Memory reporting, questionable cast to int

[Matthew Knepley]

Fixed.

  Matt

On Fri, Jun 5, 2009 at 6:50 AM, Jed Brown <jed at 59a2.org> wrote:

> plog.c:1557:
>
> |  ierr = PetscFPrintf(comm, fd, "%20s %5d          %5d  %9d     %g\n",
> stageLog->classLog->classInfo[oclass].name,
> |                      classInfo[oclass].creations,
> classInfo[oclass].destructions, (int) classInfo[oclass].mem,
> |                      classInfo[oclass].descMem);CHKERRQ(ierr);
>
> Why cast classInfo[oclass].mem to int?  Any reason not to change to %9f
> and remove the cast?
>
> Note that memory usage can be much larger than 2^31 before 64-bit
> indices are needed so casting to PetscInt isn't right either.
>
> | --- Event Stage 0: Main Stage
> |
> |                  Vec    38             38  176360064     0
> |          Vec Scatter     1              1        460     0
> |               Matrix     2              2  -2147483648     0
> |        Krylov Solver     1              1      18056     0
> |       Preconditioner     1              1        760     0
> |               Viewer     1              1        552     0
> |            Index Set     4              4    9281568     0
>
>
> Jed
>
>


-- 
What most experimenters take for granted before they begin their experiments
is infinitely more interesting than any results to which their experiments
lead.
-- Norbert Wiener
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20090605/90dfe578/attachment.html>