GCC build failed for mips-elf with your patch on 2002-05-04T08:57:11Z.

Geoff Keating geoffk@geoffk.org
Sat May 4 12:01:00 GMT 2002


> Date: Sat, 4 May 2002 18:53:11 +0100
> From: Neil Booth <neil@daikokuya.demon.co.uk>
> Cc: gcc-regression@gcc.gnu.org, echristo@redhat.com, gcc-patches@gcc.gnu.org
> Content-Disposition: inline
> User-Agent: Mutt/1.3.27i
> 
> Geoff Keating wrote:-
> 
> > It doesn't follow, though, that cpp should be allowing a wide
> > character to fill all 64 bits.  There should probably be a macro
> > saying what the largest sensible wchar_t value can be, or we could
> > just arbitrarily clip it at 32 bits.
> 
> The macro MAX_WCHAR_TYPE_SIZE is supposed to do that I think; at
> least CPP is the only user.  I've posted another patch so that
> CPP uses "unsigned long" or "unsigned long long" internally to
> handle the cases where an unsigned int isn't enough.  If you don't
> think that's the right answer, let me know.

That sounds fine.  Certainly, if we're cross-compiling to a machine
where 'int' is 64 bits, there ought to be a 64-bit type available on
the host.

-- 
- Geoffrey Keating <geoffk@geoffk.org> <geoffk@redhat.com>



More information about the Gcc-regression mailing list