kernel-hacking-2024-linux-s.../net
Eric Dumazet e64e469b9a ipv6: addrconf: break critical section in addrconf_verify_rtnl()
Heiner reported a lockdep splat [1]

This is caused by attempting GFP_KERNEL allocation while RCU lock is
held and BH blocked.

We believe that addrconf_verify_rtnl() could run for a long period,
so instead of using GFP_ATOMIC here as Ido suggested, we should break
the critical section and restart it after the allocation.

[1]
[86220.125562] =============================
[86220.125586] WARNING: suspicious RCU usage
[86220.125612] 4.15.0-rc7-next-20180110+ #7 Not tainted
[86220.125641] -----------------------------
[86220.125666] kernel/sched/core.c:6026 Illegal context switch in RCU-bh read-side critical section!
[86220.125711]
               other info that might help us debug this:

[86220.125755]
               rcu_scheduler_active = 2, debug_locks = 1
[86220.125792] 4 locks held by kworker/0:2/1003:
[86220.125817]  #0:  ((wq_completion)"%s"("ipv6_addrconf")){+.+.}, at: [<00000000da8e9b73>] process_one_work+0x1de/0x680
[86220.125895]  #1:  ((addr_chk_work).work){+.+.}, at: [<00000000da8e9b73>] process_one_work+0x1de/0x680
[86220.125959]  #2:  (rtnl_mutex){+.+.}, at: [<00000000b06d9510>] rtnl_lock+0x12/0x20
[86220.126017]  #3:  (rcu_read_lock_bh){....}, at: [<00000000aef52299>] addrconf_verify_rtnl+0x1e/0x510 [ipv6]
[86220.126111]
               stack backtrace:
[86220.126142] CPU: 0 PID: 1003 Comm: kworker/0:2 Not tainted 4.15.0-rc7-next-20180110+ #7
[86220.126185] Hardware name: ZOTAC ZBOX-CI321NANO/ZBOX-CI321NANO, BIOS B246P105 06/01/2015
[86220.126250] Workqueue: ipv6_addrconf addrconf_verify_work [ipv6]
[86220.126288] Call Trace:
[86220.126312]  dump_stack+0x70/0x9e
[86220.126337]  lockdep_rcu_suspicious+0xce/0xf0
[86220.126365]  ___might_sleep+0x1d3/0x240
[86220.126390]  __might_sleep+0x45/0x80
[86220.126416]  kmem_cache_alloc_trace+0x53/0x250
[86220.126458]  ? ipv6_add_addr+0xfe/0x6e0 [ipv6]
[86220.126498]  ipv6_add_addr+0xfe/0x6e0 [ipv6]
[86220.126538]  ipv6_create_tempaddr+0x24d/0x430 [ipv6]
[86220.126580]  ? ipv6_create_tempaddr+0x24d/0x430 [ipv6]
[86220.126623]  addrconf_verify_rtnl+0x339/0x510 [ipv6]
[86220.126664]  ? addrconf_verify_rtnl+0x339/0x510 [ipv6]
[86220.126708]  addrconf_verify_work+0xe/0x20 [ipv6]
[86220.126738]  process_one_work+0x258/0x680
[86220.126765]  worker_thread+0x35/0x3f0
[86220.126790]  kthread+0x124/0x140
[86220.126813]  ? process_one_work+0x680/0x680
[86220.126839]  ? kthread_create_worker_on_cpu+0x40/0x40
[86220.126869]  ? umh_complete+0x40/0x40
[86220.126893]  ? call_usermodehelper_exec_async+0x12a/0x160
[86220.126926]  ret_from_fork+0x4b/0x60
[86220.126999] BUG: sleeping function called from invalid context at mm/slab.h:420
[86220.127041] in_atomic(): 1, irqs_disabled(): 0, pid: 1003, name: kworker/0:2
[86220.127082] 4 locks held by kworker/0:2/1003:
[86220.127107]  #0:  ((wq_completion)"%s"("ipv6_addrconf")){+.+.}, at: [<00000000da8e9b73>] process_one_work+0x1de/0x680
[86220.127179]  #1:  ((addr_chk_work).work){+.+.}, at: [<00000000da8e9b73>] process_one_work+0x1de/0x680
[86220.127242]  #2:  (rtnl_mutex){+.+.}, at: [<00000000b06d9510>] rtnl_lock+0x12/0x20
[86220.127300]  #3:  (rcu_read_lock_bh){....}, at: [<00000000aef52299>] addrconf_verify_rtnl+0x1e/0x510 [ipv6]
[86220.127414] CPU: 0 PID: 1003 Comm: kworker/0:2 Not tainted 4.15.0-rc7-next-20180110+ #7
[86220.127463] Hardware name: ZOTAC ZBOX-CI321NANO/ZBOX-CI321NANO, BIOS B246P105 06/01/2015
[86220.127528] Workqueue: ipv6_addrconf addrconf_verify_work [ipv6]
[86220.127568] Call Trace:
[86220.127591]  dump_stack+0x70/0x9e
[86220.127616]  ___might_sleep+0x14d/0x240
[86220.127644]  __might_sleep+0x45/0x80
[86220.127672]  kmem_cache_alloc_trace+0x53/0x250
[86220.127717]  ? ipv6_add_addr+0xfe/0x6e0 [ipv6]
[86220.127762]  ipv6_add_addr+0xfe/0x6e0 [ipv6]
[86220.127807]  ipv6_create_tempaddr+0x24d/0x430 [ipv6]
[86220.127854]  ? ipv6_create_tempaddr+0x24d/0x430 [ipv6]
[86220.127903]  addrconf_verify_rtnl+0x339/0x510 [ipv6]
[86220.127950]  ? addrconf_verify_rtnl+0x339/0x510 [ipv6]
[86220.127998]  addrconf_verify_work+0xe/0x20 [ipv6]
[86220.128032]  process_one_work+0x258/0x680
[86220.128063]  worker_thread+0x35/0x3f0
[86220.128091]  kthread+0x124/0x140
[86220.128117]  ? process_one_work+0x680/0x680
[86220.128146]  ? kthread_create_worker_on_cpu+0x40/0x40
[86220.128180]  ? umh_complete+0x40/0x40
[86220.128207]  ? call_usermodehelper_exec_async+0x12a/0x160
[86220.128243]  ret_from_fork+0x4b/0x60

