[Bug 687998] New: 2.6.39-rc3: suspend to RAM crashes after a few tries (WiFi related?)

newer
[Bug 731043] New: cannot shut down...

older
[Bug 734321] New: Missing header...

bugzilla_noreply＠novell.com

16 Apr 2011 16 Apr '11

08:23

https://bugzilla.novell.com/show_bug.cgi?id=687998 https://bugzilla.novell.com/show_bug.cgi?id=687998#c0 Summary: 2.6.39-rc3: suspend to RAM crashes after a few tries (WiFi related?) Classification: openSUSE Product: openSUSE 11.5 Version: Factory Platform: Other OS/Version: Other Status: NEW Severity: Major Priority: P5 - None Component: Kernel AssignedTo: kernel-maintainers@forge.provo.novell.com ReportedBy: seife@novell.slipkontur.de QAContact: qa@suse.de CC: rjw@novell.com Found By: Third Party Developer/Partner Blocker: --- Created an attachment (id=425263) --> (http://bugzilla.novell.com/attachment.cgi?id=425263) photograph of a panic On a ThinkPad X200s, suspend to RAM crashes reproducibly after a few cycles (often on the second try). It seems to panic, sometimes the capslock LED is blinking, sometimes I get a panic on the framebuffer, but sometimes it just hangs solid when going into suspend. This happens only on a full boot, init=/bin/bash works fine. It *seems* to be related to WiFi, bluetooth or WWAN, since switching off the killswitch before suspending and switching it on after resume made it survive more cycles than usual (>> 5), but on the next try without switching off the killswitch I got the panic. It does not matter if I run X with fbdev (because of the recently broken Intel X driver) or with the intel driver. I did take a photograph of the panic, but it most likely shows only the aftermath of the catastrophe. Maybe it is helpful anyway. Suspend has been rock solid on this machine before for a long time. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

Show replies by date

bugzilla_noreply＠novell.com

18 Apr 18 Apr

22:28

New subject: [Bug 687998] 2.6.39-rc3: suspend to RAM crashes after a few tries (WiFi related?)

https://bugzilla.novell.com/show_bug.cgi?id=687998 https://bugzilla.novell.com/show_bug.cgi?id=687998#c1 Rafael Wysocki changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO InfoProvider| |seife@novell.slipkontur.de AssignedTo|kernel-maintainers@forge.pr |rjw@novell.com |ovo.novell.com | --- Comment #1 from Rafael Wysocki 2011-04-18 22:28:30 UTC --- The kernel is based on 2.6.39-rc, right? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

bugzilla_noreply＠novell.com

19 Apr 19 Apr

06:24

New subject: [Bug 687998] 2.6.39-rc3: suspend to RAM crashes after a few tries (WiFi related?)

https://bugzilla.novell.com/show_bug.cgi?id=687998 https://bugzilla.novell.com/show_bug.cgi?id=687998#c2 Stefan Seyfried changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |NEW InfoProvider|seife@novell.slipkontur.de | --- Comment #2 from Stefan Seyfried 2011-04-19 08:24:02 CEST --- Yes, it's 2.6.39-rc3 Even switching off WiFi before suspend has not been reliable (the kernel seemed to lock up hard on plain WiFi activity), so it might not necessarily be a suspend problem. I just thought it would be good to report it early in the cycle :-) I'm back to 2.6.38 for now, but will retest once Kernel:HEAD gets a new build. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

bugzilla_noreply＠novell.com

30 Apr 30 Apr

13:28

New subject: [Bug 687998] 2.6.39-rc3: suspend to RAM crashes after a few tries (WiFi related?)

https://bugzilla.novell.com/show_bug.cgi?id=687998 https://bugzilla.novell.com/show_bug.cgi?id=687998#c3 Stefan Seyfried changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |oneukum@novell.com, | |vbotka@novell.com --- Comment #3 from Stefan Seyfried 2011-04-30 15:28:12 CEST --- Still there with 2.6.39-rc5-14, both -desktop and -vanilla (KOTD of today). I found out that it is most probably the bluetooth stuff that's to blame. If I "killall bluetoothd" before suspend (which brings btusb usage count to zero and would allow its unloading), then it never crashes. Otherwise, I get a pretty reproducible kernel panic (visible by blinking capslock, nothing on console). But the bug is not that simple ;-) * unloading btusb, then loading it and starting bluetoothd, then suspending => no crash * then virtually un-/replugging the bluetooth adapter (via thinkpad-acpi module facilities), then suspending => crash Basically the workaround for me is to make sure that bluetooth is disabled (I have configured a hotkey for that), disabling via the gnome bluetooth applet is enough, before suspending. I'm suspecting that something in the btusb autosuspend code fails during suspend to RAM. Unfortunately, I'm not able to get any backtrace or similar at all. This is all with the vanilla kernel, with the SUSE desktop kernel I don't even get the blinking capslock led... To add another data point: if I disable intel modesetting (i915.modeset=0 vga=0), which I tried in order to get a backtrace on plain VGA, then the VESA driver is used by X - and then it does not crash. So I really suspect it's a nasty race condition. Again - this is all with a vanilla kernel, so it is a bad upstream regression. Should I also complain loudly on LKML to avoid 2.6.39 being a show stopper on x200s? Adding Oliver to CC who might know something about btusb suspend code :-) -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.