[petsc-users] Segmentation Violation error using SuperLU_DIST in ex 19.c

[maitri ksh]

Hi Satish,
Yes, that simple build works with no error. I configured petsc again with
the configure options:

 PETSC_ARCH=linux-gnu-c-debug -start-in-debugger[noxterm]
--with-cc=/usr/local/gcc11/bin/gcc --with-cxx=/usr/local/gcc11/bin/g++
--with-fc=gfortran --with-debugging=1 --with-scalar-type=complex
--download-mpich --download-fblaslapack --with-matlab-dir=/usr/local/matlab
--download-superlu --download-superlu_dist --download-hdf5 --download-mumps
--download-scalapack --download--parmetis --download-metis
--download-ptscotch --download-bison --download-cmake --download-make

Now, it runs the superLU_dist test successfully but it gives an error with
MATLAB engine 'Possible error running C/C++ src/vec/vec/tutorials/ex31 with
MATLAB engine' and also an error with MAKE check.



On Thu, Nov 23, 2023 at 10:26 PM Satish Balay <balay at mcs.anl.gov> wrote:

> Can you do a simple build with only superlu-dist and see if the error
> persists?
>
> ./configure PETSC_ARCH=linux-slu --with-cc=/usr/local/gcc11/bin/gcc
> --with-cxx=/usr/local/gcc11/bin/g++ --with-fc=gfortran --with-debugging=1
> --with-scalar-type=complex --download-mpich --download-fblaslapack
> --download-superlu_dist
> make
> make PETSC_ARCH=linux-slu check
>
> Satish
>
> On Thu, 23 Nov 2023, maitri ksh wrote:
>
> > Hi,
> > I ran into an error while using SuperLU_DIST in ex 19.c, I am not sure
> how
> > to debug this, can anyone please help. The 'configure.log' file is
> attached
> > for your reference.
> > Thanks,
> > Maitri
> >
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20231124/fb191a95/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: configure.log
Type: application/octet-stream
Size: 9984976 bytes
Desc: not available
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20231124/fb191a95/attachment-0001.obj>