net: do not consume a cacheline for system_page_pool
authorEric Dumazet <edumazet@google.com>
Thu, 28 Mar 2024 17:34:48 +0000 (17:34 +0000)
committerJakub Kicinski <kuba@kernel.org>
Fri, 29 Mar 2024 19:27:05 +0000 (12:27 -0700)
There is no reason to consume a full cacheline to store system_page_pool.

We can eventually move it to softnet_data later for full locality control.

Fixes: 2b0cfa6e4956 ("net: add generic percpu page_pool allocator")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Lorenzo Bianconi <lorenzo@kernel.org>
Cc: Toke Høiland-Jørgensen <toke@redhat.com>
Acked-by: Jesper Dangaard Brouer <hawk@kernel.org>
Link: https://lore.kernel.org/r/20240328173448.2262593-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
net/core/dev.c

index 9a67003e49db87f3f92b6c6296b3e7a5ca9d9171..984ff8b9d0e1aa5646a7237a8cf0b0a21c2aa559 100644 (file)
@@ -429,7 +429,7 @@ EXPORT_PER_CPU_SYMBOL(softnet_data);
  * PP consumers must pay attention to run APIs in the appropriate context
  * (e.g. NAPI context).
  */
-static DEFINE_PER_CPU_ALIGNED(struct page_pool *, system_page_pool);
+static DEFINE_PER_CPU(struct page_pool *, system_page_pool);
 
 #ifdef CONFIG_LOCKDEP
 /*