calling nf90_put_var with count=0

[Jim Edwards]

Rob,

Setting the environment variable works,

Thanks,
Jim

On 7/11/06, Robert Latham <robl at mcs.anl.gov> wrote:
>
> On Tue, Jul 11, 2006 at 12:02:02PM -0600, Jim Edwards wrote:
> > Hi Rob,
> >
> > In order to avoid using the independent mode I am trying to call
> > nfmpi_put_vara_all with count=0 on some processors.  This appears to
> > work fine under the AIX implementation but causes a malloc error in
> > romio on BG/L.   Here is a core dump referring to source code in
> > romio, I don't have that source code but I have some hope that this
> > will be enough info for you to pinpoint the problem...
>
> Hi Jim
> We talk about this a little bit in our README.bgl.  If you set the
> BGLMPIO_TUNEBLOCKING environment variable to '0', you should no longer
> get a seg fault
>
> They just rolled out a new version of IBM's MPI-IO for BGL on our
> machine that fixes this problem:  with driver version V1R3M0 I no
> longer see this segfault with our test cases.
>
> Let us know how those workarounds perform for you.
>
> ==rob
>
> --
> Rob Latham
> Mathematics and Computer Science Division    A215 0178 EA2D B059 8CDF
> Argonne National Labs, IL USA                B29D F333 664A 4280 315B
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/parallel-netcdf/attachments/20060712/31de4475/attachment.htm>