<html xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<meta name="Generator" content="Microsoft Word 15 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
{font-family:DengXian;
panose-1:2 1 6 0 3 1 1 1 1 1;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
{font-family:Menlo-Regular;}
@font-face
{font-family:"\@DengXian";
panose-1:2 1 6 0 3 1 1 1 1 1;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
font-size:11.0pt;
font-family:"Calibri",sans-serif;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
span.apple-converted-space
{mso-style-name:apple-converted-space;}
.MsoChpDefault
{mso-style-type:export-only;}
@page WordSection1
{size:8.5in 11.0in;
margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
{page:WordSection1;}
--></style>
</head>
<body lang="EN-US" link="blue" vlink="#954F72" style="word-wrap:break-word">
<div class="WordSection1">
<p class="MsoNormal">Thanks Jacob, </p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Right, I forgot to change the clone -b release to main - my mistake. The c++ dialect option now works without problem.
</p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">My laptop system is indeed configured with WSL2, with Linux kernel of “5.10.60.1-microsoft-standard-WSL2”. And I have a windows 11 Version 21H2 (OS Build 22000.438) as the host system. The Nvidia driver version is “510.06” with cuda 11.4.
<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Interestingly, my ex11fc can now get pass the second petscfinalize. The code can now get to the third loop of kspsolve and reach the mpifinalize without a problem. So changing to main branch solves the petscfinalize error problem. However,
it still complains with an error like: <o:p></o:p></p>
<p class="MsoNormal" style="margin-left:.5in">--------------------------------------------------------------------------</p>
<p class="MsoNormal" style="margin-left:.5in">The call to cuEventDestory failed. This is a unrecoverable error and will</p>
<p class="MsoNormal" style="margin-left:.5in">cause the program to abort.</p>
<p class="MsoNormal" style="margin-left:.5in"> cuEventDestory return value: 400</p>
<p class="MsoNormal" style="margin-left:.5in">Check the cuda.h file for what the return value means.</p>
<p class="MsoNormal" style="margin-left:.5in">--------------------------------------------------------------------------</p>
<p class="MsoNormal">The full log files are also attached. I also noticed there are other event-management related errors like cuEventCreate and cuIpcGetEventHandle in the log. Does it give any insights on why we have the problem?
</p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Cheers, </p>
<p class="MsoNormal">Hao</p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Sent from <a href="https://go.microsoft.com/fwlink/?LinkId=550986">
Mail</a> for Windows</p>
<p class="MsoNormal"><o:p> </o:p></p>
<div style="mso-element:para-border-div;border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in 0in 0in">
<p class="MsoNormal" style="border:none;padding:0in"><b>From: </b><a href="mailto:jacob.fai@gmail.com">Jacob Faibussowitsch</a><br>
<b>Sent: </b>Tuesday, January 25, 2022 11:19 PM<br>
<b>To: </b><a href="mailto:dong-hao@outlook.com">Hao DONG</a><br>
<b>Cc: </b><a href="mailto:petsc-users@mcs.anl.gov">petsc-users</a>; <a href="mailto:junchao.zhang@gmail.com">
Junchao Zhang</a><br>
<b>Subject: </b>Re: [petsc-users] Strange CUDA failure with a second petscfinalize with PETSc 3.16</p>
</div>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Hi Hao,<o:p></o:p></p>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<p class="MsoNormal">I have tried to git pull on the laptop from main and re-config as you suggested.<o:p></o:p></p>
</div>
</blockquote>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<p class="MsoNormal">It looks like you’re still on the release branch. Do <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">```<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">$ git checkout main<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">$ git pull<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">```<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">Then reconfigure. This is also why the cxx dialect flag did not work, I forgot that this change had not made it to release yet.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<p class="MsoNormal">my laptop setup is based on WSL<o:p></o:p></p>
</div>
</blockquote>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<p class="MsoNormal">What version of windows do you have? And what version of WSL? And what version is the linux kernel? You will need at least WSL2 and both your NVIDIA driver, windows version, and linux kernel version are required to be fairly new AFAIK to
