From: sandmann AT clio DOT rice DOT edu (Charles Sandmann)
Message-Id: <10108130150.AA16986@clio.rice.edu>
Subject: Windows 2000 io testing
To: djgpp-workers AT delorie DOT com (DJGPP developers)
Date: Sun, 12 Aug 2001 20:50:15 -0500 (CDT)
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

Windows 2000 notes:

1) 0x7160 cl=1 to get short truename fails on devices like con and ./nul
2) lfn open succeeds on both nul and ./nul
3) sfn _truename fails on ./nul but works on nul (returns C:/NUL)
4) sfn open fails on ./nul but works on nul (consistent with _truename)
5) _chmod (__file_exists) is true with lfn active on devices, returns 0x20
   (this is different than w95 for example).
6) _get_dev_info always returns 0 on valid lfn handles
7) truncate (write 0) fails on lfn device handles, ok on sfn handles
8) _get_dev_info looks ok for pre-opened handles

So far I've been unable to get an _open working reliably on all cases.
Using lfn opens is actually more compatible (you can test directory
existence).  The downside is _get_dev_info is bogus and write 0 fails.
The write 0 is an easy fix (ignore) - we'll probably just have to accept
this.

So far of the library routines I've checked the get_dev_info failure
on handles we open would not be critical.