[patch] bug in cpp's #ident handling in gcc 3.4 [Was: ccache with buildworld]

[David O'Brien]

On Thu, Aug 26, 2004 at 07:59:53PM +0300, Maxim Sobolev wrote:
> Steve Kargl wrote:
> >On Thu, Aug 26, 2004 at 07:10:20PM +0300, Maxim Sobolev wrote:
> >
> >>Pointless. Instead of fixing the problem we will remove condition when 
> >>this problem evidences itself. I am not advocating about keeping #ident, 
> >>since it is completely orthogonal question. gcc claims that it supports 
> >>#ident, there are problems in this support - they have to be fixed.
> >>
> >>Remember, there are lot of software out of /usr/src over which we have 
> >>no control (e.g. /usr/ports), so that removing #ident will really be 
> >>only half-measure, especially considering that the patch is readily 
> >>available.
> >>-Maxim
> >
> >Isn't src/contrib/gcc/c-ppoutput.c on the vendor branch?
> >Have you sent your patch to GCC?
> 
> I wonder if it would be better if kan will do it, since he obviously 
> knows better inner circles of gcc team.

No, it would not be better.   If all the GCC team ever sees on the their
mailing lists are Kan and myself that doesn't show them that FreeBSD is a
community that they need to be concerned with its needs.

Please email this patch to 'gcc-patches at gcc.gnu.org'.

Thanks,
-- 
-- David  (obrien at FreeBSD.org)