My guess is somehow you are passing the wrong integer for the option<br>
because I am looking at the code and it is fine. However, it is easy to<br>
clear up. Go to<br>
<br>
$PETSC_DIR/src/mat/impls/rowbs/mpi/mpirowbs.c:1411<br>
<br>
and replace it with<br>
<br>
SETERRQ1(PETSC_ERR_SUP,"unknown option %d",op);<br>
<br>
and then in that directory<br>
<br>
make<br>
make shared<br>
<br>
This will tell us what integer was passed in for the option. I have already made<br>
this fix in the dev version.<br>
<br>
Matt<br><br><div><span class="gmail_quote">On 5/17/06, <b class="gmail_sendername">Boulahya Faïza</b> <<a href="mailto:F.Boulahya@brgm.fr">F.Boulahya@brgm.fr</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div>
<div>
<p><font size="2">Ok but the error does not come from parallell ICC. It's from the MatSetOption().</font>
</p>
<p><font size="2">Moreover, when I tried with MATMPIROWBS , it's the same problem :</font>
<br><font size="2">[0]PETSC ERROR: MatSetOption_MPIRowbs() line 1411 in src/mat/impls/rowbs/mpi/mpirowbs.c</font>
</p></div><div><span class="q"><br><font size="2">[0]PETSC ERROR: No support for this operation for this object type!</font>
<br><font size="2">[0]PETSC ERROR: unknown option!</font>
<br><font size="2">[0]PETSC ERROR: MatSetOption() line 4137 in src/mat/interface/matrix.c</font>
</span></div><div><p></p>
<br>
<p><font size="2">Faïza</font>
</p>
<p><font size="2">-----Message d'origine-----</font>
<br><font size="2">De : Satish Balay [<a href="mailto:balay@mcs.anl.gov" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)">mailto:balay@mcs.anl.gov</a>] </font>
<br><font size="2">Envoyé : mardi 16 mai 2006 15:39</font>
</p></div><div><span class="q"><br><font size="2">À : '<a href="mailto:petsc-users@mcs.anl.gov" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)">petsc-users@mcs.anl.gov</a>'</font>
<br></span></div><div><font size="2">Objet : RE: Petsc + BlockSolve95</font>
<p></p></div><div><span class="e" id="q_10b41ff7d5e86adf_5">
<p><font size="2">> [0]PETSC ERROR: MatSetOption_MPIAIJ() line 1251 in </font>
<br><font size="2">> src/mat/impls/aij/mpi/mpiaij.c [0]PETSC ERROR: No support for this operation for this object type!</font>
</p>
<p><font size="2">AIJ matrix does not support parallell ICC. BlockSolve provides parallel ICC - and to use it one must use MatType MATMPIROWBS</font></p>
<p><font size="2">Satish</font>
</p>
<p><font size="2">On Tue, 16 May 2006, Boulahya Faïza wrote:</font>
</p>
<p><font size="2">> Is there something new about it?</font>
</p>
<p><font size="2">> De : Vaz, Guilherme [<a href="mailto:G.Vaz@marin.nl" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)">mailto:G.Vaz@marin.nl</a>]</font>
</p>
<p><font size="2">> I have exactly the same problem as Faiza... </font>
<br><font size="2">> In sequential it runs ok but in parallel not.</font>
</p>
<p><font size="2">> -----Original Message-----</font>
<br><font size="2">> From: <a href="mailto:owner-petsc-users@mcs.anl.gov" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)">owner-petsc-users@mcs.anl.gov</a> </font>
<br><font size="2">> [<a href="mailto:owner-petsc-users@mcs.anl.gov" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)">mailto:owner-petsc-users@mcs.anl.gov</a>]</font>
<br><font size="2">> On Behalf Of Boulahya Faïza</font>
</p>
<p><font size="2">> I tried something else :</font>
<br><font size="2">> </font>
<br><font size="2">> - creation of the matrix with MatCreateMPIAIJ</font>
<br><font size="2">> </font>
<br><font size="2">> - initialization</font>
<br><font size="2">> </font>
<br><font size="2">> - adding of the options MAT_SYMMETRIC and MAT_SYMMETRY_ETERNAL</font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> When solving in sequential CG + ICC, everything is ok. When I tried in </font>
<br><font size="2">> parallel the same code the options lead to the same error :</font>
<br><font size="2">> </font>
<br><font size="2">> [0]PETSC ERROR: MatSetOption_MPIAIJ() line 1251 in </font>
<br><font size="2">> src/mat/impls/aij/mpi/mpiaij.c [0]PETSC ERROR: No support for this </font>
<br><font size="2">> operation for this object type!</font>
<br><font size="2">> [0]PETSC ERROR: unknown option!</font>
<br><font size="2">> [0]PETSC ERROR: MatSetOption() line 4137 in src/mat/interface/matrix.c</font>
<br><font size="2">> </font>
<br><font size="2">> Can MatSetOption be only used in sequential?</font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> Faïza</font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> _____</font>
<br><font size="2">> </font>
<br><font size="2">> De : Matthew Knepley [<a href="mailto:knepley@gmail.com" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)">mailto:knepley@gmail.com</a>] Envoyé : mardi 9 mai </font>
<br><font size="2">> 2006 18:19 À : <a href="mailto:petsc-users@mcs.anl.gov" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)">petsc-users@mcs.anl.gov</a> Objet : Re: Petsc + </font>
<br><font size="2">> BlockSolve95</font>
<br><font size="2">> </font>
<br><font size="2">> I believe there is a problem with the option that you specified. All </font>
<br><font size="2">> these are integers, and it is complaining that the integer does not </font>
<br><font size="2">> match MAT_SYMMETRIC. I will fix the error message to print the </font>
<br><font size="2">> offending option, but please check the code.</font>
<br><font size="2">> </font>
<br><font size="2">> Thanks,</font>
<br><font size="2">> </font>
<br><font size="2">> Matt</font>
<br><font size="2">> </font>
<br><font size="2">> On 5/9/06, Boulahya Faïza <<a href="mailto:F.Boulahya@brgm.fr" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)">F.Boulahya@brgm.fr</a> </font>
