Hi Jon,
You can use MTT to do performance testing only. Here is a good
starting point for using MTT:
https://svn.open-mpi.org/trac/mtt/wiki/OMPITesting
Have you tried running trivial tests with your MPI installation? Were
you planning on submitting the results to the open-mpi.org database?
t's coming from Test::Analyze::Correctness. Somehow $pass came
> out to be 0, so it fell into the "fail" path...?
>
>
> On Apr 14, 2010, at 10:51 AM, Ethan Mallove wrote:
>
> > This loop_spawn test appears to have passed, though it's marked
> > as FAIL:
&
This loop_spawn test appears to have passed, though it's marked
as FAIL:
http://www.open-mpi.org/mtt/index.php?do_redir=1233
Its pass condition is set to:
pass = &and(&test_wifexited(), &eq(&test_wexitstatus(), 0))
-Ethan
I can do the below stuff in a seperate commit.
-Ethan
On Fri, Mar/05/2010 09:26:21PM, Jeff Squyres wrote:
> On Mar 5, 2010, at 2:05 PM, Ethan Mallove wrote:
>
> > The check is there now. Ready for review.
>
> Did you mean to include these parts?
>
> --- lib/MTT/Mess
On Fri, Feb/19/2010 12:00:55PM, Ethan Mallove wrote:
> On Thu, Feb/18/2010 04:13:15PM, Jeff Squyres wrote:
> > On Feb 18, 2010, at 10:48 AM, Ethan Mallove wrote:
> >
> > > To ensure there is never a collision between $a->{k} and $b->{k}, the
> > > user ca
On Thu, Feb/18/2010 04:13:15PM, Jeff Squyres wrote:
> On Feb 18, 2010, at 10:48 AM, Ethan Mallove wrote:
>
> > To ensure there is never a collision between $a->{k} and $b->{k}, the
> > user can have two MTT clients share a $scratch, but they cannot both
> > run the s
of the .dump files for way too long.
E.g., MTT::MPI::LoadInstalls happens very early in client/mtt, and an
hour could elapse before MTT::MPI::SaveInstalls is called in
Install.pm.
-Ethan
>
>
> On Feb 11, 2010, at 12:09 PM, Ethan Mallove wrote:
>
> > This apparently got lost in t
This apparently got lost in the shuffle a few months ago. The fix
allows one to kick off all of their MPI Installs and Test Builds in
parallel. Give it a try when you have a chance.
-Ethan
> On Sat, Nov/07/2009 04:15:42PM, Jeff Squyres wrote:
> > On Nov 6, 2009, at 5:18 PM, Etha
Hi Vishal,
The debug output shows that your MTT run aborted because you do not
have a username/passwd for the open-mpi.org/mtt/submit URL. To skip
the IU Database part, try running with --no-section iu.database, e.g.,
# cat ompi-core-perf-testing.ini | ../client/mtt --no-section iu.database
--
Hi Vishal,
This is an MTT question for mtt-us...@open-mpi.org (see comments
below).
On Tue, Dec/22/2009 03:54:08PM, vishal shorghar wrote:
>Hi All,
>
>I have one issue with MTT trivial tests.All tests are not getting
>passed,Please read below for detailed description.
>
>Today I
should be the next developer.ini change that I should be making ?
>
> The complete output and the developer.ini attached.
Oops. &join() should be &stringify().
"svn up" and try again.
-Ethan
>
> Thanks!
>
> Venkat
>
> -Original Message-
>
r.ini
-Ethan
>
> Attached are the new developer.ini and the /tmp/UqPeD61ett.ini files.
>
> Venkat
>
> -Original Message-
> From: mtt-users-boun...@open-mpi.org
> [mailto:mtt-users-boun...@open-mpi.org] On Behalf Of Ethan Mallove
> Sent: Thursday, November 19, 2009 8:
-
> From: mtt-users-boun...@open-mpi.org
> [mailto:mtt-users-boun...@open-mpi.org] On Behalf Of Ethan Mallove
> Sent: Wednesday, November 18, 2009 1:41 PM
> To: General user list for the MPI Testing Tool
> Subject: Re: [MTT users] MTT trivial tests fails to completeon
> Centos5.3-x8
Could you run with --debug (instead of --verbose), and send the
output.
