linux/include/net/sch_generic.h, branch v3.8

pkt_sched: avoid requeues if possible

2012-12-12T05:16:47Z

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 Cc: Jamal Hadi Salim Cc: John Fastabend Signed-off-by: David S. Miller

net sched: Pass the skb into change so it can access NETLINK_CB

2012-08-15T04:55:28Z

cls_flow.c plays with uids and gids. Unless I misread that code it is possible for classifiers to depend on the specific uid and gid values. Therefore I need to know the user namespace of the netlink socket that is installing the packet classifiers. Pass in the rtnetlink skb so I can access the NETLINK_CB of the passed packet. In particular I want access to sk_user_ns(NETLINK_CB(in_skb).ssk). Pass in not the user namespace but the incomming rtnetlink skb into the the classifier change routines as that is generally the more useful parameter. Cc: Jamal Hadi Salim Acked-by: David S. Miller Acked-by: Serge Hallyn Signed-off-by: Eric W. Biederman

net: rename bond_queue_mapping to slave_dev_queue_mapping

2012-07-20T18:07:00Z

As this is going to be used not only by bonding. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller

bonding: Fix corrupted queue_mapping

2012-06-12T22:29:21Z

In the transmit path of the bonding driver, skb->cb is used to stash the skb->queue_mapping so that the bonding device can set its own queue mapping. This value becomes corrupted since the skb->cb is also used in __dev_xmit_skb. When transmitting through bonding driver, bond_select_queue is called from dev_queue_xmit. In bond_select_queue the original skb->queue_mapping is copied into skb->cb (via bond_queue_mapping) and skb->queue_mapping is overwritten with the bond driver queue. Subsequently in dev_queue_xmit, __dev_xmit_skb is called which writes the packet length into skb->cb, thereby overwriting the stashed queue mappping. In bond_dev_queue_xmit (called from hard_start_xmit), the queue mapping for the skb is set to the stashed value which is now the skb length and hence is an invalid queue for the slave device. If we want to save skb->queue_mapping into skb->cb[], best place is to add a field in struct qdisc_skb_cb, to make sure it wont conflict with other layers (eg : Qdiscc, Infiniband...) This patchs also makes sure (struct qdisc_skb_cb)->data is aligned on 8 bytes : netem qdisc for example assumes it can store an u64 in it, without misalignment penalty. Note : we only have 20 bytes left in (struct qdisc_skb_cb)->data[]. The largest user is CHOKe and it fills it. Based on a previous patch from Tom Herbert. Signed-off-by: Eric Dumazet Reported-by: Tom Herbert Cc: John Fastabend Cc: Roland Dreier Acked-by: Neil Horman Signed-off-by: David S. Miller

net: Make qdisc_skb_cb upper size bound explicit.

2012-02-09T18:50:34Z

Just like skb->cb[], so that qdisc_skb_cb can be encapsulated inside of other data structures. This is intended to be used by IPoIB so that it can remember addressing information stored at hard_header_ops->create() time that it can fetch when the packet gets to the transmit routine. Signed-off-by: David S. Miller

net: sch_generic remove redundant use of

2011-10-31T23:32:25Z

This file has modular references, but they are limited to those which are covered by the simple "struct module;" declaration used in dozens of other places. In fact that declaration is already there (just outside of the context of this commit) so simply remove the include line. Signed-off-by: Paul Gortmaker

net: constify skbuff and Qdisc elements

2011-10-20T21:45:43Z

Preliminary patch before tcp constification Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller

net: sched: constify tcf_proto and tc_action

2011-07-06T09:52:16Z

Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller

net_sched: fix THROTTLED/RUNNING race

2011-03-24T07:13:14Z

commit fd245a4adb52 (net_sched: move TCQ_F_THROTTLED flag) added a race. qdisc_watchdog() is run from softirq, so special care should be taken or we can lose one state transition (THROTTLED/RUNNING) Prior to fd245a4adb52, we were manipulating q->flags (qdisc->flags &= ~TCQ_F_THROTTLED;) and this manipulation could only race with qdisc_warn_nonwc(). Since we want to avoid atomic ops in qdisc fast path - it was the meaning of commit 371121057607e (QDISC_STATE_RUNNING dont need atomic bit ops) - fix is to move THROTTLE bit into 'state' field, this one being manipulated with SMP and IRQ safe operations. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller

Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6

2011-03-04T05:27:42Z

Conflicts: drivers/net/bnx2x/bnx2x.h