SCO port update - what now?

[Adam Jackson]

On Friday 10 June 2005 13:52, Kean Johnston wrote:
> > If we really don't want to require #warning, then we need to fix Mesa,
> > fontconfig, and the i810 and sis drivers as well.
>
> I dont know why those didn't break. Perhaps they are more correctly
> surrounded with #ifdef __GNUC__ ?

All the cases I just mentioned have guardian ifdefs, though they're not #ifdef 
__GNUC__.

> > I'm a bit dismayed that there exist compilers that don't simply throw out
> > unrecognized preprocessor directives (since you basically have to do that
> > for #pragma).
>
> But #pragma is compiler control, and #warning is a pre-processor thing.
> I can see compilers throwing away unrecognised pragmas but barfing on
> unrecognised pre-processor directives.

There were one or two other examples I probably should have used instead of 
#pragma, #sccs for example.  I can see your point, of course, I'm just saying 
preprocessors that don't discard unrecognized directives are needlessly 
fragile.

Not that it matters, of course.  If #warning makes things break on a platform 
someone cares about, then we can't use it.  Life is hard.

- ajax
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.x.org/archives/xorg/attachments/20050610/fab8ea90/attachment.pgp>