On Mon, Jul 27, 2009 at 3:55 PM, Pavel Machek<pavel@xxxxxx> wrote: > On Mon 2009-07-27 15:45:48, Arve Hj?nnev?g wrote: >> On Mon, Jul 27, 2009 at 9:06 AM, <gregkh@xxxxxxx> wrote: >> ... >> > This is original version from Arve, fixed only to compile; I do have >> > cleaner version, but I broken something inside, and this will preserve >> > authorship better. I originally tried to push the driver directly to >> > input/touchscreen, but the driver has some issues with interrupt >> > handling that are more difficult to fix than I expected at first. >> > >> >> While it would be nice to use the threaded interrupt support, I don't >> see why that change cannot be postponed until the the threaded >> interrupt code supports this class of devices where the interrupt >> needs to be masked before the thread can run. > > That's up to linux-input people, I guess. But... > > From the discussion, it seems that the driver has more issues. I have > quite long list. > > 1) either it needs to switch to threaded interrupt supports, or all > the races need to be fixed. > > [Comments I remember here is:] > > ( >> > + ts->input_dev->name, ts->use_irq ? "interrupt" : > "polling"); >> > + >> > + return 0; >> > + >> > + err_input_register_device_failed: >> > + input_free_device(ts->input_dev); >> > + >> >> Don't see canceling timer nor shutting off WQ here. Also, maybe >> implement open() and close() so we don't reschedule WQ while polling?_ > ) Unless you changed something there should be no failure path after starting the timer or enabling the interrupt. synaptics_ts_remove is missing a flush_work call though. > > 2) it uses non-standard multitouch support. Either multitouch needs to > be disabled for merge, or it needs to switch to standard config. > I have a patch that adds support for the new protocol, but it does not yet remove the old hack. > 3) poll mode should probably be removed. > > 4) minor stuff, like properly filling identification fields > (vendor/device/product) etc. > > Yes, ideally it should be 3 or 4 patches and it is out of staging and > in the kernel proper. But as I already got patches for people trying > to help with cleaning, I believe staging is right place for > it. > Pavel > -- > (english) http://www.livejournal.com/~pavelmachek > (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html > -- Arve Hjønnevåg -- 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
