Satish, Thanks, we'll have to live with this fix. Barry On Thu, 15 Dec 2005, Satish Balay wrote:

Barry,

I'm attaching the relavent logfiles for this failure case. Looks like there is an exception raised for this case - so I can trap it in MPI.py - and error out. [other case returns a '0' - which is ok for PETSc sharedlibs]

So with this change to MPI.py, configure give a nice error message:

********************************************************************************* UNABLE to CONFIGURE with GIVEN OPTIONS (see configure.log for details): --------------------------------------------------------------------------------------- PETSc shared libraries cannot be built using MPI provided. Either rebuild PETSc with --with-shared=0 or rebuild MPI with shared library support *********************************************************************************

I'll push this change.

Satish

On Wed, 14 Dec 2005, Barry Smith wrote:

...

...

This also succeeds, unless I misunderstand greatly the error.

The libraries get built, fine. Now you generate the executable. That FAILS! As I have been saying all the time! It fails NOT because the library is not shared but because one cannot use that library WITH shared libraries!

The new test is suppose to check that it can link and generate the executable, BUT not run the executable to see what message you get.

The new test could be called checkCanBeUsedWithSharedLibrary() and it would shared almost all the code with checkShared() except actually running the executable and checking the result.

Is it clear now, or am I still messed up?