1

I'm currently dual booting Ubuntu 14.04 LTS with Windows 8.1. On Ubuntu, when I wake up my system after a suspend, I'm taken to the login screen, as I'd expect. However, the system is consistently unresponsive for a period of time ranging roughly between 16 and 24 seconds. During this period, I am unable to move the cursor or to enter anything using the keyboard. After the unresponsive period, I am able to move the cursor and interact with the system using the keyboard normally; also, if I type during the unresponsive period, up to 5 characters will appear in the password field once the system resumes normal function (if I typed x characters during the unresponsive period, then if x<5, x characters will appear. Otherwise, 5 characters will appear). What might be causing this behaviour, and how can I fix it?

A copy of my pm-suspend.log and Xorg.0.log files, and the output of sudo dmesg can be found here: https://drive.google.com/folderview?id=0B1YKLXOjqaPATWR6VGF2WDJlZ2s&usp=sharing

6
  • Could it be related to disks spinning up? Oct 6, 2014 at 11:52
  • That's possible. How would I go about investigating whether that's the issue? Note, though, that on Windows 8.1 my system is responsive after suspend with virtually no unresponsive period. Wouldn't I expect similar behaviour on Windows 8.1 if it were my hard drive causing the problem?
    – anon1802
    Oct 6, 2014 at 11:55
  • Welcome to AU! Can you please edit your question to include or link to copies of the files /var/log/pm-suspend.log and /var/log/Xorg.0.log and the output of sudo dmesg just after resuming from suspension? They should contain relevant info about your problem. Oct 6, 2014 at 11:58
  • Maybe you can listen to the disks making different noises when spinning up? It should not block, certainly, independent of whether it's related to disks - but would be good to know as a start. Oct 6, 2014 at 11:59
  • @VolkerSiegel At the point when I resume my system, I only hear the fan spinning up, but about 0.5s before the system becomes responsive again, I am able to hear a short burst of hard drive activity. This behaviour seems to be repeatable.
    – anon1802
    Oct 6, 2014 at 12:18

1 Answer 1

2

According to your dmesg output there's a bug in the WiFi driver.

Let's take a closer look: the most recent wakeup event is at

[22500.127146] ACPI: Waking up from system sleep state S3

and the wakeup finishes ~5 s later:

[22505.833530] PM: Finishing wakeup.

The system then takes another ~5s to re-initiate the PS/2 interface (for keyboard and touch pad):

[22507.823851] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
[22515.585161] i8042: Can't write CTR while closing AUX port

When it tries to re-initiate the WiFi, there are 3 subsequent traces of identical errors in the WiFi subsystem over the course of ~10s:

