FS#44790 - [linux] kernel 4.0.1-1-ARCH: issues with not-locked & not-suspended system, while closing lid

Attached to Project: Arch Linux
Opened by Ruben Kelevra (RubenKelevra) - Thursday, 30 April 2015, 12:58 GMT
Last edited by Doug Newgard (Scimmia) - Tuesday, 03 October 2017, 02:43 GMT
Task Type Bug Report
Category Kernel
Status Closed
Assigned To Tobias Powalowski (tpowa)
Architecture x86_64
Severity Medium
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 5
Private No

Details

Description:
Note: This issue appear together with #44779 so they might sticked together.

When I close the lid of my notebook, the system sometimes just run on. No suspend and NO LOCK OF THE SCREEN. I think the second one is a critical security flaw!


Additional info:
* package version(s)
Linux 4.0.1-1
* config and/or log files etc.
Default config, Gnome.

Journal:

Apr 30 01:25:15 i3 systemd-logind[409]: Lid closed.
Apr 30 01:25:15 i3 root[1290]: LID closed
Apr 30 01:25:15 i3 kernel: i915 0000:00:02.0: BAR 6: [??? 0x00000000 flags 0x2] has bogus alignment
Apr 30 01:25:15 i3 kernel: i915 0000:00:02.0: BAR 6: [??? 0x00000000 flags 0x2] has bogus alignment
Apr 30 01:25:15 i3 kernel: i915 0000:00:02.0: BAR 6: [??? 0x00000000 flags 0x2] has bogus alignment
Apr 30 01:25:15 i3 kernel: i915 0000:00:02.0: BAR 6: [??? 0x00000000 flags 0x2] has bogus alignment
Apr 30 01:25:15 i3 kernel: pci_bus 0000:02: Allocating resources
Apr 30 01:25:15 i3 kernel: pci_bus 0000:03: Allocating resources
Apr 30 01:25:15 i3 kernel: i915 0000:00:02.0: BAR 6: [??? 0x00000000 flags 0x2] has bogus alignment
Apr 30 01:25:15 i3 kernel: i915 0000:00:02.0: BAR 6: [??? 0x00000000 flags 0x2] has bogus alignment
Apr 30 01:25:27 i3 systemd-logind[409]: Lid opened.
Apr 30 01:25:27 i3 root[1293]: LID opened

Steps to reproduce:
Close the lid several times, sometimes it doesn't work. When you wait some minutes the issue seems to disapear and the lid-close do it's work (when retry to close the lid, not when you closed the lid and just wait.

Expected behavoir:
-The screen should be locked immediately!
-Suspend should work.
This task depends upon

Closed by  Doug Newgard (Scimmia)
Tuesday, 03 October 2017, 02:43 GMT
Reason for closing:  Fixed
Comment by Ruben Kelevra (RubenKelevra) - Thursday, 30 April 2015, 13:00 GMT
System is an SandyBridge:

Intel(R) Core(TM) i3-2330M CPU @ 2.20GHz
Comment by Iván Perdomo (katratxo) - Thursday, 30 April 2015, 13:33 GMT
I'm experiencing the same issue. This worked just fine on previous kernel version.

(Edit: Adding hardware specs)
   lshw.txt (15.6 KiB)
Comment by Ruben Kelevra (RubenKelevra) - Thursday, 30 April 2015, 15:35 GMT
adding hwspecs
   lshw.txt (13.5 KiB)
Comment by Iván Perdomo (katratxo) - Tuesday, 05 May 2015, 13:24 GMT
Quick update:
After my recent update, the suspend/resume after closing/opening the lid is working fine. I'm not sure what package is the one affecting the functionality. If is a real kernel issue, a systemd one (or any other package)
Comment by Michał Gawron (olimp) - Thursday, 14 May 2015, 21:29 GMT
I have the same issue - https://bugs.archlinux.org/task/44962 (I requested to close that bug, since it's a duplicate)
Comment by Ruben Kelevra (RubenKelevra) - Sunday, 02 August 2015, 12:21 GMT
I think the bug has been fixed, never had this issue again after some sub-updates on 4.0-Kernel
Comment by Dave Kennedy (Ocypheum) - Monday, 23 November 2015, 02:50 GMT
I don't think it's been fixed as it's happened to me several times today. This is the first time I've noticed a problem suspending on lid close.

Furthermore, attempting to suspend the system with `systemctl` fails with the following error:

~ $ systemctl suspend
Failed to suspend system via logind: There's already a shutdown or sleep operation in progress

I've attached the relevant section from journal and the output of `modinfo i915`.