be able to run CUDA on them. See here <a href="https://docs.nvidia.com/cuda/wsl-user-guide/index.html">
https://docs.nvidia.com/cuda/wsl-user-guide/index.html</a>. <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">To get your windows version:<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">1. Press Windows key+R<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">2. Type winver in the box, and press enter<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">3. You should see a line with Version and a build number. For example on my windows machine I see “Version 21H2 (OS Build 19044.1496)”<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">To get WSL version:<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">1. Open WSL<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">2. Type uname -r, for example I get "5.10.60.1-microsoft-standard-wsl2"<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">To get NVIDIA driver version:<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">1. Open up the NVIDIA control panel<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">2. Click on “System Information” in the bottom left corner<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">3. You should see a dual list, “Items” and “Details”. In the details column. You should see “Driver verion”. For example on my machine I see “511.23”<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
<div>
<div>
<div>
<div>
<p class="MsoNormal"><span style="color:black">Best regards,<br>
<br>
Jacob Faibussowitsch<br>
(Jacob Fai - booss - oh - vitch)<o:p></o:p></span></p>
</div>
</div>
</div>
</div>
<div>
<p class="MsoNormal"><br>
<br>
<o:p></o:p></p>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<p class="MsoNormal">On Jan 25, 2022, at 03:42, Hao DONG <<a href="mailto:dong-hao@outlook.com">dong-hao@outlook.com</a>> wrote:<o:p></o:p></p>
</div>
<p class="MsoNormal"><o:p> </o:p></p>
<div>
<div>
<p class="MsoNormal">Hi Jacob,<span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">Thanks for the comments – silly that I have overlooked the debugging flag so far. Unfortunately, I am out of office for a couple of days so I cannot confirm the result on my workstation, for now.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">However, I have a laptop with nvidia graphic card (old gtx1050, which is actually slower than the cpu in terms of double precision calculation), I have tried to git pull on the laptop from main and re-config as you suggested.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">However, using “--with-cxx-dialect=14” throws up an error like:<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div style="margin-left:.5in">
<p class="MsoNormal">Unknown C++ dialect: with-cxx-dialect=14<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">And omitting the “--with-cuda-dialect=cxx14” also gives me a similar complaint with:<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div style="margin-left:.5in">
<p class="MsoNormal">CUDA Error: Using CUDA with PetscComplex requires a C++ dialect at least cxx11. Use --with-cxx-dialect=xxx and --with-cuda-dialect=xxx to specify a suitable compiler.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">Eventually I was able to configure and compile with the following config setup:<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div style="margin-left:.5in">
<p class="MsoNormal">./configure --prefix=/opt/petsc/debug --with-cc=mpicc --with-cxx=mpicxx --with-fc=mpif90 --with-scalar-type=complex --with-precision=double --with-cuda-dir=/usr/local/cuda --with-fortran-kernels=1 --with-cxx-dialect=cxx14 --with-cuda-dialect=cxx14
--with-debugging=1<o:p></o:p></p>
</div>
<div style="margin-left:.5in">
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">But still, I got the same error regarding cuda (still the error code 97 thing). I attached the configure and output log of my ex11 on my laptop – is there anything that can help pinpoint the problem? I can also confirm that PETSc 3.15.2
works well with my ex11fc code with cuda, on my laptop. Sadly, my laptop setup is based on WSL, which is far from an ideal environment to test CUDA. I will let you know once I get my hands on my workstations.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">Cheers,<span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">Hao<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">Sent from<span class="apple-converted-space"> </span><a href="https://go.microsoft.com/fwlink/?LinkId=550986">Mail</a><span class="apple-converted-space"> </span>for Windows<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div style="border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in 0in 0in">
