<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content="text/html; charset=gb2312" http-equiv=Content-Type>
<META name=GENERATOR content="MSHTML 8.00.6001.18876"></HEAD>
<BODY>
<DIV dir=ltr align=left><FONT color=#0000ff size=2 face=Arial><SPAN
class=640584413-22032010>Not sure if this message went out on the
list...</SPAN></FONT></DIV><BR>
<DIV dir=ltr lang=en-us class=OutlookMessageHeader align=left>
<HR tabIndex=-1>
<FONT size=2 face=Tahoma><B>From:</B> ??? [mailto:limu713@gmail.com]
<BR><B>Sent:</B> Sunday, March 21, 2010 10:25 PM<BR><B>To:</B>
mpich-discuss@mcs.anl.gov; Brice.Goglin@inria.fr;
mercier@labri.fr<BR><B>Subject:</B> Re: [mpich-discuss] Regarding MPICH2-1.1.1p1
testing basing on open-mx<BR></FONT><BR></DIV>
<DIV></DIV>hello,<BR><BR>We used rhel server5.4. There is two cores per node,
but only one core was used. <BR>Details are attached. Other nodes have the same
configuration.<BR><BR>When testing, we have this error message:<BR>[root@cu02
~]# mpiexec -n 4 /usr/lib64/mpich2/bin/mpitests-IMB-EXT Unidir_Get<BR>
<DIV>>> rank 0 in job 8 cu02.hpc.com_54277 caused collective abort of
all<BR>>> ranks exit status of rank 0: killed by signal 9<BR><BR>
And the same wrong comes with "mpiexec -n 4<BR>
/usr/lib64/mpich2/bin/mpitests-IMB-EXT Bidir_Get "<BR><BR></DIV><BR> <BR>
<DIV class=gmail_quote>2010/3/20 Brice Goglin <SPAN dir=ltr><<A
href="mailto:Brice.Goglin@inria.fr"
target=_blank>Brice.Goglin@inria.fr</A>></SPAN><BR>
<BLOCKQUOTE
style="BORDER-LEFT: rgb(204,204,204) 1px solid; MARGIN: 0pt 0pt 0pt 0.8ex; PADDING-LEFT: 1ex"
class=gmail_quote>Some bugs were reported in the past about some MPICH2 tests
not working,<BR>but we never reproduced them with recent MPICH2 and Open-MX
versions.<BR>I'd like to know what kind of interfaces, hosts and kernels were
used<BR>here. And also how many processes per node were used.<BR><FONT
color=#888888><BR>Brice<BR></FONT>
<DIV>
<DIV></DIV>
<DIV><BR><BR><BR>Dave Goodell wrote:<BR>> I don't think that we have tested
OpenMX with the mx netmod, so I'm<BR>> not sure if there are any bugs
there. I've CCed the primary<BR>> developers of both OpenMX and our
mx netmod in case they have any<BR>> information on this.<BR>><BR>>
Do simpler tests work? The "examples/cpi" program in your MPICH2<BR>>
build directory is a good simple sanity test.<BR>><BR>>
-Dave<BR>><BR>> On Mar 19, 2010, at 3:31 AM, Àî¿¡Àö
wrote:<BR>><BR>>> Hello,<BR>>><BR>>> Just do:<BR>>>
./configure --with-device=ch3:nemesis:mx<BR>>>
--with-mx-lib=/opt/open-mx/lib/
--with-mx-include=/opt/open-mx/include/<BR>>><BR>>>
make<BR>>><BR>>> make install<BR>>><BR>>> Then, I
start open-omx service, and test mpich2 based on
open-mx.<BR>>><BR>>><BR>>><BR>>> [root@cu02 ~]#
mpiexec -n 4 /usr/lib64/mpich2/bin/mpitests-IMB-EXT<BR>>>
Unidir_Get<BR>>><BR>>><BR>>><BR>>> It has this error
message:<BR>>><BR>>> rank 0 in job 8 cu02.hpc.com_54277 caused
collective abort of all<BR>>> ranks exit status of rank 0: killed by
signal 9<BR>>><BR>>> And the same wrong comes with "mpiexec -n
4<BR>>> /usr/lib64/mpich2/bin/mpitests-IMB-EXT Bidir_Get
"<BR>>><BR>>> Is there any way to solve this
problem?<BR>>><BR>>> Thanks!<BR>>><BR>>>
Regards<BR>>><BR>>> Lily<BR>>>
_______________________________________________<BR>>> mpich-discuss
mailing list<BR>>> <A href="mailto:mpich-discuss@mcs.anl.gov"
target=_blank>mpich-discuss@mcs.anl.gov</A><BR>>> <A
href="https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss"
target=_blank>https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss</A><BR>><BR><BR>_______________________________________________<BR>mpich-discuss
mailing list<BR><A href="mailto:mpich-discuss@mcs.anl.gov"
target=_blank>mpich-discuss@mcs.anl.gov</A><BR><A
href="https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss"
target=_blank>https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss</A><BR></DIV></DIV></BLOCKQUOTE></DIV><BR></BODY></HTML>