From: Eugene Leitl <eugene DOT leitl AT lrz DOT uni-muenchen DOT de>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Date: Mon, 10 May 1999 13:31:02 -0700 (PDT)
To: pgcc AT delorie DOT com
Subject: Re: Optimization question
In-Reply-To: <19990510192104.C27542@vim.org>
References: <Pine DOT SOL DOT 3 DOT 96 DOT 990510125208 DOT 12696D-100000 AT ursa DOT cus DOT cam DOT ac DOT uk>
	<m10gpft-001mHUC AT Dirac DOT Chemie DOT FU-Berlin DOT DE>
	<14135 DOT 5025 DOT 220697 DOT 966833 AT lrz DOT de>
	<19990510192104 DOT C27542 AT vim DOT org>
X-Mailer: VM 6.43 under 20.4 "Emerald" XEmacs  Lucid
Message-ID: <14135.16678.291781.569805@lrz.de>
Reply-To: pgcc AT delorie DOT com
X-Mailing-List: pgcc AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

Felix von Leitner writes:

 > Huh?
 > What makes you claim that?
 > It works quite good in commercial compilers for >10 years now, what is
 > your problem with it.  gcc uses it successfully, too.
 
According to what I've read in Kevin Dowd's High Performance computing 
and have seen as an embedded developer, I would exactly call it 
'successfull'. Optimization never works very well. You have to tweak
stuff and look at the machine code dumped until you get it right. Any
additional smartness buried in the compiler only makes the process
less straightforward. Ymmv, though.

Eugene
 
 > Felix