On 08/16/2011 10:41 AM, Chase Douglas wrote:

On 08/12/2011 10:16 AM, Daniel Kurtz wrote:

quoted

Hello,

----
For v3:
  Merged Patch 5 & 6:
     BTN_TOOL_*TAP always reports correct number of fingers.
     During certain ambiguous number-of-finger-transitions, there may be less
     than 2 valid MT-B slots, even if the BTN_TOOL_*TAP value indicates two or
     more fingers are present on the pad.

I am happy with the patch series. I'm not sure if anyone will actually
use the MT slots because of all the caveats :), but this driver appears
to do the best thing I can think of with the data from the device. It
looks like it was a lot of work to figure it out, so cheers to Daniel!

At this time I see Henrik has provided two replies and Daniel has
followed up. The first reply set will cause a small amount of axis
handling changes, and the second reply about the amount of code changes
and implementation details of the tracking ids has been rebuffed. If
this is the end result, you can add my Acked-by to all the patches:

Acked-by: Chase Douglas <redacted>

I forgot about the change in protocol aspect of this. We still need
documentation stating that when BTN_TOOL_*TAP > max mt slots, the
BTN_TOOL_*TAP events should be used to get the current number of active
touches instead of the number of open MT slots. Would you be able to
send a separate patch for that?

(I'm trying to coerce you into updating the docs partly because you're
touching this area and partly because I'm a bit weary of writing input
protocol docs myself :).

Thanks!

-- Chase