[CIG-LONG] new install

Matty Mookerjee matty.mookerjee at sonoma.edu
Mon Feb 4 20:33:21 PST 2013


Walter,

That seems entirely possible considering through this process I have compiled petsc several/many times along with and without the two main dependencies.  

I certainly can delete the folder that had Petsc installed in it, but will that completely remove it?  I suspect there is a more complete way to uninstall it, but without using Synamptic (which doesn't recognize this version of Petsc) I don't know the best way to completely uninstall.

Advice?

Matty
________________________________________
From: cig-long-bounces at geodynamics.org [cig-long-bounces at geodynamics.org] on behalf of Walter Landry [wlandry at caltech.edu]
Sent: Monday, February 04, 2013 8:12 PM
To: cig-long at geodynamics.org
Subject: Re: [CIG-LONG] new install

Matty Mookerjee <matty.mookerjee at sonoma.edu> wrote:
> Walter,
>
> OK, I think I did what you wanted.  Here are the results:
>
> matty at matty-ThinkPad-T530:~/Programs$ mpicc -c hello.c
> matty at matty-ThinkPad-T530:~/Programs$ mpicc -o hello hello.o
> matty at matty-ThinkPad-T530:~/Programs$ mpirun hello
> Hello world from process 0 of 1
> matty at matty-ThinkPad-T530:~/Programs$ mpirun -np 2 hello
> Hello world from process 0 of 2
> Hello world from process 1 of 2
> matty at matty-ThinkPad-T530:~/Programs$
>
> How does that look?

It looks good.  My guess is that there is an old build of MPI lurking
in the petsc directory.  Could you try completely removing the petsc
directory and configuring from scratch?  I believe that the correct
configuration is something like

  ./configure --without-x

Run "make" and "make test".

Cheers,
Walter Landry
_______________________________________________
CIG-LONG mailing list
CIG-LONG at geodynamics.org
http://geodynamics.org/cgi-bin/mailman/listinfo/cig-long



More information about the CIG-LONG mailing list