From: sandmann AT clio DOT rice DOT edu (Charles Sandmann)
Message-Id: <10108090324.AA15396@clio.rice.edu>
Subject: Re: Fw: Windows 2000 /dev/null permission query
To: acottrel AT ihug DOT com DOT au (Andrew Cottrell)
Date: Wed, 8 Aug 2001 22:24:08 -0500 (CDT)
Cc: eliz AT is DOT elta DOT co DOT il (Eli Zaretskii), djgpp-workers AT delorie DOT com
In-Reply-To: <006801c11ff8$1a39ca70$0a02a8c0@acceleron> from "Andrew Cottrell" at Aug 08, 2001 08:51:37 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

> What results do you get on XP for the code I sent last night for eaach of
> the tests? This may give some help in finding differences between 2000 and
> XP.

First of all, I modified the test program to display the results from all
of the 7143 calls, which shows what's happening.

Second, I build the image (from cvs lib + your modules) on W2K.  I created
the old.txt on the W2K box and copied it to the XP box when I copied the
image.  

W2K:
5700 call time cx = 0x4F53 date dx = 0x2B08 r.x.flags = 0x0002 r.x.ax = 0x5700
7143-8 time cx = 0x4F54 date di = 0x2B08 r.x.flags = 0x0002 r.x.ax = 0x7143
Creation time 10 milliseconds (si)= 0x00C6
7143-6 time cx = 0x0000 date di = 0x2B08 r.x.flags = 0x0002 r.x.ax = 0x7143
7143-4 time cx = 0x4F53 date di = 0x2B08 r.x.flags = 0x0002 r.x.ax = 0x7143
NEW date (di)= 0x2B08
7143 call time cx = 0x4F53 date di = 0x2B08 r.x.flags = 0x0002 r.x.ax = 0x7143
Creation time 10 milliseconds (si)= 0x00C7

XP RC1 (file created by echo >new.txt):
5700 call time cx = 0x4855 date dx = 0x2B08 r.x.flags = 0x0002 r.x.ax = 0x5700
7143-8 time cx = 0x4856 date di = 0x2B08 r.x.flags = 0x0002 r.x.ax = 0x7143
Creation time 10 milliseconds (si)= 0x00C6
7143-6 time cx = 0x0000 date di = 0x2B08 r.x.flags = 0x0002 r.x.ax = 0x7143
7143-4 time cx = 0x4855 date di = 0x2B08 r.x.flags = 0x0002 r.x.ax = 0x7143
NEW date (di)= 0x2B08
7143 call time cx = 0x4855 date di = 0x2B08 r.x.flags = 0x0002 r.x.ax = 0x7143
Creation time 10 milliseconds (si)= 0x00C7

XP RC1 (file created by copy from W2K system):
5700 call time cx = 0x5753 date dx = 0x2B08 r.x.flags = 0x0002 r.x.ax = 0x5700
7143-8 time cx = 0x4856 date di = 0x2B08 r.x.flags = 0x0002 r.x.ax = 0x7143
Creation time 10 milliseconds (si)= 0x00C6
7143-6 time cx = 0x0000 date di = 0x2B08 r.x.flags = 0x0002 r.x.ax = 0x7143
7143-4 time cx = 0x5753 date di = 0x2B08 r.x.flags = 0x0002 r.x.ax = 0x7143
NEW date (di)= 0x2B08
7143 call time cx = 0x5753 date di = 0x2B08 r.x.flags = 0x0002 r.x.ax = 0x7143
Creation time 10 milliseconds (si)= 0x00C7

First, not that we do have consistency between the 5700 call and 7143-4 call.
So, the 5700 call returns the last write time.
The 7143-8 time you and Eli were comparing was the creation time, which
if you note can be quite different from the last write time - and is actually
after the last write time (bug).

So, I hope this clears things up a bit - that the 7143 call seems to work 
on XP, and 7143 is consistent with 5700 on both platforms.