UNSUB
t for any MPI
implementation. For example, Open MPI uses our MPI I/O library, so they
have the same problem too.
-- Pavan
--
Pavan Balaji
http://www.mcs.anl.gov/~balaji
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". T
should work
for all MPI implementations including MPICH2, LAM, OpenMPI, MVAPICH,
what-have-you.
-- Pavan
--
Pavan Balaji
http://www.mcs.anl.gov/~balaji
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
nks,
-- Pavan
--
Pavan Balaji
http://www.mcs.anl.gov/~balaji
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
9 12:41 AM, Lucas Nussbaum wrote:
On 15/06/09 at 16:32 -0500, Pavan Balaji wrote:
Hello,
I'm one of the developers of MPICH at Argonne National Laboratory. As
some of you might be aware, our current focus is mainly on the second
generation of MPICH, i.e., MPICH2. I noticed that curren
31 matches
Mail list logo
