Re: GCC 3.1 / Ada / Cygwin / Win2k ?

["nicolas" <n.brunot@cadwin.com>]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1993 bytes --]

I think it is better to use only Mingw and not mix things with cygwin
I build 3.3 with Mingw tools available on
http://sourceforge.net/project/mingw   (I use msys 1.07)

some patches for gcc sources are necessary for mingw, especially in
gcc/gcc/ada
I can send you my diff if you are interested
I just have the bare minimum diff to get a usable Gnat 3.3

Mingw folks use more patches
their patches have been ready for monthes now, if not years, but nobody
seems willing to allow them to be commited in the main tree ...
The first reason is that very few people submit patches for Mingw
The other is a kind of ABM (anything but MS) attitude ...

I have your ICE when compiling g-spipat.adb, I just recompile this file
in -O1 which works fine, and make gnatlib_and_tools again


"Rolf Ebert" <rolf.ebert@gmx.net> a �crit dans le message de news:
d2b11234.0210162346.2b3b1a31@posting.google.com...
> Wes Groleau <wesgroleau@despammed.com> wrote in message
news:<3DAC4528.81F57BBD@despammed.com>...
> > Would like to hear from anyone who successfully installed
> > the whole combo of Cygwin 1.3.13 & GCC 3.1.1 (enable Ada,
> > C, C++) on Windows 2K
> >
>
> I haven't installed it successfully yet, but I think I am quite close.
>  There are open issues, though.  Perhaps someone can provide some
> hints.
>
> I am using the current CVS head (from 2002-10-15), which will
> eventually become gcc-3.3.  The gnat1 and gnatbind binaries seem to
> work correctly and "make bootstrap" runs flawlessly.
>
> 1) I get an ICE when compiling g-spipat.adb.  My workaround was to
> completely remove the XMatch implementation and make it call XMatchD.
>
> 2) The pragma Import in a-except was replaced from
> _Unwind_RaiseException to
> _Unwind_SjLj_RaiseException.  I am not sure that this is the correct
> solution; does the EH really use setjump/longjump on cygwin?
>
> 3) I get an unresolved symbol '_nanosleep' from
> system__os_primitives__timed_delay.  No idea how to solve this one.
>
>     Rolf