Ah, yes - I see what you mean. Thanks for catching it!
On Oct 22, 2009, at 7:22 PM, George Bosilca wrote:
If you go at the top of the configure output, basically just after
the detection of the revision, you will see this warning. Somehow,
it doesn't stop configure from working, but it preve
If you go at the top of the configure output, basically just after the
detection of the revision, you will see this warning. Somehow, it
doesn't stop configure from working, but it prevent it from doing what
it was supposed to do as the two conditionals are having the wrong
values.
geor
I just fixed it in r22130.
On Thu, Oct 22, 2009 at 9:16 PM, Ralph Castain wrote:
> Most interesting - I have been building on Mac OSX both yesterday and today
> with those changes without problem. I am on Snow Leopard, but I checked and
> "true" is indeed in /usr/bin.
>
> I'm not seeing any warni
Most interesting - I have been building on Mac OSX both yesterday and
today with those changes without problem. I am on Snow Leopard, but I
checked and "true" is indeed in /usr/bin.
I'm not seeing any warnings or problems. Perhaps a difference in
configuration? Though I did also test it wit
There seems to be an issue with the latest changes on the configure
scripts. A careful reading of the output of configure on MAC OS X
shows a lot of errors/warnings, which leads to undefined
AM_CONDITIONALS (PROJECT_OMPI_*). This apparently comes from
configure.ac line 62 where the path to
Chris Samuel wrote:
- "Ashley Pittman" wrote:
$ grep Cpus_allowed_list /proc/$$/status
Useful, ta!
Does this imply the default is to report on processes
in the current cpuset rather than the entire system?
Does anyone else feel that violates the principal of
least surprise?
Not reall
- "Ashley Pittman" wrote:
> $ grep Cpus_allowed_list /proc/$$/status
Useful, ta!
> Does this imply the default is to report on processes
> in the current cpuset rather than the entire system?
> Does anyone else feel that violates the principal of
> least surprise?
Not really, I feel that
FYI.
Begin forwarded message:
From: "Kim, DongInn"
Date: October 22, 2009 1:46:47 PM EDT
Subject: SVN/trac server needs to reboot, Oct 23, 8am
We need to reboot SVN/trac server(sourcehaven) to replace RAM on Oct
23, Friday.
Date: Friday Oct 23, 2009.
Time:
- 5:00am-5:30am Pacific US time
-
Thanks! Ethan has committed the code to svn if other sites want to
enable it, it might be worth considering having one of the test sites
running debug code to be able to see variable values inside the MPI
library.
It also appears that message queue support is broken on the head, the
DLL is compl
That rocks!!
On Oct 22, 2009, at 6:19 AM, Ashley Pittman wrote:
Thanks to Sun who have integrated padb with MTT there are now padb
traces in the automated testing logs of OpenMPI for cases where the
tests have timed out. These traces show the contents of MPI message
queues, stack traces and w
The code for the Job to Core Binding (aka. thread binding, or CPU
binding) feature was checked into the Grid Engine project cvs. It uses
OpenMPI's Portable Linux Processor Affinity (PLPA) library, and is
topology and NUMA aware.
The presentation from HPC Software Workshop '09:
http://wikis.sun.com
I added tickets #21 and #22 about these features.
https://svn.open-mpi.org/trac/hwloc/ticket/21
https://svn.open-mpi.org/trac/hwloc/ticket/22
Thanks!
On Oct 22, 2009, at 5:54 AM, Ashley Pittman wrote:
On Thu, 2009-10-22 at 11:05 +0200, Brice Goglin wrote:
> Ashley Pittman wrote:
> >
Paul H. Hargrove wrote:
Sorry if this has been fixed for 1.3.4, but in the manpge for mpirun
in 1.3.3 I noticed the following in the "MCA" section of the manpage:
Note: The -mca switch is simply a shortcut for setting
environment variables. The same
effect may be accomplis
Thanks to Sun who have integrated padb with MTT there are now padb
traces in the automated testing logs of OpenMPI for cases where the
tests have timed out. These traces show the contents of MPI message
queues, stack traces and where available local variables and parameters
to functions.
An exam
On Thu, 2009-10-22 at 11:05 +0200, Brice Goglin wrote:
> Ashley Pittman wrote:
> > Does this imply the default is to report on processes in the current
> > cpuset rather than the entire system? Does anyone else feel that
> > violates the principal of least surprise?
> Yes, by default, it's the c
Ashley Pittman wrote:
>> [csamuel@tango069 ~]$ ~/local/hwloc/0.9.1rc2/bin/lstopo
>> System(31GB)
>> Node#0(15GB) + Socket#0 + L3(6144KB) + L2(512KB) + L1(64KB) + Core#0 + P#0
>> Node#1(16GB) + Socket#1 + L3(6144KB)
>> L2(512KB) + L1(64KB) + Core#0 + P#4
>> L2(512KB) + L1(64KB) + Core#1
On Thu, 2009-10-22 at 10:37 +1100, Chris Samuel wrote:
> - "Chris Samuel" wrote:
>
> > Some sample results below for configs not represented
> > on the current website.
>
> A final example of a more convoluted configuration with
> a Torque job requesting 5 CPUs on a dual Shanghai node
> and
- "Tony Breeds" wrote:
> Powerpc kernels that old do not have the topology information needed
> (in /sys or /proc/cpuinfo) So for the short term that's be best we
> can do.
That's fine, I quite understand. I'm trying to get that
cluster replaced anyway.. ;-)
> FWIW I'm looking at how we
Sorry if this has been fixed for 1.3.4, but in the manpge for mpirun in
1.3.3 I noticed the following in the "MCA" section of the manpage:
Note: The -mca switch is simply a shortcut for setting
environment variables. The same
effect may be accomplished by setting correspon
19 matches
Mail list logo