Thanks,
Ethan
On Wed, Nov/18/2009 11:08:18AM, Venkat Venkatsubra wrote:
>
>
>
>
>From: Venkat Venkatsubra
>Sent: Wednesday, November 18, 2009 12:54 PM
>To: 'mtt-us...@open-mpi.org'
>Subject: MTT trivial te
On Sat, Nov/07/2009 04:15:42PM, Jeff Squyres wrote:
> On Nov 6, 2009, at 5:18 PM, Ethan Mallove wrote:
>
>> I'm running multiple MTT clients out of the same scratch directory
>> using SGE. I'm running into race conditions between the multiple
>> clients, where o
On Sat, Nov/07/2009 04:15:42PM, Jeff Squyres wrote:
> On Nov 6, 2009, at 5:18 PM, Ethan Mallove wrote:
>
>> I'm running multiple MTT clients out of the same scratch directory
>> using SGE. I'm running into race conditions between the multiple
>> clients, where o
On Mon, Mar/05/2007 09:42:53PM, MTT wrote:
> #212: Generic network locking server
> --+-
> Reporter: jsquyres | Owner: jsquyres
> Type: defect | Status: closed
On Wed, Nov/04/2009 03:56:10PM, Jeff Squyres wrote:
> On Nov 4, 2009, at 3:19 PM, Ethan Mallove wrote:
>
>> Hmmm. QLogic and IU are also running into the same issue:
>>
>> http://www.open-mpi.org/mtt/index.php?do_redir=1170
>>
>> I just filed a Trac ticket:
&
On Wed, Nov/04/2009 02:22:28PM, Jeff Squyres wrote:
> On Nov 4, 2009, at 1:51 PM, Ethan Mallove wrote:
>
>> I recently ran into the error "./configure: No such file or directory"
>> in my nightly test build for the ibm, mpicxx, and onesided test
>> suites:
&
I recently ran into the error "./configure: No such file or directory"
in my nightly test build for the ibm, mpicxx, and onesided test
suites:
http://www.open-mpi.org/mtt/index.php?do_redir=1169
This problem had something to do with autogen.sh being run in the Test
get section (different enviro
On Mon, Apr/13/2009 09:45:09AM, Barrett, Brian W wrote:
> Hi all -
>
> I can't find documentation on the test run block, but is there a way to set
> the max time a test can run?
>
In our test run sections we set the "timeout" param, e.g.,
timeout = &max(60, &multiply(20, &test_np()))
That w
On Mon, Apr/13/2009 09:38:27AM, Rafael Folco wrote:
> On Thu, 2009-04-09 at 16:05 -0400, Ethan Mallove wrote:
> > On Wed, Apr/08/2009 05:05:30PM, Rafael Folco wrote:
> > > Ethan, Jeff,
> > >
> > > Sorry for the perl confusion, nevermind...
> > >
>
t = MTT::Values::Functions::find_executables(".")
> &find_executables got .
> &find_exectuables returning: ./c_ring ./f77_ring ./f90_hello ./cxx_ring
>./f77_hello ./cxx_hello ./f90_ring ./c_hello
> *** ERROR: Module aborted: MTT::Test::Specify::Simple:Specify: Can
On Wed, Apr/08/2009 11:36:05AM, Rafael Folco wrote:
> Well, I took a look at /tmp/ompi-core-testers/lib/MTT/Values.pm line 75.
>
> This piece of code looks wrong to me:
>
> if ($#{@$ret} < 0) {
>
> $ret references an array
> @$ret points to the first element of this array
> $# returns the numbe
On Wed, Apr/08/2009 11:51:31AM, Rafael Folco wrote:
>
> I just forgot to mention that removing @ from "if ($#{@$ret} < 0) {" it
> worked fine.
>
That might be masking a problem in your INI file, but it is hard to
know for sure without seeing your INI file.