<div>
<p class="MsoNormal"><b>From:<span class="apple-converted-space"> </span></b><a href="mailto:jacob.fai@gmail.com">Jacob Faibussowitsch</a><br>
<b>Sent:<span class="apple-converted-space"> </span></b>Tuesday, January 25, 2022 1:22 AM<br>
<b>To:<span class="apple-converted-space"> </span></b><a href="mailto:dong-hao@outlook.com">Hao DONG</a><br>
<b>Cc:<span class="apple-converted-space"> </span></b><a href="mailto:petsc-users@mcs.anl.gov">petsc-users</a>;<span class="apple-converted-space"> </span><a href="mailto:junchao.zhang@gmail.com">Junchao Zhang</a><br>
<b>Subject:<span class="apple-converted-space"> </span></b>Re: [petsc-users] Strange CUDA failure with a second petscfinalize with PETSc 3.16<o:p></o:p></p>
</div>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">Hi Hao,<o:p></o:p></p>
</div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
<div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<div>
<div>
<div>
<div>
<p class="MsoNormal"><span style="font-family:"Menlo-Regular",serif">Any luck reproducing the CUDA problem?</span><o:p></o:p></p>
</div>
</div>
</div>
</div>
</div>
</blockquote>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal">Sorry for the long radio silence, I still have not been able to reproduce the problem unfortunately. I have tried on a local machine, and a few larger clusters and all return without errors both with and without cuda…<o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal">Can you try pulling the latest version of main, reconfiguring and trying again?<o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal">BTW, your configure arguments are a little wonky:<o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal">1. --with-clanguage=c - this isn’t needed, PETSc will default to C<o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal">2. --with-cxx-dialect=cxx14 --with-cuda-dialect=cxx14 - use --with-cxx-dialect=14 instead, PETSc will detect that you have gnu compilers and enable gnu extensions<o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal">3. -with-debugging=1 - this is missing an extra dash, but you also have optimization flags set so maybe just leave this one out<o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<div>
<div>
<div>
<div>
<p class="MsoNormal">Best regards,<br>
<br>
Jacob Faibussowitsch<br>
(Jacob Fai - booss - oh - vitch)<o:p></o:p></p>
</div>
</div>
</div>
</div>
</div>
<div>
<div>
<p class="MsoNormal"><br>
<br>
<br>
<o:p></o:p></p>
</div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<div>
<p class="MsoNormal">On Jan 23, 2022, at 21:29, Hao DONG <<a href="mailto:dong-hao@outlook.com">dong-hao@outlook.com</a>> wrote:<o:p></o:p></p>
</div>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<div>
<div>
<div>
<div>
<p class="MsoNormal"><span style="font-size:9.0pt;font-family:"Menlo-Regular",serif">Dear Jacob, </span><o:p></o:p></p>
</div>
<div>
<div>
<p class="MsoNormal"><span style="font-size:9.0pt;font-family:"Menlo-Regular",serif"> </span><o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"><span style="font-size:9.0pt;font-family:"Menlo-Regular",serif">Any luck reproducing the CUDA problem? - just write to check in, in case somehow the response did not reach me (this happens to my colleagues abroad sometimes, probably due
to the Wall). </span><o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"><span style="font-size:9.0pt;font-family:"Menlo-Regular",serif"> </span><o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"><span style="font-size:9.0pt;font-family:"Menlo-Regular",serif">All the best,</span><o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"><span style="font-size:9.0pt;font-family:"Menlo-Regular",serif">Hao</span><o:p></o:p></p>
</div>
<div>
<div>
<p class="MsoNormal"><span style="font-size:9.0pt;font-family:"Menlo-Regular",serif"><br>
<br>
<br>
</span><o:p></o:p></p>
</div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<p class="MsoNormal" style="margin-bottom:12.0pt"><span style="font-size:9.0pt;font-family:"Menlo-Regular",serif">On Jan 19, 2022, at 3:01 PM, Hao DONG <</span><a href="mailto:dong-hao@outlook.com"><span style="font-size:9.0pt;font-family:"Menlo-Regular",serif">dong-hao@outlook.com</span></a><span style="font-size:9.0pt;font-family:"Menlo-Regular",serif">>
wrote:</span></p>
</blockquote>
</div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<div>
<p class="MsoNormal"><span style="font-size:9.0pt"></span><o:p></o:p></p>
</div>
<div>
<div>
