From: mdruiter AT cs DOT vu DOT nl (Ruiter de M)
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Pointer to ... and a question
Date: 1 Dec 1997 12:46:11 GMT
Organization: Fac. Wiskunde & Informatica, VU, Amsterdam
Lines: 22
Message-ID: <65ubij$13s$1@star.cs.vu.nl>
References: <199711281711 DOT JAA10706 AT adit DOT ap DOT net>
NNTP-Posting-Host: sloep66.cs.vu.nl
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Precedence: bulk

Nate Eldredge (eldredge AT ap DOT net) wrote:
> I've never heard that. `-O3' is just `-O2' + `-finline-functions'.

Is this documented to be the *only* difference?

> I don't know of any problems with inlining, except that it isn't
> always a Good Thing. The only thing I know of is a bug with
> `-fstrength-reduce' which is turned on by `-O2', and is fixed in GCC
> 2.7.2.1.

I also once (trying to compile ghostscript 403) noticed some sort of
bug, which went away by using `-fno-cse-follow-jumps'.

I think it was a bug, at least code was generated that operated
differently. Maybe I will further investigate this some time, but
right now I haven't got enough time for that.

-- 
Groeten, Michel.        http://www.cs.vu.nl/~mdruiter
 \----/==\----/
  \  /    \  /          "You know, Beavis, you need things that suck,
   \/      \/           to have things that are cool", Butt-Head.