Hi Joerg, >>>>>> On Fri, Apr 17, 2015 at 5:36 AM, Jörg Otte <jrg.otte@xxxxxxxxx> wrote: >>>>>>> The BT mouse is "death" in v4.1. >>>>>>> The BT mouse has been working in 4.0 and previous kernels, so this >>>>>>> is a regression. >>>>>> >>>>>> Any chance of bisecting it? >>>>>> >>>>>> Linus >>>>> I will try that. >>>>> >>>>> Thanks, Jörg >>>> >>>> I first tried to bisect over all. But than I got an unbootable kernel. >>>> Then I did a bisect over net/bluetooth and I get the following fist bad >>>> commit: >>>> >>>> 5f5da99f1da5b01c7c45473a500c7dbb77a00958 is the first bad commit >>>> commit 5f5da99f1da5b01c7c45473a500c7dbb77a00958 >>>> Author: Marcel Holtmann <marcel@xxxxxxxxxxxx> >>>> Date: Wed Apr 1 13:51:53 2015 -0700 >>>> >>>> Bluetooth: Restrict HIDP flags to only valid ones >>>> >>>> The HIDP flags should be clearly restricted to valid ones. So this puts >>>> extra checks in place to ensure this. >>>> >>>> Signed-off-by: Marcel Holtmann <marcel@xxxxxxxxxxxx> >>>> Signed-off-by: Johan Hedberg <johan.hedberg@xxxxxxxxx> >>>> >>>> :040000 040000 b51ac3634c9d44f4d9df0e7f548b524954b99c76 >>>> 63bfb47283609849f1b3b8f05fe61743ccddfee6 M net >>> >>> thanks for bi-secting this. I looked at our existing userspace and restricted it to the flags that are currently in use. However it seems that I made a mistake. What version of BlueZ userspace are you using (bluetoothd --version). >>> >>> >> bluetoothd --version >> 4.98 > > okay. I only looked at BlueZ 5.x and that might have been my mistake. Let me check this and fix this properly. I think that I know what I screwed up here. I sent you a patch to fix this. Can you please test it and report back. If that fixes it for you, then I will send it to Linus for inclusion. Regards Marcel -- To unsubscribe from this list: send the line "unsubscribe linux-bluetooth" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
