<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
<title></title>
</head>
<body bgcolor="#ffffff" text="#000000">
Hi,<br>
<br>
I put warnings in CMake and autoconf about these versions, although
no actual checks are being done. We should probably fix that for the
final release.<br>
<br>
Rossen<br>
<br>
On 8/11/10 11:58 PM, Roland Schulz wrote:
<blockquote
cite="mid:AANLkTik7wo1+wKJxi-rAS6f4At12LXsDhwvp0qt6rEwx@mail.gmail.com"
type="cite">Hi,
<div><br>
</div>
<div>should we add configure errors/warning for MPI versions?</div>
<div><br>
</div>
<div>Problems known to me:</div>
<div>MVAPICH2 <1.4.1 crashes</div>
<div>OpenMPI <1.4.x is slow</div>
<div>OpenMPI 1.4.0 hangs on GCC 4.4.x on x86/x64</div>
<div><br>
</div>
<div>Thus I would suggest we give an error for MVAPICH2 <1.4.1
and OpenMPI 1.4.0 and a warning for OpenMPI <1.4.x </div>
<div><br>
</div>
<div>Roland</div>
<div><br>
<div class="gmail_quote">---------- Forwarded message ----------<br>
From: <b class="gmail_sendername">Da-Wei Li</b> <span
dir="ltr"><<a moz-do-not-send="true"
href="mailto:lidawei@gmail.com">lidawei@gmail.com</a>></span><br>
Date: Wed, Aug 11, 2010 at 4:34 PM<br>
Subject: Re: [gmx-users] New beta release: 4.5-beta3<br>
To: Discussion list for GROMACS users <<a
moz-do-not-send="true" href="mailto:gmx-users@gromacs.org">gmx-users@gromacs.org</a>><br>
<br>
<br>
According to our HPC manage, our MVAPICH2's version is 1.0.3.<br>
<br>
thanks for the information.<br>
<br>
dawei<br>
<div>
<div class="h5"><br>
On Wed, Aug 11, 2010 at 2:52 PM, Roland Schulz <<a
moz-do-not-send="true" href="mailto:roland@utk.edu">roland@utk.edu</a>>
wrote:<br>
> MVAPICH2 >=1.4.1 should be fine. Did you use a
version <1.4.1?<br>
> On Wed, Aug 11, 2010 at 2:45 PM, Da-Wei Li <<a
moz-do-not-send="true" href="mailto:lidawei@gmail.com">lidawei@gmail.com</a>>
wrote:<br>
>><br>
>> Yes it is. Now everything is fine with openmpi
instead of mvapich2.<br>
>><br>
>> dawei<br>
>><br>
>> On Wed, Aug 11, 2010 at 12:59 AM, Mark Abraham
<<a moz-do-not-send="true"
href="mailto:mark.abraham@anu.edu.au">mark.abraham@anu.edu.au</a>><br>
>> wrote:<br>
>> ><br>
>> ><br>
>> > ----- Original Message -----<br>
>> > From: Da-Wei Li <<a
moz-do-not-send="true" href="mailto:lidawei@gmail.com">lidawei@gmail.com</a>><br>
>> > Date: Wednesday, August 11, 2010 6:23<br>
>> > Subject: Re: [gmx-users] New beta release:
4.5-beta3<br>
>> > To: Discussion list for GROMACS users <<a
moz-do-not-send="true"
href="mailto:gmx-users@gromacs.org">gmx-users@gromacs.org</a>><br>
>> ><br>
>> >> Hi,all<br>
>> >><br>
>> >> Unfortunately, it still crash. Here it
is the output of mdrun. Both<br>
>> >> Gromacs-4.0.7 and non-parallel 4.5b3
work fine on this system. I use<br>
>> >> mvapich2 and intel compiler.<br>
>> ><br>
>> > MPICH variants are known to cause problems.
Try OpenMPI.<br>
>> ><br>
>> > Mark<br>
>> ><br>
>> >> ***************************<br>
>> >> Getting Loaded...<br>
>> >> Reading file em.tpr, VERSION 4.5-beta3
(single precision)<br>
>> >> Loaded with Money<br>
>> >><br>
>> >><br>
>> >> Will use 9 particle-particle and 7 PME
only nodes<br>
>> >> This is a guess, check the performance
at the end of the log file<br>
>> >> Making 1D domain decomposition 9 x 1 x 1<br>
>> >><br>
>> >> Back Off! I just backed up em.trr to
./#em.trr.4#<br>
>> >><br>
>> >> Back Off! I just backed up em.edr to
./#em.edr.4#<br>
>> >><br>
>> >> Steepest Descents:<br>
>> >> Tolerance (Fmax) = 1.00000e+03<br>
>> >> Number of steps<br>
>> >> = 50000<br>
>> >> rank 9 in job 1 hpc-8-6.local_58777
caused<br>
>> >> collective abort of all ranks<br>
>> >> exit status of rank 9: killed by
signal 9<br>
>> >><br>
>> >> *******************************<br>
>> >><br>
>> >> dawei<br>
>> >><br>
</div>
</div>
</div>
</div>
</blockquote>
<br>
</body>
</html>