Multiple devices on USB hub -> sporadic freeze

Soukyuu · 2016-04-02 09:05:47

I'm using a USB 3.0 hub that connects via the internal connector, providing 4 USB 3.0 ports to my front panel.
Sometimes when connecting multiple devices to the hub, everything freezes.

I tried disconnecting all USB components, and it seems the hub is the key point. It looks like the USB controller(s) freeze completely, re-connecting the keyboard for example doesn't make the lights on it go on again, so it seems completely dead. REISUB etc can't be executed and there is nothing in dmesg/journalctl, either because I have to force restart the PC, or because the system doesn't notice anything. So far, it hasn't happened on windows - yet - but I'm not running that often.

Is there any way to debug this, or should I just replace the USB hub and hope it was just a hardware defect?

swalladge · 2016-04-02 11:05:10

Are you sure there is no output in the journal?
Note that you can check the journal output from the previous boot with `journalctl -b -1` if your system freezes and you have to force restart.

Soukyuu · 2016-04-02 14:53:14

Well, now that I am attempting to reproduce the issue, it's of course not occurring...

But yes, I'm quite sure, because I looked at the logs and couldn't find anything suspicious - they just stop midway.
I've connected an old PS/2 keyboard, so in case the USB freeze happens again, I should be able to still control the system, unless the freeze is complete.

R00KIE · 2016-04-03 11:54:05

If you have a second machine it might help, when things hang you can try to ping it and see if it's still alive and try to ssh in and check the logs/dmesg/journal.

Soukyuu · 2016-04-05 17:14:58

It doesn't seem like it's caused by multiple devices, only one has just sent me to oblivion. The connected PS/2 keyboard had a flashing NumLock LED, which indicates a kernel panic, afaik.

And, interestingly, this time, the logs have a kernel BUG message!

