Urban,<div><br></div><div> Awesome ! I just got the latest CVS working with gfortran/mvapich2 on x86-64/Linux, and will test it with Intel compilers.</div><div><br></div><div> Thanks.</div><div><br></div><div>Hanning<br>
<br><div class="gmail_quote">On Wed, Oct 28, 2009 at 10:21 AM, Urban <span dir="ltr"><<a href="mailto:bors...@pci.uzh.ch">bors...@pci.uzh.ch</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<br>
There was a known issue that resulted in an error in<br>
cp_fm_cholesky_decompose when run with certain processor numbers. That<br>
issue, along with the Intel compiler issues, should have been fixed<br>
with the commits of Oct. 27.<br>
<div><div></div><div class="h5"><br>
On Oct 26, 4:46 pm, NUCP2K <<a href="mailto:chenh...@gmail.com">chenh...@gmail.com</a>> wrote:<br>
> Dear cp2k community,<br>
><br>
> I just downloaded the latest CVS snapshot and tried to compile the<br>
> codes using my once working ARCH files for PGI and Intel compilers on<br>
> Linux-x86-64 MPI machines. Unfortunately, none of the two compilers is<br>
> working though both can get the codes compiled successfully. It seems<br>
> to me that the "cp_fm_cholesky_decompose" is the cause of the runtime<br>
> trouble since my testing job always exit abnormally when the<br>
> subroutine is called.<br>
><br>
> Anyone has the experience on successfully compiling and running the<br>
> latest cp2k code on Linux PC MPI clusters? The sharing of your ARCH<br>
> files is highly appreciated.<br>
><br>
> Thanks.<br>
><br>
> Hanning<br>
<br>
</div></div></blockquote></div><br></div>