<p class="MsoNormal">Thanks Jacob for looking into this –<span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal">You can see the updated source code of ex11fc in the attachment – although there is not much that I modified (except for the jabbers I outputted). I also attached the full output (ex11fc.log) along with the configure.log file. It’s an old
dual Xeon workstation (one of my “production” machines) with Linux kernel 5.4.0 and gcc 9.3.<o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal">I simply ran the code with<span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal">mpiexec -np 2 ex11fc -usecuda<span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal">for GPU test. And as stated before, calling without the “-usecuda” option shows no errors.<o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal">Please let me know if you find anything wrong with the configure/code. <o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal">Cheers,<span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal">Hao<o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
<div style="border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in 0in 0in">
<div>
<div>
<p class="MsoNormal"><b>From:<span class="apple-converted-space"> </span></b><a href="mailto:jacob.fai@gmail.com">Jacob Faibussowitsch</a><br>
<b>Sent:<span class="apple-converted-space"> </span></b>Wednesday, January 19, 2022 3:38 AM<br>
<b>To:<span class="apple-converted-space"> </span></b><a href="mailto:dong-hao@outlook.com">Hao DONG</a><br>
<b>Cc:<span class="apple-converted-space"> </span></b><a href="mailto:junchao.zhang@gmail.com">Junchao Zhang</a>;<span class="apple-converted-space"> </span><a href="mailto:petsc-users@mcs.anl.gov">petsc-users</a><br>
<b>Subject:<span class="apple-converted-space"> </span></b>Re: [petsc-users] Strange CUDA failure with a second petscfinalize with PETSc 3.16<o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal">Apologies, forgot to mention in my previous email but can you also include a copy of the full printout of the error message that you get? It will include all the command-line flags that you ran with (if any) so I can exactly mirror your
environment.<o:p></o:p></p>
</div>
</div>
<div>
<div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
<div>
<div>
<div>
<div>
<div>
<div>
<p class="MsoNormal">Best regards,<br>
<br>
Jacob Faibussowitsch<br>
(Jacob Fai - booss - oh - vitch)<o:p></o:p></p>
</div>
</div>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal"><br>
<br>
<br>
<br>
<o:p></o:p></p>
</div>
</div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<div>
<div>
<p class="MsoNormal">On Jan 18, 2022, at 14:06, Jacob Faibussowitsch <<a href="mailto:jacob.fai@gmail.com">jacob.fai@gmail.com</a>> wrote:<o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal" style="margin-bottom:12.0pt"><span style="font-size:9.0pt;font-family:"Menlo-Regular",serif">Can you send your updated source file as well as your configure.log (should be $PETSC_DIR/configure.log). I will see if I can reproduce the error
on my end.</span></p>
<div>
<div>
<div>
<p class="MsoNormal"><span style="font-size:9.0pt;font-family:"Menlo-Regular",serif">Best regards,<br>
<br>
Jacob Faibussowitsch<br>
(Jacob Fai - booss - oh - vitch)</span><o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal"><span style="font-size:9.0pt;font-family:"Menlo-Regular",serif"><br>
<br>
<br>
<br>
</span><o:p></o:p></p>
</div>
</div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<p class="MsoNormal" style="margin-bottom:12.0pt"><span style="font-size:9.0pt;font-family:"Menlo-Regular",serif">On Jan 17, 2022, at 23:06, Hao DONG <</span><a href="mailto:dong-hao@outlook.com"><span style="font-size:9.0pt;font-family:"Menlo-Regular",serif">dong-hao@outlook.com</span></a><span style="font-size:9.0pt;font-family:"Menlo-Regular",serif">>
wrote:</span></p>
</blockquote>
</div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<div>
<div>
<p class="MsoNormal"><span style="font-size:9.0pt"></span><o:p></o:p></p>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal">Dear Junchao and Jacob,<span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal">Thanks a lot for the response – I also don’t understand why this is related to the device, especially on why the procedure can be successfully finished for *<b>once</b>* – As instructed, I tried to add a CHKERRA() macro after (almost) every
petsc line – such as the initialization, mat assemble, ksp create, solve, mat destroy, etc. However, all other petsc commands returns with error code 0. It only gives me a similar (still not very informative) error after I call the petscfinalize (again for
