Eric Dumazet
ebda37c27d
rps: avoid one atomic in enqueue_to_backlog ...
If CONFIG_SMP=y, then we own a queue spinlock, we can avoid the atomic
test_and_set_bit() from napi_schedule_prep().
We now have same number of atomic ops per netif_rx() calls than with
pre-RPS kernel.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2010-05-17 17:18:50 -07:00
2010-04-05 13:42:54 -07:00
2010-04-11 14:53:53 -07:00
2010-05-16 00:42:15 -07:00
2010-04-11 14:53:53 -07:00
2010-05-01 15:00:15 -07:00
2010-04-20 16:37:13 -07:00
2010-05-10 11:33:10 +02:00
2010-05-15 23:10:15 -07:00
2010-04-28 12:55:15 -07:00
2010-04-13 03:03:14 -07:00
2010-05-17 17:18:50 -07:00
2010-03-30 22:02:32 +09:00
2010-05-01 15:00:15 -07:00
2010-05-10 18:39:28 +02:00
2010-04-11 14:53:53 -07:00
2010-03-30 22:02:32 +09:00
2010-05-02 02:21:44 -07:00
2010-04-26 11:20:32 -07:00
2010-05-16 22:26:58 -07:00
2010-05-17 17:08:08 -07:00
2010-03-30 22:02:32 +09:00
2010-04-20 16:37:13 -07:00
2010-05-01 15:00:15 -07:00
2010-03-30 22:02:32 +09:00
2010-04-23 16:37:33 -07:00
2010-03-30 22:02:32 +09:00
2010-05-12 00:05:35 -07:00
2010-05-12 00:05:35 -07:00
2010-05-14 13:52:30 -07:00
2010-04-06 08:34:06 -07:00
2010-04-11 14:53:53 -07:00
2010-04-20 16:37:13 -07:00
2010-04-21 01:14:25 -07:00
2010-05-01 15:00:15 -07:00
2010-04-27 12:49:13 -07:00
2010-04-11 14:53:53 -07:00
2010-04-20 16:37:13 -07:00
2010-05-01 15:00:15 -07:00
2010-05-11 18:59:21 +02:00
2010-05-16 22:26:58 -07:00
2010-05-12 00:05:35 -07:00
2010-05-12 23:02:29 -07:00
2010-05-03 15:39:58 -07:00
2010-05-11 14:09:10 -07:00
2010-05-11 14:24:55 -04:00
2010-04-27 12:49:13 -07:00
2010-05-15 23:49:26 -07:00
2010-03-30 22:02:32 +09:00
2010-04-03 14:56:02 -07:00
2010-04-03 14:56:02 -07:00
2010-05-01 15:00:15 -07:00
2010-01-17 19:16:02 -08:00
ebda37c27d