[22918.219140] ERROR @wl_cfg80211_get_station : Wrong Mac address
[22918.219460] ERROR @wl_cfg80211_get_station : Wrong Mac address
[22920.320875] cfg80211: Calling CRDA to update world regulatory domain
[22920.323141] cfg80211: World regulatory domain updated:
[22920.323145] cfg80211:   (start_freq - end_freq @ bandwidth), (max_antenna_gain, max_eirp)
[22920.323148] cfg80211:   (2402000 KHz - 2472000 KHz @ 40000 KHz), (300 mBi, 2000 mBm)
[22920.323149] cfg80211:   (2457000 KHz - 2482000 KHz @ 40000 KHz), (300 mBi, 2000 mBm)
[22920.323151] cfg80211:   (2474000 KHz - 2494000 KHz @ 20000 KHz), (300 mBi, 2000 mBm)
[22920.323152] cfg80211:   (5170000 KHz - 5250000 KHz @ 40000 KHz), (300 mBi, 2000 mBm)
[22920.323154] cfg80211:   (5735000 KHz - 5835000 KHz @ 40000 KHz), (300 mBi, 2000 mBm)
[22923.987050] ------------[ cut here ]------------
[22923.987110] WARNING: CPU: 0 PID: 484 at /build/buildd/linux-3.13.0/net/wireless/sme.c:797 cfg80211_roamed+0x89/0x90 [cfg80211]()
[22923.987113] Modules linked in: michael_mic arc4 dm_crypt rfcomm bnep bluetooth microread_mei microread mei_phy crc_ccitt hci nfc binfmt_misc snd_hda_codec_hdmi snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_pcm snd_page_alloc intel_rapl x86_pkg_temp_thermal intel_powerclamp coretemp snd_seq_midi snd_seq_midi_event kvm uvcvideo lib80211_crypt_tkip videobuf2_vmalloc videobuf2_memops crct10dif_pclmul videobuf2_core videodev snd_rawmidi crc32_pclmul ghash_clmulni_intel aesni_intel aes_x86_64 lrw snd_seq gf128mul glue_helper ablk_helper snd_seq_device cryptd snd_timer joydev serio_raw wl(POF) lib80211 cfg80211 snd rtsx_pci_ms lpc_ich memstick soundcore mei_me mei parport_pc ppdev nls_iso8859_1 sony_laptop lp parport mac_hid rtsx_pci_sdmmc psmouse i915 ahci r8169 libahci rtsx_pci mii i2c_algo_bit video drm_kms_helper drm
[22923.987211] CPU: 0 PID: 484 Comm: wl_event_handle Tainted: PF       W  O 3.13.0-35-generic #62-Ubuntu
[22923.987216] Hardware name: Sony Corporation SVF15A1C5E/VAIO, BIOS R0220DA 11/18/2013
[22923.987220]  0000000000000009 ffff88023fa0bd88 ffffffff8171e320 0000000000000000
[22923.987228]  ffff88023fa0bdc0 ffffffff8106775d ffff8800a6e1a000 ffff8800a37bdf80
[22923.987235]  000000000000006a ffff88003398c700 0000000000000001 ffff88023fa0bdd0
[22923.987241] Call Trace:
[22923.987254]  [<ffffffff8171e320>] dump_stack+0x45/0x56
[22923.987263]  [<ffffffff8106775d>] warn_slowpath_common+0x7d/0xa0
[22923.987270]  [<ffffffff8106783a>] warn_slowpath_null+0x1a/0x20
[22923.987298]  [<ffffffffa024a509>] cfg80211_roamed+0x89/0x90 [cfg80211]
[22923.987374]  [<ffffffffa0416528>] wl_bss_connect_done.isra.21+0x98/0x1a0 [wl]
[22923.987441]  [<ffffffffa041682c>] wl_notify_connect_status+0x1fc/0x410 [wl]
[22923.987504]  [<ffffffffa0415135>] wl_event_handler+0x55/0x220 [wl]
[22923.987566]  [<ffffffffa04150e0>] ? wl_get_assoc_ies+0x240/0x240 [wl]
[22923.987574]  [<ffffffff8108b4a2>] kthread+0xd2/0xf0
[22923.987581]  [<ffffffff8108b3d0>] ? kthread_create_on_node+0x1c0/0x1c0
[22923.987588]  [<ffffffff8172ecbc>] ret_from_fork+0x7c/0xb0
[22923.987594]  [<ffffffff8108b3d0>] ? kthread_create_on_node+0x1c0/0x1c0
[22923.987599] ---[ end trace a8bec1fd8bc015b5 ]---

The system typically freezes while it recovers from kernel errors (if possible), so this would explain the freeze-up you see. I recommend that you file a bug report against the Broadcom driver package, include the dmesg output, and maybe link to this answer. They know better how to help you and bugs are off topic on AU. [Edit] There's a similar bug already: Broadcom Sta/wl Drivers causes (more) kernel panic.

If you want to poke around more, you can try to (temporarily) disable the WiFi subsystem to see, if the problem persists. The bug may be fixed in newer kernels and/or driver versions. You can try out the most recent (until yesterday) kernel series 3.16 with the package linux-generic-lts-utopic from the Canonical Kernel Team PPA. The repositories for the upcoming Ubuntu Utopic contain a newer version of the Broadcom WiFi driver.

[Edit] A commenter to a similar bug on the RPM Fusion bug tracker suggests to manually set your radio frequency spectrum regulation area in the module parameters (replace EU with one of US or JP depending on your region):

sudo tee -a /etc/modprobe.d/cfg80211.conf > /dev/null <<< 'options cfg80211 ieee80211_regdom="EU"'
sudo update-initramfs -u
sudo reboot

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .