提交 · 79cf79abce71eb7dbc40e2f3121048ca5405cb47 · openeuler / raspberrypi-kernel

29 9月, 2014 4 次提交

net: sched: cls_rcvp, complete rcu conversion · 53dfd501

由 John Fastabend 提交于 9月 26, 2014

This completes the cls_rsvp conversion to RCU safe
copy, update semantics.

As a result all cases of tcf_exts_change occur on
empty lists now.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

53dfd501

net_sched: fix another regression in cls_tcindex · 68f6a7c6

由 WANG Cong 提交于 9月 25, 2014

Clearly the following change is not expected:

	-       if (!cp.perfect && !cp.h)
	-               cp.alloc_hash = cp.hash;
	+       if (!cp->perfect && cp->h)
	+               cp->alloc_hash = cp->hash;

Fixes: commit 331b7292 ("net: sched: RCU cls_tcindex")
Cc: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Acked-by: NJohn Fastabend <john.r.fastabend@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

68f6a7c6

net_sched: fix errno in tcindex_set_parms() · 02c5e844

由 WANG Cong 提交于 9月 25, 2014

When kmemdup() fails, we should return -ENOMEM.

Cc: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Acked-by: NJohn Fastabend <john.r.fastabend@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

02c5e844

net_sched: remove the first parameter from tcf_exts_destroy() · 18d0264f

由 WANG Cong 提交于 9月 25, 2014

Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Acked-by: NJamal Hadi Salim <hadi@mojatatu.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

18d0264f

26 9月, 2014 1 次提交

net: sched: use pinned timers · 4a8e320c

由 Eric Dumazet 提交于 9月 20, 2014

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: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

4a8e320c

23 9月, 2014 4 次提交

net: sched: fix compile warning in cls_u32 · a2aeb02a

由 Eric Dumazet 提交于 9月 22, 2014

$ grep CONFIG_CLS_U32_MARK .config
# CONFIG_CLS_U32_MARK is not set

net/sched/cls_u32.c: In function 'u32_change':
net/sched/cls_u32.c:852:1: warning: label 'errout' defined but not used
[-Wunused-label]
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

a2aeb02a

net: sched: cls_u32 changes to knode must appear atomic to readers · de5df632

由 John Fastabend 提交于 9月 19, 2014

Changes to the cls_u32 classifier must appear atomic to the
readers. Before this patch if a change is requested for both
the exts and ifindex, first the ifindex is updated then the
exts with tcf_exts_change(). This opens a small window where
a reader can have a exts chain with an incorrect ifindex. This
violates the the RCU semantics.

Here we resolve this by always passing u32_set_parms() a copy
of the tc_u_knode to work on and then inserting it into the hash
table after the updates have been successfully applied.

Tested with the following short script:

#tc filter add dev p3p2 parent 8001:0 protocol ip prio 99 handle 1: \
	       u32 divisor 256

#tc filter add dev p3p2 parent 8001:0 protocol ip prio 99 \
	       u32 link 1: hashkey mask ffffff00 at 12    \
	       match ip src 192.168.8.0/2

#tc filter add dev p3p2 parent 8001:0 protocol ip prio 102    \
	       handle 1::10 u32 classid 1:2 ht 1: 	      \
	       match ip src 192.168.8.0/8 match ip tos 0x0a 1e

#tc filter change dev p3p2 parent 8001:0 protocol ip prio 102 \
		 handle 1::10 u32 classid 1:2 ht 1:        \
		 match ip src 1.1.0.0/8 match ip tos 0x0b 1e

CC: Eric Dumazet <edumazet@google.com>
CC: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

de5df632

net: cls_u32: fix missed pcpu_success free_percpu · a1ddcfee

由 John Fastabend 提交于 9月 19, 2014

This fixes a missed free_percpu in the unwind code path and when
keys are destroyed.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

a1ddcfee

net: sched: shrink struct qdisc_skb_cb to 28 bytes · 25711786

由 Eric Dumazet 提交于 9月 18, 2014

We cannot make struct qdisc_skb_cb bigger without impacting IPoIB,
or increasing skb->cb[] size.

