kernel.git/net/sched/sch_generic.c, branch linux-3.8.y

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: qdisc busylock needs lockdep annotations

2012-09-05T21:49:27Z

It seems we need to provide ability for stacked devices to use specific lock_class_key for sch->busylock We could instead default l2tpeth tx_queue_len to 0 (no qdisc), but a user might use a qdisc anyway. (So same fixes are probably needed on non LLTX stacked drivers) Noticed while stressing L2TPV3 setup : ====================================================== [ INFO: possible circular locking dependency detected ] 3.6.0-rc3+ #788 Not tainted ------------------------------------------------------- netperf/4660 is trying to acquire lock: (l2tpsock){+.-...}, at: [] l2tp_xmit_skb+0x172/0xa50 [l2tp_core] but task is already holding lock: (&(&sch->busylock)->rlock){+.-...}, at: [] dev_queue_xmit+0xd75/0xe00 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (&(&sch->busylock)->rlock){+.-...}: [] lock_acquire+0x90/0x200 [] _raw_spin_lock_irqsave+0x4c/0x60 [] __wake_up+0x32/0x70 [] tty_wakeup+0x3e/0x80 [] pty_write+0x73/0x80 [] tty_put_char+0x3c/0x40 [] process_echoes+0x142/0x330 [] n_tty_receive_buf+0x8fb/0x1230 [] flush_to_ldisc+0x142/0x1c0 [] process_one_work+0x198/0x760 [] worker_thread+0x186/0x4b0 [] kthread+0x93/0xa0 [] kernel_thread_helper+0x4/0x10 -> #0 (l2tpsock){+.-...}: [] __lock_acquire+0x1628/0x1b10 [] lock_acquire+0x90/0x200 [] _raw_spin_lock+0x41/0x50 [] l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [] l2tp_eth_dev_xmit+0x32/0x60 [l2tp_eth] [] dev_hard_start_xmit+0x502/0xa70 [] sch_direct_xmit+0xfe/0x290 [] dev_queue_xmit+0x1e5/0xe00 [] ip_finish_output+0x3d0/0x890 [] ip_output+0x59/0xf0 [] ip_local_out+0x2d/0xa0 [] ip_queue_xmit+0x1c3/0x680 [] tcp_transmit_skb+0x402/0xa60 [] tcp_write_xmit+0x1f4/0xa30 [] tcp_push_one+0x30/0x40 [] tcp_sendmsg+0xe82/0x1040 [] inet_sendmsg+0x125/0x230 [] sock_sendmsg+0xdc/0xf0 [] sys_sendto+0xfe/0x130 [] system_call_fastpath+0x16/0x1b Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(&(&sch->busylock)->rlock); lock(l2tpsock); lock(&(&sch->busylock)->rlock); lock(l2tpsock); *** DEADLOCK *** 5 locks held by netperf/4660: #0: (sk_lock-AF_INET){+.+.+.}, at: [] tcp_sendmsg+0x2c/0x1040 #1: (rcu_read_lock){.+.+..}, at: [] ip_queue_xmit+0x0/0x680 #2: (rcu_read_lock_bh){.+....}, at: [] ip_finish_output+0x135/0x890 #3: (rcu_read_lock_bh){.+....}, at: [] dev_queue_xmit+0x0/0xe00 #4: (&(&sch->busylock)->rlock){+.-...}, at: [] dev_queue_xmit+0xd75/0xe00 stack backtrace: Pid: 4660, comm: netperf Not tainted 3.6.0-rc3+ #788 Call Trace: [] print_circular_bug+0x1fb/0x20c [] __lock_acquire+0x1628/0x1b10 [] ? check_usage+0x9b/0x4d0 [] ? __lock_acquire+0x2e4/0x1b10 [] lock_acquire+0x90/0x200 [] ? l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [] _raw_spin_lock+0x41/0x50 [] ? l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [] l2tp_xmit_skb+0x172/0xa50 [l2tp_core] [] l2tp_eth_dev_xmit+0x32/0x60 [l2tp_eth] [] dev_hard_start_xmit+0x502/0xa70 [] ? dev_hard_start_xmit+0x5e/0xa70 [] ? dev_queue_xmit+0x141/0xe00 [] sch_direct_xmit+0xfe/0x290 [] dev_queue_xmit+0x1e5/0xe00 [] ? dev_hard_start_xmit+0xa70/0xa70 [] ip_finish_output+0x3d0/0x890 [] ? ip_finish_output+0x135/0x890 [] ip_output+0x59/0xf0 [] ip_local_out+0x2d/0xa0 [] ip_queue_xmit+0x1c3/0x680 [] ? ip_local_out+0xa0/0xa0 [] tcp_transmit_skb+0x402/0xa60 [] ? tcp_md5_do_lookup+0x18e/0x1a0 [] tcp_write_xmit+0x1f4/0xa30 [] tcp_push_one+0x30/0x40 [] tcp_sendmsg+0xe82/0x1040 [] inet_sendmsg+0x125/0x230 [] ? inet_create+0x6b0/0x6b0 [] ? sock_update_classid+0xc2/0x3b0 [] ? sock_update_classid+0x130/0x3b0 [] sock_sendmsg+0xdc/0xf0 [] ? fget_light+0x3f9/0x4f0 [] sys_sendto+0xfe/0x130 [] ? trace_hardirqs_on+0xd/0x10 [] ? _raw_spin_unlock_irq+0x30/0x50 [] ? finish_task_switch+0x83/0xf0 [] ? finish_task_switch+0x46/0xf0 [] ? sysret_check+0x1b/0x56 [] system_call_fastpath+0x16/0x1b Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller

