[Bug 532374] Re: Lenovo Thinkpads with Core i5 and i7 suspend/resume (with kernel oops) once then fail horribly on next suspend

[Colin King]

Just to clarify what the "notsc"  kernel parameter addresses:

The CPU#0 stuck messages in comment #26 are because of a math error in
the softlockup code which gets confused when the TSC warps backwards
when coming out of S3:

[1266874885.662888] BUG: soft lockup - CPU#0 stuck for 0s! [pm-suspend:1512]
[1266874885.662889] Modules linked in: binfmt_misc ppdev joydev snd_hda_codec_intelhdmi fbcon tileblit font bitblit softcursor vga16fb vgastate arc4 snd_hda_intel snd_hda_codec snd_hwdep snd_pcm_oss snd_mixer_oss snd_pcm snd_seq_dummy snd_seq_oss snd_seq_midi snd_rawmidi snd_seq_midi_event snd_seq snd_timer snd_seq_device iwlagn iwlcore thinkpad_acpi i915 drm_kms_helper led_class snd mac80211 nvram drm i2c_algo_bit psmouse serio_raw video output cfg80211 intel_agp soundcore snd_page_alloc agpgart lp parport e1000e
[1266874885.662915] 
[1266874885.662917] Pid: 1512, comm: pm-suspend Not tainted (2.6.32-16-generic #25-Ubuntu) 3626F8U
[1266874885.662919] EIP: 0060:[<c059ddf5>] EFLAGS: 00000292 CPU: 0
[1266874885.662925] EIP is at _spin_unlock_irqrestore+0x15/0x20
[1266874885.662926] EAX: 00000292 EBX: 00000292 ECX: 00000100 EDX: 00000292
[1266874885.662928] ESI: f7002cbd EDI: 00000000 EBP: f47f1d10 ESP: f47f1d0c
[1266874885.662929]  DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068
[1266874885.662930] CR0: 8005003b CR2: 08346b18 CR3: 33097000 CR4: 000006d0
[1266874885.662932] DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000
[1266874885.662933] DR6: ffff0ff0 DR7: 00000400
[1266874885.662934] Call Trace:

Using the kernel parameter "notsc" will disable the TSC and stop these
false-positives.  I have a patch (http://linux.derkeiler.com/Mailing-
Lists/Kernel/2010-03/msg08854.html) that should get into the mainline
kernel very soon now that addresses the math overflow error that trips
these false-positive errors.

-- 
Lenovo Thinkpads with Core i5 and i7 suspend/resume (with kernel oops) once then fail horribly on next suspend
https://bugs.launchpad.net/bugs/532374
You received this bug notification because you are a member of Kernel
Bugs, which is subscribed to Linux.