This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: gcc compile-time performance
- From: Marc Espie <espie at quatramaran dot ens dot fr>
- To: gcc at gcc dot gnu dot org
- Cc:
- Date: Sun, 19 May 2002 14:16:18 +0200
- Subject: Re: gcc compile-time performance
- Organization: Ecole Normale Superieure (quatramaran)
- References: <20020519111849.D908FF28CC@nile.gnat.com>
In article <20020519113946.GA1476@daikokuya.demon.co.uk> you write:
>Robert Dewar wrote:-
>
>> I disagree, unless you have some perverse definiteion of "properly".
>
>Well, a call to the mb functions or iconv is considerably more
>expensive than
> c = *buffer->cur++;
>I'm not sure, but if we want caret-positioned diagnostics in future
>(I certainly do) then it may require more book-keeping for that too.
>Why do you think multibyte char support is cheap?
If multibyte support is that expensive, I believe that providing two
versions of the preprocessor, one multibyte, the other not, selectable
at runtime, would be a very good idea.
Most code out there does not need multibyte support in the preprocessor.