Fixed petsc_linear_solver for PETSc-master

[fdkong]

No description provided.

[jwpeterson]

Looks good to me... I attached "Test with PETSc master" as well.

[jwpeterson]

I think it's been a while since the CIVET recipe for testing PETSc master on libMesh PRs was updated. I'm sure you guys have already run into this before, though so it shouldn't be too hard to fix.

*******************************************************************************
         UNABLE to CONFIGURE with GIVEN OPTIONS    (see configure.log for details):
-------------------------------------------------------------------------------
Cannot use SuperLU_DIST without enabling C++11, see --with-cxx-dialect=C++11
*******************************************************************************

[fdkong]

I think it's been a while since the CIVET recipe for testing PETSc master on libMesh PRs was updated. I'm sure you guys have already run into this before, though so it shouldn't be too hard to fix.

*******************************************************************************
         UNABLE to CONFIGURE with GIVEN OPTIONS    (see configure.log for details):
-------------------------------------------------------------------------------
Cannot use SuperLU_DIST without enabling C++11, see --with-cxx-dialect=C++11
*******************************************************************************

Yes we knew this. Just did not updater the script in civet yet. I am going to do this.

[moosebuild]

Job Test with PETSc master on cc3f190 : invalidated by @fdkong

Configure script should be right now

[fdkong]

@jwpeterson I think we could merge the PR. The falling test is moose. I have a fix here idaholab/moose#12897

[jwpeterson]

This will also go in v1.4.0 (#2029).

[fdkong]

Thanks, @jwpeterson