Re: Testing GLib cross compilation

[Owen Taylor <otaylor redhat com>]

"J. Ali Harlow" <ali optosun7 city ac uk> writes:

> On 2002.12.13 04:36 Owen Taylor wrote:
> > Note the mention of "partially working" in my mail....
> > I've never successfully gotten a win32 build past GLib either
> > cross or natively, honestly.
> > Supposedly J. Ali Harlowe was having some success getting everything
> > cross-compiled from Linux to Windows; I haven't seen full details
> > of what he was doing, however.
> 
> It's been a while since I last looked at it, but I had gtk building
> pretty well (although not properly tested). I haven't got glib to
> build cleanly yet; I just fixed a few obvious problems. Last tested
> with version 2.0.6.
> 
> I'm aware that I really ought to have a go with the latest 2.1
> series and see where we are, but I don't see any way that I can
> fit in with Owen's timetable.

I wanted to get the cross-compilation changes in before 2.2.0 to
prevent the patches from further getting out of sync with GLib,
but I don't really see making particular cross-compiles work as
a goal for this release.

IMO, docs are really the most important missing element at this
point; there's a lot of stuff going on in the win32 build
that isn't at all obvious for someone coming from Unix.

Regards,
                                        Owen

[
 How interesting is cross-compilation to win32? Probably not
 *that* interesting, since I'd expect most development on the
 win32 backend to be done by people with only win32. 

 But for someone coming from Linux/Unix, like me, there is
 certain appeal to something that gets cygwin, multiple 
 interacting path systems, and 5 minute configure runs out
 of the equation.
]