Re: [petsc-users] Setting step acceptance criteria and/or domain validity using TS module

Pierre Barbier de Reuille writes: > Hello, > > Looking for methods to ensure negative values are rejected, I found this in > the archives: > > http://lists.mcs.anl.gov/pipermail/petsc-users/2014-June/021978.html > > The answer gives two options: > 1 - Set a function for the step acceptance crite

Re: [petsc-users] Installing petsc on Linux with Intel mpi

Wee Beng Tay writes: > Hi, > > I'm using the latest version 5 update 1,just installed 1 wk ago. Why are you installing such an old version now? Satish wasn't joking around when he said you need at least Update 2. Earlier versions return 0 for failure making it impossible to test for features.

Re: [petsc-users] Issue with window SF type using derived datatypes for reduction

Lawrence Mitchell writes: > Having just tried a build with --download-mpich, I notice this problem > does not occur. So should I shout at the OpenMPI team? Open MPI has many long-standing bugs with one-sided and datatypes. I have pleaded with them to error instead of corrupting memory or return

Re: [petsc-users] Installing petsc on Linux with Intel mpi

You need always to send configure.log when a configure fails. Barry > On Feb 9, 2015, at 5:30 PM, Wee Beng Tay wrote: > > Hi, > > I'm trying to install petsc on Linux with Intel mpi 5 and compiler. What > should be the configure > Command line to use? Anyone has experience? > > I tried

Re: [petsc-users] Installing petsc on Linux with Intel mpi

Hi, I'm using the latest version 5 update 1,just installed 1 wk ago. Compiling of code uses mpiicc, mpiifort.  Compiling a hello mpi code works. Sent using CloudMagic On Tue, Feb 10, 2015 at 7:43 am, Satish Balay wrote:Which version of IMPI do you have? Make sure you have t

Re: [petsc-users] Installing petsc on Linux with Intel mpi

Which version of IMPI do you have? Make sure you have the following (or newer?) - as it has the required fixes. https://software.intel.com/en-us/articles/intel-mpi-library-50-update-2-readme Satish On Mon, 9 Feb 2015, Wee Beng Tay wrote: > > Hi, > > I'm trying to install petsc on Linux with

[petsc-users] Installing petsc on Linux with Intel mpi

Hi, I'm trying to install petsc on Linux with Intel mpi 5 and compiler. What should be the configure Command line to use? Anyone has experience? I tried the usual options but they all can't work. Thanks Sent using CloudMagic

[petsc-users] Issue with window SF type using derived datatypes for reduction

Hi all, I'm trying to use an SFReduce to update some data using a contiguous derived type: sketch: MPI_Type_contiguous(bs, MPI_DOUBLE, &dtype); MPI_Type_commit(&dtype); PetscSFReduceBegin(sf, dtype, rootdata, leafdata, MPI_SUM); PetscSFReduceEnd(sf, dtype, rootdata, leafdata, MPI_SUM); All is

Re: [petsc-users] AllReduce function for mat

Use MatDenseGetArray() and call MPI_Allreduce() on the resulting array point as the output buffer. Past MPI_IN_PLACE as the input buffer. Barry > On Feb 9, 2015, at 11:12 AM, Ghosh, Swarnava wrote: > > The matrix is dense and each process has the same number of nonzeros. > > - Origi

Re: [petsc-users] AllReduce function for mat

The matrix is dense and each process has the same number of nonzeros. - Original Message - From: "Barry Smith" To: "Swarnava Ghosh" Cc: "PETSc users list" Sent: Monday, February 9, 2015 12:11:07 PM Subject: Re: [petsc-users] AllReduce function for mat > On Feb 9, 2015, at 10:57 AM, Gh

Re: [petsc-users] AllReduce function for mat

> On Feb 9, 2015, at 10:57 AM, Ghosh, Swarnava wrote: > > Hi, > > I have sequential matrix on each process. I wanted to do something like > MPI_Allreduce to this matrix to add entries from all processes. The resulting > matrix is sequential. Is the matrix dense or sparse? If sparse

Re: [petsc-users] PCMG and DMMG

Looks like you are looking at a very old PETSc. You should be using version 3.5.3 and nothing earlier. DMMG has been gone from PETSc for a long time. Here is the easiest way to provide the information: Loop over the levels yourself and provide the matrices, function pointers etc. See for e

[petsc-users] AllReduce function for mat

Hi, I have sequential matrix on each process. I wanted to do something like MPI_Allreduce to this matrix to add entries from all processes. The resulting matrix is sequential. I wanted to know if there is a PETSc function to do this? -- SG

Re: [petsc-users] Full blas-lapack on Windows

I think we had this conversation before. --download-f2cblaslapack will give you a full blas/lapack. And you can't use MUMPS without a fortran compiler [as far as I know] You should be able to use superlu_dist Satish On Mon, 9 Feb 2015, Fabien RAPHEL wrote: > Hello, > > I can configure, compi

[petsc-users] PCMG and DMMG

Good morning, everyone: I have an existing general CFD solver with multigrid implemented. All functions (initialization, restriction, prolong/interpolation, coarse/fine grids solver..) are working correctly. Now I am trying to rewrite it with PETSc. I found that the manual provides very little

[petsc-users] Setting step acceptance criteria and/or domain validity using TS module

Hello, Looking for methods to ensure negative values are rejected, I found this in the archives: http://lists.mcs.anl.gov/pipermail/petsc-users/2014-June/021978.html The answer gives two options: 1 - Set a function for the step acceptance criteria 2 - Set a domain violation for the function H

[petsc-users] Setting step acceptance criteria and/or domain validity using TS module

Hello, Looking for methods to ensure negative values are rejected, I found this in the archives: http://lists.mcs.anl.gov/pipermail/petsc-users/2014-June/021978.html The answer gives two options: 1 - Set a function for the step acceptance criteria 2 - Set a domain violation for the function H

Re: [petsc-users] MatMult inside a for loop

On Sun, Feb 8, 2015 at 9:09 PM, Jed Brown wrote: > Matthew Knepley writes: > > > On Sun, Feb 8, 2015 at 7:20 PM, Ronal Celaya > > wrote: > > > >> I know that. I want to have all the vector x replicated in all processes > >> and update it in each iteration, so I don't need to communicate the > v

Re: [petsc-users] passing information to TSIFunction

Hi Thanks for that. I have since resolved my issue for purpose. I will however implement this suggestion by Jed as it seems more efficient and more consistent with the Petsc paradigm. cheers Sanjay From: Jed Brown [j...@jedbrown.org] Sent: 09 Februar