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

Re: [bluetooth-dev] btduser: hci receive error with 115200 !



Matthias Fuchs wrote:

> That means I have firmware version P9A and therefore
> CONFIG_BLUETOOTH_SET_BAUDRATE_BLOCKING must not be defined (default in
> cvs). I think that is not my problem ???
>

That's my understanding as well. Ok.

> >
> > Another suggestion: this is related to my recent email about syncing. If you
> > modify hci.c hci_receive_data() to simply return when it gets bad uart data
> > then your stack might successfully resync on the next buffer from the uart &
> > proceed ok from there after a command timeout.
> >
> I put your resyncing code into hci.c. It seems to be a good idea, but
> its a bit dangerous to increase the buffer pointer again ang again
> without any boundary checking !

I thought the count check at the top of the loop would take care of the boundary
checking (that's how it normally seems to work, anyway).

>
> In the end it did not help ! I though that changing the baudrate was
> already solved since a long time. Do you have any further idea ? What
> debug information might be helpful ?
>

Not too sure, Matthias. It seems like there's an unavoidable race condition with
the P9A firmware (i.e. we must switch our serial baud rate before it sends the
reply at the new rate). But I was hoping we'd just lose the reply and resync on
subsequent traffic.

--gmcnutt


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