Re: latest -git: hibernate: possible circular locking dependency detected

["Rafael J. Wysocki" <rjw@sisk.pl>]

On Thursday, 21 of August 2008, Vegard Nossum wrote:
> On Thu, Aug 21, 2008 at 8:31 PM, Vegard Nossum <vegard.nossum@gmail.com> wrote:
> > On Thu, Aug 21, 2008 at 8:22 PM, Rafael J. Wysocki <rjw@sisk.pl> wrote:
> >> On Thursday, 21 of August 2008, Vegard Nossum wrote:
> >>>
> >>> I just got this on v2.6.27-rc4 (+ unrelated fix):
> >>>
> >>> ACPI: Preparing to enter system sleep state S5
> >>
> >> This looks fishy.
> >>
> >> Please check what's there in /sys/power/disk .
> >>
> 
> Sorry, forgot this one:
> 
> # cat /sys/power/disk
> test testproc [shutdown] reboot

Hm, I wonder why 'platform' isn't there.

Perhaps the dmesg log will tell.

[/me looks]

No, it doesn't.  The BIOS evidently supports S4.

[/me looks into the source code]

I have no idea whatsoever how it is possible to have

"ACPI: (supports S0 S1 S3 S4 S5)"

in the log and not to have "platform" in the output of "cat /sys/power/disk".

Can you please add

printk(KERN_CRIT "PM: hibernation_ops = %p\n", hibernation_ops);

at the end of kernel/power/disk.c:hibernation_set_ops() and see if it gets
printed during boot?

Thanks,
Rafael