kernel-packages team mailing list archive

[Guilhem Marchand <guilhem.marchand@xxxxxxxxx>]

Hi,

I am affected by this bug.

My Computer is an Asus UX52VS running an all fresh Ubuntu 13.10 Install.

Running Kernel 3.11 (current 3.11.0-13), if a let the alx module loaded and go to suspend (closing my laptop)
When opening up, i get back the X11 authentication, i can log in then the system totally hangs

Ventirad begins to go in high load (so i guess because of an high CPU
load) , sometimes i can access to the terminal (first console) and i can
see the message repeated very fast:

 alx 0000:04:00.0: invalid PHY speed/duplex: 0xffff

The only solution is to hard power off the system to recover.

I can reproduce that everytime, if a unload the alx module before
suspending everthing works fine...

I can see traces in kern.log:

Nov 18 21:01:59 guilhem-UX52VS kernel: [   70.009079] iwlwifi 0000:03:00.0: L1 Disabled; Enabling L0S
Nov 18 21:01:59 guilhem-UX52VS kernel: [   70.015965] iwlwifi 0000:03:00.0: Radio type=0x2-0x1-0x0
Nov 18 21:01:59 guilhem-UX52VS kernel: [   70.105370] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready
Nov 18 21:01:59 guilhem-UX52VS kernel: [   70.108113] alx 0000:04:00.0: irq 49 for MSI/MSI-X
Nov 18 21:01:59 guilhem-UX52VS kernel: [   70.108348] IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready
Nov 18 21:01:59 guilhem-UX52VS kernel: [   70.109175] alx 0000:04:00.0: invalid PHY speed/duplex: 0xffff
Nov 18 21:01:59 guilhem-UX52VS kernel: [   70.110913] alx 0000:04:00.0: invalid PHY speed/duplex: 0xffff
Nov 18 21:01:59 guilhem-UX52VS kernel: [   70.112513] alx 0000:04:00.0: invalid PHY speed/duplex: 0xffff
Nov 18 21:01:59 guilhem-UX52VS kernel: [   70.114340] alx 0000:04:00.0: invalid PHY speed/duplex: 0xffff
Nov 18 21:01:59 guilhem-UX52VS kernel: [   70.115916] alx 0000:04:00.0: invalid PHY speed/duplex: 0xffff
Nov 18 21:01:59 guilhem-UX52VS kernel: [   70.117484] alx 0000:04:00.0: invalid PHY speed/duplex: 0xffff


The system hanged when began the message about alx...

Is there any solution or workaround ?

I was expecting to test with kernel 3.12 if the same bug was existing,
unfortunately under kernel 3.12 for me th situation is even worst and
suspend simply does not work (alx module loaded or unloaded) all i get a
black screen but this it an other story -:-)

Blacklisting the alx module "resolves" the issue as far as you don't
need any wired connection, but this is a dirty workaround not a
solution...

Thank you in advance  for your help!

-- 
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1173952

Title:
  Suspend and hibernate not working in 13.04/13.10 with alx module

Status in “linux” package in Ubuntu:
  Confirmed
Status in “pm-utils” package in Ubuntu:
  Fix Released
Status in “linux” source package in Precise:
  Won't Fix
Status in “pm-utils” source package in Precise:
  Invalid
Status in “linux” source package in Raring:
  Won't Fix
Status in “pm-utils” source package in Raring:
  Fix Released
Status in “linux” source package in Saucy:
  Confirmed
Status in “pm-utils” source package in Saucy:
  Fix Released

Bug description:
  Description:	Ubuntu 13.04
  Release:	13.04

  After upgrading to 13.04 suspend and hibernate do not work.
  In terminal window following error is displayed: "PM: Some devices failed to suspend". Then the suspend fails and the screensaver gets activated, requiring a password to unlock.

  dmesg

  [  893.962025] PM: Syncing filesystems ... done.
  [  894.030726] Freezing user space processes ... (elapsed 0.01 seconds) done.
  [  894.047580] Freezing remaining freezable tasks ... (elapsed 0.01 seconds) done.
  [  894.064140] Suspending console(s) (use no_console_suspend to debug)
  [  894.071511] sd 0:0:0:0: [sda] Synchronizing SCSI cache
  [  894.085881] alx 0000:03:00.0: PHY SPD/DPLX unresolved :ffff
  [  894.085884] alx 0000:03:00.0 eth1: shutown err(ffffffea)
  [  894.085886] alx 0000:03:00.0: shutdown fail in suspend -5
  [  894.085901] pci_pm_suspend(): alx_suspend+0x0/0x90 [alx] returns -5
  [  894.085906] dpm_run_callback(): pci_pm_suspend+0x0/0x140 returns -5
  [  894.085908] PM: Device 0000:03:00.0 failed to suspend async: error -5
  [  894.111147] sd 0:0:0:0: [sda] Stopping disk
  [  894.684692] PM: Some devices failed to suspend
  [  894.684903] ath: phy0: ASPM enabled: 0x43
  [  894.685198] sd 0:0:0:0: [sda] Starting disk
  [  895.442853] PM: resume of devices complete after 758.396 msecs
  [  895.443224] Restarting tasks ... done.
  [  895.445509] video LNXVIDEO:00: Restoring backlight state
  ---
  ApportVersion: 2.9.2-0ubuntu8
  Architecture: amd64
  DistroRelease: Ubuntu 13.04
  InstallationDate: Installed on 2012-12-20 (166 days ago)
  InstallationMedia: Ubuntu 12.10 "Quantal Quetzal" - Release amd64 (20121017.5)
  MarkForUpload: True
  Package: pm-utils 1.4.1-9git1
  PackageArchitecture: all
  ProcVersionSignature: Ubuntu 3.8.0-19.29-generic 3.8.8
  Tags:  raring
  Uname: Linux 3.8.0-19-generic x86_64
  UpgradeStatus: Upgraded to raring on 2013-04-28 (37 days ago)
  UserGroups: adm cdrom dip lpadmin plugdev sambashare sudo

  SRU INFORMATION
  ---------------
  * Impact: suspend regression in 13.04 with Qualcomm Atheros Gigabit ethernet devices
  * Test case: Suspend is broken in current raring when the alx module is loaded, should work again with the fix; needs to be tested on actual hardware, see comment 56 for positive testing result that this pm-utils workaround works.
  * Regression potential: If the script is malformed, it could potentially break suspend on other devices; this should generate an error in /var/log/pm-suspend.log. Please test suspend on devices without the alx module as well (for the record, Martin Pitt did that on his ThinkPad X201)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1173952/+subscriptions