[haiku-development] Re: GCC 2.95 and certain #define macros
- From: "Duane Ryan" <bailey.d.r@xxxxxxxxx>
- To: haiku-development@xxxxxxxxxxxxx
- Date: Sun, 19 Aug 2007 18:48:14 -0400
I'm afraid you're screwed here; apple hasn't used 2.95 for, what, four
years? The same's true for the kde folks who (??) also contribute to the
code. Quite frankly, noone has. You're going to run into these problems left
and right with code as modern as webkit
Sorry, -Duane.
On 8/19/07, Ryan Leavengood <leavengood@xxxxxxxxx> wrote:
>
> On 8/19/07, Axel Dörfler <axeld@xxxxxxxxxxxxxxxx> wrote:
> >
> > Even if it sounds very hacky: wouldn't it be possible to use the
> > preprocessor of GCC4 with GCC2?
>
> Maybe. I've been learning a lot about the GCC tool chain lately out of
> necessity.
>
> > Alternatively, there might be patches to enhance the GCC2 preprocessor
> > to support this?
>
> I've since looked over some code from Murai in Japan who got the
> Syllable port of JavaScriptCore compiling and there is another way to
> fix this: just define all the possible values for the various WTF_
> constants that would be checked by the problematic macro. It is a bit
> hacky but may work.
>
> I haven't done that yet because, frankly, I don't want what could be a
> relatively simple port turning into a huge mess because of the old
> crusty GCC 2.95. Though I suppose BeOS programmers have been doing
> this for years, I think it is time we stop.
>
> I suggest we see about implementing what I suggested in the first
> email. Unless there is some huge show stopping reason why GCC 2.95 and
> GCC 4.x binaries cannot exist on the same system. I don't see why not.
> Apple even got PPC apps running on Intel through emulation when they
> moved to x86. I'm sure we can get x86 apps running on an x86 system ;)
>
> By the way I will look into implementing this, as I'm not usually in
> the habit of trying to suggest work for other volunteers in this
> project :)
>
> Ryan
>
>
Other related posts:
