On 31/01/17 18:14, Anoob Soman wrote:
On 31/01/17 18:00, Eric Dumazet wrote:
On Tue, 2017-01-31 at 17:03 +0000, Anoob Soman wrote:
On 30/01/17 19:44, Eric Dumazet wrote:
On Mon, 2017-01-30 at 19:08 +0000, Anoob Soman wrote:
On 30/01/17 17:26, Eric Dumazet wrote:
On Thu, 2016-10-06 at 20:50 -0400, David Miller wrote:
From: Anoob Soman <[email protected]>
Date: Wed, 5 Oct 2016 15:12:54 +0100
If a socket has FANOUT sockopt set, a new proto_hook is registered
as part of fanout_add(). When processing a NETDEV_UNREGISTER
event in
af_packet, __fanout_unlink is called for all sockets, but
prot_hook which was
registered as part of fanout_add is not removed. Call
fanout_release, on a
NETDEV_UNREGISTER, which removes prot_hook and removes fanout
from the
fanout_list.
This fixes BUG_ON(!list_empty(&dev->ptype_specific)) in
netdev_run_todo()
Signed-off-by: Anoob Soman <[email protected]>
Applied and queued up for -stable, thanks.
This commit (6664498280cf "packet: call fanout_release, while
UNREGISTERING a netdev")
looks buggy :
We end up calling fanout_release() while holding a spinlock
( spin_lock(&po->bind_lock); )
But fanout_release() grabs a mutex ( mutex_lock(&fanout_mutex) ),
and
this is absolutely not valid while holding a spinlock.
Yes, that is wrong.
Anoob, can you cook a fix, I guess you have a way to reproduce
the thing
that wanted a kernel patch ?
(Please build your test kernel with CONFIG_LOCKDEP=y)
Sure, I am planning to move fanout_release(sk) after
spin_unlock(bind_lock). Something like this.
}
if (msg == NETDEV_UNREGISTER) {
packet_cached_dev_reset(po);
- fanout_release(sk);
po->ifindex = -1;
if (po->prot_hook.dev)
dev_put(po->prot_hook.dev);
po->prot_hook.dev = NULL;
}
spin_unlock(&po->bind_lock);
+ if (msg == NETDEV_UNREGISTER) {
+ fanout_release(sk);
+ }
}
break;
I will quickly test it out.
It wont be enough.
You need to also fix a race if two cpus call fanout_release(sk) at the
same time.
Thanks.
Hi Eric,
I have ran into some problem trying to enable CONFIG_LOCKDEP. I think
this particular scenario, taking mutex_lock() while holding a spin_lock
debugging, requires CONFIG_DEBUG_ATOMIC_SLEEP to be enabled.
CONFIG_DEBUG_ATOMIC_SLEEP, selects CONFIG_PREEMPT_COUNT and my kernel
doesn't behave well if PREEMPTION is enabled. I am trying to reproduce
this issue in a way that I might be able to use debug_atomic_sleep.
Meanwhile, I have modified patch fix the race.
So you can definitely have in a .config all these at the same time
(LOCKDEP, non PREEMPT, and DEBUG_ATOMIC_SLEEP)
$ egrep "DEBUG_ATOMIC_SLEEP|PREEMPT|LOCKDEP" .config
CONFIG_LOCKDEP_SUPPORT=y
CONFIG_PREEMPT_NOTIFIERS=y
CONFIG_PREEMPT_NONE=y
# CONFIG_PREEMPT_VOLUNTARY is not set
# CONFIG_PREEMPT is not set
CONFIG_PREEMPT_COUNT=y
CONFIG_LOCKDEP=y
# CONFIG_DEBUG_LOCKDEP is not set
CONFIG_DEBUG_ATOMIC_SLEEP=y
yes, thats exactly what I have.
$ egrep "DEBUG_ATOMIC_SLEEP|PREEMPT|LOCKDEP" .config
CONFIG_LOCKDEP_SUPPORT=y
# CONFIG_PREEMPT_NONE is not set
CONFIG_PREEMPT_VOLUNTARY=y
# CONFIG_PREEMPT is not set
CONFIG_PREEMPT_COUNT=y
CONFIG_LOCKDEP=y
# CONFIG_DEBUG_LOCKDEP is not set
CONFIG_DEBUG_ATOMIC_SLEEP=y
I initially thought CONFIG_PREEMPT_COUNT enables CONFIG_PREEMPT, but
looks like all it does is to inc/dec preempt_count.
