[petsc-users] Unable to configure PETSc with CUDA: Problem with thrust directory

[Satish Balay]

On Fri, 29 Aug 2014, Dominic Meiser wrote:

> On 08/29/2014 08:31 AM, Matthew Knepley wrote:
> > On Fri, Aug 29, 2014 at 9:35 AM, Dominic Meiser <dmeiser at txcorp.com
> > <mailto:dmeiser at txcorp.com>> wrote:

> > >     Dominic, I think that thrust.py should depend on cuda.py. Do you
> > >     know why it does not?
> >     In principle you are right, thrust.py should depend on cuda.py.
> >
> >     However, in my opinion, thrust.py should go away as a separate
> >     package altogether. Thrust is shipped as part of any recent
> >     version of the cuda toolkit (I forget since which version, Paul
> >     might know) and it's always installed in
> >     $CUDA_TOOLKIT_ROOT/include/thrust. Thus we can automatically
> >     deduct the thrust location from the cuda location. Thrust should
> >     be considered part of cuda.

I also think it should be removed.

> > That does not mean the configure modules should be organized that way.
> > Thrust is a package included in CUDA, thus
> > it should have an independent module, but depend on the CUDA module.

Its fine for configure to have thrust tests [and organize it
appropriately]

But I think the options --with-thrust-dir--with-thrust-include etc
should be removed. We require a minimum version of cuda anyway

The primay loss of functionality is - user can't override default
thrust with a prefered version. But should we be supporting such a
thing anyway?  [it might not be compatible with the given cuda?]

Satish

> I see your point. If you keep thrust as a module it should definitely depend
> on cuda. Default guesses for its location should then be generated based on
> the cuda location. I can work on this.