Re: compiling netcdf

• To: Ed Hartnett <ed@xxxxxxxxxxxxxxxx>
• Subject: Re: compiling netcdf
• From: Warren Turkal <wt@xxxxxxxxxxxxxxxxxxx>
• Date: Thu, 5 Apr 2007 15:59:16 -0600

On Thursday 05 April 2007 15:36, Ed Hartnett wrote:
> Nope. Is there some reason that it should?

Well, I don't know if it should be default, but it enables to a different 
memory model that may be useful for large datasets. I am not familiar enough 
with how NetCDF works to make a really educated assessment. I have a user 
that is having a hard time stuffing a very large dataset into NetCDF, and I 
think it may have something to do with this limitation. I may be way off base 
though.

> The general idea is that if there are any flags that need to be set
> for your compiler, you must set them.

This seems reasonable.

> The exception is that the netCDF configure does set the correct flag
> to allow the fortran wrapper functions to work (if it can figure that
> out from the name of your fortran compiler).

Again, reasonable.

Thanks,
wt
-- 
Warren Turkal

==============================================================================
To unsubscribe netcdfgroup, visit:
http://www.unidata.ucar.edu/mailing-list-delete-form.html
==============================================================================

• Follow-Ups:
  • Re: compiling netcdf
    • From: Ed Hartnett

• References:
  • compiling netcdf
    • From: Warren Turkal
  • Re: compiling netcdf
    • From: Ed Hartnett