Re: [PATCH RFC tip/core/rcu] Add shrinker to shift to fast/inefficient GP mode

From: Paul E. McKenney
Date: Wed May 13 2020 - 09:03:32 EST


On Wed, May 13, 2020 at 03:07:26PM +1000, Dave Chinner wrote:
> On Tue, May 12, 2020 at 08:18:26PM -0700, Paul E. McKenney wrote:
> > On Wed, May 13, 2020 at 11:32:38AM +1000, Dave Chinner wrote:
> > > On Sat, May 09, 2020 at 09:09:00AM -0700, Paul E. McKenney wrote:
> > > > On Sat, May 09, 2020 at 11:54:40AM +0300, Konstantin Khlebnikov wrote:
> > > > > On 08/05/2020 17.46, Paul E. McKenney wrote:
> > > > > > Easy for me to provide "start fast and inefficient mode" and "stop fast
> > > > > > and inefficient mode" APIs for MM to call!
> > > > > >
> > > > > > How about rcu_mempressure_start() and rcu_mempressure_end()? I would
> > > > > > expect them not to nest (as in if you need them to nest, please let
> > > > > > me know). I would not expect these to be invoked all that often (as in
> > > > > > if you do need them to be fast and scalable, please let me know). >
> > > > > > RCU would then be in fast/inefficient mode if either MM told it to be
> > > > > > or if RCU had detected callback overload on at least one CPU.
> > > > > >
> > > > > > Seem reasonable?
> > > > >
> > > > > Not exactly nested calls, but kswapd threads are per numa node.
> > > > > So, at some level nodes under pressure must be counted.
> > > >
> > > > Easy enough, especially given that RCU already "counts" CPUs having
> > > > excessive numbers of callbacks. But assuming that the transitions to/from
> > > > OOM are rare, I would start by just counting them with a global counter.
> > > > If the counter is non-zero, RCU is in fast and inefficient mode.
> > > >
> > > > > Also forcing rcu calls only for cpus in one numa node might be useful.
> > > >
> > > > Interesting. RCU currently evaluates a given CPU by comparing the
> > > > number of callbacks against a fixed cutoff that can be set at boot using
> > > > rcutree.qhimark, which defaults to 10,000. When this cutoff is exceeded,
> > > > RCU becomes more aggressive about invoking callbacks on that CPU, for
> > > > example, by sacrificing some degree of real-time response. I believe
> > > > that this heuristic would also serve the OOM use case well.
> > >
> > > So one of the things that I'm not sure people have connected here is
> > > that memory reclaim done by shrinkers is one of the things that
> > > drives huge numbers of call_rcu() callbacks to free memory via rcu.
> > > If we are reclaiming dentries and inodes, then we can be pushing
> > > thousands to hundreds of thousands of objects into kfree_rcu()
> > > and/or direct call_rcu() calls to free these objects in a single
> > > reclaim pass.
> >
> > Good point!
> >
> > > Hence the trigger for RCU going into "excessive callback" mode
> > > might, in fact, be kswapd running a pass over the shrinkers. i.e.
> > > memory reclaim itself can be responsible for pushing RCU into this "OOM
> > > pressure" situation.
> > >
> > > So perhaps we've missed a trick here by not having the memory
> > > reclaim routines trigger RCU callbacks at the end of a priority
> > > scan. The shrinkers have queued the objects for freeing, but they
> > > haven't actually been freed yet and so things like slab pages
> > > haven't actually been returned to the free pool even though the
> > > shrinkers have said "freed this many objects"...
> > >
> > > i.e. perhaps the right solution here is a "rcu_run_callbacks()"
> > > function that memory reclaim calls before backing off and/or winding
> > > up reclaim priority.
> >
> > It would not be hard to make something that put RCU into fast/inefficient
> > mode for a couple of grace periods. I will also look into the possibility
> > of speeding up callback invocation.
> >
> > It might also make sense to put RCU grace periods into fast mode while
> > running the shrinkers that are freeing dentries and inodes. However,
> > kbuild test robot reports ugly regressions when putting RCU into
> > fast/inefficient mode to quickly and too often. As in 78.5% degradation
> > on one of the benchmarks.
>
> I don't think it should be dependent on what specific shrinkers
> free. There are other objects that may be RCU freed by shrinkers,
> so it really shouldn't be applied just to specific shrinker
> instances.

Plus a call_rcu() might be freeing a linked structure, so counting the
size of the argument to call_rcu() would be understating the total amount
of memory being freed.

