<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=Windows-1252">
<style type="text/css" style="display:none;"><!-- P {margin-top:0;margin-bottom:0;} --></style>
</head>
<body dir="ltr">
<div id="divtagdefaultwrapper" style="font-size:12pt;color:#000000;font-family:Calibri,Helvetica,sans-serif;" dir="ltr">
<p><br>
</p>
<meta content="text/html; charset=UTF-8">
<div dir="ltr">
<div id="x_divtagdefaultwrapper" dir="ltr" style="font-size: 12pt; color: rgb(0, 0, 0); font-family: Calibri, Helvetica, sans-serif, "EmojiFont", "Apple Color Emoji", "Segoe UI Emoji", NotoColorEmoji, "Segoe UI Symbol", "Android Emoji", EmojiSymbols;">
Ok thanks all, so with Satish test, we know it is MPI_Init call from PetscInitialize.
<p>Probably, as Matt said, something related to <span>gethostbyname()</span> ...</p>
<p><br>
</p>
<p>I will do some other test, contact MPICH support and will report any workaround,</p>
<p><br>
</p>
<p>Thanks again for your reactivity !<br>
</p>
<p><br>
</p>
<div id="x_Signature">
<div style="font-family:Tahoma; font-size:13px">
<div class="x_BodyFragment"><font size="2"><span style="font-size:10pt">
<div class="x_PlainText">Pierre LEDAC<br>
Commissariat à l’énergie atomique et aux énergies alternatives<br>
Centre de SACLAY<br>
DES/ISAS/DM2S/STMF/LGLS<br>
Bâtiment 451 – point courrier n°43<br>
F-91191 Gif-sur-Yvette<br>
+33 1 69 08 04 03<br>
+33 6 83 42 05 79</div>
</span></font></div>
</div>
</div>
</div>
<hr tabindex="-1" style="display:inline-block; width:98%">
<div id="x_divRplyFwdMsg" dir="ltr"><font style="font-size:11pt" face="Calibri, sans-serif" color="#000000"><b>De :</b> Satish Balay <balay@mcs.anl.gov><br>
<b>Envoyé :</b> mercredi 27 juillet 2022 18:57:16<br>
<b>À :</b> LEDAC Pierre<br>
<b>Cc :</b> petsc-dev@mcs.anl.gov<br>
<b>Objet :</b> Re: [petsc-dev] [Minor issue] Lag during PetscInitialize on Fedora 36 (never seen before with other OS)</font>
<div> </div>
</div>
</div>
<font size="2"><span style="font-size:10pt">
<div class="PlainText">I'm seeing this behavior on a laptop - but not desktop [both F36] (with both mpich and openmpi buids)<br>
<br>
>>>> laptop/mpich <<<<<br>
balay@p1 /home/balay/petsc/src/ksp/ksp/tests (main =)<br>
$ sleep 30; time ./ex39; time ./ex39<br>
<br>
real 0m1.709s<br>
user 0m0.079s<br>
sys 0m0.042s<br>
<br>
real 0m0.116s<br>
user 0m0.077s<br>
sys 0m0.014s<br>
<<<<<br>
<br>
However --with-mpi=0 is fine<br>
<br>
>>>><br>
balay@p1 /home/balay/petsc/src/ksp/ksp/tests (main =)<br>
$ sleep 30; time ./ex39; time ./ex39<br>
<br>
real 0m0.116s<br>
user 0m0.108s<br>
sys 0m0.008s<br>
<br>
real 0m0.081s<br>
user 0m0.072s<br>
sys 0m0.008s<br>
<<<<<<br>
<br>
Here is the desktop run (with mpich)<br>
<br>
>>><br>
[balay@pj01 tests]$ sleep 30; time ./ex39; time ./ex39<br>
<br>
real 0m0.065s<br>
user 0m0.033s<br>
sys 0m0.019s<br>
<br>
real 0m0.053s<br>
user 0m0.029s<br>
sys 0m0.013s<br>
<<<<<br>
<br>
And --with-mpi=0<br>
<br>
>>><br>
[balay@pj01 tests]$ sleep 30; time ./ex39; time ./ex39<br>
<br>
real 0m0.087s<br>
user 0m0.046s<br>
sys 0m0.015s<br>
<br>
real 0m0.024s<br>
user 0m0.018s<br>
sys 0m0.006s<br>
<<<<br>
<br>
But don't know why its behaving this way [on the laptop]<br>
<br>
Satish<br>
<br>
On Wed, 27 Jul 2022, LEDAC Pierre wrote:<br>
<br>
> Hello,<br>
> <br>
> <br>
> Recently migrated from Fedora34 to Fedora36, using PETSc I have some lag during the PetscInitialize, which<br>
> <br>
> disappeared if I run again immediately the binary. But after few seconds, the lag happens again (see below).<br>
> <br>
> <br>
> I suspected MPICH 4.0.2 on Fedora36, but a small reproducer indicated the issue is not in MPI_Init but really during PetscInitialize<br>
> <br>
> <br>
> Just annoying during testing, not very important, but did someone already see this ?<br>
> <br>
> <br>
> Thanks,<br>
> <br>
> <br>
> portable: /volatile/ledacp/petsc/src/ksp/ksp/tests (main) > time ./ex39<br>
> <br>
> real 0m3,593s<br>
> user 0m0,872s<br>
> sys 0m0,433s<br>
> portable: /volatile/ledacp/petsc/src/ksp/ksp/tests (main) > time ./ex39<br>
> <br>
> real 0m0,310s<br>
> user 0m0,884s<br>
> sys 0m0,023s<br>
> <br>
> # Wait ~20 seconds then again:<br>
> <br>
> portable: /volatile/ledacp/petsc/src/ksp/ksp/tests (main) > time ./ex39<br>
> <br>
> real 0m3,507s<br>
> user 0m0,897s<br>
> sys 0m0,025s<br>
> portable: /volatile/ledacp/petsc/src/ksp/ksp/tests (main) > time ./ex39<br>
> <br>
> real 0m0,154s<br>
> user 0m0,725s<br>
> sys 0m0,015s<br>
> portable: /volatile/ledacp/petsc/src/ksp/ksp/tests (main) > time ./ex39<br>
> <br>
> real 0m0,176s<br>
> user 0m0,875s<br>
> sys 0m0,021s<br>
> <br>
> <br>
> <br>
> Pierre LEDAC<br>
> Commissariat à l’énergie atomique et aux énergies alternatives<br>
> Centre de SACLAY<br>
> DES/ISAS/DM2S/STMF/LGLS<br>
> Bâtiment 451 – point courrier n°43<br>
> F-91191 Gif-sur-Yvette<br>
> +33 1 69 08 04 03<br>
> +33 6 83 42 05 79<br>
> <br>
</div>
</span></font></div>
</body>
</html>