netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] net: shrink struct softnet_data
@ 2014-11-02 14:00 Eric Dumazet
  2014-11-03 17:25 ` David Miller
  0 siblings, 1 reply; 2+ messages in thread
From: Eric Dumazet @ 2014-11-02 14:00 UTC (permalink / raw)
  To: David Miller; +Cc: netdev

From: Eric Dumazet <edumazet@google.com>

flow_limit in struct softnet_data is only read from local cpu
and can be moved to fill a hole, reducing softnet_data size by
64 bytes on x86_64

While we are at it, move output_queue, output_queue_tailp and
completion_queue, so that rx / tx paths touch a single cache line.

Signed-off-by: Eric Dumazet <edumazet@google.com>
---
 include/linux/netdevice.h |   15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index c85e065122460e9f077bcb6788018be38e1d7ddf..5ed05bd764dcf3699afdd9a7b17600246de22d1d 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -2329,10 +2329,7 @@ extern int netdev_flow_limit_table_len;
  * Incoming packets are placed on per-cpu queues
  */
 struct softnet_data {
-	struct Qdisc		*output_queue;
-	struct Qdisc		**output_queue_tailp;
 	struct list_head	poll_list;
-	struct sk_buff		*completion_queue;
 	struct sk_buff_head	process_queue;
 
 	/* stats */
@@ -2340,10 +2337,17 @@ struct softnet_data {
 	unsigned int		time_squeeze;
 	unsigned int		cpu_collision;
 	unsigned int		received_rps;
-
 #ifdef CONFIG_RPS
 	struct softnet_data	*rps_ipi_list;
+#endif
+#ifdef CONFIG_NET_FLOW_LIMIT
+	struct sd_flow_limit __rcu *flow_limit;
+#endif
+	struct Qdisc		*output_queue;
+	struct Qdisc		**output_queue_tailp;
+	struct sk_buff		*completion_queue;
 
+#ifdef CONFIG_RPS
 	/* Elements below can be accessed between CPUs for RPS */
 	struct call_single_data	csd ____cacheline_aligned_in_smp;
 	struct softnet_data	*rps_ipi_next;
@@ -2355,9 +2359,6 @@ struct softnet_data {
 	struct sk_buff_head	input_pkt_queue;
 	struct napi_struct	backlog;
 
-#ifdef CONFIG_NET_FLOW_LIMIT
-	struct sd_flow_limit __rcu *flow_limit;
-#endif
 };
 
 static inline void input_queue_head_incr(struct softnet_data *sd)

^ permalink raw reply related	[flat|nested] 2+ messages in thread

* Re: [PATCH] net: shrink struct softnet_data
  2014-11-02 14:00 [PATCH] net: shrink struct softnet_data Eric Dumazet
@ 2014-11-03 17:25 ` David Miller
  0 siblings, 0 replies; 2+ messages in thread
From: David Miller @ 2014-11-03 17:25 UTC (permalink / raw)
  To: eric.dumazet; +Cc: netdev

From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Sun, 02 Nov 2014 06:00:12 -0800

> From: Eric Dumazet <edumazet@google.com>
> 
> flow_limit in struct softnet_data is only read from local cpu
> and can be moved to fill a hole, reducing softnet_data size by
> 64 bytes on x86_64
> 
> While we are at it, move output_queue, output_queue_tailp and
> completion_queue, so that rx / tx paths touch a single cache line.
> 
> Signed-off-by: Eric Dumazet <edumazet@google.com>

Applied.

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2014-11-03 17:25 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2014-11-02 14:00 [PATCH] net: shrink struct softnet_data Eric Dumazet
2014-11-03 17:25 ` David Miller

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).