Let me give the test a spin again, and see why everything seems to
fall apart.
Hi Eric,
I managed to reproduce the problem consistently with LOCKDEP enabled. I have
to workaround few other problems in order to make the repro consistent.
There are 4 potential problem with the commit.
1. calling mutex_lock(&fanout_mutex) (fanout_release()) from inside
rcu_read-side
critical section. rcu_read_lock disables preemption, most often (expect if
CONFIG_PREEMPT/CONFIG_PREEMPTE_RCU are set), which prohibits calling
sleeping
functions.
[ 180.940388] include/linux/rcupdate.h:560 Illegal context switch in
RCU read-side critical section!
[ 180.940401]
[ 180.940401] other info that might help us debug this:
[ 180.940401]
[ 180.940417]
[ 180.940417] rcu_scheduler_active = 1, debug_locks = 0
[ 180.940430] 4 locks held by ovs-vswitchd/1969:
[ 180.940438] #0: (cb_lock){++++++}, at: [<ffffffff8158a6c9>]
genl_rcv+0x19/0x40
[ 180.940498] #1: (ovs_mutex){+.+.+.}, at: [<ffffffffa04878ca>]
ovs_vport_cmd_del+0x4a/0x100 [openvswitch]
[ 180.940530] #2: (rtnl_mutex){+.+.+.}, at: [<ffffffff81564157>]
rtnl_lock+0x17/0x20
[ 180.940557] #3: (rcu_read_lock){......}, at: [<ffffffff81614165>]
packet_notifier+0x5/0x3f0
[ 180.940587]
[ 180.940697] Call Trace:
[ 180.940710] [<ffffffff813770c1>] dump_stack+0x85/0xc4
[ 180.940727] [<ffffffff810c9077>] lockdep_rcu_suspicious+0x107/0x110
[ 180.940742] [<ffffffff810a2da7>] ___might_sleep+0x57/0x210
[ 180.940755] [<ffffffff810a2fd0>] __might_sleep+0x70/0x90
[ 180.940768] [<ffffffff8162e80c>] mutex_lock_nested+0x3c/0x3a0
[ 180.940785] [<ffffffff810de93f>] ? vprintk_default+0x1f/0x30
[ 180.940801] [<ffffffff81186e88>] ? printk+0x4d/0x4f
[ 180.940814] [<ffffffff816106dd>] fanout_release+0x1d/0xe0
[ 180.940828] [<ffffffff81614459>] packet_notifier+0x2f9/0x3f0
2. calling mutex_lock(&fanout_mutex) inside spin_lock(&po->bind_lock).
"sleeping function called from invalid context"
[ 181.941336] BUG: sleeping function called from invalid context at
kernel/locking/mutex.c:620
[ 181.941352] in_atomic(): 1, irqs_disabled(): 0, pid: 1969, name:
ovs-vswitchd
[ 181.941365] INFO: lockdep is turned off.
[ 181.941462] Call Trace:
[ 181.941469] [<ffffffff813770c1>] dump_stack+0x85/0xc4
[ 181.941480] [<ffffffff810a2f52>] ___might_sleep+0x202/0x210
[ 181.941492] [<ffffffff810a2fd0>] __might_sleep+0x70/0x90
[ 181.941503] [<ffffffff8162e80c>] mutex_lock_nested+0x3c/0x3a0
[ 181.941515] [<ffffffff810de93f>] ? vprintk_default+0x1f/0x30
[ 181.941526] [<ffffffff81186e88>] ? printk+0x4d/0x4f
[ 181.941537] [<ffffffff816106dd>] fanout_release+0x1d/0xe0
[ 181.941548] [<ffffffff81614459>] packet_notifier+0x2f9/0x3f0
3. calling dev_remove_pack(&fanout->prot_hook), from inside
spin_lock(&po->bind_lock)
or rcu_read-side critcial.section. dev_remove_pack() -> synchronize_net(),
which might sleep.
[ 181.942401] BUG: scheduling while atomic: ovs-vswitchd/1969/0x00000002
[ 181.942411] INFO: lockdep is turned off.