> On Wed, 2009-04-08 at 11:36 -0300,
On Tue, Apr/07/2009 03:53:48PM, Rafael Folco wrote:
> Hi,
>
> I'm trying to run MTT on SLES11, but I am getting an error message
> during the RUN phase and I can't figure out what is the problem.
>
> *** ERROR: Module aborted: MTT::Test::Specify::Simple:Specify: Can't use
> string ("183") as
On Sat, Mar/21/2009 12:36:38AM, Ji Jerry wrote:
>Hi,Admin!
>*
>I want to apply mtt testing account for test my ompi .thanks for help!
>*
>Jesse
>2008/03/10
Hi Jesse,
Just curious - what compilers and platforms were you planning on
testing Open MPI on? I need to know your
the latest version of whatami, it may break existing tests
>> they depend on.
>>
>> I'm leaning towards a).
>>
>> Thoughts?
>>
>> -Brian
>>
>>
>> Jeff Squyres wrote:
>>> Ditto:
>>>
>>> [8:23] svbu-mpi:~/svn/
Folks,
I have a fix for this ticket. The syntax to get
non-contiguous date ranges is to use space-delimited quoted
tokens (see http://tinyurl.com/6e2pqa), e.g.,
"2007-10-29 - 2007-11-30" "2008-11-07 - 2008-11-14" "by month"
Trying to view year-old results alongside week-old results in the
cur
>>
>>
>> Tim Mattox wrote:
>>> The 2008.10.31 version that Jeff forwarded works fine on SLES9 PPC64.
>>>
>>> 2008/10/31 Jeff Squyres :
>>>> Sorry; my stupid mailer didn't display it as an attachment when I
>>>> forwarded.
>&
-Ethan
On Fri, Oct/31/2008 02:01:55PM, Jeff Squyres wrote:
> Sorry; my stupid mailer didn't display it as an attachment when I
> forwarded. Try this.
>
> On Oct 31, 2008, at 1:56 PM, Ethan Mallove wrote:
>
>> On Fri, Oct/31/2008 01:01:02PM, Jeff Squyres wrote:
>&g
;>>>
>>>>>> I just reverted the whatami I am using on BigRed to MTT's r1236 and it
>>>>>> works
>>>>>> again. No rush to fix this for me, but it is bad that whatami seems
>>>>>> to
>>>>>> now
&g
-scratch into --scratch.
>>>
>>> As for names like local-scratch or fast-scratch, they don't convey
>>> what it's used for, so should it be fast-for-big-files, of
>>> fast-for-small-files?
>>> Or similarly, "local" to my cluster, my n
:08 2008 -0500
+++ b/client/whatami/whatamiThu Oct 30 14:27:57 2008 -0500
@@ -86,6 +86,8 @@
# 2008.10.14 Brian Finley
# - Add generic lsb_release support
# - includes CentOS 5.x
+# 2008.10.30 Ethan Mallove
+# - Support two different SuSE 10 lsb-release file fo
n
>
> -- Josh
>
> On Oct 30, 2008, at 11:36 AM, Ethan Mallove wrote:
>
>> On Wed, Oct/29/2008 09:15:37AM, Ethan Mallove wrote:
>>> On Tue, Oct/28/2008 06:17:12PM, Jeff Squyres wrote:
>>>> Should we set a default value of report_after_n_results to, say, 50 o
On Wed, Oct/29/2008 09:15:37AM, Ethan Mallove wrote:
> On Tue, Oct/28/2008 06:17:12PM, Jeff Squyres wrote:
> > Should we set a default value of report_after_n_results to, say, 50 or 100?
> >
>
> We should.
>
> -Ethan
>
>
> > On Oct 28, 2008, at 6:15 PM,
t;> Tim/Josh -- this is you guys...
>>
>>
>> On Oct 28, 2008, at 3:45 PM, Ethan Mallove wrote:
>>
>>> Folks,
>>>
>>> I got an alert from the http-log-checker.pl script. Somebody appears to
>>> have
>>> lost some MTT results. (Possibl
ur IP address to the log file, and then this can be
matched with the IP addr against the submit.php errors.
