Messages in this thread | | | From | Jeffrey Brown <> | Date | Fri, 25 Mar 2011 16:30:20 -0700 | Subject | Re: [PATCH 1/4] input: Set default events per packet. |
| |
On Fri, Mar 25, 2011 at 1:20 AM, Henrik Rydberg <rydberg@euromail.se> wrote: >> +static inline bool is_mt_axis(int axis) >> +{ >> + return axis == ABS_MT_SLOT || >> + (axis >= ABS_MT_FIRST && axis <= ABS_MT_LAST); >> +} > > It would be great to get this inline into input/mt.h instead.
Makes sense. I'll do that.
>> + else if (test_bit(ABS_MT_TRACKING_ID, dev->absbit)) >> + mt_slots = dev->absinfo[ABS_MT_TRACKING_ID].maximum - >> + dev->absinfo[ABS_MT_TRACKING_ID].minimum + 1; > > This one is a bit iffy - the tracking id is not limited like this in > mainline, looks like android usage. A test againts some arbitrary max > should do it.
Yeah, I'm not sure about this one. Tracking ID could effectively have any range. All of the MT Protocol A touch screen drivers I have looked at, assuming they report tracking ids at all, report a reasonable upper bound on the contact points they support.
Originally, I set an arbitrary maximum bound of 20 slots. In the interests of keeping it simple, I decided to remove that bound when I submitted the patch for review here.
How about:
mt_slots = min(MAX_MT_SLOTS_TO_INFER_FROM_TRACKING_ID_RANGE, dev->absinfo[ABS_MT_TRACKING_ID].maximum - dev->absinfo[ABS_MT_TRACKING_ID].minimum + 1);
Where MAX_MT_SLOTS_TO_INFER_FROM_TRACKING_ID_RANGE is set to 32 or something.
There's also the question of how many slots we should infer when neither ABS_MT_SLOT or ABS_MT_TRACKING_ID is available. The drivers I've seen that don't provide tracking ids, are very basic and tend to only support 2 touch points.
I guess we could add a DEFAULT_NUMBER_OF_MT_SLOTS constant to handle that case.
Please feel free to suggest better names for these constants.
Jeff. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |