linux-toradex.git/net/sched/sch_api.c, branch v3.18-rc4

net: sched: initialize bstats syncp

2014-10-22T01:45:21+00:00

Use netdev_alloc_pcpu_stats to allocate percpu stats and initialize syncp.

Fixes: 22e0f8b9322c "net: sched: make bstats per cpu and estimator RCU safe"
Signed-off-by: Sabrina Dubroca 
Acked-by: Cong Wang 
Signed-off-by: David S. Miller

net: sched: avoid costly atomic operation in fq_dequeue()

2014-10-06T04:55:10+00:00

Standard qdisc API to setup a timer implies an atomic operation on every
packet dequeue : qdisc_unthrottled()

It turns out this is not really needed for FQ, as FQ has no concept of
global qdisc throttling, being a qdisc handling many different flows,
some of them can be throttled, while others are not.

Fix is straightforward : add a 'bool throttle' to
qdisc_watchdog_schedule_ns(), and remove calls to qdisc_unthrottled()
in sch_fq.

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

net: sched: suspicious RCU usage in qdisc_watchdog

2014-10-05T00:45:54+00:00

Suspicious RCU usage in qdisc_watchdog call needs to be done inside
rcu_read_lock/rcu_read_unlock. And then Qdisc destroy operations
need to ensure timer is cancelled before removing qdisc structure.

[ 3992.191339] ===============================
[ 3992.191340] [ INFO: suspicious RCU usage. ]
[ 3992.191343] 3.17.0-rc6net-next+ #72 Not tainted
[ 3992.191345] -------------------------------
[ 3992.191347] include/net/sch_generic.h:272 suspicious rcu_dereference_check() usage!
[ 3992.191348]
[ 3992.191348] other info that might help us debug this:
[ 3992.191348]
[ 3992.191351]
[ 3992.191351] rcu_scheduler_active = 1, debug_locks = 1
[ 3992.191353] no locks held by swapper/1/0.
[ 3992.191355]
[ 3992.191355] stack backtrace:
[ 3992.191358] CPU: 1 PID: 0 Comm: swapper/1 Not tainted 3.17.0-rc6net-next+ #72
[ 3992.191360] Hardware name:                  /DZ77RE-75K, BIOS GAZ7711H.86A.0060.2012.1115.1750 11/15/2012
[ 3992.191362]  0000000000000001 ffff880235803e48 ffffffff8178f92c 0000000000000000
[ 3992.191366]  ffff8802322224a0 ffff880235803e78 ffffffff810c9966 ffff8800a5fe3000
[ 3992.191370]  ffff880235803f30 ffff8802359cd768 ffff8802359cd6e0 ffff880235803e98
[ 3992.191374] Call Trace:
[ 3992.191376]    [] dump_stack+0x4e/0x68
[ 3992.191387]  [] lockdep_rcu_suspicious+0xe6/0x130
[ 3992.191392]  [] qdisc_watchdog+0x8a/0xb0
[ 3992.191396]  [] __run_hrtimer+0x72/0x420
[ 3992.191399]  [] ? hrtimer_interrupt+0x7d/0x240
[ 3992.191403]  [] ? tc_classify+0xc0/0xc0
[ 3992.191406]  [] hrtimer_interrupt+0xff/0x240
[ 3992.191410]  [] ? __atomic_notifier_call_chain+0x5/0x140
[ 3992.191415]  [] local_apic_timer_interrupt+0x3b/0x60
[ 3992.191419]  [] smp_apic_timer_interrupt+0x45/0x60
[ 3992.191422]  [] apic_timer_interrupt+0x6f/0x80
[ 3992.191424]    [] ? cpuidle_enter_state+0x73/0x2e0
[ 3992.191432]  [] ? cpuidle_enter_state+0x6e/0x2e0
[ 3992.191437]  [] cpuidle_enter+0x17/0x20
[ 3992.191441]  [] cpu_startup_entry+0x3d1/0x4a0
[ 3992.191445]  [] ? clockevents_config_and_register+0x26/0x30
[ 3992.191448]  [] start_secondary+0x1b6/0x260

Fixes: b26b0d1e8b1 ("net: qdisc: use rcu prefix and silence sparse warnings")
Signed-off-by: John Fastabend 
Acked-by: Cong Wang 
Signed-off-by: David S. Miller

net: sched: enable per cpu qstats

2014-09-30T05:02:26+00:00

After previous patches to simplify qstats the qstats can be
made per cpu with a packed union in Qdisc struct.

Signed-off-by: John Fastabend 
Signed-off-by: David S. Miller

net: sched: restrict use of qstats qlen

2014-09-30T05:02:26+00:00

This removes the use of qstats->qlen variable from the classifiers
and makes it an explicit argument to gnet_stats_copy_queue().

The qlen represents the qdisc queue length and is packed into
the qstats at the last moment before passnig to user space. By
handling it explicitely we avoid, in the percpu stats case, having
to figure out which per_cpu variable to put it in.

It would probably be best to remove it from qstats completely
but qstats is a user space ABI and can't be broken. A future
patch could make an internal only qstats structure that would
avoid having to allocate an additional u32 variable on the
Qdisc struct. This would make the qstats struct 128bits instead
of 128+32.

