Received-SPF: pass (google.com: domain of linux-bluetooth-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
From:   Chris Murphy <lists@colorremedies.com>
Date:   Thu, 10 Feb 2022 18:44:22 -0700
Message-ID: <CAJCQCtSeUtHCgsHXLGrSTWKmyjaQDbDNpP4rb0i+RE+L2FTXSA@mail.gmail.com>
Subject: bug kernel 5.17, qualcom and intel adapters, unable to reliably
 connect to bluetooth devices
To:     Bluetooth <linux-bluetooth@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

Since kernel 5.17, some users are seeing problems connecting to
bluetooth devices, the problem doesn't happen with 5.16 series
kernels.

So far two different devices are affected, although it's possible it's
not even device related, but could be the interface between kernel and
user space.

Case 1:

Bus 001 Device 003: ID 0cf3:e301 Qualcomm Atheros Communications

Not seeing anything that stands out in dmesg, bluetoothd -d shows
various curious messages like

bluetoothd[5889]: src/service.c:change_state() 0x7f33becdebc0: device
94:65:2D:DC:F4:A7 profile avrcp-controller state changed: disconnected
-> unavailable (0)
and
bluetoothd[5889]: src/service.c:change_state() 0x7f33becd91e0: device
94:65:2D:DC:F4:A7 profile Hands-Free Voice gateway state changed:
unavailable -> disconnected (0)


Case 2:

Bus 001 Device 005: ID 8087:0aaa Intel Corp. Bluetooth 9460/9560
Jefferson Peak (JfP)
[    5.923865] kernel: Bluetooth: hci0: Found device firmware:
intel/ibt-17-16-1.sfi
[    5.930536] kernel: Bluetooth: hci0: Boot Address: 0x40800
[    5.930539] kernel: Bluetooth: hci0: Firmware Version: 60-44.21
...
[16102.640651] Bluetooth: hci0: unexpected event 0xff length: 5 > 0
[16102.640802] Bluetooth: hci0: Waiting for device to boot
[16102.654712] Bluetooth: hci0: Device booted in 13698 usecs
[16102.654753] Bluetooth: hci0: unexpected event 0xff length: 7 > 0
[16102.655907] Bluetooth: hci0: Found Intel DDC parameters:
intel/ibt-17-16-1.ddc
[16102.657821] Bluetooth: hci0: Applying Intel DDC parameters completed
[16102.658829] Bluetooth: hci0: Firmware revision 0.1 build 60 week 44 2021
[17422.210558] Bluetooth: hci0: command 0x0c03 tx timeout
[17430.338412] Bluetooth: hci0: HCI reset during shutdown failed

[15975.194153] bluetoothd[42895]:
src/adapter.c:device_found_callback() hci0 addr 75:0B:19:0C:47:F6,
rssi -82 flags 0x0000 eir_len 17
[15975.194412] bluetoothd[42895]:
src/adv_monitor.c:btd_adv_monitor_offload_supported() Manager is NULL,
get offload support failed
[15975.194510] bluetoothd[42895]: src/device.c:device_set_legacy() legacy 0
[15975.194599] bluetoothd[42895]:
src/device.c:device_set_rssi_with_delta() rssi -82 delta 1
[15975.194661] bluetoothd[42895]: src/device.c:device_set_flags() flags 26

There are many many of the "Manager is NULL, get offload support
failed" messages in this case. I'm not entirely sure the root causes
are the same for the two cases.

Downstream bug has attachments (dmesg, btmon, lsbsb, journalctl
snippet with bluetoothd in debug, for both)
https://bugzilla.redhat.com/show_bug.cgi?id=2053283

Thanks,

-- 
Chris Murphy