<br><font size="2">> <<a href="mailto:F.Boulahya@brgm.fr" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)">mailto:F.Boulahya@brgm.fr</a>> ></font>
<br><font size="2">> wrote: </font>
<br><font size="2">> </font>
<br><font size="2">> Hi All,</font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> Has anyone used Conjugate Gradient Solver + Icomplete Cholesky </font>
<br><font size="2">> Preconditionner in parallel case? I tried as said in the manual : I </font>
<br><font size="2">> use MATMPIROWBS for the storage of the matrice. However I get this message :</font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> PETSC ERROR: To use incomplete Cholesky</font>
<br><font size="2">>
preconditioning with a MATMPIROWBS matrix you </font>
<br><font size="2">> must declare it to be</font>
<br><font size="2">>
symmetric using the option </font>
<br><font size="2">> MatSetOption(A,MAT_SYMMETRIC)!</font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> So I tried adding this option (even if in the namual it is written </font>
<br><font size="2">> that it is not required). Then I obtained this message</font>
<br><font size="2">> </font>
<br><font size="2">> PETSC ERROR: MatSetOption_MPIRowbs() line 1411 in </font>
<br><font size="2">> src/mat/impls/rowbs/mpi/mpirowbs.c</font>
<br><font size="2">> PETSC ERROR: No support for this operation for this object type!</font>
<br><font size="2">> PETSC ERROR: unknown option!</font>
<br><font size="2">> PETSC ERROR: MatSetOption() line 4137 in src/mat/interface/matrix.c</font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> In advance thanks,</font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> Faïza</font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> </font>
<br><font size="2">> Pensez à visiter le site BRGM sur.... <a href="http://www.brgm.fr/" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)">http://www.brgm.fr/</a></font>
<br><font size="2">> </font>
<br><font size="2">> ***</font>
<br><font size="2">> Le contenu de cet e-mail et de ses pièces jointes est destiné à </font>
<br><font size="2">> l'usage exclusif du</font>
<br><font size="2">> (des) destinataire(s) expressément désigné(s) comme tel(s). En cas de </font>
<br><font size="2">> réception de cet e-mail par erreur, le signaler à son expéditeur et </font>
<br><font size="2">> ne pas en divulguer le contenu.</font>
<br><font size="2">> L'absence de virus a été vérifié à l'émission du message. Il convient </font>
<br><font size="2">> néanmoins de vérifier l'absence de corruption à sa réception.</font>
<br><font size="2">> </font>
<br><font size="2">> The contents of this email and any attachments are confidential. They </font>
<br><font size="2">> are intended for the named recipient(s) only. If you have received </font>
<br><font size="2">> this email in error please notify the system manager or the sender </font>
<br><font size="2">> immediately and do not disclose the contents to anyone or make copies. </font>
<br><font size="2">> eSafe scanned this email for viruses, vandals and malicious content.</font>
<br><font size="2">> ***</font>
<br><font size="2">> Pensez à visiter le site BRGM sur.... <a href="http://www.brgm.fr/" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)">http://www.brgm.fr/</a></font>
<br><font size="2">> </font>
<br><font size="2">> ***</font>
<br><font size="2">> Le contenu de cet e-mail et de ses pièces jointes est destiné à </font>
<br><font size="2">> l'usage exclusif du</font>
<br><font size="2">> (des) destinataire(s) expressément désigné(s) comme tel(s). En cas de </font>
<br><font size="2">> réception de cet e-mail par erreur, le signaler à son expéditeur et ne pas en divulguer le contenu.</font>
<br><font size="2">> L'absence de virus a été vérifié à l'émission du message. Il convient </font>
<br><font size="2">> néanmoins de vérifier l'absence de corruption à sa réception.</font>
<br><font size="2">> </font>
<br><font size="2">> The contents of this email and any attachments are confidential. They </font>
<br><font size="2">> are intended for the named recipient(s) only. If you have received </font>
<br><font size="2">> this email in error please notify the system manager or the sender </font>
<br><font size="2">> immediately and do not disclose the contents to anyone or make copies. </font>
<br><font size="2">> eSafe scanned this email for viruses, vandals and malicious content.</font>
<br><font size="2">> ***</font>
</p>
<font color="BLUE" size="3"></font><pre><font color="BLUE" size="3">Pensez à visiter le site BRGM sur.... <a href="http://www.brgm.fr/" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)">http://www.brgm.fr/
</a><br><br>***<br>Le contenu de cet e-mail et de ses pièces jointes est destiné à l'usage exclusif du <br>(des) destinataire(s) expressément désigné(s) comme tel(s). En cas de réception de cet <br> e-mail par erreur, le signaler à son expéditeur et ne pas en divulguer le contenu.
<br>L'absence de virus a été vérifié à l'émission du message. Il convient néanmoins de <br>vérifier l'absence de corruption à sa réception.<br><br>The contents of this email and any attachments are confidential. They are intended for
<br>the named recipient(s) only. If you have received this email in error please notify the <br>system manager or the sender immediately and do not disclose the contents to <br>anyone or make copies. eSafe scanned this email for viruses, vandals and malicious
<br>content.<br>***<br></font></pre>
</span></div><div></div>
</div></blockquote></div><br><br clear="all"><br>-- <br>"Failure has a thousand explanations. Success doesn't need one" -- Sir Alec Guiness