[Bug 1163486] New: Pulseaudo Bluetooth profile setting broken in plasma 5
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486 Bug ID: 1163486 Summary: Pulseaudo Bluetooth profile setting broken in plasma 5 Classification: openSUSE Product: openSUSE Distribution Version: Leap 15.1 Hardware: x86-64 OS: Other Status: NEW Severity: Normal Priority: P5 - None Component: Sound Assignee: tiwai@suse.com Reporter: stakanov@freenet.de QA Contact: qa-bugs@suse.de Found By: --- Blocker: --- This is true since the first version of plasma 5 (about 5 years now). Use any BT adapter and Leap 15.1. KDE Plasma session. Connect any BT headset that supports both(!) profiles HSF/HFP and "sink A2DP". What does work correctly: The headset is paired in automatic if given trust. The source is changed to the new available one if set accordingly, BUT a) the source is not selected at appearance if the application is already open. E.g. in VLC audio sources the BT set is not electable. If closed down and restarted it appears. Some of the VOIP apps do not see it at all and if, do not change the source (e.g. Linphone). b) The profile selection of pulseaudio is not working. It seams broken insofar that when you use a device that is capable of HSF/HFP this profile is ALWAYS selected, although the settings menu senses and offers the A2DP profile. Once set A2DP (and no matter were, in the tray or in system settings) the profile jumps back to HSF/HFP headset function. The audio capacity of this protocol is to say the minor limited, thus BT headsets and pulseaudio in combination are not usable in 15.1. I can exclude a BT hardware problem as e.g. the very headset of SE-CL761 BT do connect flawlessly with a Nokia 8110 4G and full sound capacities. The problem was verified with several BT adapters. For this report I do reference to: Bus 004 Device 003: ID 0a12:0001 Cambridge Silicon Radio, Ltd Bluetooth Dongle (HCI mode) However trying other brands and chipsets, other BTversions (2, 3, 4) do not add anything to the solution. This problem presents alike with JBC BT boxes that bring both profiles. Quality is actually voiding the function of external boxes for reproduction of music. What the software does: profile selection in pulseaudio are not honored in BT devices when selected and the device is capable of both protocols. BT device is electable only after restart of the respective program. What the software should do: leave the choice to the user if the usage is music (A2DP) or headset (HSF/HFP) when using a BT device. Honor the setting of the user. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c1
Takashi Iwai
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c2
Takashi Iwai
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c3
--- Comment #3 from Stakanov Schufter
Is it a problem of pulseaudio, bluez or KDE-specific issue?
I vaguely remember of the HFS/A2DP profile issue. Adding Seife to Cc, who should know of the mess better.
FYI Well, without giving responsibilities to anybody. When we (several users) filed during the years (yes, this one is a historic one, AFAIR, since 42.x) we were told finally, it is a pulseaudio based problem, as the BT function itself now works and pairing IS possible. Also the selection of new hardwaresource when it becomes available works ...limited to the plasma desktop, it seems not honored by applications. And in the settings the issue concerns settings in "multimedia". AFAIU the settings in the tray are only a link to the settings in the "system settings". The profiles are actually sensed correctly by BT, hardware capacities are shown correctly. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c8
--- Comment #8 from Stakanov Schufter
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c9
--- Comment #9 from Stakanov Schufter
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c11
--- Comment #11 from Stakanov Schufter
(In reply to Stakanov Schufter from comment #9)
Update: after the installation I am getting also a complaint in the dmesg. As it might be related, I think it will be of interest (so this is with the fixes I installed as of indication): sudo dmesg -r (excerpt)
<4>[ 1338.404968] ------------[ cut here ]------------ <4>[ 1338.404980] WARNING: CPU: 2 PID: 934 at ../kernel/workqueue.c:2458 check_flush_dependency+0xad/0x120
This looks like a long-standing bug in btusb driver.
Can you trigger the bug at your will, or is it just intermittent?
For what I know (...and I know little, as I catch these things when I check and, of course, I have to admit I do not check always) it is more frequently presenting when suspending to disc. However it seems also randomly there so I wouldn't wager my boots on it, if you know what I mean. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c12
--- Comment #12 from Takashi Iwai
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c13
--- Comment #13 from Stakanov Schufter
OK, if you can trigger the problem at your will, it'd be great, we can have a better chance to fix it. Other than that, you can ignore it; this doesn't look too serious but rather a warning.
Did you also update bluez-related packages from that repo, too? The connection and device enumeration issue might be rather a bluez problem.
Also please check the setup rather with pavucontrol program (the package is the same name). In Configuration tab, you should be able to choose the BT device profile.
