diff options
| author | Eric Dumazet <eric.dumazet@gmail.com> | 2011-01-20 08:27:16 +0300 |
|---|---|---|
| committer | David S. Miller <davem@davemloft.net> | 2011-01-21 03:59:32 +0300 |
| commit | fd245a4adb5288eac37250875f237c40a20a1944 (patch) | |
| tree | 1c16670c53dab9d9d05b26a7e7ae8a6a8267e847 /net/sched/sch_cbq.c | |
| parent | 817fb15dfd988d8dda916ee04fa506f0c466b9d6 (diff) | |
| download | linux-fd245a4adb5288eac37250875f237c40a20a1944.tar.xz | |
net_sched: move TCQ_F_THROTTLED flag
In commit 371121057607e (net: QDISC_STATE_RUNNING dont need atomic bit
ops) I moved QDISC_STATE_RUNNING flag to __state container, located in
the cache line containing qdisc lock and often dirtied fields.
I now move TCQ_F_THROTTLED bit too, so that we let first cache line read
mostly, and shared by all cpus. This should speedup HTB/CBQ for example.
Not using test_bit()/__clear_bit()/__test_and_set_bit allows to use an
"unsigned int" for __state container, reducing by 8 bytes Qdisc size.
Introduce helpers to hide implementation details.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: Patrick McHardy <kaber@trash.net>
CC: Jesper Dangaard Brouer <hawk@diku.dk>
CC: Jarek Poplawski <jarkao2@gmail.com>
CC: Jamal Hadi Salim <hadi@cyberus.ca>
CC: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/sched/sch_cbq.c')
| -rw-r--r-- | net/sched/sch_cbq.c | 6 |
1 files changed, 3 insertions, 3 deletions
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c index 4aaf44c95c52..25ed522b2891 100644 --- a/net/sched/sch_cbq.c +++ b/net/sched/sch_cbq.c @@ -351,7 +351,7 @@ cbq_mark_toplevel(struct cbq_sched_data *q, struct cbq_class *cl) { int toplevel = q->toplevel; - if (toplevel > cl->level && !(cl->q->flags & TCQ_F_THROTTLED)) { + if (toplevel > cl->level && !(qdisc_is_throttled(cl->q))) { psched_time_t now; psched_tdiff_t incr; @@ -625,7 +625,7 @@ static enum hrtimer_restart cbq_undelay(struct hrtimer *timer) hrtimer_start(&q->delay_timer, time, HRTIMER_MODE_ABS); } - sch->flags &= ~TCQ_F_THROTTLED; + qdisc_unthrottled(sch); __netif_schedule(qdisc_root(sch)); return HRTIMER_NORESTART; } @@ -974,7 +974,7 @@ cbq_dequeue(struct Qdisc *sch) skb = cbq_dequeue_1(sch); if (skb) { sch->q.qlen--; - sch->flags &= ~TCQ_F_THROTTLED; + qdisc_unthrottled(sch); return skb; } |