X-Authentication-Warning: delorie.com: mailnull set sender to djgpp-workers-bounces using -f
From: sandmann AT clio DOT rice DOT edu (Charles Sandmann)
Message-Id: <10201211735.AA21945@clio.rice.edu>
Subject: Re: V2.03 refresh details
To: eliz AT is DOT elta DOT co DOT il (Eli Zaretskii)
Date: Mon, 21 Jan 2002 11:35:55 -0600 (CST)
Cc: djgpp-workers AT delorie DOT com (DJGPP developers)
In-Reply-To: <Pine.SUN.3.91.1020121191536.20832C-100000@is> from "Eli Zaretskii" at Jan 21, 2002 07:16:55 PM
X-Mailer: ELM [version 2.5 PL2]
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

> I agree that we can defer the problem in stat, since you only see that on 
> one specific machine, and then only in the root directory.  (If the 
> problem turns out to be more frequent when the update is released, we can 
> always reconsider.)

The problem I was refering to wasn't this problem, but the one I patched
in cvs for stat() on devices on Win2K/XP only.  Today the v2.03 refresh
won't set the device bit when you stat() a device.  In practice it turns
out that most applications fstat() to figure this out, or already know
something is a device by name and don't check this.

(This was only discovered with the discussion on bash and /dev/null being
a normal file).

I'm not happy this bug lives in v2.03 refresh - but I don't think it's
grave enough to do anything about right now.