Eric Dumazet 1abbe1394a pkt_sched: avoid requeues if possible ... With BQL being deployed, we can more likely have following behavior : We dequeue a packet from qdisc in dequeue_skb(), then we realize target tx queue is in XOFF state in sch_direct_xmit(), and we have to hold the skb into gso_skb for later. This shows in stats (tc -s qdisc dev eth0) as requeues. Problem of these requeues is that high priority packets can not be dequeued as long as this (possibly low prio and big TSO packet) is not removed from gso_skb. At 1Gbps speed, a full size TSO packet is 500 us of extra latency. In some cases, we know that all packets dequeued from a qdisc are for a particular and known txq : - If device is non multi queue - For all MQ/MQPRIO slave qdiscs This patch introduces a new qdisc flag, TCQ_F_ONETXQUEUE to mark this capability, so that dequeue_skb() is allowed to dequeue a packet only if the associated txq is not stopped. This indeed reduce latencies for high prio packets (or improve fairness with sfq/fq_codel), and almost remove qdisc 'requeues'. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: John Fastabend <john.r.fastabend@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>		2012-12-12 00:16:47 -05:00
..
act_api.c	net: Push capable(CAP_NET_ADMIN) into the rtnl methods	2012-11-18 20:32:44 -05:00
act_csum.c
act_gact.c
act_ipt.c
act_mirred.c	act_mirred: do not drop packets when fails to mirror it	2012-08-16 14:54:44 -07:00
act_nat.c
act_pedit.c
act_police.c
act_simple.c
act_skbedit.c
cls_api.c	net: Push capable(CAP_NET_ADMIN) into the rtnl methods	2012-11-18 20:32:44 -05:00
cls_basic.c
cls_cgroup.c	cgroup: net_cls: Rework update socket logic	2012-10-26 03:40:51 -04:00
cls_flow.c	userns: Convert cls_flow to work with user namespaces enabled	2012-08-14 21:55:28 -07:00
cls_fw.c
cls_route.c
cls_rsvp.c
cls_rsvp.h
cls_rsvp6.c
cls_tcindex.c
cls_u32.c
em_canid.c
em_cmp.c
em_ipset.c
em_meta.c	net: use a per task frag allocator	2012-09-24 16:31:37 -04:00
em_nbyte.c
em_text.c
em_u32.c
ematch.c
Kconfig	net: sched: enable CAN Identifier to be build into kernel	2012-11-25 16:06:06 -05:00
Makefile
sch_api.c	pkt_sched: avoid requeues if possible	2012-12-12 00:16:47 -05:00
sch_atm.c
sch_blackhole.c
sch_cbq.c	pkt_sched: use ns_to_ktime() helper	2012-10-21 22:21:27 -04:00
sch_choke.c
sch_codel.c
sch_drr.c	pkt_sched: Fix warning false positives.	2012-09-27 18:35:47 -04:00
sch_dsmark.c
sch_fifo.c
sch_fq_codel.c	fq_codel: dont reinit flow state	2012-09-03 14:36:50 -04:00
sch_generic.c	pkt_sched: avoid requeues if possible	2012-12-12 00:16:47 -05:00
sch_gred.c	net_sched: gred: actually perform idling in WRED mode	2012-09-13 16:10:13 -04:00
sch_hfsc.c
sch_htb.c	htb: fix two bugs	2012-11-06 19:06:29 -05:00
sch_ingress.c
sch_mq.c	pkt_sched: avoid requeues if possible	2012-12-12 00:16:47 -05:00
sch_mqprio.c	pkt_sched: avoid requeues if possible	2012-12-12 00:16:47 -05:00
sch_multiq.c
sch_netem.c
sch_plug.c
sch_prio.c
sch_qfq.c	pkt_sched: QFQ Plus: fair-queueing service at DRR cost	2012-11-28 11:19:35 -05:00
sch_red.c
sch_sfb.c
sch_sfq.c
sch_tbf.c
sch_teql.c