Re: [ft-devel] freetype-2.3.7 -- ftconfig.h for biarch systems

[mpsuzuki]

Dear Peter,

# also I Cc: this message to address@hidden
# who recently asked about 64bit support of FreeType2.

I have to apologize to all people waiting for the terminal fix
for the solution. While I was waiting for the stabilization of
Mac OS X 64bit development style, I guess GNU/Linux people are
irritated very much. I will work this issue after 2008-July-03,
sorry, I don't have enough time today.

Taking a glance on FreeType2 sources, it seems that the required
information is "which type (or no type) is sufficient to store
32bit and 64bit numerical value", not the exact storage size in
memory. So, I think, using traditional ANSI C header "limits.h"
is more generic than using modern headers "bits/wordsize.h", like...

        #ifdef HAVE_LIMITS_H

        # if   ( UINT_MAX == 0xFFFFFFFFFFFFFFFF )
        #  define FT_SIZEOF_INT 8
        # elif ( UINT_MAX == 0xFFFFFFFFFFFF )
        #  define FT_SIZEOF_INT 6
        # elif ( UINT_MAX == 0xFFFFFFFF )
        #  define FT_SIZEOF_INT 4
        # else
        #  error XXX
        # endif

        # if   ( ULONG_MAX == 0xFFFFFFFFFFFFFFFF )
        #  define FT_SIZEOF_LONG 8
        # elif ( ULONG_MAX == 0xFFFFFFFFFFFF )
        #  define FT_SIZEOF_LONG 6
        # elif ( ULONG_MAX == 0xFFFFFFFF )
        #  define FT_SIZEOF_LONG 4
        # else
        #  error XXX
        # endif

        #else

        # ( sizes found by configure on systems without limits.h )

        #endif

One of my anxiety is that the sizes obtained by AC_CHECK_SIZEOF()
would be more reliable than the sizes deduced by limits.h (although
yet I've not checked the detail that how configure script checks
the sizes).

I want to check the sizes by AC_CHECK_SIZEOF() and compare them
with the sizes deduced by limits.h - if any inconsistency is found,
AC_CHECK_SIZEOF() results should be used. Werner, how do you think?

Regards,
mpsuzuki

On Mon, 30 Jun 2008 12:16:55 +0200 (CEST)
Werner LEMBERG <address@hidden> wrote:

>
>> In order to have headers usable for both 32 and 64 bit mode, that
>> ought to be replaced by something like [...]
>
>You've already reported this:
>
>  https://savannah.nongnu.org/bugs/index.php?21250
>
>unfortunately, it hasn't been handled yet.  Maybe Suzuki-san (who has
>access to a 64bit-system, IIRC) can handle this.
>
>
>    Werner
>
>
>_______________________________________________
>Freetype-devel mailing list
>address@hidden
>http://lists.nongnu.org/mailman/listinfo/freetype-devel