Bug 216683
| Summary: | [REGRESSION] HCI_QUIRK_NO_SUSPEND_NOTIFIER causes kernel panic on fake CSR 5.0 clones | ||
|---|---|---|---|
| Product: | Drivers | Reporter: | JustANormalTinkererMihir (justanormaltinkerermihir) |
| Component: | Bluetooth | Assignee: | linux-bluetooth (linux-bluetooth) |
| Status: | RESOLVED CODE_FIX | ||
| Severity: | high | CC: | dxor+kernel, justanormaltinkerermihir, lelgenio, luiz.dentz, ostroffjh, regressions |
| Priority: | P1 | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Kernel Version: | 6.0.7 | Subsystem: | |
| Regression: | Yes | Bisected commit-id: | |
| Attachments: | btmon log | ||
|
Description
JustANormalTinkererMihir
2022-11-12 06:31:37 UTC
I'm getting what may be the same issue not quite daily, without any unplug/replug of my cheap CSR clone BT dongles. I'm currently on 6.0.8 (Gentoo sources) with what I think is the same patchset mentioned above. (That link shows no current patches for me.) When it happens, the PC seems to respond for a while, but by the time I realize there is a problem, a power cycle is the only recovery that works for me. Nov 23 14:23:29 ffortso9 kernel: ------------[ cut here ]------------ Nov 23 14:23:29 ffortso9 kernel: notifier callback hci_suspend_notifier [bluetooth] already registered Nov 23 14:23:29 ffortso9 kernel: WARNING: CPU: 6 PID: 2174 at kernel/notifier.c:28 notifier_chain_register+0x3e/0x70 Nov 23 14:23:29 ffortso9 kernel: Modules linked in: fuse rfcomm auth_rpcgss nfsv4 dns_resolver nfs lockd grace fscache netfs sunrpc cmac algif_skcipher bnep vfat fat amdgpu mfd_core iommu_v2 gpu_sched drm_buddy radeon uvcvideo btusb videobuf2_vmalloc btrtl kvm_amd btbcm videobuf2_memops i2c_algo_bit btintel videobuf2_v4l2 kvm drm_ttm_helper snd_usb_audio videobuf2_common bluetooth ttm snd_usbmidi_lib irqbypass crct10dif_pclmul snd_hda_codec_realtek videodev drm_display_helper crc32_pclmul sr_mod snd_rawmidi rfkill snd_hda_codec_generic crc32c_intel mc ecdh_generic ledtrig_audio snd_hda_codec_hdmi snd_hda_intel drm_kms_helper cdrom ghash_clmulni_intel snd_intel_dspcfg sd_mod syscopyarea snd_hda_codec sysfillrect sysimgblt nct6775 snd_hwdep aic7xxx aesni_intel fb_sys_fops snd_hda_core nct6775_core ppdev hwmon_vid crypto_simd cryptd scsi_transport_spi snd_pcm drm sp5100_tco snd_timer xhci_pci pcspkr backlight rapl xhci_pci_renesas snd i2c_piix4 parport_pc cec ccp ahci efi_pstore soundcore xhci_hcd k10temp Nov 23 14:23:29 ffortso9 kernel: parport libahci i2c_core gpio_amdpt gpio_generic acpi_cpufreq efivarfs Nov 23 14:23:29 ffortso9 kernel: CPU: 6 PID: 2174 Comm: kworker/6:1 Not tainted 6.0.8-gentoo-x86_64-01 #1 Nov 23 14:23:29 ffortso9 kernel: Hardware name: Micro-Star International Co., Ltd. MS-7A34/B350 TOMAHAWK (MS-7A34), BIOS 1.M0 01/23/2019 Nov 23 14:23:29 ffortso9 kernel: Workqueue: usb_hub_wq hub_event Nov 23 14:23:29 ffortso9 kernel: RIP: 0010:notifier_chain_register+0x3e/0x70 Nov 23 14:23:29 ffortso9 kernel: Code: 10 7f 33 75 04 84 d2 75 3b 48 8d 78 08 48 8b 40 08 48 85 c0 74 20 48 39 c6 75 e0 48 8b 36 48 c7 c7 40 ad 34 8b e8 a9 f1 a7 00 <0f> 0b b8 ef ff ff ff e9 e6 17 d6 00 48 