Patrik Stridvall wrote:
> > Why can't we just install the config.h generated at Winelib
> > installation
> > time to the Wine system header directory? Maybe rename it to avoid
> > conflicts with an application's configure script?
>
> Because you might have more than one version of the compiler installed
> and indeed even different brands of compilers.
>
> Of course some things are truely architecture (install) dependent and
> compiler independant like whether the CPU is big endian.
Well, the IMO most important configure checks are CPU architecture
and OS specific (e.g. location of system headers / libraries and the
like). While there are of course also compiler / toolchain dependent
checks, those should matter only to the build process itself, and not
the header files.
> However since
> it would be good to be able to support cross compilation it is probably
> not a good idea to have a configure architecture dependent configure
> file anyway.
Cross compilation doesn't work anyway, and it does not appear easy
to set up the build process for it. Why do you think it important?
Most userland packages (with the obvious exceptions like gcc and glibc)
are not prepared for cross compilation either ...
> Unless it is truely impossible to do something with a such configure
> file we really shouldn't IMHO. It is just asking for trouble and
> inflexibillity in the future.
It simply means that we will always have problems when we try to
use either OS-dependent or CPU specific features in Windows headers.
See e.g. my other mail on the winsock.h problems ...
Bye,
Ulrich
--
Dr. Ulrich Weigand
[EMAIL PROTECTED]
