On Mon, 10 Aug 2015, Jason Gerecke wrote: > This bug was exposed by an update to libinput which makes it more > reliant on accurate resolution values. It is a regression in our code, > but one which has gone unnoticed for quite some time. I've tracked its > introduction to somewhere between 3.11 and 3.13, and having looked at > the commits in that range believe that 401d7d1 (merged in 3.12) is > likely the culprit. That commit replaced a function which calculated the > resolution as part of the input device registration process with one > that calculates it as part of the probe process after the HID descriptor > is read (which doesn't do us any good in the wireless case). > > Fixing this in 4.2 ensures that this bug doesn't tickle libinput any > longer than necessary. It would also be nice to backport this patch to > stable, but I'm not sure if it quite meets the necessary severity standards. Thanks for elaborting on this, Jason. I will be looking into this a little bit more, but am applying to for-4.2/upstream-fixes-devm-fixed in any case (my current plan is to send this to Linus this week due to devm memory corruption fix, and let this patch piggy-back on it). -- Jiri Kosina SUSE Labs -- 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
