Date: Sun, 29 Apr 2001 15:49:26 +0300 (IDT)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender: eliz AT is
To: Tim Van Holder <tim DOT van DOT holder AT pandora DOT be>
cc: djgpp-workers AT delorie DOT com, Laurynas Biveinis <lauras AT softhome DOT net>
Subject: Re: gcc-3.0 related patches for DJGPP CVS version
In-Reply-To: <CAEGKOHJKAAFPKOCLHDIAEOICCAA.tim.van.holder@pandora.be>
Message-ID: <Pine.SUN.3.91.1010429154753.28788B-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk


On Sun, 29 Apr 2001, Tim Van Holder wrote:

> > > Of course in this case these switches should go into specs too.
> > 
> > Not necessarily: building a library could involve considerations which 
> > don't apply to applications on a Joe Random User's machine.
> > 
> Don't the -march and or -mcpu switches affect alignment?

Yes, they do.  I'm afraid I don't get your point.

> Seems to me that
> the choice of what alignment is best is closely tied to the architecture
> and/or CPU you're generating code for.
> As such, it shouldn't be necessary to explicitly specify alignment unless
> you have specific needs.

Laurynas seems to say that he can suggest a better default than what 
-mcpu implies.