X-Authentication-Warning: delorie.com: mailnull set sender to djgpp-workers-bounces using -f
From: pavenis AT lanet DOT lv
To: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>, djgpp-workers AT delorie DOT com
Date: Wed, 30 Jan 2002 15:24:56 +0200
MIME-Version: 1.0
Subject: Re: Long delays running SED under Win98SE
Message-ID: <3C581048.299.106B51B@localhost>
References: <3C57E975 DOT 1307 DOT 6F0855 AT localhost>
In-reply-to: <Pine.SUN.3.91.1020130150858.17919S-100000@is>
X-mailer: Pegasus Mail for Windows (v4.01)
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
Content-description: Mail message body
Reply-To: djgpp-workers AT delorie DOT com

On 30 Jan 2002 at 15:10, Eli Zaretskii wrote:

> 
> On Wed, 30 Jan 2002 pavenis AT lanet DOT lv wrote:
> 
> > I experienced a long delays (more than 30 seconds) when tried to use SED to 
> > extract contents of .eh_frame section from GCC generated assembler source:
> > 
> > sed -n -e '/\.eh_frame/,/\.section/p' <foo.s 
> 
> What version of Sed is that?  Is it 3.02 or 3.02.80?  If the latter, does 
> it have the patch I sent a few months ago to fix slow operation?  Also, 
> did you compile Sed yourself, and if so, with what version of libc.a?
> 
> > It happens on one single LFN call (0x7143) as far as I saw with some old
> > DOS trace program.
> 
> Can you tell the details of that call?

It seems to be wildcard expansion problem at program startup before main()
is called. I'm getting it with both libc.a from DJGPP v2.03 (updated) and
current CVS version. Version of SED is not critical as one can get the same 
also by
	djecho '/\.foo/'

Andris