Commit e0f31d84 ("flow_keys: Record IP layer protocol in
skb_flow_dissect()") broke IPoIB.

Only current offender is sch_choke, and this one do not need an
absolutely precise flow key.

If we store 17 bytes of flow key, its more than enough. (Its the actual
size of flow_keys if it was a packed structure, but we might add new
fields at the end of it later)
Signed-off-by: NEric Dumazet <edumazet@google.com>
Fixes: e0f31d84 ("flow_keys: Record IP layer protocol in skb_flow_dissect()")
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

25711786

20 9月, 2014 2 次提交

net: sched: cls_u32: rcu can not be last node · 4e2840ee

由 John Fastabend 提交于 9月 17, 2014

tc_u32_sel 'sel' in tc_u_knode expects to be the last element in the
structure and pads the structure with tc_u32_key fields for each key.

 kzalloc(sizeof(*n) + s->nkeys*sizeof(struct tc_u32_key), GFP_KERNEL)

CC: Eric Dumazet <edumazet@google.com>
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

4e2840ee

net: sched: use __skb_queue_head_init() where applicable · ab34f648

由 Eric Dumazet 提交于 9月 17, 2014

pfifo_fast and htb use skb lists, without needing their spinlocks.
(They instead use the standard qdisc lock)

We can use __skb_queue_head_init() instead of skb_queue_head_init()
to be consistent.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

ab34f648

17 9月, 2014 7 次提交

net: sched: cls_cgroup need tcf_exts_init in all cases · 9f6c38e7

由 John Fastabend 提交于 9月 16, 2014

This ensures the tcf_exts_init() is called for all cases.

Fixes: 952313bd ("net: sched: cls_cgroup use RCU")
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NCong Wang <cwang@twopensource.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

9f6c38e7

net: sched: cls_fw: add missing tcf_exts_init call in fw_change() · e1f93eb0

由 John Fastabend 提交于 9月 15, 2014

When allocating a new structure we also need to call tcf_exts_init
to initialize exts.

A follow up patch might be in order to remove some of this code
and do tcf_exts_assign(). With this we could remove the
tcf_exts_init/tcf_exts_change pattern for some of the classifiers.
As part of the future tcf_actions RCU series this will need to be
done. For now fix the call here.

Fixes e35a8ee5 ("net: sched: fw use RCU")
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NCong Wang <cwang@twopensource.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

e1f93eb0

net: sched: cls_cgroup fix possible memory leak of 'new' · d14cbfc8

由 John Fastabend 提交于 9月 15, 2014

tree:   git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next.git master
head:   54996b52
commit: c7953ef2 [625/646] net: sched: cls_cgroup use RCU

net/sched/cls_cgroup.c:130 cls_cgroup_change() warn: possible memory leak of 'new'
net/sched/cls_cgroup.c:135 cls_cgroup_change() warn: possible memory leak of 'new'
net/sched/cls_cgroup.c:139 cls_cgroup_change() warn: possible memory leak of 'new'

Fixes: c7953ef2 ("net: sched: cls_cgroup use RCU")
Reported-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NCong Wang <cwang@twopensource.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

d14cbfc8

net: sched: cls_u32 add missing rcu_assign_pointer and annotation · a96366bf

由 John Fastabend 提交于 9月 15, 2014

Add missing rcu_assign_pointer and missing  annotation for ht_up
in cls_u32.c

Caught by kbuild bot,

>> net/sched/cls_u32.c:378:36: sparse: incorrect type in initializer (different address spaces)
   net/sched/cls_u32.c:378:36:    expected struct tc_u_hnode *ht
   net/sched/cls_u32.c:378:36:    got struct tc_u_hnode [noderef] <asn:4>*ht_up
>> net/sched/cls_u32.c:610:54: sparse: incorrect type in argument 4 (different address spaces)
   net/sched/cls_u32.c:610:54:    expected struct tc_u_hnode *ht
   net/sched/cls_u32.c:610:54:    got struct tc_u_hnode [noderef] <asn:4>*ht_up
>> net/sched/cls_u32.c:684:18: sparse: incorrect type in assignment (different address spaces)
   net/sched/cls_u32.c:684:18:    expected struct tc_u_hnode [noderef] <asn:4>*ht_up
   net/sched/cls_u32.c:684:18:    got struct tc_u_hnode *[assigned] ht
>> net/sched/cls_u32.c:359:18: sparse: dereference of noderef expression
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

