Hi Fande,

No particular reason for using the PETSc-dev - frankly I didn't notice I
was using it.
What would you recommend me to do? Should I download the latest or a
reasonably previous stable version?

However, I am now checking / running the installation of a previous PETSc
version (petsc-3.7.2; which I've been using until recently) with the up to
date Libmesh version. Will keep you posted if I have any problems with the
installation.

BW,
Vasileios





On 5 April 2017 at 15:55, Kong, Fande <[email protected]> wrote:

> Hi Vasileios,
>
> Any reason to use PETSc-dev? Do you look for a new functionality that is
> not available in the current release? Some of PETSc APIs may get updated
> when switch from a low version to a high one or dev.
>
> Here are updated APIs in petsc-dev
>
> https://www.mcs.anl.gov/petsc/documentation/changes/dev.html
>
> If no specific reason, I would suggest to stay with the current release.
>
> Fande,
>
> On Wed, Apr 5, 2017 at 8:34 AM, Vasileios Vavourakis <[email protected]>
> wrote:
>
>> Indeed, very strange error.
>> To answer your question about PETSc version; I employed the following
>> commit:
>>  fbfa34dc2ebfea5502b6dc1528c107aa61ca96c8
>>
>>
>> On 5 April 2017 at 15:17, John Peterson <[email protected]> wrote:
>>
>> >
>> >
>> > On Wed, Apr 5, 2017 at 2:39 AM, Vasileios Vavourakis <[email protected]>
>> > wrote:
>> >
>> >> dear all,
>> >>
>> >> after pulling the newest version of libmesh (commit
>> >> dbcd57f5f718a088a19b9be38f162b997ec8f97c) and configuring the library:
>> >>
>> >> START ==========================
>> >> ./configure --enable-everything --with-methods="opt" \
>> >> --enable-petsc PETSC_DIR=/Users/vasvav/Programs/petsc
>> >> PETSC_ARCH=arch-darwin-c-opt --with-metis=PETSc \
>> >> --with-glpk-include=/usr/local/Cellar/glpk/4.61/include
>> >> --with-glpk-lib=/usr/local/Cellar/glpk/4.61/lib \
>> >> --enable-netcdf --enable-netcdf-4 --enable-exodus \
>> >> --disable-shared --disable-complex --disable-1D-only --disable-2D-only
>> >> --disable-openmp --disable-eigen --disable-ifem --disable-pfem
>> >> --disable-slepc --disable-trilinos --disable-tecio --disable-tecplot
>> >> --disable-tetgen --disable-triangle --disable-qhull --disable-tbb
>> >> --disable-pthreads --disable-cppthreads --disable-openmp \
>> >> --prefix=/Users/vasvav/Programs/installed/libmesh-1.2.0-opt
>> >> END =============================
>> >>
>> >> I receive the following error:
>> >>
>> >> START ==========================
>> >> ...
>> >>   CXX      src/numerics/libmesh_opt_la-numeric_vector.lo
>> >>   CXX      src/numerics/libmesh_opt_la-petsc_matrix.lo
>> >> /numerics/petsc_matrix.C:811:10: error: use of undeclared identifier
>> >> 'MatGetSubMatrix'
>> >>   ierr = MatGetSubMatrix(_mat,
>> >>          ^
>> >> src/numerics/petsc_matrix.C:1131:16: note: in instantiation of member
>> >> function 'libMesh::PetscMatrix<double>::_get_submatrix' requested here
>> >> template class PetscMatrix<Number>;
>> >>                ^
>> >> 1 error generated.
>> >> make[1]: *** [src/numerics/libmesh_opt_la-petsc_matrix.lo] Error 1
>> >> make: *** [all-recursive] Error 1
>> >> END =============================
>> >>
>> >> can someone please give it a look? many thanks (:
>> >>
>> >
>> > Pretty strange error, I do not believe this code has changed in quite
>> some
>> > time.
>> >
>> > What version of PETSc are you using? Perhaps they removed or renamed
>> this
>> > function in some newer version... I will have a look at the PETSc git
>> repo.
>> >
>> > --
>> > John
>> >
>> ------------------------------------------------------------
>> ------------------
>> Check out the vibrant tech community on one of the world's most
>> engaging tech sites, Slashdot.org! https://urldefense.proofpoint.
>> com/v2/url?u=http-3A__sdm.link_slashdot&d=DwICAg&c=54IZrppPQ
>> ZKX9mLzcGdPfFD1hxrcB__aEkJFOKJFd00&r=DUUt3SRGI0_JgtN
>> aS3udV68GRkgV4ts7XKfj2opmiCY&m=SPlqHC3G8qn_LAN_WookKGCh37v7
>> XP8yXg2ajm_VAqw&s=vVpLydwR6zt4RZyBLBNGHkR5uhdq-n9538Nl1rh_Qko&e=
>> _______________________________________________
>> Libmesh-users mailing list
>> [email protected]
>> https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.s
>> ourceforge.net_lists_listinfo_libmesh-2Dusers&d=DwICAg&c=54I
>> ZrppPQZKX9mLzcGdPfFD1hxrcB__aEkJFOKJFd00&r=DUUt3SRGI0_JgtN
>> aS3udV68GRkgV4ts7XKfj2opmiCY&m=SPlqHC3G8qn_LAN_WookKGCh37v7
>> XP8yXg2ajm_VAqw&s=TuCX7-7WrFvpnG0MjRVnqlig5D4DyZQH4b_qY9unhqc&e=
>>
>
>
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Libmesh-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/libmesh-users

Reply via email to