On Thu, Dec 22, 2016 at 10:00:10PM +0100, Jiri Kosina wrote:
> We've discussed with Tomasz off-list, and it looks like that there likely
> is a better and more general way to fix it; this seems to have actually
> been caused by 6da7066906e977d4 ("HID: ignore absolute values which don't
> fit between logical min and max"), as it turns out that this check should
> be applied only in case Null state is set (refer to HID specification
> 1.11, section 6.2.2.5, p.31), bko#68621.
>
> Tomasz comitted to look into this after coming back from christmas/new
> year vacation. I'm putting the innomedia branch on hold for now.
>
> --
> Jiri Kosina
> SUSE Labs
I've tested the patch from the bugzilla and it does indeed prevent the
values being discarded. evtest reports the minimum and maximums
correctly as -1 and 1 but the events give values -2 and 1, I'm not sure
if this will be an issue as jstest correctly sees left/up presses as
-32767 and right/down presses as 32767. The device still needs the
HID_QUIRK_MULTI_INPUT quirk to correctly show up as two event/js
devices.
I will still need to do a few more tests to make sure there are no
issues in other applications and to make sure that other devices still
operate correctly too.
Once that is done I will go ahead and send two patches, one with the
authorship preserved for the bugzilla patch and one patch to add the
quirks.
--
Tomasz Kramkowski | GPG: 40B037BA0A5B8680 | Web: https://the-tk.com/
--
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
