This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Revised mips alignment patches.
- To: Theodore dot Papadopoulo at sophia dot inria dot fr
- Subject: Re: Revised mips alignment patches.
- From: Geoff Keating <geoffk at ozemail dot com dot au>
- Date: Thu, 2 Sep 1999 20:14:26 +1000
- CC: law at cygnus dot com, gavin at cygnus dot com, egcs-patches at gcc dot gnu dot org, rth at cygnus dot com, davem at redhat dot com
- References: <199909020937.LAA15653@mururoa.inria.fr>
> Date: Thu, 02 Sep 1999 11:37:34 +0200
> From: Theodore Papadopoulo <Theodore.Papadopoulo@sophia.inria.fr>
> If gcc goes towards the -fxxx=n notation shouldn't g++ be
> converted as well ?
>
> I see at least:
>
> template-depth-n
> name-mangling-version-n
>
> The main problem might be that the first one might be somewhat used.
>
> However, deferring the read of the = into the read_integral_parameter
> function would allow a simple way to ensure for backward compatibility
> for all those -fxxxx-n...
>
> Does this seems a good idea ???
There's also some -W options, like -Wid-clash-<len>. There's no
hurry, though, gcc has had a mix of ...-<n> and ...=<n> options for
some years now.
I'm not sure about read_integral_parameter, it's used in other
places (like -O). I doubt we'll be changing to -O=2 anytime soon.
For options that might be popular, it'd be polite to accept both spellings
of the options for a year or two; there's not much overhead even doing
it the way I chose for -finline-limit.
--
Geoffrey Keating <geoffk@cygnus.com>