This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.
>Date: Wed, 6 Apr 94 12:50:38 MDT >From: Joshua Goldstein <address@hidden> >To: address@hidden >Subject: Netcdf 2.3.2 on the VMS Alpha In the above message you wrote: > I have to port NetCDF 2.3.2 to the Alpha running VMS. I >have been trying to use the descrip.mms and the source as given. >Unfortunately I need error and warning free compilation and its >not seeming to compile that way. Have you tried compiling on a >VMS alpha? Do you use different source? Since we don't have access to one, we haven't built the netCDF package on an Alpha running VMS. We have, however, built it on an Alpha running OSF1 and on VAXen running VMS. We do not, however, recommend the use of the descrip.mms file for building under VMS. Rather, the make.com files should be used. > I do do one thing differently than you: > >COPTS = /NOLIST/EXTERN=STRICT_REFDEF/ANSI/FLOAT=D_FLOAT > >The htons.mar and ntohs.mar seem strikingly like VAX assembly to me. >If so, why aren't there different files for the Alpha? I'm confused. Again, we can't speak to Alpha/VMS, but under Alpha/OSF1 the operating system supplies correct implementations of ntohs() and htons(). I suspect that the current release of VMS for the Alpha does also. Consequently, You might try deleting the references to htons.mar and ntohs.mar from the relevant make.com file before attempting to build the package. Please keep this dialog open. We want to hear about Alpha/VMS experiences. -------- Steve Emmerson <address@hidden>