Re: Freezeable workqueues [Was: 2.6.22-rc1: Broken suspend on SMP with tifm]

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

On Sunday, 20 May 2007 23:06, Oleg Nesterov wrote:
> On 05/20, Rafael J. Wysocki wrote:
> >
> > On Sunday, 20 May 2007 21:54, Oleg Nesterov wrote:
> > > 
> > > I am a bit afraid of too many yes/no options for the freezer, a couple of naive
> > > questions.
> > > 
> > > 1. Can't we make all wqs freezable? I still can't see the reason to have both
> > >    freezable and not freezable wqs.
> > 
> > The reason might be the same as for having freezable and nonfreezable kernel
> > threads in general.  For example, there are some kernel threads that we need
> > for saving the image and I don't see why there shouldn't be any such
> > workqueues.
> 
> OK, I see.
> 
> > > 2. Why do we need CPU_TASKS_FROZEN? Can't we change cpu-hotplug to always
> > >    freeze tasks right now, without any additional changes?
> > 
> > In principle, we can, but for this purpose we'd have to modify all NOFREEZE
> > tasks.
> 
> Why?

Ah, sorry, I didn't understand the question correctly.

> >        That wouldn't fly, I'm afraid.
> > 
> > >    Any subsystem should handle correctly the case when _cpu_down() (say)
> > >    is called with tasks_frozen == 1 anyway. So, why can't we simplify
> > >    things and do
> > > 
> > >    	_cpu_down(int tasks_frozen)
> > > 
> > >  		if (!tasks_frozen)
> > >  			freeze_processes();
> > >  		...
> > > 
> > >   right now?

Yes, we can do this, I think.
 
> > But we call _cpu_down() after device_suspend(), so many tasks are already
> > frozen at this point.  We'd only need to freeze those that are not frozen and
> > in _cpu_up() we'd have to thaw them.
> 
> Not sure I understand. When we call _cpu_down() after device_suspend(), we
> check tasks_frozen == 1, and do not call freeze_processes(). If the task
> could be frozen, it is already frozen.
> 
> When _cpu_down() sees tasks_frozen = 0, it does freeze_processes() itself,
> and thaw_tasks() on return.
> 
> IOW, we never send (say) CPU_DEAD, always CPU_DEAD_FROZEN.

Yes, that seems reasonable.

This means that every user of freezable kernel threads who installs a CPU
hotplug notifier will have to assume that its kernel threads are frozen when
the notifier is called.

Greetings,
Rafael