From:	sfield AT microsoft DOT com (Scott Field)
Subject:	RE: ps and kill
8 Dec 1996 22:14:21 -0800 :
Sender:	daemon AT cygnus DOT com
Approved:	cygnus DOT gnu-win32 AT cygnus DOT com
Distribution:	cygnus
Message-ID:	<c=US%a=_%p=msft%l=RED-66-MSG-961209035338Z-71428.cygnus.gnu-win32@INET-01-IMC.microsoft.com>
Original-To:	"'Doug Evans'" <dje AT cygnus DOT com>,
	"'Ben Constable'"
	<s2172184 AT cse DOT unsw DOT edu DOT au>
Original-Cc:	"'gnu-win32 AT cygnus DOT com'" <gnu-win32 AT cygnus DOT com>
X-Mailer:	Microsoft Exchange Server Internet Mail Connector Version 4.0.994.63
Encoding:	51 TEXT
Original-Sender:	owner-gnu-win32 AT cygnus DOT com

Writing a device driver sounds overkill to me.  Also, these are
user-mode API calls, and I don't think kernel-mode equivalents exist.
Why not write OS dependant code and hide it in cygwin.dll ?

>----------
>From: 	Ben Constable[SMTP:s2172184 AT cse DOT unsw DOT edu DOT au]
>Sent: 	Friday, December 06, 1996 6:01 PM
>To: 	Scott Field; 'Doug Evans'
>Cc: 	'gnu-win32 AT cygnus DOT com'
>Subject: 	Re: ps and kill
>
>> The code to do this right for both platforms would be very different.
>
>This does not have to be true. Under win32, some devices are mapped to
>files. The com ports, for instance, are obtained by opening a file with the
>name of the com port. If we wrote a driver for NT that took a filename of
>utmp, and it updated everytime somebody logged on and off, then we would
>not have a problem anymore. The code for both platforms would open the same
>file and then the drivers would be the only different bit.
> 
>> For Win95, only one logged on security context is possible, so use
>> GetUserName()
>> For WinNT, multiple users can be logged on (see SU utility is NT reskit)
>> .  NetWkstaUserEnum() can give you a list of logged on users.
>
>There you go. My point proven - NT and win95 are different. However, in
>this case, it could be gotten over by a little simple driver stuff. Anybody
>out there got any driver info (I do not have the DDK)?
>
>> >Maybe `who' has been ported to NT yet.
>> >You'll find there are lots of such binaries that will compile but
>> >that no one has had time yet to see whether they do anything appropriate
>> >or not.
>
>Well, have the people at cygnus heard us? Will B17 have a nonworking who in
>it? The short term fix is to use native API's. The long term fix is the
>driver solution.
>
>> >One difference between the linux project and the gnu-win32 project
>> >is the several orders of magnitude of difference in quantity of
>> >contributed patches ....
>
>How do I contribute a patch?
>
>Ben Constable
>s2172184 AT cse DOT unsw DOT edu DOT au
>
>-
>For help on using this list, send a message to
>"gnu-win32-request AT cygnus DOT com" with one line of text: "help".
>
-
For help on using this list, send a message to
"gnu-win32-request AT cygnus DOT com" with one line of text: "help".

- Raw text -