net: move and rename netif_notify_peers()

2012-08-14T21:28:23Z

I believe net/core/dev.c is a better place for netif_notify_peers(), because other net event notify functions also stay in this file. And rename it to netdev_notify_peers(). Cc: David S. Miller Cc: Ian Campbell Signed-off-by: Cong Wang Signed-off-by: David S. Miller

net: Convert net_ratelimit uses to net__ratelimited

2012-05-15T17:45:03Z

Standardize the net core ratelimited logging functions. Coalesce formats, align arguments. Change a printk then vprintk sequence to use printf extension %pV. Signed-off-by: Joe Perches Signed-off-by: David S. Miller

pkt_sched: Stop using NLA_PUT*().

2012-04-01T22:11:37Z

These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller

net: Add queue state xoff flag for stack

2011-11-29T17:46:19Z

Create separate queue state flags so that either the stack or drivers can turn on XOFF. Added a set of functions used in the stack to determine if a queue is really stopped (either by stack or driver) Signed-off-by: Tom Herbert Acked-by: Eric Dumazet Signed-off-by: David S. Miller

net: new counter for tx_timeout errors in sysfs

2011-11-17T04:14:02Z

This adds the /sys/class/net/DEV/queues/Q/tx_timeout attribute containing the total number of timeout events on the given queue. It is always available with CONFIG_SYSFS, independently of CONFIG_RPS/XPS. Credits to Stephen Hemminger for a preliminary version of this patch. Tested: without CONFIG_SYSFS (compilation only) with sysfs and without CONFIG_RPS & CONFIG_XPS with sysfs and without CONFIG_RPS with sysfs and without CONFIG_XPS with defaults Signed-off-by: David Decotigny Signed-off-by: David S. Miller

Remove redundant variable/code in __qdisc_run

2011-07-15T15:08:26Z

Remove redundant variable "work". Signed-off-by: Krishna Kumar Signed-off-by: David S. Miller

net_sched: fix dequeuer fairness

2011-06-27T07:14:10Z

Results on dummy device can be seen in my netconf 2011 slides. These results are for a 10Gige IXGBE intel nic - on another i5 machine, very similar specs to the one used in the netconf2011 results. It turns out - this is a hell lot worse than dummy and so this patch is even more beneficial for 10G. Test setup: ---------- System under test sending packets out. Additional box connected directly dropping packets. Installed prio qdisc on the eth device and default netdev default length of 1000 used as is. The 3 prio bands each were set to 100 (didnt factor in the results). 5 packet runs were made and the middle 3 picked. results ------- The "cpu" column indicates the which cpu the sample was taken on, The "Pkt runx" carries the number of packets a cpu dequeued when forced to be in the "dequeuer" role. The "avg" for each run is the number of times each cpu should be a "dequeuer" if the system was fair. 3.0-rc4 (plain) cpu Pkt run1 Pkt run2 Pkt run3 ================================================ cpu0 21853354 21598183 22199900 cpu1 431058 473476 393159 cpu2 481975 477529 458466 cpu3 23261406 23412299 22894315 avg 11506948 11490372 11486460 3.0-rc4 with patch and default weight 64 cpu Pkt run1 Pkt run2 Pkt run3 ================================================ cpu0 13205312 13109359 13132333 cpu1 10189914 10159127 10122270 cpu2 10213871 10124367 10168722 cpu3 13165760 13164767 13096705 avg 11693714 11639405 11630008 As you can see the system is still not perfect but is a lot better than what it was before... At the moment we use the old backlog weight, weight_p which is 64 packets. It seems to be reasonably fine with that value. The system could be made more fair if we reduce the weight_p (as per my presentation), but we are going to affect the shared backlog weight. Unless deemed necessary, I think the default value is fine. If not we could add yet another knob. Signed-off-by: Jamal Hadi Salim Signed-off-by: David S. Miller

net: Rework netdev_drivername() to avoid warning.

2011-06-06T23:41:33Z

This interface uses a temporary buffer, but for no real reason. And now can generate warnings like: net/sched/sch_generic.c: In function dev_watchdog net/sched/sch_generic.c:254:10: warning: unused variable drivername Just return driver->name directly or "". Reported-by: Connor Hansen Signed-off-by: David S. Miller