Hi Kenneth, On Fri, Mar 4, 2022 at 4:06 PM Kenneth R. Crudup <kenny@xxxxxxxxx> wrote: > > > On Sat, 29 Jan 2022, Kenneth R. Crudup wrote: > > > So, it turns out that my lack of discoverability is also tied into this last > > series of BT commits > > So, after "for-net-next-2021-11-16" was pushed into Linus' master, my Laptop's > Bluetooth has been horribly broken; I can't add new devices (says "can't put > your adaptor into Discoverable mode"), and my MS ArcTouch mouse (and who knows > what else) doesn't work anymore. I also get stalls/crashes from the BT system > when trying to shutdown or reboot. > > This is a *serious* regression- I had to revert the entirety of that merge. > I keep checking the BT changes that have made it to Linus' master of late (and > even monitoring the Bluetooth git), after nearly 6 weeks, BT is still broken- > certainly I can't be the only person seeing these issues. I did see where > someone had a regression about not being able to do scans, and that was fixed, > but trying that commit (since merged to master) doesn't fix anything for me. > > My BT adaptor is an Intel AX201 on USB: > > ---- > [Fri Mar 4 15:40:52 2022] Bluetooth: hci0: Found device firmware: intel/ibt-19-32-4.sfi > [Fri Mar 4 15:40:52 2022] Bluetooth: hci0: Boot Address: 0x24800 > [Fri Mar 4 15:40:52 2022] Bluetooth: hci0: Firmware Version: 126-5.22 > ---- > > ---- > Bus 003 Device 003: ID 8087:0026 Intel Corp. > Device Descriptor: > bLength 18 > bDescriptorType 1 > bcdUSB 2.01 > bDeviceClass 224 Wireless > bDeviceSubClass 1 Radio Frequency > bDeviceProtocol 1 Bluetooth > bMaxPacketSize0 64 > idVendor 0x8087 Intel Corp. > idProduct 0x0026 > ---- > > What can I do to help you fix this? Have you tried with bluetooth.git tree: https://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth.git/ This was the latest one that fixes the MGMT commands not working: https://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth.git/commit/?h=for-net-2022-03-03&id=008ee9eb8a11bcabf12c91771dd4f470b082bd44 It seems some setups are more susceptible to it than others and CI was one of those when only one MGMT command is ever pending which masked the problem for us for a while. -- Luiz Augusto von Dentz
