Message-Id: <200106111543.f5BFhcI00366@hal.astr.lu.lv>
Content-Type: text/plain;
  charset="iso-8859-13"
From: Andris Pavenis <pavenis AT latnet DOT lv>
To: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
Subject: Re: [PATCH] First part of compiler options update for CVS version of DJGPP
Date: Mon, 11 Jun 2001 18:22:54 +0300
X-Mailer: KMail [version 1.2.2]
Cc: djgpp-workers AT delorie DOT com
References: <Pine DOT SUN DOT 3 DOT 91 DOT 1010611171201 DOT 20317G-100000 AT is>
In-Reply-To: <Pine.SUN.3.91.1010611171201.20317G-100000@is>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Reply-To: djgpp-workers AT delorie DOT com

On Monday 11 June 2001 17:13, Eli Zaretskii wrote:
> On Mon, 11 Jun 2001, Andris Pavenis wrote:
> > Here is diffs for first part of compiler options update for building
> > DJGPP runtime library to be ready for gcc-3.0:
> >
> > 1) replacing -m486 with -mcpu=pentium -march=i386
> > 2) removal of -malign-* options
> > 3) removal of -Wtraditional
>
> Looks good to me.

I committed these changes.

> > I didn't add additional warning options yet. Perhaps it can be done later
>
> Yes.

Perhaps we should think about adding also option -W.
Currently it doesn't compile without warnings when -W is specified, so
perhaps sources should be fixed at first

> What about -fno-strength-reduce -- can you ask GCC maintainers about that?

I think it can be removed unless we have some real reason why to keep it
(I don't have one myself). Perhaps we don't need fine tuning of optimization 
(as it's called in GCC docs, see fragment below) unless we know we're really 
getting something from it (or fixing some breakage)

Currently I left possible removal of -fno-strength-reduce to next updates

Andris

--------------------  fragment from gcc.info (gcc-3.0 20010610) ----------
   The following options control specific optimizations.  The `-O2'
option turns on all of these optimizations except `-funroll-loops' and
`-funroll-all-loops'.  On most machines, the `-O' option turns on the
`-fthread-jumps' and `-fdelayed-branch' options, but specific machines
may handle it differently.

   You can use the following flags in the rare cases when "fine-tuning"
of optimizations to be performed is desired.

`-fstrength-reduce'
     Perform the optimizations of loop strength reduction and
     elimination of iteration variables.

`-fthread-jumps'
     Perform optimizations where we check to see if a jump branches to a
     location where another comparison subsumed by the first is found.
     If so, the first branch is redirected to either the destination of
     the second branch or a point immediately following it, depending
     on whether the condition is known to be true or false.

`-fcse-follow-jumps'
     In common subexpression elimination, scan through jump instructions
 .......
---------------------------------------------------------------------