<div class="gmail_quote">On Mon, May 28, 2012 at 9:40 AM, Hui Zhang <span dir="ltr"><<a href="mailto:mike.hui.zhang@hotmail.com" target="_blank">mike.hui.zhang@hotmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div style="word-wrap:break-word"><br><div><div><div class="h5"><div>On May 28, 2012, at 3:42 PM, Jed Brown wrote:</div><br><blockquote type="cite"><div class="gmail_quote">On Mon, May 28, 2012 at 8:40 AM, Hui Zhang <span dir="ltr"><<a href="mailto:mike.hui.zhang@hotmail.com" target="_blank">mike.hui.zhang@hotmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div style="word-wrap:break-word"><br><div><div><div>On May 28, 2012, at 3:20 PM, Jed Brown wrote:</div><br><blockquote type="cite"><div class="gmail_quote">On Mon, May 28, 2012 at 5:19 AM, Hui Zhang <span dir="ltr"><<a href="mailto:mike.hui.zhang@hotmail.com" target="_blank">mike.hui.zhang@hotmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Sorry for repost the following question, but after reading composite.c and asm.c,<br>
I can not understand. I observed that my subroutine for modifying submatrices of PCASM was<br>
called twice. In composite.c it seems there is no calling of PCSetUp on its componential PC's.<br>
In asm.c, the PCSetUp_ASM first determines whether pc->setupcalled is true.<br>
<br>
I think this is a general question for PCComposite say pc. If it consists of two pc's pc0, pc1<br>
and I have PCSetUp(pc0), I would not expect pc0 to be SetUp again. I can not understand<br>
in my case why PCASM seems SetUp twice.<br></blockquote><div><br></div><div>Does the routine do anything or just return immediately? If nothing has been changed since the last time a PC was set up, it should return immediately. If you have changed some configuration, it may need to be set up again.</div>
</div></blockquote><div><br></div></div><div>Thanks. But I said something wrong. The source code asm.c does call the ModifySubMatrices</div><div>whenever PCSetUp is called on PCASM, no matter whether it was called before. So this </div>
<div>explains my observation (two calls of ModifySubMatrices). My new question is: can you</div><div>add a user control to determine whether <span style="font-family:Menlo;font-size:11px">MatGetSubMatrices and </span><span style="font-family:Menlo;font-size:11px">PCModifySubMatrices </span></div>
<div><font face="Menlo"><span style="font-size:11px">should be called inside PCSetUp_ASM ? But maybe this is too invasive and not</span></font></div><div><font face="Menlo"><span style="font-size:11px">very useful to most people (although useful to me).</span></font></div>
</div></div></blockquote><div><br></div><div>It does not reach PCSetUp_ASM unless something has changed. Run with -info and/or in a debugger to see why.</div><div><br></div><div><div>PetscErrorCode PCSetUp(PC pc)</div><div>
{</div><div> PetscErrorCode ierr;</div><div> const char *def;</div><div><br></div><div> PetscFunctionBegin;</div><div> PetscValidHeaderSpecific(pc,PC_CLASSID,1);</div><div> if (!pc->mat) SETERRQ(((PetscObject)pc)->comm,PETSC_ERR_ARG_WRONGSTATE,"Matrix must be set first");</div>
<div><br></div><div> if (pc->setupcalled > 1) {</div><div> ierr = PetscInfo(pc,"Setting PC with identical preconditioner\n");CHKERRQ(ierr);</div><div> PetscFunctionReturn(0);</div></div><div><br></div>
</div></blockquote><div><br></div><div><br></div></div></div><div>Thanks for the hints. I found why (I hope). We have KSP named ksp and its PC is pc,</div><div>and pc is PCComposite with pc0, pc1. The program runs like KSPSetUp on ksp, which</div>
<div>calls PCSetUp on pc, i.e. PCSetUp_Composite on pc. The source codes read</div><div><br></div><div><pre width="80" style="line-height:normal;text-indent:0px;letter-spacing:normal;text-align:-webkit-auto;font-variant:normal;text-transform:none;font-style:normal;font-weight:normal;word-spacing:0px">
