I've asked about the STACKSIZE issue on the binutils mailing list.
Danny
On Sat, 2009-04-18 at 01:32 +0200, Ismail Khatib wrote:
> Am 16.04.2009 um 19:52 schrieb Danny Backx:
> > I have questions about your submissions. With some answers, I can
> > commit
> > stuff to the SVN.
> >
> > Would it work on your target if we said something like
> > #ifdef __DARWIN_UNIX03
> > %token NAME LIBRARY DESCRIPTION LEOPARD_STACKSIZE
> > %token HEAPSIZE CODE DATAU DATAL
> > #else
> > %token NAME LIBRARY DESCRIPTION STACKSIZE HEAPSIZE CODE DATAU DATAL
> > #endif
> >
> > and obviously a similar change to the other line ?
>
> Hmm I haven't tried this before, because I supposed that yacc does not
> support #ifdefs...
> I tried it now it and does not work:
>
> /bin/sh /Users/cerial/Workspace/cegcc/src/binutils/ld/../ylwrap /Users/
> cerial/Workspace/cegcc/src/binutils/ld/deffilep.y y.tab.c deffilep.c
> y.tab.h deffilep.h y.output deffilep.output -- bison -y -d
> /Users/cerial/Workspace/cegcc/src/binutils/ld/deffilep.y:107.1:
> invalid character: `#'
> /Users/cerial/Workspace/cegcc/src/binutils/ld/deffilep.y:107.2-6:
> syntax error, unexpected identifier
> make[4]: *** [deffilep.c] Error 1
> make[3]: *** [all-recursive] Error 1
> make[2]: *** [all] Error 2
> make[1]: *** [all-ld] Error 2
> make: *** [all] Error 2
>
> I don't have enough knowledge about yacc to fix this correctly so that
> this patch does
> not affect other platforms; if someone here knows more please help!
>
> >> +#ifndef __DARWIN_UNIX03
> >> extern int sigaltstack(const struct sigaltstack *, struct sigaltstack
> >> *);
> >> +#endif
> >
> > Can you explain what the __DARWIN_UNIX03 symbol means ? Is it one
> > specific release of the operating system ? Is it better to use
> > something
> > like __DARWIN_UNIX or __DARWIN or are there no such things ?
>
> The __DARWIN_UNIX03 symbol is defined from Mac OSX 10.5 Leopard onwards
> according to several sources in the internet and after a brief look
> in /usr/include/sys/cdefs.h.
>
> The patch is not required on pre-Leopard platforms (according to:
> http://mail-index.netbsd.org/netbsd-bugs/2008/02/13/msg001324.html)
> , but I don't have any older installations of Mac OSX here to check :(
> So I thought it would be correct to use this symbol here. What do you
> think?
>
>
> >> -WINUSERAPI HMENU WINAPI GetMenu(HWND);
> >> +/* WINUSERAPI HMENU WINAPI GetMenu(HWND); */
> >> +#define GetMenu(x) (NULL)
> >
> > Shouldn't this be
> >
> > + #ifndef UNDER_CE
> > WINUSERAPI HMENU WINAPI GetMenu(HWND);
> > + #endif
> >
> > Any CE application that tries to use GetMenu is invalid so it would be
> > better not to allow it to compile. Right ?
>
> You are completely right, I have fixed this now (I will supply a new
> patch as soon the above problems
> are solved).
>
> Cerial
>
> --
> Ismail "Cerial" Khatib, http://pocketinsanity.org
>
--
Danny Backx ; danny.backx - at - scarlet.be ; http://danny.backx.info
------------------------------------------------------------------------------
Stay on top of everything new and different, both inside and
around Java (TM) technology - register by April 22, and save
$200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
300 plus technical and hands-on sessions. Register today.
Use priority code J9JMT32. http://p.sf.net/sfu/p
_______________________________________________
Cegcc-devel mailing list
Cegcc-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/cegcc-devel