a96366bf

net: sched: fix unsued cpu variable · 80aab73d

由 John Fastabend 提交于 9月 15, 2014

kbuild test robot reported an unused variable cpu in cls_u32.c
after the patch below. This happens when PERF and MARK config
variables are disabled

Fix this is to use separate variables for perf and mark
and define the cpu variable inside the ifdef logic.

Fixes: 459d5f62 ("net: sched: make cls_u32 per cpu")'
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NCong Wang <cwang@twopensource.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

80aab73d

net_sched: fix a null pointer dereference in tcindex_set_parms() · 69301eaa

由 WANG Cong 提交于 9月 15, 2014

This patch fixes the following crash:

[   42.199159] BUG: unable to handle kernel NULL pointer dereference at 0000000000000018
[   42.200027] IP: [<ffffffff817e3fc4>] tcindex_set_parms+0x45c/0x526
[   42.200027] PGD d2319067 PUD d4ffe067 PMD 0
[   42.200027] Oops: 0000 [#1] SMP DEBUG_PAGEALLOC
[   42.200027] CPU: 0 PID: 541 Comm: tc Not tainted 3.17.0-rc4+ #603
[   42.200027] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
[   42.200027] task: ffff8800d22d2670 ti: ffff8800ce790000 task.ti: ffff8800ce790000
[   42.200027] RIP: 0010:[<ffffffff817e3fc4>]  [<ffffffff817e3fc4>] tcindex_set_parms+0x45c/0x526
[   42.200027] RSP: 0018:ffff8800ce793898  EFLAGS: 00010202
[   42.200027] RAX: 0000000000000001 RBX: ffff8800d1786498 RCX: 0000000000000000
[   42.200027] RDX: ffffffff82114ec8 RSI: ffffffff82114ec8 RDI: ffffffff82114ec8
[   42.200027] RBP: ffff8800ce793958 R08: 00000000000080d0 R09: 0000000000000001
[   42.200027] R10: ffff8800ce7939a0 R11: 0000000000000246 R12: ffff8800d017d238
[   42.200027] R13: 0000000000000018 R14: ffff8800d017c6a0 R15: ffff8800d1786620
[   42.200027] FS:  00007f4e24539740(0000) GS:ffff88011a600000(0000) knlGS:0000000000000000
[   42.200027] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   42.200027] CR2: 0000000000000018 CR3: 00000000cff38000 CR4: 00000000000006f0
[   42.200027] Stack:
[   42.200027]  ffff8800ce0949f0 0000000000000000 0000000200000003 ffff880000000000
[   42.200027]  ffff8800ce7938b8 ffff8800ce7938b8 0000000600000007 0000000000000000
[   42.200027]  ffff8800ce7938d8 ffff8800ce7938d8 0000000600000007 ffff8800ce0949f0
[   42.200027] Call Trace:
[   42.200027]  [<ffffffff817e4169>] tcindex_change+0xdb/0xee
[   42.200027]  [<ffffffff817c16ca>] tc_ctl_tfilter+0x44d/0x63f
[   42.200027]  [<ffffffff8179d161>] rtnetlink_rcv_msg+0x181/0x194
[   42.200027]  [<ffffffff8179cf9d>] ? rtnl_lock+0x17/0x19
[   42.200027]  [<ffffffff8179cfe0>] ? __rtnl_unlock+0x17/0x17
[   42.200027]  [<ffffffff817ee296>] netlink_rcv_skb+0x49/0x8b
[   43.462494]  [<ffffffff8179cfc2>] rtnetlink_rcv+0x23/0x2a
[   43.462494]  [<ffffffff817ec8df>] netlink_unicast+0xc7/0x148
[   43.462494]  [<ffffffff817ed413>] netlink_sendmsg+0x5cb/0x63d
[   43.462494]  [<ffffffff810ad781>] ? mark_lock+0x2e/0x224
[   43.462494]  [<ffffffff817757b8>] __sock_sendmsg_nosec+0x25/0x27
[   43.462494]  [<ffffffff81778165>] sock_sendmsg+0x57/0x71
[   43.462494]  [<ffffffff81152bbd>] ? might_fault+0x57/0xa4
[   43.462494]  [<ffffffff81152c06>] ? might_fault+0xa0/0xa4
[   43.462494]  [<ffffffff81152bbd>] ? might_fault+0x57/0xa4
[   43.462494]  [<ffffffff817838fd>] ? verify_iovec+0x69/0xb7
[   43.462494]  [<ffffffff817784f8>] ___sys_sendmsg+0x21d/0x2bb
[   43.462494]  [<ffffffff81009db3>] ? native_sched_clock+0x35/0x37
[   43.462494]  [<ffffffff8109ab53>] ? sched_clock_local+0x12/0x72
[   43.462494]  [<ffffffff810ad781>] ? mark_lock+0x2e/0x224
[   43.462494]  [<ffffffff8109ada4>] ? sched_clock_cpu+0xa0/0xb9
[   43.462494]  [<ffffffff810aee37>] ? __lock_acquire+0x5fe/0xde4
[   43.462494]  [<ffffffff8119f570>] ? rcu_read_lock_held+0x36/0x38
[   43.462494]  [<ffffffff8119f75a>] ? __fcheck_files.isra.7+0x4b/0x57
[   43.462494]  [<ffffffff8119fbf2>] ? __fget_light+0x30/0x54
[   43.462494]  [<ffffffff81779012>] __sys_sendmsg+0x42/0x60
[   43.462494]  [<ffffffff81779042>] SyS_sendmsg+0x12/0x1c
[   43.462494]  [<ffffffff819d24d2>] system_call_fastpath+0x16/0x1b

