Re: MSWindows CVS install problem

[Marc Brünink]

On Tuesday, Mar 22, 2005, at 13:33 Europe/Berlin, Marko Riedel wrote:

> --- Marc Brünink <mbruen@smartsoft.de> wrote:
> > On Monday, Mar 21, 2005, at 20:44 Europe/Berlin, Marc Brünink wrote:
> [...]
> > a reproduceable error concerning conftest.exe. This error occurs while
> > the configure script of ffcall checks for the long long type. Actually
> > the configure script runs through (I just ignored the error pop-ups),
> > but therefore the make of gnustep-base failed in GSFFCall* (<- cannot
> > remember the name exactly, but it was line 1004 and it bleated about
> > va_return_long_long or something like this)
> > Is someone able to reproduce this error? Does someone have a hint for
> > me?
> [...]
>
> > Occured on a MS Windows 2000 SP4 machine.
>
> I got exactly the same error, also on a Windows 2000 machine:
>
>
> GSFFCallInvocation.m: In function `GSInvocationCallback':
> GSFFCallInvocation.m:1003: error: union has no member named `_longlong'
> GSFFCallInvocation.m:1004: error: union has no member named 
> `_ulonglong'
> GSFFCallInvocation.m:1003: warning: value computed is not used
> GSFFCallInvocation.m:1004: warning: value computed is not used
> make[2]: *** [shared_obj/GSFFCallInvocation.o] Error 1
> make[1]: *** [libgnustep-base.all.library.variables] Error 2
> make: *** [internal-all] Error 2

If I use the binary installer the ffcall configure runs through without 
error.
Anyway. If you force the configure of ffcall to set  long_long=yes its 
possible to compile GNUstep CVS on a Windows 2000 machine (as Tom 
pointed out). But the install is not as   clean as it looked like. I'm 
getting errrors while running eg GFractal:

abnormal program termination
objc runtime: cannot find class NSAutoreleasePool

So. Now we have a problem with the objc runtime. gosh! This is no fun! 
Anyone a clue?


Thanks
Marc