[ 181.942751] Call Trace:
[ 181.942760] [<ffffffff813770c1>] dump_stack+0x85/0xc4
[ 181.942771] [<ffffffff81186274>] __schedule_bug+0x64/0x73
[ 181.942782] [<ffffffff8162b8cb>] __schedule+0x6b/0xd10
[ 181.942794] [<ffffffff8162c5db>] schedule+0x6b/0x80
[ 181.942805] [<ffffffff81630b1d>] schedule_timeout+0x38d/0x410
[ 181.942820] [<ffffffff810efac0>] ? internal_add_timer+0x80/0x80
[ 181.942835] [<ffffffff810ea3fd>] synchronize_sched_expedited+0x53d/0x810
[ 181.942850] [<ffffffff810bf650>] ? prepare_to_wait_event+0x110/0x110
[ 181.942862] [<ffffffff810ea6de>] synchronize_rcu_expedited+0xe/0x10
[ 181.942873] [<ffffffff8154eab5>] synchronize_net+0x35/0x50
[ 181.942884] [<ffffffff8154eae3>] dev_remove_pack+0x13/0x20
[ 181.942896] [<ffffffff8161077e>] fanout_release+0xbe/0xe0
[ 181.942909] [<ffffffff81614459>] packet_notifier+0x2f9/0x3f0
4. fanout_release() races with calls from different CPU.
Moving mutex_lock(&fanout_mutex) out of spin_lock(&po->bind_lock),
if (msg == NETDEV_UNREGISTER) {
packet_cached_dev_reset(po);
- fanout_release(sk);
po->ifindex = -1;
if (po->prot_hook.dev)
dev_put(po->prot_hook.dev);
po->prot_hook.dev = NULL;
}
spin_unlock(&po->bind_lock);
+ if (msg == NETDEV_UNREGISTER)
+ fanout_release(sk);
}
break;
case NETDEV_UP:
will solve 2 and part (spin_lock) of 3, but definetly will not solve 1 and
(rcu_read-side cs) other part of 3 (if CONFIG_PREEMPT is enabled).
Inside packet_notifier packet.sklist is traversed under rcu_read_lock,
which meant any calls which might sleep (dev_remove_pack(),
mutex_lock()) cannot
be done. Instead of traversing sklist under rcu_read_lock, we can traverse
sklist using mutex_lock(packet.sklist_lock). This would fix 1 and 3, but
adds
additional overhead of blocking modifications, while traversing sklist.
Another way to fix, all the above problem, would be not to call
fanout_release()
under rcu_read_lock(), instead call
__dev_remove_pack(&fanout->prot_hook) and
netdev_run_todo is happy that &dev->ptype_specific list in empty. In order
to make this work, I had to move dev_{add,remove}_pack() out of
fanout_{add,release} to __fanout_{link,unlink}. So, call to
{,__}unregister_prot_hook() will make sure fanout->prot_hook is removed
as well.
@@ -1498,6 +1498,8 @@ static void __fanout_link(struct sock *sk, struct
packet_sock *po)
f->arr[f->num_members] = sk;
smp_wmb();
f->num_members++;
+ if (f->num_members == 1)
+ dev_add_pack(&f->prot_hook);
spin_unlock(&f->lock);
}
@@ -1514,6 +1516,8 @@ static void __fanout_unlink(struct sock *sk,
struct packet_sock *po)
BUG_ON(i >= f->num_members);
f->arr[i] = f->arr[f->num_members - 1];
f->num_members--;
+ if (f->num_members == 0)
+ __dev_remove_pack(&f->prot_hook);
spin_unlock(&f->lock);
}
@@ -1692,7 +1696,6 @@ static int fanout_add(struct sock *sk, u16 id, u16
type_flags)
match->prot_hook.func = packet_rcv_fanout;
match->prot_hook.af_packet_priv = match;
match->prot_hook.id_match = match_fanout_group;
- dev_add_pack(&match->prot_hook);
list_add(&match->list, &fanout_list);
}
err = -EINVAL;
@@ -1731,7 +1734,6 @@ static void fanout_release(struct sock *sk)
if (atomic_dec_and_test(&f->sk_ref)) {
list_del(&f->list);
- dev_remove_pack(&f->prot_hook);
fanout_release_data(f);
kfree(f);
}
@@ -3855,7 +3857,6 @@ static int packet_notifier(struct notifier_block
*this,
}
if (msg == NETDEV_UNREGISTER) {
packet_cached_dev_reset(po);
- fanout_release(sk);
po->ifindex = -1;
if (po->prot_hook.dev)
dev_put(po->prot_hook.dev);
I have tested both the approaches and LOCKDEP doesn't seem to catch any
problem with the test I was doing.
Thanks,
Anoob.
