Re: [RFC-PATCH 1/2] mm: Add __GFP_NO_LOCKS flag

[Michal Hocko]

On Thu 13-08-20 15:22:00, Thomas Gleixner wrote:
> Uladzislau Rezki <urezki@xxxxxxxxx> writes:
> > On Thu, Aug 13, 2020 at 09:50:27AM +0200, Michal Hocko wrote:
> >> On Wed 12-08-20 02:13:25, Thomas Gleixner wrote:
> >> [...]
> >> > I can understand your rationale and what you are trying to solve. So, if
> >> > we can actually have a distinct GFP variant:
> >> > 
> >> >   GFP_I_ABSOLUTELY_HAVE_TO_DO_THAT_AND_I_KNOW_IT_CAN_FAIL_EARLY
> >> 
> >> Even if we cannot make the zone->lock raw I would prefer to not
> >> introduce a new gfp flag. Well we can do an alias for easier grepping
> >> #define GFP_RT_SAFE	0
> 
> Just using 0 is sneaky but yes, that's fine :)
> 
> Bikeshedding: GFP_RT_NOWAIT or such might be more obvious.

Sounds goood.

> >> that would imply nowait semantic and would exclude waking up kswapd as
> >> well. If we can make wake up safe under RT then the alias would reflect
> >> that without any code changes.
> 
> It basically requires to convert the wait queue to something else. Is
> the waitqueue strict single waiter?

I would have to double check. From what I remember only kswapd should
ever sleep on it.

> >> The second, and the more important part, would be to bail out anytime
> >> the page allocator is to take a lock which is not allowed in the current
> >> RT context. Something like
> 
> >> +	/*
> >> +	 * Hard atomic contexts are not supported by the allocator for
> >> +	 * anything but pcp requests
> >> +	 */
> >> +	if (!preemtable())
> 
> If you make that preemtible() it might even compile, but that still wont
> work because if CONFIG_PREEMPT_COUNT=n then preemptible() is always
> false.

It would be nice to hide all that behind a helper and guarded by
PREEMPT_RT. That would imply PREEMPT_COUNT automatically, right?

> 
> So that should be:
> 
> 	if (!preemptible() && gfp == GFP_RT_NOWAIT)
> 
> which is limiting the damage to those callers which hand in
> GFP_RT_NOWAIT.
> 
> lockdep will yell at invocations with gfp != GFP_RT_NOWAIT when it hits
> zone->lock in the wrong context. And we want to know about that so we
> can look at the caller and figure out how to solve it.

Yes, that would have to somehow need to annotate the zone_lock to be ok
in those paths so that lockdep doesn't complain.
-- 
Michal Hocko
SUSE Labs