89 46 08 31 c0 48 89 37 e9 Nov 23 14:23:29 ffortso9 kernel: RSP: 0018:ffffb3e0d420f7d8 EFLAGS: 00010286 Nov 23 14:23:29 ffortso9 kernel: RAX: 0000000000000000 RBX: ffffffffc0a91170 RCX: 0000000000000000 Nov 23 14:23:29 ffortso9 kernel: RDX: 0000000000000001 RSI: ffffffff8b39ae13 RDI: 00000000ffffffff Nov 23 14:23:29 ffortso9 kernel: RBP: ffffffff8b64f0a0 R08: 0000000000000000 R09: ffffffff8bdc0da0 Nov 23 14:23:29 ffortso9 kernel: R10: 0000000000000001 R11: 0000000000000001 R12: ffff8c647188cb78 Nov 23 14:23:29 ffortso9 kernel: R13: ffffffff8b64f0c8 R14: ffff8c647188cd08 R15: ffffffffc0f1a7e0 Nov 23 14:23:29 ffortso9 kernel: FS: 0000000000000000(0000) GS:ffff8c6afa780000(0000) knlGS:0000000000000000 Nov 23 14:23:29 ffortso9 kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 Nov 23 14:23:29 ffortso9 kernel: CR2: 00007f0c4c003078 CR3: 00000001db098000 CR4: 00000000003506e0 Nov 23 14:23:29 ffortso9 kernel: Call Trace: Nov 23 14:23:29 ffortso9 kernel: <TASK> Nov 23 14:23:29 ffortso9 kernel: blocking_notifier_chain_register+0x3d/0x80 Nov 23 14:23:29 ffortso9 kernel: hci_register_dev+0x32d/0x3e0 [bluetooth] Nov 23 14:23:29 ffortso9 kernel: btusb_probe+0xc24/0xdea [btusb] Nov 23 14:23:29 ffortso9 kernel: ? recalibrate_cpu_khz+0x10/0x10 Nov 23 14:23:29 ffortso9 kernel: ? ktime_get_mono_fast_ns+0x3d/0x90 Nov 23 14:23:29 ffortso9 kernel: usb_probe_interface+0xf6/0x2d0 Nov 23 14:23:29 ffortso9 kernel: really_probe+0xe1/0x3a0 Nov 23 14:23:29 ffortso9 kernel: ? pm_runtime_barrier+0x61/0xb0 Nov 23 14:23:29 ffortso9 kernel: __driver_probe_device+0x78/0x180 Nov 23 14:23:29 ffortso9 kernel: driver_probe_device+0x2c/0xb0 Nov 23 14:23:29 ffortso9 kernel: __device_attach_driver+0x8c/0x100 Nov 23 14:23:29 ffortso9 kernel: ? driver_allows_async_probing+0x60/0x60 Nov 23 14:23:29 ffortso9 kernel: ? driver_allows_async_probing+0x60/0x60 Nov 23 14:23:29 ffortso9 kernel: bus_for_each_drv+0x7e/0xd0 Nov 23 14:23:29 ffortso9 kernel: __device_attach+0xca/0x230 Nov 23 14:23:29 ffortso9 kernel: bus_probe_device+0x8e/0xb0 Nov 23 14:23:29 ffortso9 kernel: device_add+0x45c/0x970 Nov 23 14:23:29 ffortso9 kernel: ? preempt_count_add+0x70/0xa0 Nov 23 14:23:29 ffortso9 kernel: usb_set_configuration+0x483/0x890 Nov 23 14:23:29 ffortso9 kernel: usb_generic_driver_probe+0x50/0x70 Nov 23 14:23:29 ffortso9 kernel: usb_probe_device+0x47/0x110 Nov 23 14:23:29 ffortso9 kernel: really_probe+0xe1/0x3a0 Nov 23 14:23:29 ffortso9 kernel: ? pm_runtime_barrier+0x61/0xb0 Nov 23 14:23:29 ffortso9 kernel: __driver_probe_device+0x78/0x180 Nov 23 14:23:29 ffortso9 kernel: driver_probe_device+0x2c/0xb0 Nov 23 14:23:29 ffortso9 kernel: __device_attach_driver+0x8c/0x100 Nov 23 14:23:29 ffortso9 kernel: ? driver_allows_async_probing+0x60/0x60 Nov 23 14:23:29 ffortso9 kernel: ? driver_allows_async_probing+0x60/0x60 Nov 23 14:23:29 ffortso9 kernel: bus_for_each_drv+0x7e/0xd0 Nov 23 14:23:29 ffortso9 kernel: __device_attach+0xca/0x230 Nov 23 14:23:29 ffortso9 kernel: bus_probe_device+0x8e/0xb0 Nov 23 14:23:29 ffortso9 kernel: device_add+0x45c/0x970 Nov 23 14:23:29 ffortso9 kernel: ? blake2s_update+0x5c/0xe0 Nov 23 