FWIW, I created a repo containing a downstream version of pulseaudio BT module that supports more codecs like aptX, and also applied the downstream patches for HFP support. They are in OBS home:tiwai:bluetooth repo. If you're interested, give it a try. https://build.opensuse.org/project/show/home:tiwai:bluetooth
I did update also the BT stack, but I will now change the repo to the one you indicated and try. I will also see that I have pavucontrol installed and use it. I will report then back here about the results. I will then try if I can trigger the bug. A word to all people here. I would like to take the occasion to thank everybody that continues to work on software bugs while real "bugs" make our life quite difficult currently. This is noted, noteworthy and I thank you all sincerely for this. Stay healthy. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c14
--- Comment #14 from Stakanov Schufter
From sudo dmesg:
[ 17.362046] lpc_ich: Resource conflict(s) found affecting gpio_ich [ 17.362298] i801_smbus 0000:00:1f.3: SMBus using PCI interrupt [ 17.365807] snd_hda_intel 0000:00:1b.0: bound 0000:00:02.0 (ops i915_audio_component_bind_ops [i915]) [ 17.371776] kvm: disable TXT in the BIOS or activate TXT before enabling KVM [ 17.371779] kvm: disabled by bios [ 17.422365] kvm: disable TXT in the BIOS or activate TXT before enabling KVM [ 17.422367] kvm: disabled by bios [ 17.425011] snd_hda_codec_conexant hdaudioC0D0: CX20585: BIOS auto-probing. [ 17.425533] snd_hda_codec_conexant hdaudioC0D0: autoconfig for CX20585: line_outs=1 (0x1f/0x0/0x0/0x0/0x0) type:speaker [ 17.425536] snd_hda_codec_conexant hdaudioC0D0: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0) [ 17.425538] snd_hda_codec_conexant hdaudioC0D0: hp_outs=2 (0x1c/0x19/0x0/0x0/0x0) [ 17.425539] snd_hda_codec_conexant hdaudioC0D0: mono: mono_out=0x0 [ 17.425540] snd_hda_codec_conexant hdaudioC0D0: inputs: [ 17.425542] snd_hda_codec_conexant hdaudioC0D0: Internal Mic=0x23 [ 17.425544] snd_hda_codec_conexant hdaudioC0D0: Mic=0x1b [ 17.425546] snd_hda_codec_conexant hdaudioC0D0: Dock Mic=0x1a [ 17.442989] input: HDA Intel MID Mic as /devices/pci0000:00/0000:00:1b.0/sound/card0/input10 [ 17.443073] input: HDA Intel MID Dock Mic as /devices/pci0000:00/0000:00:1b.0/sound/card0/input11 [ 17.443147] input: HDA Intel MID Dock Headphone as /devices/pci0000:00/0000:00:1b.0/sound/card0/input12 [ 17.443223] input: HDA Intel MID Headphone as /devices/pci0000:00/0000:00:1b.0/sound/card0/input13 [ 17.443294] input: HDA Intel MID HDMI/DP,pcm=3 as /devices/pci0000:00/0000:00:1b.0/sound/card0/input14 [ 17.446315] iTCO_vendor_support: vendor-support=0 [ 17.447066] iTCO_wdt: Intel TCO WatchDog Timer Driver v1.11 [ 17.447131] iTCO_wdt: Found a QM57 TCO device (Version=2, TCOBASE=0x1060) [ 17.447218] iTCO_wdt: initialized. heartbeat=30 sec (nowayout=0) [ 17.448203] e1000e 0000:00:19.0 eth1: renamed from eth0 [ 17.635148] Adding 7950332k swap on /dev/mapper/system-swap. Priority:-1 extents:1 across:7950332k SSFS [ 17.740356] EXT4-fs (dm-3): mounted filesystem with ordered data mode. Opts: acl,user_xattr,data=ordered [ 18.262829] Bluetooth: Core ver 2.22 [ 18.262856] NET: Registered protocol family 31 [ 18.262858] Bluetooth: HCI device and connection manager initialized [ 18.262863] Bluetooth: HCI socket layer initialized [ 18.262867] Bluetooth: L2CAP socket layer initialized [ 18.262875] Bluetooth: SCO socket layer initialized [ 19.088827] usbcore: registered new interface driver btusb [ 19.102888] ip_tables: (C) 2000-2006 Netfilter Core Team [ 19.123628] ip6_tables: (C) 2000-2006 Netfilter Core Team [ 19.157670] Ebtables v2.0 registered [ 19.197336] Bluetooth: BNEP (Ethernet Emulation) ver 1.3 [ 19.197338] Bluetooth: BNEP filters: protocol multicast [ 19.197343] Bluetooth: BNEP socket layer initialized [ 19.221623] nf_conntrack version 0.5.0 (65536 buckets, 262144 max) [ 19.224410] rtl8192cu: Chip version 0x10 [ 19.315112] Loading iSCSI transport class v2.0-870. [ 19.361924] rtl8192cu: Board Type 0 [ 19.362164] rtl_usb: rx_max_size 15360, rx_urb_num 8, in_ep 1 [ 19.362236] rtl8192cu: Loading firmware rtlwifi/rtl8192cufw_TMSC.bin [ 19.369772] bridge: filtering via arp/ip/ip6tables is no longer available by default. Update your scripts to load br_netfilter if you need this. [ 