From:	pavenis AT lanet DOT lv
Message-ID:	<B0000107412@stargate.astr.lu.lv>
To:	Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>, djgpp-workers AT delorie DOT com
Date:	Tue, 2 Nov 1999 15:13:41 +0200
MIME-Version:	1.0
Subject:	Re: -g vs -s
References:	<Pine DOT A41 DOT 4 DOT 05 DOT 9911010946420 DOT 132160-100000 AT ieva01 DOT lanet DOT lv>
In-reply-to:	<Pine.SUN.3.91.991102095358.19809B-100000@is>
X-mailer:	Pegasus Mail for Win32 (v3.12a)
Reply-To:	djgpp-workers AT delorie DOT com

On 2 Nov 99, at 9:54, Eli Zaretskii wrote:

> 
> On Mon, 1 Nov 1999, Andris Pavenis wrote:
> 
> > For example I specially building gcc without -g and I'm not running
> > strip on binaries as this:
> 
> That's why DJ said that people who need this can use -g0.  Won't it
> solve your problem in this case?
> 
> Alternatively, I think "strip --strip-debug" will remove debugging
> symbols added by -g, but leave enough to have meaningful traceback.
> 
> The automatic implication of -s is for those who don't know how to use
> command-line options, but do know how to ask questions.  Keeping the
> flood of those questions as low as we can is always a good idea,
> IMHO.
> 

Once more about this topic:

It seems that this idea is even worse as I thought before as it will for 
example break debugging support in rhide:

Let's assume we have too files foo.c and bar.c in project. Rhide will 
generate commands like (if options -g and -O2 are requested in project)
	gcc -g -O2 -c foo.c -o foo.o
	gcc -g -O2 -c bar.c -o bar.o
	gcc foo.o bar.o -o foo.exe

As result we'll not have debugging info in executable as gcc would 
default to use option -s. I'm afraid these problems will generate even 
more questions in mailing list than slight bloating of executables.
We'll have similar problem also with many makefiles.

Andris

- Raw text -