[FLASH-BUGS] Re: evrard setup

Robi Banerjee banerjee at physics.mcmaster.ca
Sat Jan 18 12:55:41 CST 2003


Hallo,

  thanks for your efforts. It seems also not to be a general
problem because other setups work fine. 
  Working with 'periodic' boundary conditions we don't get
the multigrid convergence warnings but still the floating
point exception error. So the warning and the error should 
not be related.
  I also will look a little deeper to the problem.

  Thanks,
Robi


On Fri, 17 Jan 2003, Andrew Siegel wrote:

> 
> For the record, the intel, pgf, lahey, and sgi compilers all seem to run
> this setup fine (there is a multigrid convergence warning, but I think the
> default criteria are overly strict and ultimately it does converge --
> there is no floating point exception). We have somewhat more limited
> access to Compaq machines, but we'll continue to work on this as much as
> we can with the accounts that we have until we find something. -andrew
> 
> 
> 




More information about the flash-bugs mailing list