[ofw] RE: [PATCH] Support for ND over WV (incomplete)

[Sean Hefty]

>So today, when the user calls Disconnect, it will complete NotifyDisconnect?

no - NotifyDisconnect only completes on DREQ/DREP received.  Disconnect merely
issues a DREQ (or DREP if the DREQ has been received already).

>So this requires the use of NotifyDisconnect then, to know when it's safe to
>Modify?

yep

>I think anything clever is going to actually be harder to maintain than either
>using two overlapped structures, or cancelling the existing NotifyDisconnect
>request.

We already maintain state, and we're being forced to be clever about when Modify
is called.  In some cases, it's during the Disconnect downcall.  In other cases,
its not until a DREQ/DREP has been received, which corresponds to completing
NotifyDisconnect.  The kernel really doesn't care what IOCTL gets completed when
Modify is done.  I'm more worried about clearly defining when NotifyDisconnect
completes, relative to the current state of the connection.

- Sean