<strong><font color="#4169E1">static <a href="http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/Sys/PetscErrorCode.html#PetscErrorCode" target="_blank">PetscErrorCode</a> PCSetUp_Composite(<a href="http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/PC/PC.html#PC" target="_blank">PC</a> pc)</font></strong>
<a name="13793e4a89fa8ede_line168">168: </a>{
<a name="13793e4a89fa8ede_line169">169: </a> <a href="http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/Sys/PetscErrorCode.html#PetscErrorCode" target="_blank">PetscErrorCode</a> ierr;
<a name="13793e4a89fa8ede_line170">170: </a> PC_Composite *jac = (PC_Composite*)pc->data;
<a name="13793e4a89fa8ede_line171">171: </a> PC_CompositeLink next = jac->head;
<a name="13793e4a89fa8ede_line174">174: </a> <font color="#4169E1">if</font> (!jac->work1) {
<a name="13793e4a89fa8ede_line175">175: </a> <a href="http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/Mat/MatGetVecs.html#MatGetVecs" target="_blank">MatGetVecs</a>(pc->pmat,&jac->work1,0);
<a name="13793e4a89fa8ede_line176">176: </a> }
<a name="13793e4a89fa8ede_line177">177: </a> <font color="#4169E1">while</font> (next) {
<a name="13793e4a89fa8ede_line178">178: </a> <a href="http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/PC/PCSetOperators.html#PCSetOperators" target="_blank">PCSetOperators</a>(next->pc,pc->mat,pc->pmat,pc->flag);
<a name="13793e4a89fa8ede_line179">179: </a> next = next->next;
<a name="13793e4a89fa8ede_line180">180: </a> }
<a name="13793e4a89fa8ede_line181">181: </a> <font color="#4169E1">return</font>(0);
<a name="13793e4a89fa8ede_line182">182: </a>}</pre><div>Note the 178 line which calls PCSetOperators on pc0 and pc1. The source codes of PCSetOperators</div><div>read</div><div><br></div><div><pre width="80" style="line-height:normal;text-indent:0px;letter-spacing:normal;text-align:-webkit-auto;font-variant:normal;text-transform:none;font-style:normal;font-weight:normal;word-spacing:0px">
1050: <font color="#4169E1">if</font> (pc->setupcalled == 2 && flag != SAME_PRECONDITIONER) {
<a name="13793e4a89fa8ede_line1051">1051: </a> pc->setupcalled = 1;
<a name="13793e4a89fa8ede_line1052">1052: </a> }</pre><div>Note the 1051 line reset pc->setupcalled to 1, here the pc from PCSetUp_Composite is pc0 and pc1.</div><div>So although if pc0 has been set up, this will cause pc0 be set up again. </div>
</div></div><div><br></div><div>Is my reading correct?</div></div></div></blockquote><div><br></div><div>When you change the operators, the preconditioner has to be set up again (perhaps not a "complete" setup, but some setup is required).</div>
<div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word"><div><div><br></div><div>Thanks a lot!</div><span class="HOEnZb"><font color="#888888"><div>
Hui</div></font></span><div class="im"><div><br></div><br><blockquote type="cite"><div class="gmail_quote">
<div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word"><div><div><div><br></div><br><blockquote type="cite"><div class="gmail_quote">
<div><br></div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
Thanks!<br>
Hui<br>
<br>
<br>
> I have got a new question. I'm now constructing a PCComposite from PCASM and another PCKSP (coarse<br>
> problem). And construction of PCKSP needs to use the subksp's of PCASM. So I need to PCSetUp on the<br>
> PCASM, right? But the KSP using PCComposite would setup PCASM again. How can I avoid twice setup<br>
> of PCASM ?<br>
<br>
</blockquote></div><br>
</blockquote></div></div><br></div></blockquote></div><br>
</blockquote></div></div><br></div></blockquote></div><br>