<html><head></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">I think this is definitely an issue with setting the affinities for threads, i.e., the assignment of threads to cores. Ideally each thread should be assigned to a distinct core but in your case all the 4 threads are getting pinned to the same core resulting in such a massive slowdown. Unfortunately, the thread affinities for OpenMP are set through environment variables. For Intel's OpenMP one needs to define the thread affinities through the environment variable KMP_AFFINITY. See this document here <a href="http://software.intel.com/sites/products/documentation/studio/composer/en-us/2011Update/compiler_c/optaps/common/optaps_openmp_thread_affinity.htm">http://software.intel.com/sites/products/documentation/studio/composer/en-us/2011Update/compiler_c/optaps/common/optaps_openmp_thread_affinity.htm</a>. Try setting the affinities via KMP_AFFINITY and let us know if it works.<div><br></div><div>Shri<br><div><div>On Sep 21, 2013, at 11:06 PM, Danyang Su wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite">
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
<div text="#000000" bgcolor="#FFFFFF">
<div class="moz-cite-prefix">Hi Shri,<br>
<br>
Thanks for your info. It can work with the option -threadcomm_type
openmp. But another problem arises, as described as follows.<br>
<br>
The sparse matrix is 53760*53760 with 1067392 non-zero entries.
If the codes is compiled using PETSc-3.4.2, it works fine, the
equations can be solved quickly and I can see the speedup. But if
the code is compiled using PETSc-dev with OpenMP option, it takes
a long time in solving the equations and I cannot see any speedup
when more processors are used.<br>
<br>
For PETSc-3.4.2, run by "mpiexec -n 4 ksp_inhm_d -log_summary
log_mpi4_petsc3.4.2.log", the iteration and runtime are:<br>
Iterations 6 <b><font color="#ff0000">time_assembly
0.4137E-01</font></b> time_ksp 0.9296E-01<br>
<br>
For PETSc-dev, run by "mpiexec -n 1 ksp_inhm_d -threadcomm_type
openmp -threadcomm_nthreads 4 -log_summary
log_openmp_petsc_dev.log", the iteration and runtime are:<br>
Iterations 6 <b><font color="#ff0000">time_assembly
0.3595E+03</font></b> time_ksp 0.2907E+00<br>
<br>
Most of the time <b><font color="#ff0000">'time_assembly
0.3595E+03</font></b>' is spent on the following codes<br>
do i = istart, iend - 1<br>
ii = ia_in(i+1)<br>
jj = ia_in(i+2)<br>
call MatSetValues(a, ione, i, jj-ii,
ja_in(ii:jj-1)-1, a_in(ii:jj-1), Insert_Values, ierr)<br>
end do <br>
<br>
The log files for both PETSc-3.4.2 and PETSc-dev are attached.<br>
<br>
Is there anything wrong with my codes or with running option? The
above codes works fine when using MPICH.<br>
<br>
Thanks and regards,<br>
<br>
Danyang<br>
<br>
On 21/09/2013 2:09 PM, Shri wrote:<br>
</div>
<blockquote cite="mid:933EF7F4-598A-4F88-8902-0613E27F0679@mcs.anl.gov" type="cite">
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<div>There are three thread communicator types in PETSc. The
default is "no thread" which is basically a non-threaded
version. The other two types are "openmp" and "pthread". If you
want to use OpenMP then use the option -threadcomm_type openmp.</div>
<div><br>
</div>
<div>Shri</div>
<div><br>
On Sep 21, 2013, at 3:46 PM, Danyang Su <<a moz-do-not-send="true" href="mailto:danyang.su@gmail.com">danyang.su@gmail.com</a>>
wrote:<br>
<br>
</div>
<blockquote type="cite">
<div>
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
<div class="moz-cite-prefix">Hi Barry,<br>
<br>
Thanks for the quick reply.<br>
<br>
After changing<br>
<font color="#ff0000"><i>#if
defined(PETSC_HAVE_PTHREADCLASSES) || defined
(PETSC_HAVE_OPENMP) </i></font><br>
to <br>
<font color="#ff0000"><i>#if
defined(PETSC_HAVE_PTHREADCLASSES)</i></font><br>
and comment out<br>
<font color="#ff0000"><i>#elif defined(PETSC_HAVE_OPENMP)</i><i><br>
</i><i>PETSC_EXTERN PetscStack *petscstack;</i></font><br>
<br>
It can be compiled and validated with "make test".<br>
<br>
But I still have questions on running the examples. After
rebuild the codes (e.g., ksp_ex2f.f), I can run it with
"mpiexec -n 1 ksp_ex2f", or "mpiexec -n 4 ksp_ex2f", or
"mpiexec -n 1 ksp_ex2f -threadcomm_nthreads 1", but if I run
it with "mpiexec -n 1 ksp_ex2f -threadcomm_nthreads 4",
there will be a lot of error information (attached).<br>
<br>
The codes is not modified and there is no OpenMP routines in
it. For the current development in my project, I want to
keep the OpenMP codes in calculating matrix values, but want
to solve it with PETSc (OpenMP). Is it possible? <br>
<br>
Thanks and regards,<br>
<br>
Danyang<br>
<br>
<br>
<br>
On 21/09/2013 7:26 AM, Barry Smith wrote:<br>
</div>
<blockquote cite="mid:101244E8-FFFA-4F3F-A4BC-CC47F68D93B7@mcs.anl.gov" type="cite">
<pre wrap=""> Danyang,
I don't think the || defined (PETSC_HAVE_OPENMP) belongs in the code below.
/* Linux functions CPU_SET and others don't work if sched.h is not included before
including pthread.h. Also, these functions are active only if either _GNU_SOURCE
or __USE_GNU is not set (see /usr/include/sched.h and /usr/include/features.h), hence
set these first.
*/
#if defined(PETSC_HAVE_PTHREADCLASSES) || defined (PETSC_HAVE_OPENMP)
Edit include/petscerror.h and locate these lines and remove that part and then rerun make all. Let us know if it works or not.
Barry
i.e. replace
#if defined(PETSC_HAVE_PTHREADCLASSES) || defined (PETSC_HAVE_OPENMP)
with
#if defined(PETSC_HAVE_PTHREADCLASSES)
On Sep 21, 2013, at 6:53 AM, Matthew Knepley <a moz-do-not-send="true" class="moz-txt-link-rfc2396E" href="mailto:petsc-maint@mcs.anl.gov"><petsc-maint@mcs.anl.gov></a> wrote:
</pre>
<blockquote type="cite">
<pre wrap="">On Sat, Sep 21, 2013 at 12:18 AM, Danyang Su <a moz-do-not-send="true" class="moz-txt-link-rfc2396E" href="mailto:danyang.su@gmail.com"><danyang.su@gmail.com></a> wrote:
Hi All,
I got error information in compiling petsc-dev with openmp in cygwin. Before, I have successfully compiled petsc-3.4.2 and it works fine.
The log files have been attached.
The OpenMP configure test is wrong. It clearly fails to find pthread.h, but the test passes. Then in petscerror.h
we guard pthread.h using PETSC_HAVE_OPENMP. Can someone who knows OpenMP fix this?
Matt
Thanks,
Danyang
--
What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.
-- Norbert Wiener
</pre>
</blockquote>
</blockquote>
<br>
</div>
</blockquote>
<blockquote type="cite">
<div><error.txt></div>
</blockquote>
</blockquote>
<br>
</div>
<span><log_mpi4_petsc3.4.2.log></span><span><log_openmp_petsc_dev.log></span></blockquote></div><br></div></body></html>