On 08/01/2010 01:00 AM, Éric Piel wrote: > Op 31-07-10 14:49, Henrik Rydberg schreef: >> On 07/31/2010 11:33 AM, Dmitry Torokhov wrote: >> [...] >> >>>> This sounds like a misunderstanding. To rephrase: the suggestion is to always >>>> send the last known one-finger touch point, and to update that point only when >>>> there is a single finger on the pad. >>>> >>> >>> This would break two-finger scrolling mode of synaptics X driver. >> >> >> I envision synaptics being MT-aware at some point, but alright, the idea is too >> radical at this stage. >> >> Back to the original problem. When adding the second finger, the first finger >> will actually be in one of the four corners of the detected rectangle, correct? >> So how about keeping as a state what corner of the rectangle represents the >> original touch, and continue to report that corner as ABS_X/Y during a >> two-finger touch? > That could work, but in general the policy is to not do finger tracking > in the kernel. And still, if you leave this first finger before the > second finger, you'll necessarily see this "jump" as well. The difference being that all positions reported correspond to real fingers. The problem with this particular device is that it is more primitive than type A, so some method to produce correct MT information would be desirable. Keeping track of what corners of the rectangle correspond to fingers does not tell which finger is which (thus not finger tracking), but makes this device comply with type A, which is capable of detecting an unordered set of actual touch points. Henrik -- To unsubscribe from this list: send the line "unsubscribe linux-input" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