-Ethan
- Forwarded message from Ethan Mallove -
From: Ethan Mallove
List-Post: mtt-users@lists.open-mpi.org
Date: Tue, 28 Oct 2008 08:00:41 -0400
To: ethan.mall...@sun.com, http-
I saw this last night in an MTT run:
Cannot fork Not enough space at .../lib/MTT/Util.pm line 100
Util.pm:
97 # Setup min_disk_free to be a number of bytes
98 $df_handle = new Filesys::DiskFree;
99 if ($MTT::Globals::Values->{min_disk_free} =~ m/([0-9]{1,2})\%/
On Thu, Sep/18/2008 05:35:13PM, Jeff Squyres wrote:
> On Sep 18, 2008, at 10:45 AM, Ethan Mallove wrote:
>
>>> Ah, yeah, ok, now I see why you wouldl call it --mpi-install-scratch, so
>>> that it matches the MTT ini section name. Sure, that works for me.
>>
>>
On Thu, Sep/18/2008 10:11:19AM, Tim Mattox wrote:
> I Guess I should comment on Jeff's comments too.
>
> On Thu, Sep 18, 2008 at 9:00 AM, Jeff Squyres wrote:
> > On Sep 16, 2008, at 12:07 PM, Ethan Mallove wrote:
> >
> >> What happens if one uses
On Tue, Sep/16/2008 12:07:31PM, Ethan Mallove wrote:
> Nice! A few thoughts ...
>
> What happens if one uses --local-scratch, but leaves out the
> --scratch option? In this case, I think MTT should assume
> --scratch is the same as --local-scratch.
>
> Could the "loca
Nice! A few thoughts ...
What happens if one uses --local-scratch, but leaves out the
--scratch option? In this case, I think MTT should assume
--scratch is the same as --local-scratch.
Could the "local" in --local-scratch ever be misleading?
Couldn't a user ever use a mounted filesystem that's
IU folks,
Word on the street is that you guys are creating an INI file
on-the-fly. You might find the "include_files" and
"include_section" INI parameters useful:
https://svn.open-mpi.org/trac/mtt/wiki/INI%20Includes
Maybe I should've brought this up eight months ago when the
feature was added
se...
>>> MTTDatabase client trying proxy: / Default (none)
>>> MTTDatabase proxy successful / not 500
>>> MTTDatabase response is a success
>>> MTTDatabase client got response:
>>> *** WARNING: MTTDatabase client did not get a serial; phases will be
>&g
get a serial; phases will be
>isolated from each other in the reports
> MTTDatabase client submit complete
>
> And I can not find these results in DB.
> Is it any progress with this issue ?
>
>
> Regards.
> Pasha
>
> Ethan Mallove wrote:
>> On Wed, May/
I'm CC-ing the list, since this is a good question :-)
On Tue, Jun/10/2008 11:48:04AM, Phuong Nguyen wrote:
> Ethan,
> Is there anyway that I could specify the SVN user/password in the INI
> file or pass them in command line of "mtt"?
>
This should work. E.g.,
$ client/mtt ... svn_username= s
).
>>
>> Pasha
>>
>> Jeff Squyres wrote:
>>
>>> FWIW: I think we have at least one open ticket on this issue (break up
>>> submits so that we don't overflow PHP and/or apache).
https://svn.open-mpi.org/trac/mtt/ticket/221
-Ethan
>>&g
On Wed, May/21/2008 06:46:06PM, Pavel Shamis (Pasha) wrote:
> I sent it directly to your email. Please check.
> Thanks,
> Pasha
Got it. Thanks. It's a PHP memory overload issue.
(Apparently I didn't look far back enough in the httpd
error_logs.) See below.
>
> Ethan
On Wed, May/21/2008 05:19:44PM, Pavel Shamis (Pasha) wrote:
> Jeff Squyres wrote:
> > Are we running into http max memory problems or http max upload size
> > problems again?
