R help - Dec 2002 - R on the Zaurus link

Hello All,

The link to the binary & installation instructions (tar.gz binary not an ipk
I'm afraid) is as follows: http://students.bath.ac.uk/enpsgp/Zaurus/#R

It eventually dawned on me that the WORDS_BIGENDIAN define (or lack thereof)
was causing the problems (after testing ieee NaN compliance that is).

When cross-compiling it's probably fair enough that the configure script
can't work out what the endianness should be, however the same thing
happened (WORDS_BIGENDIAN not defined) when compiling on an ARM machine (one
of the iPAQ skiff cluster machines. See
http://www.handhelds.org/z/wiki/SkiffCluster)

The following is the section of the config.log pertaining to determining
endianness, for those who may want to try to fix/alter the configure script
(from the ARM iPAQ machine):

============================================================configure:12298:
checking whether byte ordering is bigendian
configure:12328: gcc -c -g -O2 -I/usr/local/include conftest.c >&5
configure:12331: $? = 0
configure:12334: test -s conftest.o
configure:12337: $? = 0
configure:12364: gcc -c -g -O2 -I/usr/local/include conftest.c >&5
configure: In function `main':
configure:12354: `not' undeclared (first use in this function)
configure:12354: (Each undeclared identifier is reported only once
configure:12354: for each function it appears in.)
configure:12354: parse error before `big'
configure:12367: $? = 1
configure: failed program was:
#line 12340 "configure"
#include "confdefs.h"
#include <sys/types.h>
#include <sys/param.h>

#ifdef F77_DUMMY_MAIN
#  ifdef __cplusplus
     extern "C"
#  endif
   int F77_DUMMY_MAIN() { return 1; }
#endif
int
main ()
{
#if BYTE_ORDER != BIG_ENDIAN
 not big endian
#endif

  ;
  return 0;
}
configure:12481: result: no
============================================================
Might be something to do with the /sys/ tacked on the front of the types.h
and param.h includes?

In any case if you just want it to compile then use something like the
following:

$ ./configure --host=arm-linux --build=i686-mandrake-linux
ac_cv_c_bigendian=yes

I've not tested the X parts of R yet, nor have I done anything else other
than making sure that NAs are handled correctly.

Happy new year to you all,


Simon

--------------------------------------------
Simon Pickering MEng
Dept. of Mechanical Engineering
University of Bath
Bath, BA2 7AY, UK

Tel: +44 (0)1225 383314
Fax: +44 (0)1225 386928
--------------------------------------------

Simon Pickering <enpsgp at bath.ac.uk> writes:
> #include "confdefs.h"
> #include <sys/types.h>
> #include <sys/param.h>
> 
> #ifdef F77_DUMMY_MAIN
> #  ifdef __cplusplus
>      extern "C"
> #  endif
>    int F77_DUMMY_MAIN() { return 1; }
> #endif
> int
> main ()
> {
> #if BYTE_ORDER != BIG_ENDIAN
>  not big endian
> #endif
> 
>   ;
>   return 0;
> }
> configure:12481: result: no
> ============================================================> 
> Might be something to do with the /sys/ tacked on the front of the types.h
> and param.h includes?
I don't think so, but the test looks dubious: at least on i386-Linux, it is 
__BYTE_ORDER and __BIG_ENDIAN, unless you include endian.h.

-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907