Re: [PATCH 1/1] rcu/tree: support kfree_bulk() interface in kfree_rcu()

[Paul E. McKenney]

On Thu, Jan 16, 2020 at 06:27:53PM +0100, Uladzislau Rezki wrote:
> On Wed, Jan 15, 2020 at 06:41:26PM -0800, Paul E. McKenney wrote:
> > On Wed, Jan 15, 2020 at 08:14:10PM -0500, Joel Fernandes wrote:
> > > On Tue, Dec 31, 2019 at 01:22:41PM +0100, Uladzislau Rezki (Sony) wrote:
> > > > kfree_rcu() logic can be improved further by using kfree_bulk()
> > > > interface along with "basic batching support" introduced earlier.
> > > > 
> > > > The are at least two advantages of using "bulk" interface:
> > > > - in case of large number of kfree_rcu() requests kfree_bulk()
> > > >   reduces the per-object overhead caused by calling kfree()
> > > >   per-object.
> > > > 
> > > > - reduces the number of cache-misses due to "pointer chasing"
> > > >   between objects which can be far spread between each other.
> > > > 
> > > > This approach defines a new kfree_rcu_bulk_data structure that
> > > > stores pointers in an array with a specific size. Number of entries
> > > > in that array depends on PAGE_SIZE making kfree_rcu_bulk_data
> > > > structure to be exactly one page.
> > > > 
> > > > Since it deals with "block-chain" technique there is an extra
> > > > need in dynamic allocation when a new block is required. Memory
> > > > is allocated with GFP_NOWAIT | __GFP_NOWARN flags, i.e. that
> > > > allows to skip direct reclaim under low memory condition to
> > > > prevent stalling and fails silently under high memory pressure.
> > > > 
> > > > The "emergency path" gets maintained when a system is run out
> > > > of memory. In that case objects are linked into regular list
> > > > and that is it.
> > > > 
> > > > In order to evaluate it, the "rcuperf" was run to analyze how
> > > > much memory is consumed and what is kfree_bulk() throughput.
> > > > 
> > > > Testing on the HiKey-960, arm64, 8xCPUs with below parameters:
> > > > 
> > > > CONFIG_SLAB=y
> > > > kfree_loops=200000 kfree_alloc_num=1000 kfree_rcu_test=1
> > > > 
> > > > 102898760401 ns, loops: 200000, batches: 5822, memory footprint: 158MB
> > > > 89947009882  ns, loops: 200000, batches: 6715, memory footprint: 115MB
> > > > 
> > > > rcuperf shows approximately ~12% better throughput(Total time)
> > > > in case of using "bulk" interface. The "drain logic" or its RCU
> > > > callback does the work faster that leads to better throughput.
> > > 
> > > Tested-by: Joel Fernandes (Google) <joel@xxxxxxxxxxxxxxxxx>
> > > 
> > > (Vlad is going to post a v2 which fixes a debugobjects bug but that should
> > > not have any impact on testing).
> > 
> > Very good!  Uladzislau, could you please add Joel's Tested-by in
> > your next posting?
> > 
> I will add for sure, with the a V2 version. Also, i will update the
> commit message by adding the results related to different slab cache
> usage, i mean with Joel's recent patch.

Sounds good, looking forward to it!

							Thanx, Paul