On 8/8/17 12:23 PM, Vincent Bernat wrote:
> If the user hasn't installed any custom rules, don't go through the
> whole FIB rules layer. This is pretty similar to f4530fa574df (ipv4:
> Avoid overhead when no custom FIB rules are installed).
> Using a micro-benchmark module , timing ip6_route_output() with
> get_cycles(), with 40,000 routes in the main routing table, before this
> At the frequency of the host during the bench (~ 3.7 GHz), this is
> about a 100 ns difference on the median value.
> A next step would be to collapse local and main tables, as in
> 0ddcf43d5d4a (ipv4: FIB Local/MAIN table collapse).
> Signed-off-by: Vincent Bernat <vinc...@bernat.im>
> Reviewed-by: Jiri Pirko <j...@mellanox.com>
> include/net/netns/ipv6.h | 1 +
> net/ipv6/fib6_rules.c | 40 +++++++++++++++++++++++++++-------------
> net/ipv6/route.c | 1 +
> 3 files changed, 29 insertions(+), 13 deletions(-)
Acked-by: David Ahern <dsah...@gmail.com>