[MPICH] thanks RE: src node found for mpitest/io tests
Liwei Peng
liweip at windows.microsoft.com
Fri Jun 30 12:29:42 CDT 2006
Thanks Rajeev.
I got the code. It is under mpich2 code base src\mpi\romio\test,
liwei
-----Original Message-----
From: Rajeev Thakur [mailto:thakur at mcs.anl.gov]
Sent: Friday, June 30, 2006 10:08 AM
To: Liwei Peng; 'William Gropp'
Cc: mpich-discuss at mcs.anl.gov
Subject: RE: src node found for mpitest/io tests
Those are in the src/mpi/romio/test directory in MPICH2 or
src/romio/test
directory in MPICH-1.
Rajeev
> -----Original Message-----
> From: Liwei Peng [mailto:liweip at windows.microsoft.com]
> Sent: Thursday, June 29, 2006 6:52 PM
> To: William Gropp; Rajeev Thakur
> Cc: mpich-discuss at mcs.anl.gov
> Subject: src node found for mpitest/io tests
>
> Hi Bill,
>
> Thanks for the help.
>
> I couldn't find the test sources for mpitest\io posted on
> http://www-unix.mcs.anl.gov/mpi/mpich2/todo/runs/IA32-Linux-GN
U-mpd-ch3-
> 2006-06-27-22-00-testsumm.xml
>
> in the testsuite
> http://www-unix.mcs.anl.gov/mpi/mpi-test/downloads/mpich-test.tar.gz
>
> can you tell me where I can download them?
>
> Thanks,
>
> Liwei
>
>
> -----Original Message-----
> From: William Gropp [mailto:gropp at mcs.anl.gov]
> Sent: Wednesday, June 28, 2006 11:28 AM
> To: Liwei Peng
> Cc: Rajeev Thakur; mpich-discuss at mcs.anl.gov
> Subject: Re: [MPICH] MPICH1 test error report: what's the # of
> processes?
>
> At 01:14 PM 6/28/2006, Liwei Peng wrote:
> >Hi,
> >
> >The latest MPICH error report from web site
> >http://www-unix.mcs.anl.gov/mpi/mpich2/todo/runs/IA32-Linux-G
NU-mpd-ch3
> -
> >2006-06-27-22-00-testsumm.xml
> >
> >doesn't show how many processes were used on most of the test cases,
> >e.g. sendrecv.
> >
> >Can you give me a list of the # processes for each test case? Or tell
> me
> >where I can get them?
>
> Most of the tests just use two processes. For the MPICH1 tests, the
> number
> of processes is embedded in the runtests file in each
> directory . For
> MPICH2, the file "testlist" indicates how many processes for each test
> in
> the second token.
>
> Bill
>
> >Thanks,
> >
> >Liwei
>
> William Gropp
> http://www.mcs.anl.gov/~gropp
>
>
>
More information about the mpich-discuss
mailing list