[GRASS5] CRAY T3E compile errors

[John Huddleston]

Markus,

The gcc compiler can find the warnings with the -Wall
option.  Plus, you can use the -pedantic option which
causes the compile to fail if the types do not match.

John Huddleston

----- Original Message -----
From: Markus Neteler <neteler at geog.uni-hannover.de>
To: <grass5 at geog.uni-hannover.de>
Sent: Monday, September 25, 2000 10:49 AM
Subject: Re: [GRASS5] CRAY T3E compile errors


> Hi James,
>
> On Sun, Sep 24, 2000 at 11:15:05PM +1100, James Cameron wrote:
> > On Tue, Sep 19, 2000 at 05:34:35PM +0100, Markus Neteler wrote:
> > > On Tue, Sep 19, 2000 at 05:36:00PM +0200, Bernhard Reiter wrote:
> > > > Even when compatible with C standards,
> > > > there can be significant differences.
> > > Yes, very true. But I have no admin rights there... So I will have
> > > to live with this compiler currently.
> > > > What version of make are you using?
> > > > (It might be an idea to build gnu make first.)
> > > It is a CRAY make. As only the libes/digitizer/OBJ*/make.rules fails
> > > but all others are o.k. it might be a problem in
> > > libes/digitizer/OBJ*/make.rules (only I can't see it).
> >
> > You have no admin rights, but you have shell and disk space?  If so,
> > install gcc and make in your own home directory ...
> [...]
> > Then I'd remake make with gcc.  ;-)
>
> thanks for your help. But :-)
> the libes/digitizer/ directory is the *only* directory causing such
> problem. Therefore I think it is a bug which should be fixed.
> The entire src/ directories are compiled now.
>
> Currently I am jumping into the G3D stuff encountering further
> bugs. It seems that the CRAY compiler is nice to find them as
> gcc is more tolerant (at least it prevents us from further mails
> of other users later):
>
> src/libes/g3d:
>         cc -g    -I/home/t3e/fsn/nhdc/nhdcmark/grass/src/include  -c
> filecompare.c -o OBJ.alpha-cray-unicosmk2.0.4.X/filecompare.o
> CC-167 cc: ERROR File = filecompare.c, Line = 25
>   Argument of type "float *" is incompatible with parameter of type
> "caddr_t".
>
>     xdrmem_create (&xdrEncodeStream, xdrf, 4, XDR_ENCODE);
>                                      ^
>
> CC-167 cc: ERROR File = filecompare.c, Line = 46
>   Argument of type "double *" is incompatible with parameter of type
> "caddr_t".
>
>     xdrmem_create (&xdrEncodeStream, xdrd, 8, XDR_ENCODE);
>                                      ^
> On Linux I have:
> grep caddr_t /usr/include/rpc/xdr.h
>         bool_t (*x_getbytes) __PMT ((XDR *__xdrs, caddr_t __addr,
>     caddr_t x_public;           /* users' data */
>     caddr_t x_private;          /* pointer to private data */
>     caddr_t x_base;             /* private used for position info */
>  * bool_t       (*xdrproc_t)(XDR *, caddr_t *);
>  * caddr_t       addr;
> extern bool_t xdr_array __P ((XDR * _xdrs, caddr_t *__addrp, u_int
*__sizep,
> extern bool_t xdr_opaque __P ((XDR *__xdrs, caddr_t __cp, u_int __cnt));
> extern bool_t xdr_reference __P ((XDR *__xdrs, caddr_t *__xpp, u_int
__size,
> extern void xdrmem_create __P ((XDR *__xdrs, __const caddr_t __addr,
>                                 u_int __recvsize, caddr_t __tcp_handle,
>
> On CRAY I have:
> grep caddr_t /usr/include/rpc/xdr.h
>  * bool_t       (*xdrproc_t)(XDR *, caddr_t *);
>         caddr_t         x_public;       /* users' data */
>         caddr_t         x_private;      /* pointer to private data */
>         caddr_t         x_base;         /* private used for position info
*/
>  * caddr_t       addr;
> extern bool_t xdr_array __((XDR *_Xdrs, caddr_t *_Addrp, u_int *_Sizep,
> extern bool_t xdr_opaque __((XDR *_Xdrs, caddr_t _Cp, u_int _Cnt));
> extern bool_t xdr_reference __((XDR *_Xdrs, caddr_t *_Pp, u_int _Size,
> extern void xdrmem_create __((XDR *_Xdrs, caddr_t _Addr, u_int _Size,
>
> As this is pretty complex (for me), I would be glad to receive
> an advice.
>
> I hope it's not too annoying to post these problems here. If so, I
> will stop it. Just complain.
>
> Thanks again
>
>  Markus
>
> ----------------------------------------
> If you want to unsubscribe from GRASS Development Team mailing list write
to:
> minordomo at geog.uni-hannover.de with
> subject 'unsubscribe grass5'


---------------------------------------- 
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo at geog.uni-hannover.de with
subject 'unsubscribe grass5'