[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: 20050426:NetCDF 3.6.0_p1 on NEC SX

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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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.

  • Subject: Re: 20050426:NetCDF 3.6.0_p1 on NEC SX
  • Date: Tue, 26 Apr 2005 14:14:10 -0600
Unidata Support <address@hidden> writes:

> ------- Forwarded Message
>
>>To: address@hidden
>>From: Rene Redler <address@hidden>
>>Subject: NetCDF 3.6.0_p1 on NEC SX
>>Organization: UCAR/Unidata
>>Keywords: 200504260902.j3Q92Lv2005237
>
> hi,
>
>     when installing the netcdf lib on a NEC SX6 I have some trouble
> with the int1 and int2 f90 interfaces. On the SX integer*1 and
> integer(kind=1) are not supported. Therefore no int1 interfaces
> should be built.
>
> However, the way how the f90 interface descriptions are built
> lead to identical interfaces for the int1 and int2 routines,
> since OneByteInt and TwoByteInt have identical values in our case.
> Those two interfaces cannot be distiguished and the compiler terminates
> with an error, which I think is a correct behaviour.
>
> A workaround would be to edit the f90 files manually to exclude
> the int1 interfaces via some cpp directive.
>
> You determine EightByteInt with selected_int_kind(18). Although
> a value of n in the range -10^18 < n < 10^18 can well be represented
> with an 8 Byte integer on the SX the selected_int_kind returns
> a -1 here. I think that this is a failure in the SX implementation
> of this intrinsic function. Nevertheless, would it be possible
> to either use selected_int_kind(15) to determine EightByteInt
> via some SX specifc cpp key, or have some comment added to the
> installation guide?
>
> One the other hand, if we assume that integer*8 is not supported
> selected_int_kind(18) = -1 this will not be properly reflected
> in the f90 interface description as well, since declarations such as
> integer(kind=-1) will probably fail on most compilers.
>
> Rene
>
>
> _______________________________________________________________
>
>    René Redler
>    C&C Research Laboratories
>    NEC Europe Ltd.                 Tel: +49 (0)2241 925240
>    Rathausallee 10                 Fax: +49 (0)2241 925299
>    53757 Sankt Augustin            URL: www.ccrl-nece.de/~redler
> _______________________________________________________________
>
> --
> NOTE: All email exchanges with Unidata User Support are recorded in the
> Unidata inquiry tracking system and then made publicly available
> through the web.  If you do not want to have your interactions made
> available in this way, you must let us know in each email you send to us.
>
> ------- End of Forwarded Message
>
>

Rene,

Sorry, but I don't have access to that system. However, take a look
here and see if it helps:

http://www.unidata.ucar.edu/packages/netcdf/other-builds.html#nec-sx5
-- 
Ed Hartnett  -- address@hidden