> I guess it is some server side issue, you need to check the
> /var/log/httpd/* log on the server.
The only thing I foun
On Tue, May/13/2008 02:54:46PM, Rafael Folco wrote:
> Hi,
>
> I appreciate if anyone can give me some help.
>
> How to run MTT tests just against one specific MPI version ? It is
> running against all MPI versions: ompi-nightly-v1.1, ompi-nightly-trunk,
> and ompi-nightly-v1.2.
>
> I've already
ote:
> > Ethan -- are you referring to the sample INI file on the core-testers
> > branch?
Yes. I just comitted the fix.
-Ethan
> >
> >
> > On May 13, 2008, at 9:59 AM, Ethan Mallove wrote:
> >
> > > Are you running on the ompi-core-testers bran
sed |
>| | POST_EPOCH: passed |
>| | t_win: FAILED 1 |
>| start_timestamp | Tue May 13 04:54:59 2008 |
>| test_name | t_win |
>| variant | 2 |
>
>Wen Hao Wang (**)
>
>Software Engineer
> IBM China Software Development Laboratory
>E
It is odd that a "require" failed like this in the Test Run
phase. I would have expected an earlier "require" to fail
(e.g., in the MPI Install phase). I've never run MTT as root
before. Does root have read-access to Specify/Simple.pm? If
not, can you do the below command in the top-level MTT
direc
Cheers,
> > Josh
> >
> > On May 7, 2008, at 4:49 AM, Pavel Shamis (Pasha) wrote:
> >
> >> Hi,
> >> I upgraded the server side (the mtt is still running , so don't know
> >> if the problem was resolved)
> >> During upgrade I had some p
there some nightly (DB
> > pruning/cleaning/vacuuming?) cron jobs that Pasha should be
> > running?
> >
> > -Ethan
> >
> >
> >> Thanks.
> >>
> >> Ethan Mallove wrote:
> >>
> >>> Hi Pasha,
> >>>
>
there are the regular
mtt-resu...@open-mpi.org email alerts and some out-of-date
DB monitoring junk. You can ignore that stuff.
Josh, are there some nightly (DB
pruning/cleaning/vacuuming?) cron jobs that Pasha should be
running?
-Ethan
>
> Thanks.
>
> Ethan Mallove wrote:
> > H
Hi Pasha,
I thought this issue was solved in r1119 (see below). Do you
have the latest mtt/server scripts?
https://svn.open-mpi.org/trac/mtt/changeset/1119/trunk/server/php/submit
-Ethan
On Tue, May/06/2008 03:26:43PM, Pavel Shamis (Pasha) wrote:
> About the issue:
>
> 1. On client side I s
Works for me.
-Ethan
On Tue, May/06/2008 09:16:57AM, Jeff Squyres wrote:
> What do you think of reducing the default number of saved stdout lines
> in MPI install sections from 100 to 50?
>
> The main use for saving the stdout in MPI Install is to give context
> for when there's stderr outpu
Line 386 of MTT/Test/Run.pm sets LD_LIBRARY_PATH. This
appears to have masked a problem in my wrapper compiler
(missing -R/foo/lib flags).
if ($mpi_install->{libdir}) {
if (exists($ENV{LD_LIBRARY_PATH})) {
$ENV{LD_LIBRARY_PATH} = "$mpi_install->{libdir}:" .
Folks,
I added another graphing preference that might make the
performance graphs easier to grok. If you click the
[Preferences] link now, there's a "Split graphs" option that
allows you to split graphs by Halves, Thirds, or Quarters.
Here's an example of some performance graphs split into
"Quarte
Check this page out to get started using MTT:
https://svn.open-mpi.org/trac/mtt/wiki/MTTOverview
-Ethan
On Sat, Apr/05/2008 10:02:49PM, Davi Vercillo C. Garcia wrote:
> Hi all,
>
> My name is Davi Vercillo and I'm from Brazil. I'm starting right now
> to study MPI and I'll use the implementa
I turned off the "test run failures" email (see r1172).