14:23:29 ffortso9 kernel: usb_new_device.cold+0x148/0x36a Nov 23 14:23:29 ffortso9 kernel: hub_event+0xfa8/0x1950 Nov 23 14:23:29 ffortso9 kernel: process_one_work+0x1e5/0x3b0 Nov 23 14:23:29 ffortso9 kernel: ? rescuer_thread+0x390/0x390 Nov 23 14:23:29 ffortso9 kernel: worker_thread+0x50/0x3b0 Nov 23 14:23:29 ffortso9 kernel: ? rescuer_thread+0x390/0x390 Nov 23 14:23:29 ffortso9 kernel: kthread+0xe8/0x110 Nov 23 14:23:29 ffortso9 kernel: ? kthread_complete_and_exit+0x20/0x20 Nov 23 14:23:29 ffortso9 kernel: ret_from_fork+0x22/0x30 Nov 23 14:23:29 ffortso9 kernel: </TASK> Nov 23 14:23:29 ffortso9 kernel: ---[ end trace 0000000000000000 ]--- Nov 23 14:23:29 ffortso9 kernel: btusb: probe of 1-1:1.0 failed with error -17 Nov 23 14:23:31 ffortso9 kernel: general protection fault, probably for non-canonical address 0xff4c6734ff4f2703: 0000 [#1] PREEMPT SMP NOPTI Nov 23 14:23:31 ffortso9 kernel: CPU: 6 PID: 2174 Comm: kworker/6:1 Tainted: G W 6.0.8-gentoo-x86_64-01 #1 Nov 23 14:23:31 ffortso9 kernel: Hardware name: Micro-Star International Co., Ltd. MS-7A34/B350 TOMAHAWK (MS-7A34), BIOS 1.M0 01/23/2019 Nov 23 14:23:31 ffortso9 kernel: Workqueue: pm pm_runtime_work Nov 23 14:23:31 ffortso9 kernel: RIP: 0010:queued_spin_lock_slowpath+0x25b/0x2a0 Nov 23 14:23:31 ffortso9 kernel: Code: ff f3 90 48 8b 13 48 85 d2 74 f6 eb d6 c1 ea 12 83 e0 03 ff ca 48 c1 e0 04 48 63 d2 48 05 00 c0 02 00 48 03 04 d5 20 18 43 8b <48> 89 18 8b 43 08 85 c0 75 09 f3 90 8b 53 08 85 d2 74 f7 48 8b 13 Nov 23 14:23:31 ffortso9 kernel: RSP: 0018:ffffb3e0d420fcf8 EFLAGS: 00010086 Nov 23 14:23:31 ffortso9 kernel: RAX: ff4c6734ff4f2703 RBX: ffff8c6afa7ac000 RCX: 0000000000000007 Nov 23 14:23:31 ffortso9 kernel: RDX: 0000000000003ffe RSI: ffffffff8b39ae13 RDI: ffffffff8b368d04 Nov 23 14:23:31 ffortso9 kernel: RBP: ffff8c64692be18c R08: ffff8c6a62350800 R09: ffff8c64692be8a8 Nov 23 14:23:31 ffortso9 kernel: R10: 0000000000000003 R11: 000000000018e656 R12: 00000000001c0000 Nov 23 14:23:31 ffortso9 kernel: R13: 00000000001c0000 R14: 0000000000000002 R15: ffff8c6afa7b2105 Nov 23 14:23:31 ffortso9 kernel: FS: 0000000000000000(0000) GS:ffff8c6afa780000(0000) knlGS:0000000000000000 Nov 23 14:23:31 ffortso9 kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 Nov 23 14:23:31 ffortso9 kernel: CR2: 0000562aa8908098 CR3: 0000000103c8a000 CR4: 00000000003506e0 Nov 23 14:23:31 ffortso9 kernel: Call Trace: Nov 23 14:23:31 ffortso9 kernel: <TASK> Nov 23 14:23:31 ffortso9 kernel: btusb_suspend+0x87/0x1c0 [btusb] Nov 23 14:23:31 ffortso9 kernel: usb_suspend_both+0xaa/0x220 Nov 23 14:23:31 ffortso9 kernel: usb_runtime_suspend+0x2b/0x70 Nov 23 14:23:31 ffortso9 kernel: ? usb_autoresume_device+0x60/0x60 Nov 23 14:23:31 ffortso9 kernel: __rpm_callback+0x5b/0x140 Nov 23 14:23:31 ffortso9 kernel: ? usb_autoresume_device+0x60/0x60 Nov 23 14:23:31 ffortso9 kernel: rpm_callback+0x79/0x90 Nov 23 14:23:31 ffortso9 kernel: ? usb_autoresume_device+0x60/0x60 Nov 23 14:23:31 ffortso9 kernel: rpm_suspend+0x14a/0x730 Nov 23 14:23:31 ffortso9 kernel: ? vtime_task_switch_generic+0x8d/0xf0 Nov 23 14:23:31 ffortso9 kernel: ? _raw_spin_unlock+0x12/0x40 Nov 23 14:23:31 ffortso9 kernel: ? finish_task_switch.isra.0+0x96/0x2d0 Nov 23 14:23:31 ffortso9 kernel: pm_runtime_work+0x94/0xa0 Nov 23 14:23:31 ffortso9 kernel: process_one_work+0x1e5/0x3b0 Nov 23 14:23:31 ffortso9 kernel: ? rescuer_thread+0x390/0x390 Nov 23 14:23:31 ffortso9 kernel: worker_thread+0x50/0x3b0 Nov 23 14:23:31 ffortso9 kernel: ? rescuer_thread+0x390/0x390 Nov 23 14:23:31 ffortso9 kernel: kthread+0xe8/0x110 Nov 23 14:23:31 ffortso9 kernel: ? kthread_complete_and_exit+0x20/0x20 Nov 23 14:23:31 ffortso9 kernel: ret_from_fork+0x22/0x30 Nov 23 14:23:31 ffortso9 kernel: </TASK> Nov 23 14:23:31 ffortso9 kernel: Modules linked in: fuse rfcomm auth_rpcgss nfsv4 dns_resolver nfs lockd grace fscache netfs sunrpc cmac algif_skcipher bnep vfat fat amdgpu mfd_core iommu_v2 gpu_sched drm_buddy radeon uvcvideo btusb videobuf2_vmalloc btrtl kvm_amd btbcm videobuf2_memops i2c_algo_bit btintel videobuf2_v4l2 kvm drm_ttm_helper snd_usb_audio videobuf2_common bluetooth ttm snd_usbmidi_lib irqbypass crct10dif_pclmul snd_hda_codec_realtek videodev drm_display_helper crc32_pclmul sr_mod snd_rawmidi rfkill snd_hda_codec_generic crc32c_intel mc ecdh_generic ledtrig_audio snd_hda_codec_hdmi snd_hda_intel drm_kms_helper cdrom ghash_clmulni_intel snd_intel_dspcfg sd_mod syscopyarea snd_hda_codec sysfillrect sysimgblt nct6775 snd_hwdep aic7xxx aesni_intel fb_sys_fops snd_hda_core nct6775_core ppdev hwmon_vid crypto_simd cryptd scsi_transport_spi snd_pcm drm sp5100_tco snd_timer xhci_pci pcspkr backlight rapl xhci_pci_renesas snd i2c_piix4 parport_pc cec ccp ahci efi_pstore soundcore xhci_hcd k10temp Nov 23 14:23:31 ffortso9 kernel: parport libahci i2c_core gpio_amdpt gpio_generic acpi_cpufreq efivarfs Nov 23 14:23:31 ffortso9 kernel: ---[ end trace 0000000000000000 ]--- Nov 23 14:23:31 ffortso9 kernel: RIP: 0010:queued_spin_lock_slowpath+0x25b/0x2a0 Nov 23 14:23:31 ffortso9 kernel: Code: ff f3 90 48 8b 13 48 85 d2 74 f6 eb d6 c1 ea 12 83 e0 03 ff ca 48 c1 e0 04 48 63 d2 48 05 00 c0 02 00 48 03 04 d5 20 18 43 8b <48> 89 18 8b 43 08 85 c0 75 09 f3 90 8b 53 08 85 d2 74 f7 48 8b 13 Nov 23 14:23:31 ffortso9 kernel: RSP: 0018:ffffb3e0d420fcf8 EFLAGS: 00010086 Nov 23 14:23:31 ffortso9 kernel: RAX: ff4c6734ff4f2703 RBX: ffff8c6afa7ac000 RCX: 0000000000000007 Nov 23 14:23:31 ffortso9 kernel: RDX: 0000000000003ffe RSI: ffffffff8b39ae13 RDI: ffffffff8b368d04 Nov 23 14:23:31 ffortso9 kernel: RBP: ffff8c64692be18c R08: ffff8c6a62350800 R09: ffff8c64692be8a8 Nov 23 14:23:31 ffortso9 kernel: R10: 0000000000000003 R11: 000000000018e656 R12: 00000000001c0000 Nov 23 14:23:31 ffortso9 kernel: R13: 00000000001c0000 R14: 0000000000000002 R15: ffff8c6afa7b2105 Nov 23 14:23:31 ffortso9 kernel: FS: 0000000000000000(0000) GS:ffff8c6afa780000(0000) knlGS:0000000000000000 Nov 23 14:23:31 ffortso9 kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 Nov 23 14:23:31 ffortso9 kernel: CR2: 0000562aa8908098 CR3: 0000000103c8a000 CR4: 00000000003506e0 Nov 23 14:23:31 ffortso9 kernel: note: kworker/6:1[2174] exited with preempt_count 1 It seems for me the USB device reconnects every