Apr 05 19:03:22 Takemikazuchi kernel: ------------[ cut here ]------------
Apr 05 19:03:22 Takemikazuchi kernel: kernel BUG at mm/slub.c:3517!
Apr 05 19:03:22 Takemikazuchi kernel: invalid opcode: 0000 [#1] PREEMPT SMP 
Apr 05 19:03:22 Takemikazuchi kernel: Modules linked in: cdc_acm uas usb_storage sha256_generic hmac drbg ansi_cprng dm_crypt xts gf128mul algif_skcipher af_alg dm_mod des_generic ecb md4 nls_utf8 cifs dns_resolver fscache it87 adt7475 hwmon_vid ext4 nvidia(PO) crc16 mbcache jbd2 joydev mousedev kvm_amd fuse kvm snd_hda_codec_realtek evdev mxm_wmi input_leds snd_hda_codec_generic led_class mac_hid serio_raw edac_mce_amd pcspkr edac_core snd_hda_intel k10temp snd_hda_codec snd_hda_core snd_hwdep r8169 snd_pcm snd_timer mii snd soundcore sp5100_tco i2c_piix4 drm tpm_infineon acpi_cpufreq tpm_tis tpm fjes shpchp wmi button processor sch_fq_codel vboxnetflt(O) vboxnetadp(O) pci_stub vboxpci(O) vboxdrv(O) overlay ip_tables x_tables crc32c_generic btrfs xor hid_generic usbhid hid raid6_pq ata_generic pata_acpi sr_mod cdrom sd_mod
Apr 05 19:03:22 Takemikazuchi kernel:  atkbd libps2 ohci_pci ohci_hcd ehci_pci xhci_pci ahci pata_jmicron libahci xhci_hcd ehci_hcd libata usbcore scsi_mod usb_common i8042 serio
Apr 05 19:03:22 Takemikazuchi kernel: CPU: 2 PID: 531 Comm: Xorg Tainted: P           O    4.3.6-1-ck #1
Apr 05 19:03:22 Takemikazuchi kernel: Hardware name: Gigabyte Technology Co., Ltd. GA-990FXA-UD3/GA-990FXA-UD3, BIOS FE 10/26/2012
Apr 05 19:03:22 Takemikazuchi kernel: task: ffff88031def9f80 ti: ffff88031dfa0000 task.ti: ffff88031dfa0000
Apr 05 19:03:22 Takemikazuchi kernel: RIP: 0010:[<ffffffff811b214c>]  [<ffffffff811b214c>] kfree+0x14c/0x160
Apr 05 19:03:22 Takemikazuchi kernel: RSP: 0018:ffff88031dfa3c10  EFLAGS: 00010246
Apr 05 19:03:22 Takemikazuchi kernel: RAX: 0200000000000000 RBX: ffff8802b5a43000 RCX: 0000000000000000
Apr 05 19:03:22 Takemikazuchi kernel: RDX: ffff880335a43000 RSI: ffff8802b5a4301c RDI: ffff8802b5a43000
Apr 05 19:03:22 Takemikazuchi kernel: RBP: ffff88031dfa3c28 R08: ffff8802b5a43014 R09: ffffea000ad690c0
Apr 05 19:03:22 Takemikazuchi kernel: R10: ffff88031a5b0008 R11: ffffffffa0a72300 R12: 0000000000000000
Apr 05 19:03:22 Takemikazuchi kernel: R13: ffffffffa0cff7db R14: 00007fff109d3f80 R15: ffff88031afc2e80
Apr 05 19:03:22 Takemikazuchi kernel: FS:  00007fca3c767940(0000) GS:ffff88032fc80000(0000) knlGS:0000000000000000
Apr 05 19:03:22 Takemikazuchi kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Apr 05 19:03:22 Takemikazuchi kernel: CR2: 00003ae1bf0a8008 CR3: 00000000ce15b000 CR4: 00000000000006e0
Apr 05 19:03:22 Takemikazuchi kernel: Stack:
Apr 05 19:03:22 Takemikazuchi kernel:  ffff88031afc2e80 0000000000000000 ffff88031afc2ea0 ffff88031dfa3c38
Apr 05 19:03:22 Takemikazuchi kernel:  ffffffffa0cff7db ffff88031afc2e70 ffffffffa0c84ea6 0000000000000000
Apr 05 19:03:22 Takemikazuchi kernel:  0000000000000000 ffff88031afc2ed8 ffffffffa0904d48 ffff88031afc2ed8
Apr 05 19:03:22 Takemikazuchi kernel: Call Trace:
Apr 05 19:03:22 Takemikazuchi kernel:  [<ffffffffa0cff7db>] os_free_mem+0x1b/0x40 [nvidia]
Apr 05 19:03:22 Takemikazuchi kernel:  [<ffffffffa0c84ea6>] _nv001373rm+0x36/0x60 [nvidia]
Apr 05 19:03:22 Takemikazuchi kernel:  [<ffffffffa0904d48>] ? _nv004331rm+0xc8/0xf0 [nvidia]
Apr 05 19:03:22 Takemikazuchi kernel:  [<ffffffffa08ee3ea>] ? _nv004326rm+0xca/0x650 [nvidia]
Apr 05 19:03:22 Takemikazuchi kernel:  [<ffffffffa0ca0146>] ? _nv015126rm+0x576/0x5c0 [nvidia]
Apr 05 19:03:22 Takemikazuchi kernel:  [<ffffffffa0c8656e>] ? _nv000694rm+0x2e/0x60 [nvidia]
Apr 05 19:03:22 Takemikazuchi kernel:  [<ffffffffa0cddcb7>] ? _nv000789rm+0x5e7/0x8a0 [nvidia]
Apr 05 19:03:22 Takemikazuchi kernel:  [<ffffffffa0ce7ee3>] ? rm_ioctl+0x73/0x100 [nvidia]
Apr 05 19:03:22 Takemikazuchi kernel:  [<ffffffffa0cf64d2>] ? nvidia_ioctl+0x1d2/0x440 [nvidia]
Apr 05 19:03:22 Takemikazuchi kernel:  [<ffffffff810cdc9d>] ? enqueue_hrtimer+0x3d/0xa0
Apr 05 19:03:22 Takemikazuchi kernel:  [<ffffffffa0d026af>] ? nvidia_frontend_ioctl+0x2f/0x60 [nvidia]
Apr 05 19:03:22 Takemikazuchi kernel:  [<ffffffffa0d026fd>] ? nvidia_frontend_compat_ioctl+0x1d/0x40 [nvidia]
Apr 05 19:03:22 Takemikazuchi kernel:  [<ffffffffa0d0272e>] ? nvidia_frontend_unlocked_ioctl+0xe/0x10 [nvidia]
Apr 05 19:03:22 Takemikazuchi kernel:  [<ffffffff811e2718>] ? do_vfs_ioctl+0x298/0x480
Apr 05 19:03:22 Takemikazuchi kernel:  [<ffffffff8147b991>] ? __sys_recvmsg+0x51/0xa0
Apr 05 19:03:22 Takemikazuchi kernel:  [<ffffffff811e2979>] ? SyS_ioctl+0x79/0xa0
Apr 05 19:03:22 Takemikazuchi kernel:  [<ffffffff815a182e>] ? entry_SYSCALL_64_fastpath+0x12/0x71
Apr 05 19:03:22 Takemikazuchi kernel: Code: 49 8b 01 31 f6 f6 c4 40 74 04 41 8b 71 68 4c 89 cf e8 c9 86 fa ff eb 86 4c 89 d1 48 89 da 4c 89 ce e8 79 f8 ff ff e9 73 ff ff ff <0f> 0b 0f 1f 84 00 00 00 00 00 66 2e 0f 1f 84 00 00 00 00 00 0f 
Apr 05 19:03:22 Takemikazuchi kernel: RIP  [<ffffffff811b214c>] kfree+0x14c/0x160

