first win64 patch

[James Lacey]

I am not sure because I am not at home and can't check, but I believe
MS added a <sys\types.h> header and intptr_t is defined in there. BTW,
you can't define intprt_t as int on MSVC. It won't work on x64.

James

On 2/18/06, Shane Isley <shane at it.uts.edu.au> wrote:
> Hi,
> I've a process that is doing a hourly Win32 build from the head.
> http://www.promode.org/files/ioq3.zip
> I believe zakk wanted it linked on icculus.org/quake3
> Anyhow commit 550 broke my Win32 build (I'm using VS 2005 and also
> vcbuild.exe).
> <inttypes.h> was the problem, in that Windows or MSVC do not come with it.
> I began fixing my problen by adding my cygwin version of inttypes.h, but
> it depended on another header and so on and so forth.
>
> So it ended up simpler to do this:
>
> Index: q_shared.h
> ===================================================================
> --- q_shared.h    (revision 553)
> +++ q_shared.h    (working copy)
> @@ -104,7 +104,7 @@
>
>  #ifdef Q3_VM
>  typedef int intptr_t;
> -#else
> +#elif !WIN32
>  #include <inttypes.h>
>  #endif
>
> Index: vm.c
> ===================================================================
> --- vm.c    (revision 553)
> +++ vm.c    (working copy)
> @@ -886,6 +886,11 @@
>          f = fopen("syscalls.log", "w" );
>      }
>      callnum++;
> +#ifndef WIN32
>      fprintf(f, "%i: %"PRIiPTR" (%i) = %i %i %i %i\n", callnum,
> (intptr_t)(args - (int *)currentVM->dataBase),
>          args[0], args[1], args[2], args[3], args[4] );
> +#else
> +    fprintf(f, "%i: %li (%i) = %i %i %i %i\n", callnum, (intptr_t)(args
> - (int *)currentVM->dataBase),
> +        args[0], args[1], args[2], args[3], args[4] );
> +#endif
>  }
>
> I had to also change the printf(). The compiler kept bombing with
> regards to the argument type (PRIiPTR or li) being outside the quotes"".
>
> If this could be improved upon and added to the repository, I'd be greatful.
>
> Rregards,
> Shane
>
>
>
>
> Ludwig Nussel wrote:
>
> >James Lacey wrote:
> >
> >
> >>Here is my first Win64 Quake 3 patch. A couple of notes:
> >>
> >>
> >
> >I've now changed the code to use intptr_t instead of long where I
> >think it needs to. Since long works fine on Linux I can't check
> >whether I catched everything though. Would be nice if you could try
> >to create a minimal patch (ie without the size_t changes) based on
> >that to make it work on 64bit Windows.
> >
> >cu
> >Ludwig
> >
> >
> >
>
>