so often on my computer I have tracked the issue down to the HCI_QUIRK_NO_SUSPEND_NOTIFIER, can you remove the line set_bit(HCI_QUIRK_NO_SUSPEND_NOTIFIER); from btusb.c and recompile the module to test if it doesn't occur to you anymore I'll resubmit this regression as it looks like it didn't go to the https://lore.kernel.org/regressions/ Marking the resubmition as a duplicate, it doesn't look like bugs submitted on bugzilla go to the regressions mailing list, sorry I don't really know what I am doing this is confusing as hell. And adding regressions@lists.linux.dev to CC just doesn't work *** Bug 216742 has been marked as a duplicate of this bug. *** Per Comment#4 commenting out that one line does seem to prevent the crash. In 6.0.8, with the patch from Comment #0 applied, and also in 6.0.9 without that patch applied. In 6.0.9 (Gentoo sources) the dongle is never recognized, but the crash happens anyway. (In reply to JustANormalTinkererMihir from comment #4) > I don't really know what I am doing this is confusing as hell. Well, ideally you would have followed these guides: https://docs.kernel.org/admin-guide/reporting-issues.html https://docs.kernel.org/admin-guide/reporting-regressions.html But whatever, it's not important. Thing is: when you patch your kernel (or Gentoo might have -- don't known) you are own your own. And that seems to be the case here. But TBH: the hole situation wrt to the problems recently discussed in Bug 60824 are confusing for outsiders like me (and developers likely as well). But I'm willing to help there. Could you maybe please provide me with an update? All adapters that used to work really should work again without any external patches -- and ideally without any new module parameters or similar tricks, unless that can't be avoided (for example due to security concerns or because it would cause regressions for others). I have reconfirmed with a vanilla 6.0.9 kernel. I got the crash in less than an hour, without even unplug/replug. Comment out that one line in btusb.c and no crash with multiple unplug/replug of the dongle. When I get more time, I'll try applying the patch in Comment #0 to confirm it allows the dongle to work. (In reply to The Linux kernel's regression tracker (Thorsten Leemhuis) from comment #7) What sort of update do you require? As far as I can tell HCI_QUIRK_NO_SUSPEND_NOTIFIER causes this regression, and the dongle continues to work fine after just removing the line. One thing I have noticed in the dmesg log is that the dongle sometimes disconnects and reconnects, this is largely the reason why the kernel fault even occurs, it disconnects and reconnects like this: [ 53.494546] Bluetooth: RFCOMM TTY layer initialized [ 53.494558] Bluetooth: RFCOMM socket layer initialized [ 53.494561] Bluetooth: RFCOMM ver 1.11 [ 60.757711] rfkill: input handler disabled [ 73.717144] rfkill: input handler enabled [ 80.462376] rfkill: input handler disabled [ 80.468335] EXT4-fs (sdc1): mounted filesystem with ordered data mode. Quota mode: none. [ 93.275986] input: Airdopes 121v2 (AVRCP) as /devices/virtual/input/input23 [ 