<div id="geary-body" dir="auto">Dear Klaus,<div><br></div><div>thank you for the suggestion. </div><div><br></div><div>I have tried it and of course another error show up. </div><div>It complains with an unterminated '#ifdef' at row 216 of thsc_ThomsonFormFactor.F90. I have replaced '##endif' with '#endif'.</div><div>After that the compilation end with 'SUCCESS'</div><div><br></div><div>Now when I run flash the result is:</div><div>"DRIVER_ABORT: ThomsonScattering_init: No Thomson scattering beam defined!"</div><div>and I haven't turn on the setup variable ThscDemo</div><div><br></div><div><br></div></div><div id="geary-signature" dir="auto"><div>Francesco Barbato</div></div><div id="geary-quote" dir="auto"><br>Il giorno mer 4 nov 2020 alle 10:21, Klaus Weide <klaus@flash.uchicago.edu> ha scritto:<br><blockquote type="cite"><div class="plaintext" style="white-space: pre-wrap;">On Wed, 4 Nov 2020, Francesco Barbato wrote:
<blockquote> Dear All,
after a system update when I try to setup a LaserSlab simulation I get an
error (enclosed there is a print of the error).
Could be related to the new version of MPI
(<<a href="https://www.open-mpi.org/faq/?category=mpi-removed">https://www.open-mpi.org/faq/?category=mpi-removed</a>>)?
</blockquote>
Yes, it appears that that page points to the reason.
As recommended there, calls of MPI_Type_extent in the FLASH code should be
replaced with MPI_Type_get_extent.
This will include adding an additional argument (for 'lb') in second place
in the argument list in the calls. It can be an otherwise unused (but
defined) integer variable.
In the specific file IO/IOTypes/io_set_grid_mpi_types.c, the obsolete
function is only called in a block subject to 'if (debugIO)'. That could
be commented out without great loss in functionality.
Some other files will also require changes.
Klaus
</div></blockquote></div>