[petsc-dev] remove ftime from PetscErrorCode TSSolve(TS ts, Vec u, PetscReal *ftime)

Barry Smith bsmith at mcs.anl.gov
Mon Oct 29 21:36:05 CDT 2012


On Oct 29, 2012, at 9:27 PM, Jed Brown <jedbrown at mcs.anl.gov> wrote:

> Sure, you could do that. Do you want to get rid of the solution vector too?

   We still have that silly argument with KSP and SNES so I think I'll keep it for now.

   Barry

> 
> On Oct 29, 2012 6:55 PM, "Barry Smith" <bsmith at mcs.anl.gov> wrote:
> 
>    I don't understand, in TSSolve() I see
> 
>     if (ftime) *ftime = ts->prime;
> and       if (ftime) *ftime = ts->max_time;
> 
>   why not just stash ftime in the TS object and provide a TSGetFinalTime(ts,PetscInt *ftime)?
> 
>     Barry
> 
> 
> 
> 
> On Oct 29, 2012, at 8:51 PM, Jed Brown <jedbrown at mcs.anl.gov> wrote:
> 
> > In general, there is the internal state which may be later than the final time, plus the interpolated time. If you get rid of that argument, the user probably has to use TSInterpolate to get the state to coincide with the final time.
> >
> > On Oct 29, 2012 6:46 PM, "Barry Smith" <bsmith at mcs.anl.gov> wrote:
> >
> >   Can we remove the ftime argument from TSSolve()?
> >
> >    Thanks
> >
> >     Barry
> >
> 




More information about the petsc-dev mailing list