I'm using systemd v227-1 and polkit v0.113-4.

Other recently installed/upgraded packages:

[2015-11-18 19:55] [ALPM] installed xf86dgaproto (2.1-3)
[2015-11-18 19:55] [ALPM] installed libxxf86dga (1.1.4-1)
[2015-11-18 19:55] [ALPM] installed dmxproto (2.3.1-3)
[2015-11-18 19:55] [ALPM] installed libdmx (1.1.3-1)
[2015-11-18 19:55] [ALPM] installed xorg-xdpyinfo (1.3.2-1)
[2015-11-19 00:02] [ALPM] installed giflib (5.1.1-1)
[2015-11-19 00:02] [ALPM] installed libid3tag (0.15.1b-8)
[2015-11-19 00:02] [ALPM] installed imlib2 (1.4.7-1)
[2015-11-19 00:02] [ALPM] installed libexif (0.6.21-2)
[2015-11-19 00:02] [ALPM] installed feh (2.14-1)
[2015-11-20 20:29] [ALPM] upgraded gmp (6.0.0-2 -> 6.1.0-1)
[2015-11-20 20:29] [ALPM] upgraded harfbuzz (1.0.6-2 -> 1.1.0-1)
[2015-11-20 20:29] [ALPM] upgraded w3m (0.5.3.git20151109-1 -> 0.5.3.git20151119-1)
[2015-11-20 21:30] [ALPM] installed xcompmgr (1.1.7-1)
[2015-11-20 21:31] [ALPM] installed transset-df (6-4)
[2015-11-21 11:05] [ALPM] upgraded glib2 (2.46.1-1 -> 2.46.2-1)
[2015-11-21 11:05] [ALPM] upgraded mesa (11.0.5-1 -> 11.0.6-1)
[2015-11-21 11:05] [ALPM] upgraded mesa-libgl (11.0.5-1 -> 11.0.6-1)
Comment by Dave Kennedy (Ocypheum) - Wednesday, 25 November 2015, 03:03 GMT
Today, my laptop froze after it failed to resume from suspend. After a hard reboot, I used this command to see how long this has been happening:

journalctl --catalog | grep -i "bogus alignment"

The errors started on 2015-10-25 at 21:49, which was a busy day as far as package upgrades because the 4.2.3-1 kernel was released. I also installed xorg-server that day, but the errors started slightly before that - just after I upgraded cryptsetup (at 21:20) and installed abs (at 21:21).

Next, I combed through journalctl and spotted a recurring warning from intel_display:

arch kernel: WARNING: CPU: 3 PID: 735 at drivers/gpu/drm/i915/intel_display.c:9149 hsw_enable_pc8+0x5c1/0x700 [i915]()

I used this command to see when those started occurring:

journalctl --catalog | grep -i "intel_display.c"

Oddly enough the first one is on 2015-10-25 at 22:18.

I used this command to see how often that warning occurs relative to the bogus alignment errors and lid close and open events:

journalctl --catalog --since="3 days ago" | grep -i "lid closed\|lid opened\|bogus alignment\|intel_display.c"

I attached the results (after deleting duplicate messages and inserting line breaks between time chunks) as well as the pacman log from that day. I also attached the output of modinfo i915.
Comment by John Henderson (jwhendy) - Saturday, 02 January 2016, 17:16 GMT
Is there any update on this bug? I've been getting this for some time on an HP Zbook 15 with nvidia/intel graphics:

Cards (via lspci):
- 00:02.0 VGA compatible controller: Intel Corporation 4th Gen Core Processor Integrated Graphics Controller (rev 06
- 01:00.0 VGA compatible controller: NVIDIA Corporation GK106GLM [Quadro K2100M] (rev a1)

Also seems potentially related to sleep (attached dmesg from pre/post suspend cycle). The lines specifically related to this bug are:

[ 3954.580177] i915 0000:00:02.0: BAR 6: [??? 0x00000000 flags 0x2] has bogus alignment
[ 3954.581290] i915 0000:00:02.0: BAR 6: [??? 0x00000000 flags 0x2] has bogus alignment
Comment by mattia (nTia89) - Monday, 02 October 2017, 18:08 GMT
is this issue still valid?
Comment by John Henderson (jwhendy) - Monday, 02 October 2017, 20:27 GMT
I cannot recall the last time I had the issue. I forgot I even had it until this comment brought it back up! It's fixed for me.

Loading...