Eric Dumazet 2359a47671 codel: refine one condition to avoid a nul rec_inv_sqrt
One condition before codel_Newton_step() was not good if
we never left the dropping state for a flow. As a result
rec_inv_sqrt was 0, instead of the ~0 initial value.

codel control law was then set to a very aggressive mode, dropping
many packets before reaching 'target' and recovering from this problem.

To keep codel_vars_init() as efficient as possible, refine
the condition to make sure rec_inv_sqrt initial value is correct

Many thanks to Anton Mich for discovering the issue and suggesting
a fix.

Reported-by: Anton Mich <lp2s1h@gmail.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-08-10 16:52:54 -07:00
..
2011-03-31 11:26:23 -03:00
2012-07-30 14:53:22 -07:00
2012-07-18 11:28:46 -07:00
2012-06-08 14:27:23 -07:00
2012-04-15 12:44:40 -04:00
2012-04-15 12:44:40 -04:00
2010-06-03 03:21:52 -07:00
2012-04-15 12:44:40 -04:00
2012-07-19 10:43:03 -07:00
2011-03-31 11:26:23 -03:00
2012-04-15 12:44:40 -04:00
2012-08-06 13:33:21 -07:00
2012-07-18 08:59:58 -07:00
2012-08-10 14:08:57 -07:00
2012-04-14 14:47:05 -04:00
2012-07-18 11:28:46 -07:00
2011-03-31 11:26:23 -03:00
2012-05-17 18:45:20 -04:00
2011-07-26 16:49:47 -07:00
2010-06-03 03:21:52 -07:00
2009-01-06 11:38:15 -07:00
2011-05-17 14:16:58 -04:00
2012-07-26 15:50:39 -07:00
2009-03-21 19:06:50 -07:00
2012-05-19 01:08:16 -04:00
2012-07-22 23:58:00 +04:00
2009-10-07 16:39:43 -04:00
2012-04-15 12:44:40 -04:00
2012-05-16 15:17:08 -04:00
2012-04-15 12:44:40 -04:00