On Sat, Feb 27, 2010 at 12:31 PM, Daniel Åkerud <daniel.akerud@...:
> On Fri, Feb 26, 2010 at 10:48 PM, Jarod Wilson <jarod@...> wrote:
>> If you can get a trace, I can try to come up with a fix. I do have
>> hardware I can test with, but no Windows anywhere atm that I could do
>> any snooping under myself.
>> Jarod Wilson
> A new day of pure unadulterated hacking. Here is a trace containing
> connecting the Knob followed by pressing and holding UP:
> Note that I needed to sniff the whole buss due to connecting a new device,
> so inevitably there is a bit of mouse data in the end due to turning off the
> log, and also more importantly there might be other data interleaved from
> other devices, although it looks very clean.
> Interesting events at a specific SEQ number:
> 0002: Plugging in the Knob
> 0067: Idle data starting to flow
> 0204: Interesting data sent to the Knob??
> 0400: Another initialization sequence?? I see this every time I plug it in.
> 0594: I start pressing UP. We can see that during this period there is no
> data sent to the Knob.
> 1600: I let go of the UP button.
> 2170: This is where I start to move the mouse to stop the trace.
> I did an extra trace just in case, which looks pretty much identical in my
> eyes, but just to be sure:
This is weird, but the amount of time I can keep it pressed before
garbage-data arrives depends on where I point the remote. Pointing it in the
floor, or pointing it directly at the knob, etc.
Don't know if it helps. Makes me think of some kind of synchronization
problem between driver and hw?