What do I do with that? And (hopefully before I boot again), do you need any more info?

R00KIE · 2016-04-05 17:24:51

Flashing keyboard leds does mean kernel panic so it's properly dead alright.

Regarding the log, if you can reproduce this any time you like and it didn't use to happen before then the kernel devs will be very interested to hear about this (regressions are not welcome there). Even if this is not a regression you want to take this problem to the kernel bugzilla, they are the ones that can better help you debug and solve this.

Soukyuu · 2016-04-05 17:45:04

It happens sporadically, and the USB hub is rather new so I could still RMA it. I'll post it on the kernel bugzilla and see what they say. The only question would be, if I should be trying to reproduce it with a non-ck patched kernel first...

ewaller · 2016-04-05 17:56:58

Microcode up to date?

Soukyuu · 2016-04-05 18:22:29

Yes. It's an old AMD CPU (the "main" system in my signature)

Last edited by Soukyuu (2016-04-05 18:22:51)

R00KIE · 2016-04-05 23:52:15

Soukyuu wrote:

The only question would be, if I should be trying to reproduce it with a non-ck patched kernel first...

A big yes to that. Try with the mainline kernel in the repos, that should be a good starting place as the only patches included should be the ones already accepted upstream to fix any bugs/problems.

Arch Linux

#1 2016-04-02 09:05:47

Multiple devices on USB hub -> sporadic freeze

#2 2016-04-02 11:05:10

Re: Multiple devices on USB hub -> sporadic freeze

#3 2016-04-02 14:53:14

Re: Multiple devices on USB hub -> sporadic freeze

#4 2016-04-03 11:54:05

Re: Multiple devices on USB hub -> sporadic freeze

#5 2016-04-05 17:14:58

Re: Multiple devices on USB hub -> sporadic freeze

#6 2016-04-05 17:24:51

Re: Multiple devices on USB hub -> sporadic freeze

#7 2016-04-05 17:45:04

Re: Multiple devices on USB hub -> sporadic freeze

#8 2016-04-05 17:56:58

Re: Multiple devices on USB hub -> sporadic freeze

#9 2016-04-05 18:22:29

Re: Multiple devices on USB hub -> sporadic freeze

#10 2016-04-05 23:52:15

Re: Multiple devices on USB hub -> sporadic freeze

Board footer