From f1415f71c18707ea2526f1fd633ee97acd29d1ef Mon Sep 17 00:00:00 2001 From: Patrick Sanan Date: Mon, 29 Apr 2013 18:43:29 -0700 Subject: [PATCH] Minor fixes to the website faq, mostly uncontroversial typo fixes. Updated the note on gfortran for OSX, which to my knowledge is still required as an extra step on top of installing XCode, as of 10.7 (Lion). --- src/docs/website/documentation/faq.html | 37 ++++++++++++++----------------- 1 file changed, 17 insertions(+), 20 deletions(-) diff --git a/src/docs/website/documentation/faq.html b/src/docs/website/documentation/faq.html index 1784456..c983672 100644 --- a/src/docs/website/documentation/faq.html +++ b/src/docs/website/documentation/faq.html @@ -473,7 +473,7 @@

- The directory ${PETSC_DIR}/docs contains a set of HTML manual pages in + The directory ${PETSC_DIR}/src/docs contains a set of HTML manual pages in for use with a browser. You can delete these pages to save about .8 Mbyte of space.

What Fortran compiler do you recommend for the Apple Mac OS X?

- (as of 11/6/2010) We recommend installing gfortran from http://hpc.sourceforge.net. They - have gfortran-4.6.0 (experimental) for Snow Leopard (10.6) and gfortran - 4.4.1 (prerelease) for Leopard (10.5). -

http://hpc.sourceforge.net

Please contact Apple at You have AIJ and BAIJ matrix formats, and SBAIJ for symmetric storage, how come no SAIJ? - Just for historical reasons, the SBAIJ format with blocksize one is just as + Just for historical reasons; the SBAIJ format with blocksize one is just as efficient as an SAIJ would be.

How do I use PETSc for Domain Decomposition?

The When solving a system with Dirichlet boundary conditions I can use MatZeroRows() to eliminate the Dirichlet rows but this results in a non-symmetric system. How can I apply Dirichlet boundary conditions but keep the matrix symmetric?

When solving a system with Dirichlet boundary conditions I can use MatZeroRows() to eliminate the Dirichlet rows but this results in a non-symmetric system. How can I apply Dirichlet boundary conditions but keep the matrix symmetric?

For nonsymmetric systems put the appropriate boundary solutions in the - x vector and use MatZeroRows() followed by KSPSetOperators(). For symmetric - problems use MatZeroRowsColumns() instead. If you have many Dirichlet + x vector and use MatZeroRows() followed by KSPSetOperators(). For symmetric + problems use MatZeroRowsColumns() instead. If you have many Dirichlet locations you can use MatZeroRows() (not MatZeroRowsColumns()) and - -ksp_type preonly -pc_type redistribute, see the manual page for - PCREDISTRIBUTE) and PETSc will repartition the parallel matrix for load + -ksp_type preonly -pc_type redistribute; see + PCREDISTRIBUTE) and PETSc will repartition the parallel matrix for load balancing; in this case the new matrix solved remains symmetric even though MatZeroRows() is used.

- An alternative approach is when assemblying the matrix, (generating values - and passing them to the matrix), never include locations for the Dirichlet - grid points in the vector and matrix, instead take them into account as you + An alternative approach is, when assemblying the matrix (generating values + and passing them to the matrix), never to include locations for the Dirichlet + grid points in the vector and matrix, instead taking them into account as you put the other values into the load.

- Using the MATLAB Engine, this allows PETSc to automatically call MATLAB - to perform some specific computations. It does not allow MATLAB to be + Using the MATLAB Engine, allowing PETSc to automatically call MATLAB + to perform some specific computations. This does not allow MATLAB to be used interactively by the user. See the PetscMatlabEngine.
- To save PETSc Mat and Vecs to files that can be read from MATLAB use PetscViewerBinaryOpen() viewer and VecView() or MatView() to save objects for MATLAB and VecLoad() and MatLoad() to get the objects that MATLAB has saved. See @@ -1103,7 +1100,7 @@ PETSC_COMM_WORLD = NewComm; viewer that saves .mat files can then be loaded into MATLAB.
- We are just being to develop in petsc-dev an API to call most of the PETSc function directly from MATLAB; we could use help in developing this. See bin/matlab/classes/PetscInitialize.m @@ -1777,7 +1774,7 @@ source /opt/intel/idb/10.1.012/bin/idbvars.sh matrix-vector product via differencing, the residual norms computed by GMRES start to "drift" from the correct values. At the restart, we compute the residual norm directly, hence the "strange stuff," the difference - printed. The drifting, if it remains small, is harmless (doesn't effect the + printed. The drifting, if it remains small, is harmless (doesn't affect the accuracy of the solution that GMRES computes).
-- 1.7.10.2 (Apple Git-33)