-Ethan
On Sat, Mar/15/2008 08:19:40AM, Jeff Squyres wrote:
> Do we still need the OMPI MTT test run failure e-mail twice a day?
>
> Personally, I look at the "summary" mail and click through the graph
> to see the run failures. Having a
; http://www.open-mpi.org/mtt/index.php?do_redir=426
>
> MX_RCACHE=2
> BigRed for 4 nodes all benchmarks:
> http://www.open-mpi.org/mtt/index.php?do_redir=431
>
> Reran with MX_RCACHE=2
> http://www.open-mpi.org/mtt/index.php?do_redir=433
>
> On Fri, Feb 29, 2008 at 12:12
eed to
> do some text searches during that time period.
>
>
> On Feb 26, 2008, at 11:45 AM, Ethan Mallove wrote:
>
> > Does anyone have some permalinks they gathered from the
> > collectives bakeoff? I'm specifically interes
Folks,
We were seeing NetPIPE runs timing around 45sec the other
day, so I set the timeout value to -1 (which is infinity),
and last night a v1.2 NetPIPE run took 3 hrs! It looks like
the test got stuck:
0: burl-ct-v440-2
1: burl-ct-v440-2
2: burl-ct-v440-2
3: burl-ct-v440-2
Now starti
Does anyone have some permalinks they gathered from the
collectives bakeoff? I'm specifically interested in MVAPICH
and OMPI comparisons.
-Ethan
008 03:14:20PM, Josh Hursey wrote:
> I'm seeing between 12 and 20 seconds on a fairly idle machine. We can
> likely do better. I'll dig into it this week[end] and see what I can do.
>
> 12 - 20 isn't too bad though considering the amount of data that query
> is ret
I don't remember a "past 24 hour" summary taking "24
seconds". Are we seeing a slow down due to an accumulation
of results? I thought the week-long table partitions would
prevent this type of effect?
-Ethan
On Wed, Jan/30/2008 11:00:46AM, Josh Hursey wrote:
> This maintenance is complete. The re
Hi Jennis,
FWIW, I run MTT using an MTT source code checkout. (This is
admittedly not the Right Way to distribute software, but it
works.) Try this:
$ svn checkout https://svn.open-mpi.org/svn/mtt/branches/ompi-core-testers
$ cd ompi-core-testers
$ client/mtt --file samples/ompi-core-templa
On Wed, Jan/23/2008 03:36:26PM, Jennis Pruett wrote:
>
>To followup on this, I think I got it right by just using:
>
>perl Makefile.PL LIB=/usr/projects/hpctools/jennyp/mtt
>PREFIX=/usr/projects/hpctools/jennyp/mtt
>setenv PERLIB /usr/projects/hpctools/jennyp/mtt
>
>I then co
On Thu, Jan/10/2008 06:27:03PM, Pavel Shamis (Pasha) wrote:
> Ethan Mallove wrote:
> > Hi Pasha,
> >
> > I confess that I don't use the Email reporter. I think the
> > email_subject is out-of-date. Try this:
> >
> > email_subject = MPI test result
Works for me.
-Ethan
On Thu, Jan/10/2008 08:48:20AM, Jeff Squyres wrote:
> Mellanox told me that the MTT &hostlist() funclet is returning a
> comma-delimited list of hosts (and &hostlist_hosts()). That is fine for
> Open MPI, but it is not for MVAPICH -- MVAPICH requires a space-delimited
> l
Hi Pasha,
I confess that I don't use the Email reporter. I think the
email_subject is out-of-date. Try this:
email_subject = MPI test results: ¤t_section()
I'm unsure about the empty email body, but have an idea.
What does this output for you?
$ which Mail mailx mail rmail mutt elm
I'm gue
Thanks Oleg! It looks like Jeff already applied the
multilib.patch. The other one looks good to me.
