[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [bluetooth-dev] dirty code in bt_disconnect



> -----Original Message-----
> From: Ilguiz Latypov [mailto:ilatypov@xxxxxxx.com]
> Sent: 05 October 2001 16:47
> To: Peter Kjellerstedt
> Cc: OpenBT Development List
> Subject: RE: [bluetooth-dev] dirty code in bt_disconnect
> 
> Peter,
> 
> thank you for reply.  Here is what I figured out:
> 
> On Fri, 5 Oct 2001, Peter Kjellerstedt wrote:
> 
> > > c) The other issue: the handling of events from the
> > > descriptor 0 is empty when the "IP assign" feature is disabled.
> >
> > Well, it does just as much for stdin when the IP assigner 
> > is enabled, i.e., nothing...
> 
> Can this prevent btdm from calling discover_connections() once the
> standard input produced an event in select()?

Actually, yes. What happens if something would appear on stdin, is
that select() would return. But since we do not do anything about
it, the next time select() is called, it would return immediately.
Thus it would create an endless loop, and the timeout would never
occur.

I have removed stdin from the set now, so this should not
happen any longer.

> > > Besides, the "IP assign" Unix socket contains a for-loop
> > > where the body doesn't depend on the loop count.  We are
> > > not using the "IP assign" feature.
> >
> > Where is this code?
> 
> I was referring to the default case of the switch statement 
> in the main() function of experimental/btd.c.

Ok.

> Ilguiz

//Peter
-
To unsubscribe from this list: send the line "unsubscribe bluetooth-dev" in
the body of a message to majordomo@xxxxxxx.com