Signed-off-by: John Fastabend 
Signed-off-by: David S. Miller

net: sched: implement qstat helper routines

2014-09-30T05:02:26+00:00

This adds helpers to manipulate qstats logic and replaces locations
that touch the counters directly. This simplifies future patches
to push qstats onto per cpu counters.

Signed-off-by: John Fastabend 
Signed-off-by: David S. Miller

net: sched: make bstats per cpu and estimator RCU safe

2014-09-30T05:02:26+00:00

In order to run qdisc's without locking statistics and estimators
need to be handled correctly.

To resolve bstats make the statistics per cpu. And because this is
only needed for qdiscs that are running without locks which is not
the case for most qdiscs in the near future only create percpu
stats when qdiscs set the TCQ_F_CPUSTATS flag.

Next because estimators use the bstats to calculate packets per
second and bytes per second the estimator code paths are updated
to use the per cpu statistics.

Signed-off-by: John Fastabend 
Signed-off-by: David S. Miller

net: sched: use pinned timers

2014-09-26T04:26:48+00:00

While using a MQ + NETEM setup, I had confirmation that the default
timer migration ( /proc/sys/kernel/timer_migration ) is killing us.

Installing this on a receiver side of a TCP_STREAM test, (NIC has 8 TX
queues) :

EST="est 1sec 4sec"
for ETH in eth1
do
 tc qd del dev $ETH root 2>/dev/null
 tc qd add dev $ETH root handle 1: mq
 tc qd add dev $ETH parent 1:1 $EST netem limit 70000 delay 6ms
 tc qd add dev $ETH parent 1:2 $EST netem limit 70000 delay 8ms
 tc qd add dev $ETH parent 1:3 $EST netem limit 70000 delay 10ms
 tc qd add dev $ETH parent 1:4 $EST netem limit 70000 delay 12ms
 tc qd add dev $ETH parent 1:5 $EST netem limit 70000 delay 14ms
 tc qd add dev $ETH parent 1:6 $EST netem limit 70000 delay 16ms
 tc qd add dev $ETH parent 1:7 $EST netem limit 80000 delay 18ms
 tc qd add dev $ETH parent 1:8 $EST netem limit 90000 delay 20ms
done

We can see that timers get migrated into a single cpu, presumably idle
at the time timers are set up.
Then all qdisc dequeues run from this cpu and huge lock contention
happens. This single cpu is stuck in softirq mode and cannot dequeue
fast enough.

    39.24%  [kernel]          [k] _raw_spin_lock
     2.65%  [kernel]          [k] netem_enqueue
     1.80%  [kernel]          [k] netem_dequeue
     1.63%  [kernel]          [k] copy_user_enhanced_fast_string
     1.45%  [kernel]          [k] _raw_spin_lock_bh

By pinning qdisc timers on the cpu running the qdisc, we respect proper
XPS setting and remove this lock contention.

     5.84%  [kernel]          [k] netem_enqueue
     4.83%  [kernel]          [k] _raw_spin_lock
     2.92%  [kernel]          [k] copy_user_enhanced_fast_string

Current Qdiscs that benefit from this change are :

	netem, cbq, fq, hfsc, tbf, htb.

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

net: rcu-ify tcf_proto

2014-09-13T16:30:25+00:00

rcu'ify tcf_proto this allows calling tc_classify() without holding
any locks. Updaters are protected by RTNL.

This patch prepares the core net_sched infrastracture for running
the classifier/action chains without holding the qdisc lock however
it does nothing to ensure cls_xxx and act_xxx types also work without
locking. Additional patches are required to address the fall out.

Signed-off-by: John Fastabend 
Acked-by: Eric Dumazet 
Signed-off-by: David S. Miller

net_sched: drr: warn when qdisc is not work conserving

2014-06-11T22:50:59+00:00

The DRR scheduler requires that items on the active list are work
conserving, i.e. do not hold on to skbs for throttling purposes, etc.
Attaching e.g. tbf renders DRR useless because all other classes on the
active list are delayed as well.

So, warn users that this configuration won't work as expected; we
already do this in couple of other qdiscs, see e.g.

commit b00355db3f88d96810a60011a30cfb2c3469409d
('pkt_sched: sch_hfsc: sch_htb: Add non-work-conserving warning handler')

The 'const' change is needed to avoid compiler warning ("discards 'const'
qualifier from pointer target type").

tested with:
drr_hier() {
        parent=$1
        classes=$2
        for i in  $(seq 1 $classes); do
                classid=$parent$(printf %x $i)
                tc class add dev eth0 parent $parent classid $classid drr
		tc qdisc add dev eth0 parent $classid tbf rate 64kbit burst 256kbit limit 64kbit
        done
}
tc qdisc add dev eth0 root handle 1: drr
drr_hier 1: 32
tc filter add dev eth0 protocol all pref 1 parent 1: handle 1 flow hash keys dst perturb 1 divisor 32

Signed-off-by: Florian Westphal 
Signed-off-by: David S. Miller