This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [C++ Patch] PR 4093
- From: Alex Rosenberg <alexr at spies dot com>
- To: Richard Henderson <rth at redhat dot com>, Jason Merrill <jason at redhat dot com>
- Cc: Craig Rodrigues <rodrigc at attbi dot com>, <gcc-patches at gcc dot gnu dot org>
- Date: Fri, 01 Mar 2002 13:50:57 -0800
- Subject: Re: [C++ Patch] PR 4093
on 2/25/02 2:53 PM, Richard Henderson at rth@redhat.com wrote:
> On Sun, Feb 24, 2002 at 06:23:06PM +0000, Jason Merrill wrote:
>> You shouldn't be building the C++ frontend with HP's compiler ...
>
> Why don't we put
>
> #ifndef __GNUC__
> #error "you should be using make bootstrap"
> #endif
>
> in cp-tree.h and be done with it?
I totally disagree. I build a cross-compiler hosted under MPW. The host
compiler produces faster code than GCC. I don't have any interest in
targeting Mac OS and MPW and building all the libraries needed to do so. I
don't want to bootstrap since that would add work for me and slow down the
resulting compiler.
Back on 12/13, I submitted fixes for some non-ANSIisms in the C++ front-end
as part of a patch to improve MPW host support. Other than saying that
bootstrapping was required, my patches were summarily ignored. (Clue: not
everybody uses GNU/Linux on x86.)
Some of these non-ANSIisms serve only to make the code more twisty and
harder to maintain. Not only is (a ? b : c) = d a less-used extension, but
the places that used it hid the LHS in a macro that looked like it might be
a global. Ewww. This stuff should be cleaned up.
+------------------------------------------------------------+
| Alexander M. Rosenberg <mailto:alexr@_spies.com> |
| Nobody cares what I say. Remove the underscore to mail me. |