Re: [OMPI devel] configure patch

Hi Jeff, * Jeff Squyres wrote on Fri, Jan 25, 2008 at 01:00:23AM CET: > On Jan 19, 2008, at 3:38 AM, Ralf Wildenhues wrote: > > >This patch will not work with Solaris sh. Please consider using this > >instead, it should avoid the issues with both shells. (Untested.) > > Ralf -- the original pa

Re: [OMPI devel] Hierarchical Collectives Query

Rolf, Whoowh! That's actually good news, since in our own tests hierarch is always slower. But this might be due to various reasons, including the fact, that we only have two cores per node. BTW: I actually would expect IMB test to have worse performance for hierarch compared to many other be

Re: [OMPI devel] configure patch

On Jan 19, 2008, at 3:38 AM, Ralf Wildenhues wrote: This patch will not work with Solaris sh. Please consider using this instead, it should avoid the issues with both shells. (Untested.) Ralf -- the original patch changed both ompi_get_version.m4 and ompi_get_version.sh; your patch only ch

[OMPI devel] Hierarchical Collectives Query

I am curious if anyone is doing any work currently on the hierarchical collectives. I ask this because I just did some runs on a cluster made up of 4 servers with 4 processors per server. I used TCP over IB. I was running with np=16 and using the IMB benchmark to test MPI_Bcast. What I am

Re: [OMPI devel] Fwd: === CREATE FAILURE ===

Correct. Still looking into that problem -- we'll likely .ompi_ignore the v pml tonight. On Jan 24, 2008, at 6:20 PM, Rolf Vandevaart wrote: Hi Jeff: Works for me. I downloaded both the gz and bz2 formats. Unzipped and untarred using Solaris tar on sparc and all looks good. No issues

Re: [OMPI devel] Fwd: === CREATE FAILURE ===

Hi Jeff: Works for me. I downloaded both the gz and bz2 formats. Unzipped and untarred using Solaris tar on sparc and all looks good. No issues with the long filename. I also ran configure and saw this error: config.status: error: cannot find input file: ompi/mca/pml/v/vprotocol/pessimist

Re: [OMPI devel] Fwd: === CREATE FAILURE ===

Rolf -- I've made tarballs in the ustar format; can you give them a whirl? http://www.open-mpi.org/~jsquyres/unofficial/ On Jan 24, 2008, at 5:07 PM, Rolf Vandevaart wrote: Jeff Squyres wrote: On Jan 24, 2008, at 4:00 PM, Ralf Wildenhues wrote: Automake forces v7 mode so that Solaris

[OMPI devel] Svn password issues with MTT

Hi, Jeff, Ethan Mallove has been helping me with an MTT setup, but it appears that I now have an MTT svn password issue. Do you know who I should contact for this? Is it Ethan? He may be in a meeting or maybe is busy, but I'm curious. I didn't see a place to apply for a password... --Jenny

Re: [OMPI devel] Fwd: === CREATE FAILURE ===

Jeff Squyres wrote: On Jan 24, 2008, at 4:00 PM, Ralf Wildenhues wrote: Automake forces v7 mode so that Solaris tar can untar the tarball, IIRC. You can choose how it should behave:

Re: [OMPI devel] Fwd: === CREATE FAILURE ===

My question before is still relevant: can Solaris untar tarballs that are in the POSIX format? If not, finding a tar that makes POSIX tarballs is moot. On Jan 24, 2008, at 4:44 PM, Paul H. Hargrove wrote: IMHO shortening the names now just delays dealing with the problem. I see 2 possibl

Re: [OMPI devel] Fwd: === CREATE FAILURE ===

IMHO shortening the names now just delays dealing with the problem. I see 2 possible approaches to a long-term fix: 1) add some machinery on the build system to complain when a filename would exceed 99 chars (before the file gets into svn). 2) probe at configure time for a $TAR command that pro

Re: [OMPI devel] Fwd: === CREATE FAILURE ===

Ok, I'll change the filenames. Aurelien Le 24 janv. 08 à 15:43, Brian W. Barrett a écrit : Automake forces v7 mode so that Solaris tar can untar the tarball, IIRC. Brian On Thu, 24 Jan 2008, Aurélien Bouteiller wrote: According to posix, tar should not limit the file name length. Only t

Re: [OMPI devel] Fwd: === CREATE FAILURE ===

On Jan 24, 2008, at 4:00 PM, Ralf Wildenhues wrote: Automake forces v7 mode so that Solaris tar can untar the tarball, IIRC. You can choose how it should behave:

Re: [OMPI devel] Fwd: === CREATE FAILURE ===

Hello, * Brian W. Barrett wrote on Thu, Jan 24, 2008 at 09:43:12PM CET: > Automake forces v7 mode so that Solaris tar can untar the tarball, IIRC. You can choose how it should behave:

Re: [OMPI devel] Fwd: === CREATE FAILURE ===

Automake forces v7 mode so that Solaris tar can untar the tarball, IIRC. Brian On Thu, 24 Jan 2008, Aurélien Bouteiller wrote: According to posix, tar should not limit the file name length. Only the v7 implementation of tar is limited to 99 characters. GNU tar has never been limited in the num

Re: [OMPI devel] Fwd: === CREATE FAILURE ===

According to posix, tar should not limit the file name length. Only the v7 implementation of tar is limited to 99 characters. GNU tar has never been limited in the number of characters file names can have. You should check with tar --help that tar on your machine defaults to format=gnu or f

Re: [OMPI devel] Fwd: === CREATE FAILURE ===

I'm trying to replicate and getting a lot of these: tar: openmpi-1.3a1r17212M/ompi/mca/pml/v/vprotocol/mca/vprotocol/ pessimist/vprotocol_pessimist_sender_based.c: file name is too long (max 99); not dumped tar: openmpi-1.3a1r17212M/ompi/mca/pml/v/vprotocol/mca/vprotocol/ pessimist/vprotocol_

Re: [OMPI devel] Fwd: === CREATE FAILURE ===

We cannot reproduce this one. A simple "make checkdist" exit long before doing anything in the ompi directory. It is difficult to see where exactly it fails, but it is somewhere in the opal directory. I suspect the new carto framework ... Thanks, george. On Jan 24, 2008, at 7:12 AM,

Re: [OMPI devel] dropping a pls module into an Open MPI build

Appreciate the clarification. I am unaware of anyone attempting that procedure in the past, but I'm not terribly surprised to hear it would encounter problems and/or fail. Given the myriad of configuration options in the code base, it would seem almost miraculous that you could either (a) hit the s

Re: [OMPI devel] dropping a pls module into an Open MPI build

I'm sorry, but now I am totally confused. Are you saying that you are having problems with the default rsh component in the distributed 1.2.3 code?? Yes ... Or are you having a problem with your customized version? and yes. Each exhibited the same problem, a bus error. What compiler are

[OMPI devel] more problems with v pml

If these problems can't be fixed today, I think the v pml should be .ompi_ignore'd again. ./configure --prefix=/home/jsquyres/bogus --enable-mpirun-prefix-by- default make[4]: Entering directory `/home/jsquyres/svn/ompi/ompi/mca/pml/v/ vprotocol/pessimist' make[4]: Nothing to be done for `

[OMPI devel] Fwd: === CREATE FAILURE ===

Aurelien -- Can you fix please? Last night's tests didn't run because of this failure. Begin forwarded message: From: MPI Team Date: January 23, 2008 9:13:30 PM EST To: test...@open-mpi.org Subject: === CREATE FAILURE === Reply-To: de...@open-mpi.org ERROR: Command returned a non-zero