-Ethan
On Mon, Jan/07/2008 07:23:55PM, Oleg Kats wrote:
>Hello All
>
>Here is list of issues that we figure out during ourMTT usage:
>
>
>
>1. Installation of multiple MPIs f
On Thu, Dec/20/2007 11:51:30AM, Jeff Squyres wrote:
> On Dec 20, 2007, at 11:38 AM, Ethan Mallove wrote:
>
> >> FYI: per my post to the OMPI devel list, I discovered the
> >> "svnsync" command yesterday. svnsync allows you to
> >> mirror an entir
On Thu, Dec/20/2007 06:34:27AM, Jeff Squyres wrote:
> FYI: per my post to the OMPI devel list, I discovered the
> "svnsync" command yesterday. svnsync allows you to
> mirror an entire SVN repository in a read-only fashion; I
> mirrored the ompi-tests repository locally at Cisco. A
> one-line c
penoff accounts.
>
> Thanks, and sorry for the confusion.
>
> Ethan Mallove wrote:
> > I thought maybe there was a reason Karol wanted separate
> > accounts that I hadn't thought of. Karol, which accounts do
> > you plan on using?
> >
> > -Ethan
> &
I thought maybe there was a reason Karol wanted separate
accounts that I hadn't thought of. Karol, which accounts do
you plan on using?
-Ethan
On Fri, Dec/14/2007 08:04:37PM, Jeff Squyres wrote:
> Do we really need all 3 database accounts for ubc?
>
>
> On Dec 14, 2007,
Hi Karol,
I added three accounts:
* ubc
* kmroz
* penoff
OMPI MTT users use their organization name (e.g., "ubc" for
you folks), though your local UNIX username is recorded in
the results as well to help you sort through whose results
belong to who.
Cheers,
Ethan
On Fri, Dec/14/2007 02:15:
Do you have access to the ompi-tests repository?
What happens if you do this command outside of MTT?
$ svn export https://svn.open-mpi.org/svn/ompi-tests/trunk/onesided
You could also try using "http", instead of "https" in
your svn_url.
-Ethan
On Wed, Nov/14/2007 02:04:27PM, Karol Mroz wrot
rm libbar.so
$ ldd foo
libbar.so => (file not found)
...
$ ./foo
ld.so.1: foo: fatal: libbar.so: open failed: No such file or directory
Killed
$ echo $?
137
-Ethan
> On Sep 27, 2007, at 4:22 PM, Ethan Mallove wrote:
>
> > Some trivial tests outputted the f
On Wed, Oct/03/2007 09:06:16AM, Jeff Squyres wrote:
> On Oct 3, 2007, at 8:50 AM, Ethan Mallove wrote:
>
> > Josh, another item Gil and I talked about was moving a
> > subset of test results from the internal Mellanox database
> > into the general Open MPI databa
Gil,
Josh created a few HOWTO pages that should help you get
your internal Mellanox MTT server setup:
https://svn.open-mpi.org/trac/mtt/wiki/HTTPServer
https://svn.open-mpi.org/trac/mtt/wiki/ServerMaintenance
https://svn.open-mpi.org/trac/mtt/wiki/Database
Josh, another item Gil and I talk
eaner solution someday
though. Maybe an Install/Analyze/Generic.pm or something ...?
-Ethan
> On 9/28/07, Ethan Mallove wrote:
> > Hi Jelena,
> >
> > Change this line:
> >
> > alreadyinstalled_dir = /usr/local/mpich-mx/bin
> >
> > To this:
> >
> >
Hi Jelena,
Change this line:
alreadyinstalled_dir = /usr/local/mpich-mx/bin
To this:
alreadyinstalled_dir = /usr/local/mpich-mx
Does that work?
-Ethan
On Thu, Sep/27/2007 10:46:55PM, Jelena Pjesivac-Grbovic wrote:
> Hello,
>
> I am trying to test the MPICH-MX using MTT on our clusters a
Some trivial tests outputted the following:
ld.so.1: orterun: fatal: libopen-pal.so.0: open failed: No such file or
directory
See here:
http://www.open-mpi.org/mtt/index.php?do_redir=394
How could these tests result in "pass"?
I attached an mttdatabase debug file.
-Ethan
$VAR1 = {
Done!
