Message-Id: <200306091400.h59E0bsG022093@ginger.cmf.nrl.navy.mil>
To: "David S. Miller" <davem@redhat.com>
cc: baldrick@wanadoo.fr, wa@almesberger.net, linux-kernel@vger.kernel.org
Subject: Re: [PATCH][ATM] use rtnl_{lock,unlock} during device operations (take 2) 
In-reply-to: Your message of "Mon, 09 Jun 2003 06:38:25 PDT."
             <20030609.063825.123987226.davem@redhat.com> 
Date: Mon, 09 Jun 2003 09:58:45 -0400
From: chas williams <chas@cmf.nrl.navy.mil>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 910
Lines: 20

>My personal feeling is that there should be a way to zombie
>VCCs, precisely for such events.
>
>ATM should unlink them immediately, and mark them dead.
>Anything that tries to do something with a VCC should
>check that it is still alive.

i imagine marking vcc->dev = NULL would be pretty close to the above.
most operations going into the driver check vcc->dev already and it
might actually handle this correct w/o too much extra effort.  you
might also set some flags on the vcc's like ATM_VF_RELEASED etc etc.

of course, this sort of implies that the vcc's are on their own list
and not on a list that is actually part of the atm dev.  (see latest
rfc patch).
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/