'p->h' could be NULL while 'cp->h' is always update to date.

Fixes: commit 331b7292 ("net: sched: RCU cls_tcindex")
Cc: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Acked-By: NJohn Fastabend <john.r.fastabend@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

69301eaa

net_sched: fix memory leak in cls_tcindex · 44b75e43

由 WANG Cong 提交于 9月 15, 2014

Fixes: commit 331b7292 ("net: sched: RCU cls_tcindex")
Cc: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Acked-By: NJohn Fastabend <john.r.fastabend@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

44b75e43

16 9月, 2014 4 次提交

net_sched: use tcindex_filter_result_init() · 10ee1c34

由 WANG Cong 提交于 9月 15, 2014

Fixes: commit 331b7292 ("net: sched: RCU cls_tcindex")
Cc: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

10ee1c34

net_sched: fix suspicious RCU usage in tcindex_classify() · 2f9a220e

由 WANG Cong 提交于 9月 15, 2014

This patch fixes the following kernel warning:

[   44.805900] [ INFO: suspicious RCU usage. ]
[   44.808946] 3.17.0-rc4+ #610 Not tainted
[   44.811831] -------------------------------
[   44.814873] net/sched/cls_tcindex.c:84 suspicious rcu_dereference_check() usage!

Fixes: commit 331b7292 ("net: sched: RCU cls_tcindex")
Cc: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Acked-by: NJohn Fastabend <john.r.fastabend@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

2f9a220e

net_sched: fix an allocation bug in tcindex_set_parms() · a57a65ba

由 WANG Cong 提交于 9月 15, 2014

Fixes: commit 331b7292 ("net: sched: RCU cls_tcindex")
Cc: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

a57a65ba

net_sched: fix suspicious RCU usage in cls_bpf_classify() · 80dcbd12

由 WANG Cong 提交于 9月 15, 2014

Fixes: commit 1f947bf1 ("net: sched: rcu'ify cls_bpf")
Cc: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: NCong Wang <xiyou.wangcong@gmail.com>
Acked-by: NJohn Fastabend <john.r.fastabend@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

80dcbd12

14 9月, 2014 18 次提交

net: sched: rcu'ify cls_bpf · 1f947bf1

由 John Fastabend 提交于 9月 12, 2014

This patch makes the cls_bpf classifier RCU safe. The tcf_lock
was being used to protect a list of cls_bpf_prog now this list
is RCU safe and updates occur with rcu_replace.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

1f947bf1

net: sched: rcu'ify cls_rsvp · b929d86d

由 John Fastabend 提交于 9月 12, 2014

Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

b929d86d

net: sched: make cls_u32 lockless · 1ce87720

由 John Fastabend 提交于 9月 12, 2014

Make cls_u32 classifier safe to run without holding lock. This patch
converts statistics that are kept in read section u32_classify into
per cpu counters.

This patch was tested with a tight u32 filter add/delete loop while
generating traffic with pktgen. By running pktgen on vlan devices
created on top of a physical device we can hit the qdisc layer
