[petsc-dev] what happened to RAP timers?

[Mark Adams]

Barry,

THis code looks to me like it should work, but it does not seem to. What
code were you looking at?

  ierr = PetscOptionsGetBool(NULL,NULL,"-matrap",&rap,NULL);CHKERRQ(ierr);
  if (scall == MAT_INITIAL_MATRIX) {
    ierr = PetscLogEventBegin(MAT_PtAPSymbolic,A,P,0,0);CHKERRQ(ierr);
    if (rap) { /* do R=P^T locally, then C=R*A*P */
      ierr = MatPtAPSymbolic_MPIAIJ_MPIAIJ(A,P,fill,C);CHKERRQ(ierr);
    } else {       /* do P^T*A*P */
      ierr = MatPtAPSymbolic_MPIAIJ_MPIAIJ_ptap(A,P,fill,C);CHKERRQ(ierr);
    }
    ierr = PetscLogEventEnd(MAT_PtAPSymbolic,A,P,0,0);CHKERRQ(ierr);
  }
  ierr = PetscLogEventBegin(MAT_PtAPNumeric,A,P,0,0);CHKERRQ(ierr);
  ierr = (*(*C)->ops->ptapnumeric)(A,P,*C);CHKERRQ(ierr);
  ierr = PetscLogEventEnd(MAT_PtAPNumeric,A,P,0,0);CHKERRQ(ierr);

On Sun, Sep 4, 2016 at 9:05 PM, Barry Smith <bsmith at mcs.anl.gov> wrote:

>
>    Mark,
>
>    Look at the source code for MatPtAP. It depends on the approach to how
> it is computed what logging takes place. I am not sure the rational for
> this exact logging is, it would seem better to always log it so perhaps
> that could be changed.
>
>   Barry
>
>
> > On Sep 4, 2016, at 7:10 PM, Mark Adams <mfadams at lbl.gov> wrote:
> >
> > I don't seem to be seeing RAP (MatPtAP, or whatever) profile data. Did
> this get dumped?
> >
> > I see that there are two timers in GAMG that I can scrap but if RAP is
> not timed then I will add that to GAMG profile data.
> >
> > Mark
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20160905/5013fcee/attachment.html>