19.371979] Bluetooth: RFCOMM TTY layer initialized [ 19.371985] Bluetooth: RFCOMM socket layer initialized [ 19.371992] Bluetooth: RFCOMM ver 1.11 [ 19.375032] ieee80211 phy0: Selected rate control algorithm 'rtl_rc' [ 19.375396] usbcore: registered new interface driver rtl8192cu [ 19.378715] Bridge firewalling registered [ 19.382444] usbcore: registered new interface driver rtl8xxxu [ 19.492772] Netfilter messages via NETLINK v0.30. [ 19.507675] ip_set: protocol 6 [ 19.963433] IPv6: ADDRCONF(NETDEV_UP): eth1: link is not ready [ 20.188275] IPv6: ADDRCONF(NETDEV_UP): eth1: link is not ready [ 20.225722] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready [ 20.247914] rtl8192cu: MAC auto ON okay! [ 20.288794] rtl8192cu: Tx queue select: 0x05 [ 20.896269] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready [ 20.926922] rtl8192cu: MAC auto ON okay! [ 20.960167] rtl8192cu: Tx queue select: 0x05 [ 21.611567] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready [ 21.737498] NET: Registered protocol family 17 [ 21.779894] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready [ 38.181815] rtl8192cu: MAC auto ON okay! [ 38.217183] rtl8192cu: Tx queue select: 0x05 [ 38.829766] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready [ 69.003228] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready [ 69.045574] rtl8192cu: MAC auto ON okay! [ 69.078311] rtl8192cu: Tx queue select: 0x05 [ 69.715295] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready [ 70.755082] rtl8192cu: MAC auto ON okay! [ 70.788853] rtl8192cu: Tx queue select: 0x05 [ 71.429764] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready [ 77.566809] wlan0: authenticate with 00:0b:3b:d1:30:4f [ 77.578512] wlan0: send auth to 00:0b:3b:d1:30:4f (try 1/3) [ 77.583100] wlan0: authenticated [ 77.583584] wlan0: associate with 00:0b:3b:d1:30:4f (try 1/3) [ 77.593860] wlan0: RX AssocResp from 00:0b:3b:d1:30:4f (capab=0x431 status=0 aid=1) [ 77.596104] wlan0: associated [ 77.596159] IPv6: ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready [ 169.263692] fuse init (API version 7.26) [ 342.237633] input: SE-CL761BT (AVRCP) as /devices/virtual/input/input15 [ 429.694891] input: SE-CL761BT (AVRCP) as /devices/virtual/input/input16 [ 666.986664] Bluetooth: hci0: SCO packet for unknown connection handle 65 [ 666.986669] Bluetooth: hci0: SCO packet for unknown connection handle 65 [ 874.628375] input: SE-CL761BT (AVRCP) as /devices/virtual/input/input17 [ 973.794244] Bluetooth: hci0: SCO packet for unknown connection handle 63 [ 973.794248] Bluetooth: hci0: SCO packet for unknown connection handle 63 And from sudo journalctl -r apr 03 20:18:24 roadrunner.suse bluetoothd[1525]: Unable to get io data for Hands-Free Voice gateway: getpeernam> apr 03 20:18:24 roadrunner.suse rtkit-daemon[2164]: Supervising 0 threads of 0 processes of 1 users. apr 03 20:18:24 roadrunner.suse rtkit-daemon[2164]: Supervising 0 threads of 0 processes of 1 users. apr 03 20:18:24 roadrunner.suse rtkit-daemon[2164]: Supervising 0 threads of 0 processes of 1 users. apr 03 20:18:24 roadrunner.suse rtkit-daemon[2164]: Supervising 0 threads of 0 processes of 1 users. apr 03 20:18:24 roadrunner.suse rtkit-daemon[2164]: Supervising 0 threads of 0 processes of 0 users. apr 03 20:18:24 roadrunner.suse kernel: Bluetooth: hci0: SCO packet for unknown connection handle 63 apr 03 20:18:24 roadrunner.suse kernel: Bluetooth: hci0: SCO packet for unknown connection handle 63 -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c15
--- Comment #15 from Stakanov Schufter
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c17
--- Comment #17 from Stakanov Schufter
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c18
--- Comment #18 from Stakanov Schufter
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c19
--- Comment #19 from Takashi Iwai
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c20
--- Comment #20 from Stakanov Schufter
Hm, weird. Are you using Secure Boot? The kernel in that repo won't work with Secure Boot because of the lack of the proper signing. If you must use Secure Boot, you may try the official Leap 15.2 kernel instead, found in http://download.opensuse.org/distribution/leap/15.2/repo/oss/
If it's not about Secure Boot, it must be another bug and worth to track down.