Fixes: f3d9832e56 ("ipv6: addrconf: cleanup locking in ipv6_add_addr")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-01-29 14:23:38 -05:00
..
6lowpan
9p 9p: add missing module license for xen transport 2018-01-15 13:13:53 -05:00
802
8021q net: delete /proc THIS_MODULE references 2018-01-16 15:01:33 -05:00
appletalk net: delete /proc THIS_MODULE references 2018-01-16 15:01:33 -05:00
atm net: delete /proc THIS_MODULE references 2018-01-16 15:01:33 -05:00
ax25 net: delete /proc THIS_MODULE references 2018-01-16 15:01:33 -05:00
batman-adv
bluetooth Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2018-01-17 00:10:42 -05:00
bpf
bridge bridge: return boolean instead of integer in br_multicast_is_router 2018-01-22 16:13:20 -05:00
caif caif: reduce stack size with KASAN 2018-01-19 14:02:12 -05:00
can can: migrate documentation to restructured text 2018-01-26 10:46:44 +01:00
ceph
core net_sched: plug in qdisc ops change_tx_queue_len 2018-01-29 12:42:15 -05:00
dcb
dccp Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2018-01-29 10:15:51 -05:00
decnet net: delete /proc THIS_MODULE references 2018-01-16 15:01:33 -05:00
dns_resolver
dsa net: dsa: Support internal phy on 'cpu' port 2018-01-23 19:22:38 -05:00
ethernet
hsr
ieee802154
ife
ipv4 Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2018-01-29 10:15:51 -05:00
ipv6 ipv6: addrconf: break critical section in addrconf_verify_rtnl() 2018-01-29 14:23:38 -05:00
ipx net: delete /proc THIS_MODULE references 2018-01-16 15:01:33 -05:00
iucv
kcm Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2018-01-24 23:44:15 -05:00
key af_key: Fix memory leak in key_notify_policy. 2018-01-10 09:45:11 +01:00
l2tp l2tp: remove switch block in l2tp_nl_cmd_session_create() 2018-01-19 15:10:23 -05:00
l3mdev
lapb
llc net: delete /proc THIS_MODULE references 2018-01-16 15:01:33 -05:00
mac80211 debugfs_sta: Remove unneeded semicolons 2018-01-22 14:03:28 +01:00
mac802154
mpls
ncsi
netfilter netfilter: remove messages print and boot/module load time 2018-01-19 18:39:49 +01:00
netlabel
netlink Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2018-01-19 22:59:33 -05:00
netrom net: delete /proc THIS_MODULE references 2018-01-16 15:01:33 -05:00
nfc
nsh
openvswitch openvswitch: add erspan version I and II support 2018-01-25 21:39:43 -05:00
packet net: delete /proc THIS_MODULE references 2018-01-16 15:01:33 -05:00
phonet net: delete /proc THIS_MODULE references 2018-01-16 15:01:33 -05:00
psample
qrtr
rds Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2018-01-23 13:51:56 -05:00
rfkill
rose net: delete /proc THIS_MODULE references 2018-01-16 15:01:33 -05:00
rxrpc net: delete /proc THIS_MODULE references 2018-01-16 15:01:33 -05:00
sched net_sched: implement ->change_tx_queue_len() for pfifo_fast 2018-01-29 12:42:15 -05:00
sctp Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2018-01-23 13:51:56 -05:00
smc net/smc: return booleans instead of integers 2018-01-26 10:41:56 -05:00
strparser
sunrpc
switchdev
tipc tipc: fix race between poll() and setsockopt() 2018-01-19 15:12:21 -05:00
tls tls: Correct length of scatterlist in tls_sw_sendpage 2018-01-22 16:25:21 -05:00
unix net: delete /proc THIS_MODULE references 2018-01-16 15:01:33 -05:00
vmw_vsock VSOCK: set POLLOUT | POLLWRNORM for TCP_CLOSING 2018-01-26 11:16:27 -05:00
wimax
wireless lift handling of SIOCIW... out of dev_ioctl() 2018-01-24 19:13:45 -05:00
x25
xfrm Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/klassert/ipsec-next 2018-01-26 10:22:53 -05:00
compat.c
Kconfig netfilter: don't allocate space for arp/bridge hooks unless needed 2018-01-08 18:01:11 +01:00
Makefile
socket.c kill kernel_sock_ioctl() 2018-01-24 19:13:45 -05:00
sysctl_net.c