the second time), with error code 97: <span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal">[0]PETSC ERROR: --------------------- Error Message --------------------------------------------------------------<o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal">[0]PETSC ERROR: GPU error<o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal">[0]PETSC ERROR: cuda error 709 (cudaErrorContextIsDestroyed) : context is destroyed<o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal">[0]PETSC ERROR: See<span class="apple-converted-space"> </span><a href="https://petsc.org/release/faq/">https://petsc.org/release/faq/</a><span class="apple-converted-space"> </span>for trouble shooting.<o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal">[0]PETSC ERROR: Petsc Release Version 3.16.3, unknown<o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal">[0]PETSC ERROR: ./ex11f on a named stratosphere by donghao Tue Jan 18 11:39:43 2022<o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal">[0]PETSC ERROR: Configure options --prefix=/opt/petsc/complex-double-with-cuda --with-cc=mpicc --with-cxx=mpicxx --with-fc=mpif90 COPTFLAGS="-O3 -mavx2" CXXOPTFLAGS="-O3 -mavx2" FOPTFLAGS="-O3 -ffree-line-length-none -mavx2" CUDAOPTFLAGS=-O3
--with-cxx-dialect=cxx14 --with-cuda-dialect=cxx14 --with-scalar-type=complex --with-precision=double --with-cuda-dir=/usr/local/cuda --with-debugging=1<o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal">[0]PETSC ERROR: #1 PetscFinalize() at /home/donghao/packages/petsc-current/src/sys/objects/pinit.c:1638<o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal">[0]PETSC ERROR: #2 User provided function() at User file:0<o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal">I can also confirm that rolling back to petsc 3.15 will *<b>not</b>* see the problem, even with the new nvidia driver. And petsc 3.16.3 with an old nvidia driver (470.42) also get this same error. So it’s probably not connected to the
nvidia driver.<span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal">Any idea on where I should look at next?<span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal">Thanks a lot in advance, and all the best,<o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal">Hao<o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
</div>
<div style="border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in 0in 0in">
<div>
<div>
<div>
<p class="MsoNormal"><b>From:<span class="apple-converted-space"> </span></b><a href="mailto:jacob.fai@gmail.com">Jacob Faibussowitsch</a><br>
<b>Sent:<span class="apple-converted-space"> </span></b>Sunday, January 16, 2022 12:12 AM<br>
<b>To:<span class="apple-converted-space"> </span></b><a href="mailto:junchao.zhang@gmail.com">Junchao Zhang</a><br>
<b>Cc:<span class="apple-converted-space"> </span></b><a href="mailto:petsc-users@mcs.anl.gov">petsc-users</a>;<span class="apple-converted-space"> </span><a href="mailto:dong-hao@outlook.com">Hao DONG</a><br>
<b>Subject:<span class="apple-converted-space"> </span></b>Re: [petsc-users] Strange CUDA failure with a second petscfinalize with PETSc 3.16<o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<p class="MsoNormal">I don’t quite understand how it is getting to the CUDA error to be honest. None of the code in the stack trace is anywhere near the device code. Reading the error message carefully, it first chokes on PetscLogGetStageLog() from a call to
PetscClassIdRegister():<o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<div>
<p class="MsoNormal">PetscErrorCode PetscLogGetStageLog(PetscStageLog *stageLog)<o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<p class="MsoNormal">{<o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<p class="MsoNormal"> PetscFunctionBegin;<o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<p class="MsoNormal"> PetscValidPointer(stageLog,1);<o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<p class="MsoNormal"> if (!petsc_stageLog) {<o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<p class="MsoNormal"> fprintf(stderr, "PETSC ERROR: Logging has not been enabled.\nYou might have forgotten to call PetscInitialize().\n");<o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<p class="MsoNormal"> PETSCABORT(MPI_COMM_WORLD, PETSC_ERR_SUP); // Here<o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<p class="MsoNormal"> }<o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<p class="MsoNormal"> ...<o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<p class="MsoNormal">But then jumps to PetscFinalize(). You can also see the "You might have forgotten to call PetscInitialize().” message in the error message, just under the topmost level of the stack trace.<o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<p class="MsoNormal">Can you check the value of ierr of each function call (use the CHKERRA() macro to do so)? I suspect the problem here that errors occurring previously in the program are being ignored, leading to the garbled stack trace.<o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<div>
