This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Processor information document
- To: law at cygnus dot com
- Subject: Re: Processor information document
- From: Marc Espie <espie at quatramaran dot ens dot fr>
- Date: Wed, 22 Sep 1999 15:11:06 +0200
- Cc: egcs at egcs dot cygnus dot com
- Organization: Ecole Normale Superieure de Paris
In article <3890.937947598@upchuck.cygnus.com> you write:
> In message <199909212043.WAA07895@mira.isdn.cs.tu-berlin.de>you write:
> > Good Evening,
> >
> > When checking out the current tree, I found that gcc now supports a
> > new architecture, fr30. While this is certainly great news to users of
> > the fr30 processors, I found that I had no clue what the heck this is.
> >
> > I'm not at all complaining, it is reasonable to get the code working
> > first, before answering the curiosity of the public :-) I also guess
> > there will be an announcement one of these days.
> >
> > Anyway, trying to find out what fr30 is, I did a Web search. Is it
>Err, it was included in the announcement Nick sent to the gcc list. ANd
>there will be a similar blurb on the gcc home page when I get around to
>checking it in. Also note Nick's message included a URL for the fujitsu
>site.
Announcements don't last.
This definitely is the kind of thing that just needs a small addition to
the documentation that's actually bundled with the compiler... At least
the info pages, and the fr30 framework proper, should mention what is
going on.
In my opinion, the way Nick went about that is a bit backwards... Some
information doesn't belong in a separate announcement. Rather, it should
be part of the patch & code proper, so that there's no chance the information
will get lost...
It is still possible to track that information and clean it up afterwards,
but why make that difficult and/or error-prone in the first place ?