Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!sam!douglass
From: douglass@davidsys.com
Newsgroups: comp.dcom.lans
Subject: Re: NDIS mailing list/developers conference
Message-ID: <7919@davidsys.com>
Date: 7 Feb 91 14:06:05 GMT
References: <1991Feb4.231747.28012@tmsoft.uucp> <1991Feb6.221425.6029@unipalm.uucp>
Organization: DAVID Systems Inc, Sunnyvale CA
Lines: 22

In article <1991Feb6.221425.6029@unipalm.uucp>, leo@unipalm.uucp (E.J. Leoni-Smith) writes:
> mshiels@tmsoft.uucp (Michael A. Shiels) writes:
>
>[ ... ] 
> 
> As I understand it NDIS is a token passing protocol - so all incoming packets 
> get passed around . I don't know whether all packets go to all stacks tho. Or
> if a packet is claimed, whether the driver then doesn't offer it further.
> I should dig oput those NDIS specs...

Almost.  
NDIS specifies that the driver pass around the packet (or just part of it
[called lookahead]) until a protocol accepts it.
I believe that once it's accepted, it cannot be unaccepted.
Conversely, if some protocol accepts it before yours sees it, you'll
miss it. 

-- 
-{JD}-  Jeff (douglass@davidsys.com)
/*   My opinions are my own.  Who else would want them?   */
    "Never count on the inevitable until it happens. . ."
    "So therefore a pointer to dev/nul (the nul device) is a NULL pointer?"