<div>
<div>
<p class="MsoNormal">Best regards,<br>
<br>
Jacob Faibussowitsch<br>
(Jacob Fai - booss - oh - vitch)<o:p></o:p></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<p class="MsoNormal"><br>
<br>
<br>
<br>
<br>
<o:p></o:p></p>
</div>
</div>
</div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<div>
<div>
<div>
<p class="MsoNormal">On Jan 14, 2022, at 20:58, Junchao Zhang <<a href="mailto:junchao.zhang@gmail.com">junchao.zhang@gmail.com</a>> wrote:<o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<div>
<div>
<p class="MsoNormal">Jacob, <o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<p class="MsoNormal"> Could you have a look as it seems the "invalid device context" is in your newly added module?<o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<p class="MsoNormal"> Thanks<o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<div>
<div>
<p class="MsoNormal">--Junchao Zhang<o:p></o:p></p>
</div>
</div>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
</div>
<div>
<div>
<div>
<div>
<div>
<p class="MsoNormal">On Fri, Jan 14, 2022 at 12:49 AM Hao DONG <<a href="mailto:dong-hao@outlook.com">dong-hao@outlook.com</a>> wrote:<o:p></o:p></p>
</div>
</div>
</div>
</div>
<blockquote style="border:none;border-left:solid #CCCCCC 1.0pt;padding:0in 0in 0in 6.0pt;margin-left:4.8pt;margin-top:5.0pt;margin-right:0in;margin-bottom:5.0pt">
<div>
<div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">Dear All, <o:p></o:p></p>
</div>
</div>
</div>
<div>
<p class="MsoNormal" style="margin-left:667.2pt"> </p>
</div>
<div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">I have encountered a peculiar problem when fiddling with a code with PETSC 3.16.3 (which worked fine with PETSc 3.15). It is a very straight forward PDE-based optimization code which repeatedly solves a linearized
PDE problem with KSP in a subroutine (the rest of the code does not contain any PETSc related content). The main program provides the subroutine with an MPI comm. Then I set the comm as PETSC_COMM_WORLD to tell PETSC to attach to it (and detach with it when
the solving is finished each time).<o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<p class="MsoNormal" style="margin-left:667.2pt"> </p>
</div>
<div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">Strangely, I observe a CUDA failure whenever the petscfinalize is called for a *second* time. In other words, the first and second PDE calculations with GPU are fine (with correct solutions). The petsc code
just fails after the SECOND petscfinalize command is called. You can also see the PETSC config in the error message:<span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
</div>
</div>
<p class="MsoNormal" style="margin-left:667.2pt"> </p>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">[1]PETSC ERROR: --------------------- Error Message --------------------------------------------------------------<o:p></o:p></p>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">[1]PETSC ERROR: GPU error<o:p></o:p></p>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">[1]PETSC ERROR: cuda error 201 (cudaErrorDeviceUninitialized) : invalid device context<o:p></o:p></p>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">[1]PETSC ERROR: See<span class="apple-converted-space"> </span><a href="https://petsc.org/release/faq/" target="_blank">https://petsc.org/release/faq/</a><span class="apple-converted-space"> </span>for trouble
shooting.<o:p></o:p></p>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">[1]PETSC ERROR: Petsc Release Version 3.16.3, unknown<o:p></o:p></p>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">[1]PETSC ERROR: maxwell.gpu on a named stratosphere by hao Fri Jan 14 10:21:05 2022<o:p></o:p></p>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">[1]PETSC ERROR: Configure options --prefix=/opt/petsc/complex-double-with-cuda --with-cc=mpicc --with-cxx=mpicxx --with-fc=mpif90 COPTFLAGS="-O3 -mavx2" CXXOPTFLAGS="-O3 -mavx2" FOPTFLAGS="-O3 -ffree-line-length-none
