Re: [PATCH v2 2/3] Mutually exclude cpu online and suspend/hibernate

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

On Thursday, October 13, 2011, Alan Stern wrote:
> On Thu, 13 Oct 2011, Borislav Petkov wrote:
> 
> > On Thu, Oct 13, 2011 at 12:06:02PM -0400, Tejun Heo wrote:
> > > I hate to sound like a broken recorder but the above patch isn't
> > > strictly correct for hot-swap cases, right?
> > 
> > hpa is working on a ucode loading solution which will take care of your
> > hotswap case too.
> 
> You know, it would be a good idea to take a step back and think about a 
> broader picture.  Suspend and hibernate need to be mutually exclusive 
> with other things as well.  The two examples which people have been 
> talking about are firmware updates (if the computer goes to sleep 
> while the firmware is being changed, it ends up as a brick) and battery 
> recharging (the CPU needs to respond quickly to temperature and charger 
> events).
> 
> There ought to be a general-purpose "prevent system sleep" mechanism 
> usable both from within the kernel and from userspace.

I have a patch for that ready, will post it later today.

Thanks,
Rafael