[PATCH v6 3/4] mm/memcg: Improve refill_obj_stock() performance

[Waiman Long <longman@redhat.com>]

There are two issues with the current refill_obj_stock() code. First of
all, when nr_bytes reaches over PAGE_SIZE, it calls drain_obj_stock() to
atomically flush out remaining bytes to obj_cgroup, clear cached_objcg
and do a obj_cgroup_put(). It is likely that the same obj_cgroup will
be used again which leads to another call to drain_obj_stock() and
obj_cgroup_get() as well as atomically retrieve the available byte from
obj_cgroup. That is costly. Instead, we should just uncharge the excess
pages, reduce the stock bytes and be done with it. The drain_obj_stock()
function should only be called when obj_cgroup changes.

Secondly, when charging an object of size not less than a page in
obj_cgroup_charge(), it is possible that the remaining bytes to be
refilled to the stock will overflow a page and cause refill_obj_stock()
to uncharge 1 page. To avoid the additional uncharge in this case, a new
allow_uncharge flag is added to refill_obj_stock() which will be set to
false when called from obj_cgroup_charge() so that an uncharge_pages()
call won't be issued right after a charge_pages() call unless the objcg
changes.

A multithreaded kmalloc+kfree microbenchmark on a 2-socket 48-core
96-thread x86-64 system with 96 testing threads were run.  Before this
patch, the total number of kilo kmalloc+kfree operations done for a 4k
large object by all the testing threads per second were 4,304 kops/s
(cgroup v1) and 8,478 kops/s (cgroup v2). After applying this patch, the
number were 4,731 (cgroup v1) and 418,142 (cgroup v2) respectively. This
represents a performance improvement of 1.10X (cgroup v1) and 49.3X
(cgroup v2).

Signed-off-by: Waiman Long <longman@redhat.com>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
---
 mm/memcontrol.c | 48 +++++++++++++++++++++++++++++++++++-------------
 1 file changed, 35 insertions(+), 13 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index fbedfc55a248..513f3d56e89a 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -3281,10 +3281,12 @@ static bool obj_stock_flush_required(struct memcg_stock_pcp *stock,
 	return false;
 }
 
-static void refill_obj_stock(struct obj_cgroup *objcg, unsigned int nr_bytes)
+static void refill_obj_stock(struct obj_cgroup *objcg, unsigned int nr_bytes,
+			     bool allow_uncharge)
 {
 	struct memcg_stock_pcp *stock;
 	unsigned long flags;
+	unsigned int nr_pages = 0;
 
 	local_irq_save(flags);
 
@@ -3293,14 +3295,21 @@ static void refill_obj_stock(struct obj_cgroup *objcg, unsigned int nr_bytes)
 		drain_obj_stock(stock);
 		obj_cgroup_get(objcg);
 		stock->cached_objcg = objcg;
-		stock->nr_bytes = atomic_xchg(&objcg->nr_charged_bytes, 0);
+		stock->nr_bytes = atomic_read(&objcg->nr_charged_bytes)
+				? atomic_xchg(&objcg->nr_charged_bytes, 0) : 0;
+		allow_uncharge = true;	/* Allow uncharge when objcg changes */
 	}
 	stock->nr_bytes += nr_bytes;
 
-	if (stock->nr_bytes > PAGE_SIZE)
-		drain_obj_stock(stock);
+	if (allow_uncharge && (stock->nr_bytes > PAGE_SIZE)) {
+		nr_pages = stock->nr_bytes >> PAGE_SHIFT;
+		stock->nr_bytes &= (PAGE_SIZE - 1);
+	}
 
 	local_irq_restore(flags);
+
+	if (nr_pages)
+		obj_cgroup_uncharge_pages(objcg, nr_pages);
 }
 
 int obj_cgroup_charge(struct obj_cgroup *objcg, gfp_t gfp, size_t size)
@@ -3312,14 +3321,27 @@ int obj_cgroup_charge(struct obj_cgroup *objcg, gfp_t gfp, size_t size)
 		return 0;
 
 	/*
-	 * In theory, memcg->nr_charged_bytes can have enough
+	 * In theory, objcg->nr_charged_bytes can have enough
 	 * pre-charged bytes to satisfy the allocation. However,
-	 * flushing memcg->nr_charged_bytes requires two atomic
-	 * operations, and memcg->nr_charged_bytes can't be big,
-	 * so it's better to ignore it and try grab some new pages.
-	 * memcg->nr_charged_bytes will be flushed in
-	 * refill_obj_stock(), called from this function or
-	 * independently later.
+	 * flushing objcg->nr_charged_bytes requires two atomic
+	 * operations, and objcg->nr_charged_bytes can't be big.
+	 * The shared objcg->nr_charged_bytes can also become a
+	 * performance bottleneck if all tasks of the same memcg are
+	 * trying to update it. So it's better to ignore it and try
+	 * grab some new pages. The stock's nr_bytes will be flushed to
+	 * objcg->nr_charged_bytes later on when objcg changes.
+	 *
+	 * The stock's nr_bytes may contain enough pre-charged bytes
+	 * to allow one less page from being charged, but we can't rely
+	 * on the pre-charged bytes not being changed outside of
+	 * consume_obj_stock() or refill_obj_stock(). So ignore those
+	 * pre-charged bytes as well when charging pages. To avoid a
+	 * page uncharge right after a page charge, we set the
+	 * allow_uncharge flag to false when calling refill_obj_stock()
+	 * to temporarily allow the pre-charged bytes to exceed the page
+	 * size limit. The maximum reachable value of the pre-charged
+	 * bytes is (sizeof(object) + PAGE_SIZE - 2) if there is no data
+	 * race.
 	 */
 	nr_pages = size >> PAGE_SHIFT;
 	nr_bytes = size & (PAGE_SIZE - 1);
@@ -3329,14 +3351,14 @@ int obj_cgroup_charge(struct obj_cgroup *objcg, gfp_t gfp, size_t size)
 
 	ret = obj_cgroup_charge_pages(objcg, gfp, nr_pages);
 	if (!ret && nr_bytes)
-		refill_obj_stock(objcg, PAGE_SIZE - nr_bytes);
+		refill_obj_stock(objcg, PAGE_SIZE - nr_bytes, false);
 
 	return ret;
 }
 
 void obj_cgroup_uncharge(struct obj_cgroup *objcg, size_t size)
 {
-	refill_obj_stock(objcg, size);
+	refill_obj_stock(objcg, size, true);
 }
 
 #endif /* CONFIG_MEMCG_KMEM */
-- 
2.18.1