[Bug 46303] Re: [Bug 46303] Re: [Bug 46303] Re: Asus S5N fails to resume from hibernate (modprobe thermal consumes 100% CPU)

[Cameron Patrick]

On Thu, May 25, 2006 at 06:37:04AM -0000, Cameron Patrick wrote:

> On Wed, May 24, 2006 at 10:50:59AM -0000, Matthew Garrett wrote:
> 
> > If the screen is working at that point, can you do alt+sysrq+t and try
> > to get a kernel backtrace?
> 
> The screen is working at the point where it hangs, but after updating to
> the latest kernel (2.6.15-23.39) I can no longer reproduce this.  Oddly
> I can't see anything in the changelog that would be remotely relevant.
> I'll try to downgrade, confirm it breaks, then upgrade again some time
> this evening; if it's still working with the new kernel but not the old,
> I'll close the bug.

Nope, it still happens sometimes with the newer kernel too.  See the
backtrace as requested at:

    http://cameron.ucc.asn.au/files/fred.txt

Unfortunately it doesn't show anything interesting for modprobe.  Top
says it spends all its time in the R state, which I thought meant "not
in kernel space", but also lists it consuming "system" time, not "user"
time.  Attempts to strace it or attach gdb also hung.  (Incidentally,
Ubuntu seems to set up /etc/hosts brokenly - note "localhost" in all the
log files. Grr.)

FWIW, this laptop used to suspend to disc utterly reliably with Nigel's
suspend2 patches on vanilla 2.6.15.1, using the default config on
Bernard's hibernate script.

Cameron

-- 
Asus S5N fails to resume from hibernate (modprobe thermal consumes 100% CPU)
https://launchpad.net/bugs/46303