correctly. For ingress qdisc's a loopback cable was used.

for i in {1..100}; do
        q=`echo $i%8|bc`;
        echo -n "u32 tos: iteration $i on queue $q";
        tc filter add dev p3p2 parent $p prio $i u32 match ip tos 0x10 0xff \
                  action skbedit queue_mapping $q;
        sleep 1;
        tc filter del dev p3p2 prio $i;

        echo -n "u32 tos hash table: iteration $i on queue $q";
        tc filter add dev p3p2 parent $p protocol ip prio $i handle 628: u32 divisor 1
        tc filter add dev p3p2 parent $p protocol ip prio $i u32 \
                match ip protocol 17 0xff link 628: offset at 0 mask 0xf00 shift 6 plus 0
        tc filter add dev p3p2 parent $p protocol ip prio $i u32 \
                ht 628:0 match ip tos 0x10 0xff action skbedit queue_mapping $q
        sleep 2;
        tc filter del dev p3p2 prio $i
        sleep 1;
done
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

1ce87720

net: sched: make cls_u32 per cpu · 459d5f62

由 John Fastabend 提交于 9月 12, 2014

This uses per cpu counters in cls_u32 in preparation
to convert over to rcu.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

459d5f62

net: sched: RCU cls_tcindex · 331b7292

由 John Fastabend 提交于 9月 12, 2014

Make cls_tcindex RCU safe.

This patch addds a new RCU routine rcu_dereference_bh_rtnl() to check
caller either holds the rcu read lock or RTNL. This is needed to
handle the case where tcindex_lookup() is being called in both cases.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

331b7292

net: sched: RCU cls_route · 1109c005

由 John Fastabend 提交于 9月 12, 2014

RCUify the route classifier. For now however spinlock's are used to
protect fastmap cache.

The issue here is the fastmap may be read by one CPU while the
cache is being updated by another. An array of pointers could be
one possible solution.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

1109c005

net: sched: fw use RCU · e35a8ee5

由 John Fastabend 提交于 9月 12, 2014

RCU'ify fw classifier.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

e35a8ee5

net: sched: cls_flow use RCU · 70da9f0b

由 John Fastabend 提交于 9月 12, 2014

Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

70da9f0b

net: sched: cls_cgroup use RCU · 952313bd

由 John Fastabend 提交于 9月 12, 2014

Make cgroup classifier safe for RCU.

Also drops the calls in the classify routine that were doing a
rcu_read_lock()/rcu_read_unlock(). If the rcu_read_lock() isn't held
entering this routine we have issues with deleting the classifier
chain so remove the unnecessary rcu_read_lock()/rcu_read_unlock()
pair noting all paths AFAIK hold rcu_read_lock.

If there is a case where classify is called without the rcu read lock
then an rcu splat will occur and we can correct it.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

952313bd

net: sched: cls_basic use RCU · 9888faef

由 John Fastabend 提交于 9月 12, 2014

Enable basic classifier for RCU.

