Files
linux-kernel-test/include/linux
Eric Dumazet 7f5c6d4f66 netfilter: get rid of atomic ops in fast path
We currently use a percpu spinlock to 'protect' rule bytes/packets
counters, after various attempts to use RCU instead.

Lately we added a seqlock so that get_counters() can run without
blocking BH or 'writers'. But we really only need the seqcount in it.

Spinlock itself is only locked by the current/owner cpu, so we can
remove it completely.

This cleanups api, using correct 'writer' vs 'reader' semantic.

At replace time, the get_counters() call makes sure all cpus are done
using the old table.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Jan Engelhardt <jengelh@medozas.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
2011-04-04 17:04:03 +02:00
..
2011-03-27 23:34:59 -07:00
2011-03-23 10:42:00 +01:00
2011-03-18 23:38:50 -07:00
2011-03-22 17:43:59 -07:00
2011-03-22 17:44:15 -07:00
2011-03-23 19:46:17 -07:00
2011-03-22 17:44:05 -07:00
2011-03-20 14:50:53 +01:00
2011-03-28 06:05:21 -04:00
2011-03-30 00:37:41 +02:00
2011-03-29 14:48:19 +02:00
2011-03-22 04:53:54 -03:00
2011-03-22 17:43:59 -07:00
2011-03-23 15:29:04 -04:00
2011-03-22 04:53:46 -03:00
2011-03-23 19:46:43 -07:00
2011-03-22 17:44:03 -07:00
2011-03-23 07:48:42 -07:00
2011-03-22 17:44:05 -07:00