netdev - [PATCH 6/6] net: page_pool: Add a stat tracking waived pages.

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <1643237300-44904-7-git-send-email-jdamato@fastly.com>
Date:   Wed, 26 Jan 2022 14:48:20 -0800
From:   Joe Damato <jdamato@...tly.com>
To:     netdev@...r.kernel.org
Cc:     kuba@...nel.org, davem@...emloft.net, ilias.apalodimas@...aro.org,
        hawk@...nel.org, Joe Damato <jdamato@...tly.com>
Subject: [PATCH 6/6] net: page_pool: Add a stat tracking waived pages.

Track how often pages obtained from the ring cannot be added to the cache
because of a NUMA mismatch. A static inline wrapper is added for accessing
this stat.

Signed-off-by: Joe Damato <jdamato@...tly.com>
---
 include/net/page_pool.h | 11 +++++++++++
 net/core/page_pool.c    |  1 +
 2 files changed, 12 insertions(+)

diff --git a/include/net/page_pool.h b/include/net/page_pool.h
index a68d05f..cf65d78 100644
--- a/include/net/page_pool.h
+++ b/include/net/page_pool.h
@@ -88,6 +88,7 @@ struct page_pool_stats {
 			    * slow path allocation
 			    */
 		u64 refill; /* allocations via successful refill */
+		u64 waive;  /* failed refills due to numa zone mismatch */
 	} alloc;
 };
 
@@ -226,6 +227,11 @@ static inline u64 page_pool_stats_get_refill(struct page_pool *pool)
 {
 	return pool->ps.alloc.refill;
 }
+
+static inline u64 page_pool_stats_get_waive(struct page_pool *pool)
+{
+	return pool->ps.alloc.waive;
+}
 #else
 static inline void page_pool_destroy(struct page_pool *pool)
 {
@@ -270,6 +276,11 @@ static inline u64 page_pool_stats_get_refill(struct page_pool *pool)
 {
 	return 0;
 }
+
+static inline u64 page_pool_stats_get_waive(struct page_pool *pool)
+{
+	return 0;
+}
 #endif
 
 void page_pool_put_page(struct page_pool *pool, struct page *page,
diff --git a/net/core/page_pool.c b/net/core/page_pool.c
index 15f4e73..7c4ae2e 100644
--- a/net/core/page_pool.c
+++ b/net/core/page_pool.c
@@ -147,6 +147,7 @@ static struct page *page_pool_refill_alloc_cache(struct page_pool *pool)
 			 * This limit stress on page buddy alloactor.
 			 */
 			page_pool_return_page(pool, page);
+			pool->ps.alloc.waive++;
 			page = NULL;
 			break;
 		}
-- 
2.7.4