Dereferencing tp->root may look a bit strange here but it is needed
by my accounting because it is allocated at init time and needs to
be kfree'd at destroy time. However because it may be referenced in
the classify() path we must wait an RCU grace period before free'ing
it. We use kfree_rcu() and rcu_ APIs to enforce this. This pattern
is used in all the classifiers.

Also the hgenerator can be incremented without concern because it
is always incremented under RTNL.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

9888faef

net: rcu-ify tcf_proto · 25d8c0d5

由 John Fastabend 提交于 9月 12, 2014

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: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

25d8c0d5

net: qdisc: use rcu prefix and silence sparse warnings · 46e5da40

由 John Fastabend 提交于 9月 12, 2014

Add __rcu notation to qdisc handling by doing this we can make
smatch output more legible. And anyways some of the cases should
be using rcu_dereference() see qdisc_all_tx_empty(),
qdisc_tx_chainging(), and so on.

Also *wake_queue() API is commonly called from driver timer routines
without rcu lock or rtnl lock. So I added rcu_read_lock() blocks
around netif_wake_subqueue and netif_tx_wake_queue.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

46e5da40

net: sched: rcu'ify cls_bpf · d355ab09

由 John Fastabend 提交于 9月 12, 2014

This patch makes the cls_bpf classifier RCU safe. The tcf_lock
was being used to protect a list of cls_bpf_prog now this list
is RCU safe and updates occur with rcu_replace.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

d355ab09

net: sched: rcu'ify cls_rsvp · 8b21e230

由 John Fastabend 提交于 9月 12, 2014

Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

8b21e230

net: sched: make cls_u32 lockless · 8f787cd1

由 John Fastabend 提交于 9月 12, 2014

Make cls_u32 classifier safe to run without holding lock. This patch
converts statistics that are kept in read section u32_classify into
per cpu counters.

This patch was tested with a tight u32 filter add/delete loop while
generating traffic with pktgen. By running pktgen on vlan devices
created on top of a physical device we can hit the qdisc layer
correctly. For ingress qdisc's a loopback cable was used.

for i in {1..100}; do
        q=`echo $i%8|bc`;
        echo -n "u32 tos: iteration $i on queue $q";
        tc filter add dev p3p2 parent $p prio $i u32 match ip tos 0x10 0xff \
                  action skbedit queue_mapping $q;
        sleep 1;
        tc filter del dev p3p2 prio $i;

        echo -n "u32 tos hash table: iteration $i on queue $q";
        tc filter add dev p3p2 parent $p protocol ip prio $i handle 628: u32 divisor 1
        tc filter add dev p3p2 parent $p protocol ip prio $i u32 \
                match ip protocol 17 0xff link 628: offset at 0 mask 0xf00 shift 6 plus 0
        tc filter add dev p3p2 parent $p protocol ip prio $i u32 \
                ht 628:0 match ip tos 0x10 0xff action skbedit queue_mapping $q
        sleep 2;
        tc filter del dev p3p2 prio $i
        sleep 1;
done
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

8f787cd1

net: sched: make cls_u32 per cpu · f4f64050

由 John Fastabend 提交于 9月 12, 2014

This uses per cpu counters in cls_u32 in preparation
to convert over to rcu.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

f4f64050

net: sched: RCU cls_tcindex · 8332904a

由 John Fastabend 提交于 9月 12, 2014

Make cls_tcindex RCU safe.

This patch addds a new RCU routine rcu_dereference_bh_rtnl() to check
caller either holds the rcu read lock or RTNL. This is needed to
handle the case where tcindex_lookup() is being called in both cases.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

8332904a

net: sched: RCU cls_route · cc91210c

由 John Fastabend 提交于 9月 12, 2014

RCUify the route classifier. For now however spinlock's are used to
protect fastmap cache.

The issue here is the fastmap may be read by one CPU while the
cache is being updated by another. An array of pointers could be
one possible solution.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

cc91210c