Re: + do_wait-wakeup-optimization.patch added to -mm tree

[Oleg Nesterov <oleg@redhat.com>]

On 11/23, Roland McGrath wrote:
>
> > > +static int do_wait_wake_function(wait_queue_t *curr, unsigned mode, int sync,
> > > +				 void *key)
> > > +{
> > > +	struct task_struct *task = current;
> >
> > I think we can fix (and simplify) this code if we change __wake_up_parent(),
> > it should call __wake_up(key => p), so we can do
> >
> > 	struct task_struct *task = key;
>
> I don't see an exposed __wake_up* variant that both passes a "key" pointer
> through and does "sync".  For __wake_up_parent, "sync" is quite desireable.

Well, yes... and __wake_up_common() is static. Perhaps we can make a new
helper. I must admit, I don't understand what "sync" actually means nowadays.

> > > +	if (!needs_wakeup(task, w))
> > > +		return 0;
> > > +
> > > +	return default_wake_function(curr, mode, sync, key);
> >
> > perhaps autoremove_wake_function() makes more sense.
>
> Why?  The do_wait loop will have to go through again and still might just
> sleep again.  The explicit remove at the end of do_wait seems fine to me.

Yes, yes, I was wrong. I forgot about "repeat:" in do_wait().

Oleg.