-Ethan
On Wed, Sep/26/2007 11:25:30AM, Ron Brightwell wrote:
> Greetings:
>
> Does Sandia have an MTT account yet? If not, can we have one?
>
> [I'm following the instructions on MTT for the collective bakeoff...]
>
> Thanks,
>
> -Ron
>
>
> __
This issue was resolved by backslash-escaping the '&'.
So this works:
alreadyinstalled_version = &shell("tcsh -c '@alreadyinstalled_dir@/bin/mpinfo
-V |\& cut -d\" \" -f6-8'")
-Ethan
On Mon, Sep/17/2007 12:12:48PM, Ethan Mallove wrote:
On Thu, Sep/20/2007 01:55:49PM, Jelena Pjesivac-Grbovic
wrote:
> I already have that - but it did not generate anything
> when I call it second time (it generates output the first
> time when I execute mtt client with all sections). (I
> even tried adding --section reporter to the command line)
>
This is a bug. Ticket to come. A workaround is to select
"Show" from the Show/Hide menu next to the Bitness
textfield when searching on Bitness.
Thanks,
Ethan
On Thu, Sep/20/2007 08:52:59AM, Tim Prins wrote:
> Hi,
>
> I was doing a search and hit advanced, entered '32' into the bitness
> field
The warnings emitted before the performance data *were* the
problem. Fixed in r1027.
-Ethan
On Mon, Sep/17/2007 04:06:26PM, Ethan Mallove wrote:
> Jeff,
>
> Here is a run of "osu" test results that did not seem
> to register any latency_bandwidth data:
>
>
Jeff,
Here is a run of "osu" test results that did not seem
to register any latency_bandwidth data:
http://www.open-mpi.org/mtt/index.php?do_redir=340
Below is the mttdatabase debug file. Could the
Analyze/OSU.pm module is getting flummoxed by the warnings
that precede the actual performance
I am stuck trying to capture STDERR using &shell().
I am able to capture it using tcsh, but since &shell()
is sh-based, I need some sh-magic to get this working.
tcsh $ mpinfo -V |& cut -d' ' -f6-8
10 Mar 2006
bash-3.00$ /opt/SUNWhpc/HPC6.0/bin/mpinfo -V 2>&1 cut -d' ' -f6-8
mpinfo: Sun H
On Wed, Sep/05/2007 08:44:19PM, Jeff Squyres wrote:
> On Sep 4, 2007, at 12:36 PM, Ethan Mallove wrote:
>
> > reporter.php has a search field for "description", but which
> > phase is that for? MPI Install, Test build, or Test run? A
> > scenario is that Sun wil
On Tue, Sep/04/2007 12:47:29PM, Jeff Squyres wrote:
> On Sep 4, 2007, at 10:57 AM, Ethan Mallove wrote:
>
> > So if I have the following line in my [Test build: intel]
> > section, MTT will now build and run the Intel tests *twice*?
> > Once with compiler-A
reporter.php has a search field for "description", but which
phase is that for? MPI Install, Test build, or Test run? A
scenario is that Sun will have a home-grown "funclet_files"
funclet set the "description" in the Install phase (e.g., a
label identifying which ISO is being tested). It would be
u
09AM, Jeff Squyres wrote:
> I think we'll get it for everything because of the new @foo@ syntax.
>
> I don't think you'll want to use that effect everywhere, but it's now
> *possible* to do so.
>
>
> On Sep 4, 2007, at 9:57 AM, Ethan Mallove wrote:
&
So we will get a multiplicative effect on the following INI
parameters?
* setenv
* unsetenv
* env_module
Any other parameters?
-Ethan
On Fri, Aug/31/2007 10:18:30PM, MTT wrote:
> #245: Support multiplicative effect for other INI params
> --+--
On Fri, Aug/10/2007 01:02:16PM, Jeff Squyres wrote:
> On Aug 10, 2007, at 10:14 AM, Ethan Mallove wrote:
>
> > Do we still need to write out an INI file after each MPI
> > Install and Test Build? I thought we're using .dump
> > files instead? (I noticed this because
1 - 100 of 199 matches
Mail list logo