Yes, this is it, I am using secure boot. Well, whether I "must". I will try to deactivate secure boot and boot the kernel. If it will not work then the official one, but let us try first. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c21
--- Comment #21 from Stakanov Schufter
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c22
--- Comment #22 from Takashi Iwai
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c23
--- Comment #23 from Stakanov Schufter
For the completeness, please try with the official Leap 15.2 kernel in comment 19. If it still doesn't work, boot with nomodeset option for disabling the native graphics. It's often a graphics driver hang that looks like a boot problem.
Kernel from your repo: non bootable Kernel from your repo nomodeset: non bootable Kernel from official 15.2: non bootable Kernel from official 15.2 nomodeset: non bootable Kernel 4.12 original bootable in all settings (secure boot, modeset, nomodeset) None of the 5.x kernels boot currently on my X201, no matter what setting. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c24
--- Comment #24 from Stakanov Schufter
(In reply to Takashi Iwai from comment #22)
For the completeness, please try with the official Leap 15.2 kernel in comment 19. If it still doesn't work, boot with nomodeset option for disabling the native graphics. It's often a graphics driver hang that looks like a boot problem.
Kernel from your repo: non bootable Kernel from your repo nomodeset: non bootable Kernel from official 15.2: non bootable Kernel from official 15.2 nomodeset: non bootable Kernel 4.12 original bootable in all settings (secure boot, modeset, nomodeset)
None of the 5.x kernels boot currently on my X201, no matter what setting.
And: all tests on the new kernel read: with secureboot deactivated. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c25
--- Comment #25 from Stakanov Schufter
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c26
--- Comment #26 from Stakanov Schufter
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c27
--- Comment #27 from Stakanov Schufter
as this exceed the current pulseaudio issue, I opened a separate bug as Bug 1168658 -
Which triggered the comment of another user / dev and so we know now that I am bit by bug 1166664 and that, with the corresponding kernel I can boot now. uname -a Linux roadrunner.suse 5.3.18-lp152.1.g477312f-default #1 SMP Fri Mar 20 19:58:58 UTC 2020 (477312f) x86_64 x86_64 x86_64 GNU/Linux I will now start again as requested with comment 16 and then report back here. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c28
--- Comment #28 from Stakanov Schufter
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c29
--- Comment #29 from Takashi Iwai
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c30
--- Comment #30 from Stakanov Schufter
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c31
--- Comment #31 from Stakanov Schufter
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c36
Mark Wenzel
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c37
--- Comment #37 from Stakanov Schufter
I am having a similar problem with Jabra Motion Office Bluetooth headset. Device can be paired and connected but no Sound in the headset. Also no sound from the microphone. Tested under KDE Plasma Desktop. All Packages are up to date. Installation is standard Leap 15.1 with pulseaudio from Main Repository.
Computer: Dell Precision 7530 Headset: Jabra Motion https://www.jabra.com.de/supportpages/jabra-motion
I tried the supplied Patches: With the bluez Patch: No bluetooth connection is possible anymore. Bluetooth cannot be enabled via tray app. When trying to connect via bluetoothctl to the previous paired device, no connection is established
With pulseaudio patch: No sound at all - even with amarok over the internal speaker.
Did you try the patch and bluez of that repo with the KOD from Leap 15.2? Change the value of retained kernel first though. For me they worked with the NEW kernel, however only HFS and you still get some error message in the logs, but at least the microphone works with me in these settings. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486
http://bugzilla.opensuse.org/show_bug.cgi?id=1163486#c38
Stakanov Schufter
participants (2)
-
bugzilla_noreply@novell.com
-
bugzilla_noreply@suse.com