> > > > > I wonder if direct-reclaim should at some stage simply wait for RCU QS.
> > > > > I.e. call rcu_barrier() or similar somewhere before invoking OOM.
> > > >
> > > > The rcu_oom_count() function in the patch starting this thread returns the
> > > > total number of outstanding callbacks queued on all CPUs. So one approach
> > > > would be to invoke this function, and if the return value was truly
> > > > huge (taking size of memory and who knows that all else into account),
> > > > do the rcu_barrier() to wait for RCU to clear its current backlog.
> > >
> > > The shrinker scan control structure has a node mask in it to
> > > indicate what node (and hence CPUs) it should be reclaiming from.
> > > This information comes from the main reclaim scan routine, so it
> > > would be trivial to feed straight into the RCU code to have it
> > > act on just the CPUs/node that we are reclaiming memory from...
> >
> > For the callbacks, RCU can operate on CPUs, in theory anyway. The
> > grace period itself, however, is inherently global.
>
> *nod*
>
> The memory reclaim backoffs tend to be in the order of 50-100
> milliseconds, though, so we are talking multiple grace periods here,
> right? In which case, triggering a grace period expiry before a
> backoff takes place might make a lot sense...

Usually, yes, I would expect several grace periods to elapse during
a backoff.

> > > > On the NUMA point, it would be dead easy for me to supply a function
> > > > that returned the number of callbacks on a given CPU, which would allow
> > > > you to similarly evaluate a NUMA node, a cgroup, or whatever.
> > >
> > > I'd think it runs the other way around - we optimisitically call the
> > > RCU layer to do cleanup, and the RCU layer decides if there's enough
> > > queued callbacks on the cpus/node to run callbacks immediately. It
> > > would even be provided with the scan priority to indicate the level
> > > of desperation memory reclaim is under....
> >
> > Easy for RCU to count the number of callbacks. That said, it has no
> > idea which callbacks are which. Perhaps kfree_rcu() could gather that
> > information from the slab allocator, though.
> >
> > > > > All GFP_NOFAIL users should allow direct-reclaim, thus this loop
> > > > > in page_alloc shouldn't block RCU and doesn't need special care.
> > > >
> > > > I must defer to you guys on this. The main caution is the duration of
> > > > direct reclaim. After all, if it is too long, the kfree_rcu() instance
> > > > would have been better of just invoking synchronize_rcu().
> > >
> > > Individual callers of kfree_rcu() have no idea of the load on RCU,
> > > nor how long direct reclaim is taking. Calling synchronize_rcu()
> > > incorrectly has pretty major downsides to it, so nobody should be
> > > trying to expedite kfree_rcu() unless there is a good reason to do
> > > so (e.g. at unmount to ensure everything allocated by a filesystem
> > > has actually been freed). Hence I'd much prefer the decision to
> > > expedite callbacks is made by the RCU subsystem based on it's known
> > > callback load and some indication of how close memory reclaim is to
> > > declaring OOM...
> >
> > Sorry, I was unclear. There is a new single-argument kfree_rcu() under
> > way that does not require an rcu_head in the structure being freed.
> > However, in this case, kfree_rcu() might either allocate the memory
> > that is needed to track the memory to be freed on the one hand or just
> > invoke synchronize_rcu() on the other. So this decision would be taken
> > inside kfree_rcu(), and not be visible to either core RCU or the caller
> > of kfree_rcu().
>
> Ah. The need to allocate memory to free memory, and with that comes
> the requirement of a forwards progress guarantee. It's mempools all
> over again :P
>
> Personally, though, designing functionality that specifically
> requires memory allocation to free memory seems like an incredibly
> fragile thing to be doing. I don't know the use case here, though,
> but jsut the general description of what you are trying to do rings
> alarm bells in my head...

And mine as well. Hence my earlier insistence that kfree_rcu() never
block waiting for memory, but instead just invoke synchronize_rcu() and
then immediately free the memory. Others have since convinced me that
there are combinations of GFP flags that allow only limited sleeping so
as to avoid the OOM deadlocks that I fear.

> > This decision is made based on whether or not the allocator provides
> > kfree_rcu() the memory needed. The tradeoff is what GFP flags are
> > supplied.
>
> So there's a reclaim recursion problem here, too?

There was an earlier discussion as to what was safe, with one
recommendation being __GFP_NORETRY.

Thanx, Paul