[net-next v8 2/4] page_pool: Add recycle stats

[Joe Damato <jdamato@fastly.com>]

Add per-cpu stats tracking page pool recycling events:
	- cached: recycling placed page in the page pool cache
	- cache_full: page pool cache was full
	- ring: page placed into the ptr ring
	- ring_full: page released from page pool because the ptr ring was full
	- released_refcnt: page released (and not recycled) because refcnt > 1

Signed-off-by: Joe Damato <jdamato@fastly.com>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
---
 include/net/page_pool.h | 16 ++++++++++++++++
 net/core/page_pool.c    | 28 +++++++++++++++++++++++++++-
 2 files changed, 43 insertions(+), 1 deletion(-)

diff --git a/include/net/page_pool.h b/include/net/page_pool.h
index 1f27e8a4..298af95 100644
--- a/include/net/page_pool.h
+++ b/include/net/page_pool.h
@@ -95,6 +95,18 @@ struct page_pool_alloc_stats {
 	u64 refill; /* allocations via successful refill */
 	u64 waive;  /* failed refills due to numa zone mismatch */
 };
+
+struct page_pool_recycle_stats {
+	u64 cached;	/* recycling placed page in the cache. */
+	u64 cache_full; /* cache was full */
+	u64 ring;	/* recycling placed page back into ptr ring */
+	u64 ring_full;	/* page was released from page-pool because
+			 * PTR ring was full.
+			 */
+	u64 released_refcnt; /* page released because of elevated
+			      * refcnt
+			      */
+};
 #endif
 
 struct page_pool {
@@ -144,6 +156,10 @@ struct page_pool {
 	 */
 	struct ptr_ring ring;
 
+#ifdef CONFIG_PAGE_POOL_STATS
+	/* recycle stats are per-cpu to avoid locking */
+	struct page_pool_recycle_stats __percpu *recycle_stats;
+#endif
 	atomic_t pages_state_release_cnt;
 
 	/* A page_pool is strictly tied to a single RX-queue being
diff --git a/net/core/page_pool.c b/net/core/page_pool.c
index 0fa4b76..27233bf 100644
--- a/net/core/page_pool.c
+++ b/net/core/page_pool.c
@@ -29,8 +29,15 @@
 #ifdef CONFIG_PAGE_POOL_STATS
 /* alloc_stat_inc is intended to be used in softirq context */
 #define alloc_stat_inc(pool, __stat)	(pool->alloc_stats.__stat++)
+/* recycle_stat_inc is safe to use when preemption is possible. */
+#define recycle_stat_inc(pool, __stat)							\
+	do {										\
+		struct page_pool_recycle_stats __percpu *s = pool->recycle_stats;	\
+		this_cpu_inc(s->__stat);						\
+	} while (0)
 #else
 #define alloc_stat_inc(pool, __stat)
+#define recycle_stat_inc(pool, __stat)
 #endif
 
 static int page_pool_init(struct page_pool *pool,
@@ -80,6 +87,12 @@ static int page_pool_init(struct page_pool *pool,
 	    pool->p.flags & PP_FLAG_PAGE_FRAG)
 		return -EINVAL;
 
+#ifdef CONFIG_PAGE_POOL_STATS
+	pool->recycle_stats = alloc_percpu(struct page_pool_recycle_stats);
+	if (!pool->recycle_stats)
+		return -ENOMEM;
+#endif
+
 	if (ptr_ring_init(&pool->ring, ring_qsize, GFP_KERNEL) < 0)
 		return -ENOMEM;
 
@@ -410,6 +423,11 @@ static bool page_pool_recycle_in_ring(struct page_pool *pool, struct page *page)
 	else
 		ret = ptr_ring_produce_bh(&pool->ring, page);
 
+#ifdef CONFIG_PAGE_POOL_STATS
+	if (ret == 0)
+		recycle_stat_inc(pool, ring);
+#endif
+
 	return (ret == 0) ? true : false;
 }
 
@@ -421,11 +439,14 @@ static bool page_pool_recycle_in_ring(struct page_pool *pool, struct page *page)
 static bool page_pool_recycle_in_cache(struct page *page,
 				       struct page_pool *pool)
 {
-	if (unlikely(pool->alloc.count == PP_ALLOC_CACHE_SIZE))
+	if (unlikely(pool->alloc.count == PP_ALLOC_CACHE_SIZE)) {
+		recycle_stat_inc(pool, cache_full);
 		return false;
+	}
 
 	/* Caller MUST have verified/know (page_ref_count(page) == 1) */
 	pool->alloc.cache[pool->alloc.count++] = page;
+	recycle_stat_inc(pool, cached);
 	return true;
 }
 
@@ -475,6 +496,7 @@ __page_pool_put_page(struct page_pool *pool, struct page *page,
 	 * doing refcnt based recycle tricks, meaning another process
 	 * will be invoking put_page.
 	 */
+	recycle_stat_inc(pool, released_refcnt);
 	/* Do not replace this with page_pool_return_page() */
 	page_pool_release_page(pool, page);
 	put_page(page);
@@ -488,6 +510,7 @@ void page_pool_put_defragged_page(struct page_pool *pool, struct page *page,
 	page = __page_pool_put_page(pool, page, dma_sync_size, allow_direct);
 	if (page && !page_pool_recycle_in_ring(pool, page)) {
 		/* Cache full, fallback to free pages */
+		recycle_stat_inc(pool, ring_full);
 		page_pool_return_page(pool, page);
 	}
 }
@@ -636,6 +659,9 @@ static void page_pool_free(struct page_pool *pool)
 	if (pool->p.flags & PP_FLAG_DMA_MAP)
 		put_device(pool->p.dev);
 
+#ifdef CONFIG_PAGE_POOL_STATS
+	free_percpu(pool->recycle_stats);
+#endif
 	kfree(pool);
 }
 
-- 
2.7.4