diff options
| author | Eric Dumazet <eric.dumazet@gmail.com> | 2011-01-09 03:30:54 -0500 |
|---|---|---|
| committer | David S. Miller <davem@davemloft.net> | 2011-01-10 19:07:54 -0500 |
| commit | bfe0d0298f2a67d94d58c39ea904a999aeeb7c3c (patch) | |
| tree | 5a6f966ebabe8d88d6d6c78c61411e325150927f /net/sched/act_ipt.c | |
| parent | f1593d2298acca8b6680100d622911827edb8b0a (diff) | |
net_sched: factorize qdisc stats handling
HTB takes into account skb is segmented in stats updates.
Generalize this to all schedulers.
They should use qdisc_bstats_update() helper instead of manipulating
bstats.bytes and bstats.packets
Add bstats_update() helper too for classes that use
gnet_stats_basic_packed fields.
Note : Right now, TCQ_F_CAN_BYPASS shortcurt can be taken only if no
stab is setup on qdisc.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/sched/act_ipt.c')
| -rw-r--r-- | net/sched/act_ipt.c | 3 |
1 files changed, 1 insertions, 2 deletions
diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c index 8daef9632255..c2a7c20e81c1 100644 --- a/net/sched/act_ipt.c +++ b/net/sched/act_ipt.c | |||
| @@ -209,8 +209,7 @@ static int tcf_ipt(struct sk_buff *skb, struct tc_action *a, | |||
| 209 | spin_lock(&ipt->tcf_lock); | 209 | spin_lock(&ipt->tcf_lock); |
| 210 | 210 | ||
| 211 | ipt->tcf_tm.lastuse = jiffies; | 211 | ipt->tcf_tm.lastuse = jiffies; |
| 212 | ipt->tcf_bstats.bytes += qdisc_pkt_len(skb); | 212 | bstats_update(&ipt->tcf_bstats, skb); |
| 213 | ipt->tcf_bstats.packets++; | ||
| 214 | 213 | ||
| 215 | /* yes, we have to worry about both in and out dev | 214 | /* yes, we have to worry about both in and out dev |
| 216 | worry later - danger - this API seems to have changed | 215 | worry later - danger - this API seems to have changed |