-mavx2" CUDAOPTFLAGS=-O3 --with-cxx-dialect=cxx14 --with-cuda-dialect=cxx14 --with-scalar-type=complex --with-precision=double --with-cuda-dir=/usr/local/cuda --with-debugging=1<o:p></o:p></p>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">[1]PETSC ERROR: #1 PetscFinalize() at /home/hao/packages/petsc-current/src/sys/objects/pinit.c:1638<o:p></o:p></p>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">You might have forgotten to call PetscInitialize().<o:p></o:p></p>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">The EXACT line numbers in the error traceback are not available.<o:p></o:p></p>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">Instead the line number of the start of the function is given.<o:p></o:p></p>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">[1] #1 PetscAbortFindSourceFile_Private() at /home/hao/packages/petsc-current/src/sys/error/err.c:35<o:p></o:p></p>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">[1] #2 PetscLogGetStageLog() at /home/hao/packages/petsc-current/src/sys/logging/utils/stagelog.c:29<o:p></o:p></p>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">[1] #3 PetscClassIdRegister() at /home/hao/packages/petsc-current/src/sys/logging/plog.c:2376<o:p></o:p></p>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">[1] #4 MatMFFDInitializePackage() at /home/hao/packages/petsc-current/src/mat/impls/mffd/mffd.c:45<o:p></o:p></p>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">[1] #5 MatInitializePackage() at /home/hao/packages/petsc-current/src/mat/interface/dlregismat.c:163<o:p></o:p></p>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">[1] #6 MatCreate() at /home/hao/packages/petsc-current/src/mat/utils/gcreate.c:77<o:p></o:p></p>
</div>
</div>
</div>
<p class="MsoNormal" style="margin-left:667.2pt"> </p>
<div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">However, it doesn’t seem to affect the other part of my code, so the code can continue running until it gets to the petsc part again (the *<b>third</b>* time). Unfortunately, it doesn’t give me any further information
even if I set the debugging to yes in the configure file. It also worth noting that PETSC without CUDA (i.e. with simple MATMPIAIJ) works perfectly fine. <o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<p class="MsoNormal" style="margin-left:667.2pt"> </p>
</div>
<div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">I am able to re-produce the problem with a toy code modified from ex11f. Please see the attached file (ex11fc.F90) for details. Essentially the code does the same thing as ex11f, but three times with a do loop.
To do that I added an extra MPI_INIT/MPI_FINALIZE to ensure that the MPI communicator is not destroyed when PETSC_FINALIZE is called. I used the PetscOptionsHasName utility to check if you have “-usecuda” in the options. So running the code with and without
that option can give you a comparison w/o CUDA. I can see that the code also fails after the second loop of the KSP operation. Could you kindly shed some lights on this problem?<o:p></o:p></p>
</div>
</div>
</div>
</div>
<div>
<p class="MsoNormal" style="margin-left:667.2pt"> </p>
</div>
<div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">I should say that I am not even sure if the problem is from PETSc, as I also accidentally updated the NVIDIA driver (for now it is 510.06 with cuda 11.6). And it is well known that NVIDIA can give you some surprise
in the updates (yes, I know I shouldn’t have touched that if it’s not broken). But my CUDA code without PETSC (which basically does the same PDE thing, but with cusparse/cublas directly) seems to work just fine after the update. It is also possible that my
petsc code related to CUDA was not quite “legitimate” – I just use:<span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt"> MatSetType(A, MATMPIAIJCUSPARSE, ierr)<o:p></o:p></p>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">and<span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt"> MatCreateVecs(A, u, PETSC_NULL_VEC, ierr)<o:p></o:p></p>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">to make the data onto GPU. I would very much appreciate it if you could show me the “right” way to do that. <o:p></o:p></p>
</div>
</div>
</div>
<p class="MsoNormal" style="margin-left:667.2pt"> </p>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">Thanks a lot in advance, and all the best,<o:p></o:p></p>
</div>
</div>
</div>
</div>
<div style="margin-left:40.8pt">
<div>
<div style="margin-left:225.6pt">
<p class="MsoNormal" style="margin-left:974.4pt">Hao<o:p></o:p></p>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</div>
</div>
</blockquote>
</div>
</div>
</div>
</div>
</blockquote>
</div>
</div>
</blockquote>
</div>
</div>
</div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
</div>
</div>
</blockquote>
</div>
</div>
</div>
</div>
<div>
<p class="MsoNormal"><configure.log><ex11fc.F90><ex11fc.log><o:p></o:p></p>
</div>
</div>
</blockquote>
</div>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<p class="MsoNormal"><configure-xps.log><ex11fc-xps.log><o:p></o:p></p>
</div>
</blockquote>
</div>
</div>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
</body>
</html>