136.650027] usb 2-1.5: USB disconnect, device number 3 [ 136.892439] usb 2-1.5: new full-speed USB device number 4 using ehci-pci [ 136.998154] usb 2-1.5: New USB device found, idVendor=0a12, idProduct=0001, bcdDevice=88.91 [ 136.998160] usb 2-1.5: New USB device strings: Mfr=0, Product=2, SerialNumber=0 [ 136.998162] usb 2-1.5: Product: BT DONGLE10 [ 137.000158] Bluetooth: hci0: CSR: Setting up dongle with HCI ver=9 rev=0810; LMP ver=9 subver=2312; manufacturer=10 [ 137.000163] Bluetooth: hci0: CSR: Unbranded CSR clone detected; adding workarounds... [ 137.000164] Bluetooth: hci0: CSR: Unbranded CSR clone detected; force-suspending once... [ 137.000166] Bluetooth: hci0: CSR: Couldn't suspend the device for our Barrot 8041a02 receive-issue workaround [ 137.000188] Bluetooth: hci0: HCI Delete Stored Link Key command is advertised, but not supported. [ 137.000189] Bluetooth: hci0: HCI Read Default Erroneous Data Reporting command is advertised, but not supported. [ 137.000190] Bluetooth: hci0: HCI Set Event Filter command not supported. [ 137.082247] Bluetooth: MGMT ver 1.22 [ 143.561978] usb 2-1.5: USB disconnect, device number 4 This seems to only happen when idle not when there is any activity. Maybe the dongle has some built-in power saving feature that disables the dongle when not in use? Please note this also happens after disabling the HCI_QUIRK_NO_SUSPEND_NOTIFIER. These dongles seem really messed up. Also please note btusb.enable_autosuspend=0 is a kernel parameter I have set and the issue still occurs. It also occurs without it being set. Created attachment 303310 [details]
btmon log
I am now able to reproduce this issue on kernel 5.19. Same kernel fault, same thing happened again, the dongle disconnected and reconnected causing the kernel fault. (In reply to JustANormalTinkererMihir from comment #12) > I am now able to reproduce this issue on kernel 5.19. Same kernel fault, > same thing happened again, the dongle disconnected and reconnected causing > the kernel fault. I don't think this is related to HCI_QUIRK_NO_SUSPEND_NOTIFIER as it was only introduced by the following change: https://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth-next.git/commit/drivers/bluetooth/btusb.c?id=b3cf94c8b6b2f1a2b94825a025db291da2b151fd Or did it reach 5.19 via stable? @JustANormalTinkererMihir Could you try something like bellow: https://gist.github.com/Vudentz/0e0cb034a434e94daaf219f26f921aa2 (In reply to Luiz Von Dentz from comment #14) > @JustANormalTinkererMihir > > Could you try something like bellow: > > https://gist.github.com/Vudentz/0e0cb034a434e94daaf219f26f921aa2 This seems to work, I disconnected and reconnect my device several times and it seems to work now, no kernel fault can be observed as of now, I'll report back here if it happens even with this patch. By the way the linux kernel 5.19 that I was talking about was a ubuntu kernel so they might have backported the patch. Please note: I have added the set_bit(HCI_QUIRK_NO_SUSPEND_NOTIFIER) line back and it works fine with that now. https://patchwork.kernel.org/project/bluetooth/patch/20221129205413.1537851-1-luiz.dentz@gmail.com/ Please respond with Tested-by if you are not able to reproduce it anymore. If you mean does your patch (without commenting out the set_bit line) result in no more crash on unplug/resource, then yes Tested-by ostroffjh@users.sourceforge.net But I still need the earlier patch (set of three) for the dongle to function. tested with 6.0.9, both vanilla and Gentoo-sources. I also need all four patches to make it work. However, after a sleep I get the following in my dmesg output: ``` [32332.754468] Bluetooth: hci0: command 0x0405 tx timeout ``` After that I get the following repeating: ``` [36431.813695] Bluetooth: hci0: command 0x0c24 tx timeout [36431.813705] Bluetooth: hci0: Opcode 0x c24 failed: -110 [36533.323814] Bluetooth: hci0: command 0x0408 tx timeout ``` (In reply to Jack from comment #17) > If you mean does your patch (without commenting out the set_bit line) result > in no more crash on unplug/resource, then yes > > Tested-by ostroffjh@users.sourceforge.net > > But I still need the earlier patch (set of three) for the dongle to > function. tested with 6.0.9, both vanilla and Gentoo-sources. What patches are you talking about, are they already in bluetooth-next? https://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth-next.git/ (In reply to Daniel from comment #18) > I also need all four patches to make it work. However, after a sleep I get > the following in my dmesg output: > > ``` > [32332.754468] Bluetooth: hci0: command 0x0405 tx timeout > ``` > > After that I get the following repeating: > > ``` > [36431.813695] Bluetooth: hci0: command 0x0c24 tx timeout > [36431.813705] Bluetooth: hci0: Opcode 0x c24 failed: -110 > [36533.323814] Bluetooth: hci0: command 0x0408 tx timeout > > ``` Your controller is probably in a bad state, perhaps something like hciconfig reset hci0, anyway this is probably no related to this bug. (In reply to Luiz Von Dentz from comment #19) > What patches are you talking about, are they already in bluetooth-next? Apologies if I wasn't clear, but this is getting too confusing. For this specific bug, there is really only one relevant patch. I am confirming no more crash using your patch from Comment #16. The patch your mention in Comment #13 is definitely in 5.17.16, but I don't know which version it actually landed in. Commenting out one of the set_bit lines from that patch also avoids the crash, but your patch works without removing that line. The fact that my dongle is only recognized if I use the set of patches to revert the regression regarding the cheap CSR clones (I can't easily find the other bug number) is probably irrelevant to this bug, and I mentioned it only for completeness. (In reply to Luiz Von Dentz from comment #16) > https://patchwork.kernel.org/project/bluetooth/patch/20221129205413.1537851- > 1-luiz.dentz@gmail.com/ > > Please respond with Tested-by if you are not able to reproduce it anymore. Works wonderfully! Tried re-plugging multiple times and also left my system running for 3 days using this patch, no more kernel panics! Tested-by lelgenio@disroot.org |