linux/net/core
Eric Dumazet c3f9b01849 tcp: tcp_release_cb() should release socket ownership
Lars Persson reported following deadlock :

-000 |M:0x0:0x802B6AF8(asm) <-- arch_spin_lock
-001 |tcp_v4_rcv(skb = 0x8BD527A0) <-- sk = 0x8BE6B2A0
-002 |ip_local_deliver_finish(skb = 0x8BD527A0)
-003 |__netif_receive_skb_core(skb = 0x8BD527A0, ?)
-004 |netif_receive_skb(skb = 0x8BD527A0)
-005 |elk_poll(napi = 0x8C770500, budget = 64)
-006 |net_rx_action(?)
-007 |__do_softirq()
-008 |do_softirq()
-009 |local_bh_enable()
-010 |tcp_rcv_established(sk = 0x8BE6B2A0, skb = 0x87D3A9E0, th = 0x814EBE14, ?)
-011 |tcp_v4_do_rcv(sk = 0x8BE6B2A0, skb = 0x87D3A9E0)
-012 |tcp_delack_timer_handler(sk = 0x8BE6B2A0)
-013 |tcp_release_cb(sk = 0x8BE6B2A0)
-014 |release_sock(sk = 0x8BE6B2A0)
-015 |tcp_sendmsg(?, sk = 0x8BE6B2A0, ?, ?)
-016 |sock_sendmsg(sock = 0x8518C4C0, msg = 0x87D8DAA8, size = 4096)
-017 |kernel_sendmsg(?, ?, ?, ?, size = 4096)
-018 |smb_send_kvec()
-019 |smb_send_rqst(server = 0x87C4D400, rqst = 0x87D8DBA0)
-020 |cifs_call_async()
-021 |cifs_async_writev(wdata = 0x87FD6580)
-022 |cifs_writepages(mapping = 0x852096E4, wbc = 0x87D8DC88)
-023 |__writeback_single_inode(inode = 0x852095D0, wbc = 0x87D8DC88)
-024 |writeback_sb_inodes(sb = 0x87D6D800, wb = 0x87E4A9C0, work = 0x87D8DD88)
-025 |__writeback_inodes_wb(wb = 0x87E4A9C0, work = 0x87D8DD88)
-026 |wb_writeback(wb = 0x87E4A9C0, work = 0x87D8DD88)
-027 |wb_do_writeback(wb = 0x87E4A9C0, force_wait = 0)
-028 |bdi_writeback_workfn(work = 0x87E4A9CC)
-029 |process_one_work(worker = 0x8B045880, work = 0x87E4A9CC)
-030 |worker_thread(__worker = 0x8B045880)
-031 |kthread(_create = 0x87CADD90)
-032 |ret_from_kernel_thread(asm)

Bug occurs because __tcp_checksum_complete_user() enables BH, assuming
it is running from softirq context.

Lars trace involved a NIC without RX checksum support but other points
are problematic as well, like the prequeue stuff.

Problem is triggered by a timer, that found socket being owned by user.

tcp_release_cb() should call tcp_write_timer_handler() or
tcp_delack_timer_handler() in the appropriate context :

BH disabled and socket lock held, but 'owned' field cleared,
as if they were running from timer handlers.

Fixes: 6f458dfb40 ("tcp: improve latencies of timer triggered events")
Reported-by: Lars Persson <lars.persson@axis.com>
Tested-by: Lars Persson <lars.persson@axis.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-11 16:45:59 -04:00
..
datagram.c
dev_addr_lists.c net: Correctly sync addresses from multiple sources to single device 2014-01-23 13:06:34 -08:00
dev_ioctl.c
dev.c net: core: introduce netif_skb_dev_features 2014-02-13 17:17:02 -05:00
drop_monitor.c
dst.c
ethtool.c
fib_rules.c net: fix 'ip rule' iif/oif device rename 2014-02-09 19:02:52 -08:00
filter.c bpf: do not use reciprocal divide 2014-01-15 17:02:08 -08:00
flow_dissector.c netdevice: move netdev_cap_txqueue for shared usage to header 2014-02-17 00:36:34 -05:00
flow.c
gen_estimator.c
gen_stats.c
iovec.c
link_watch.c
Makefile net: netprio: rename config to be more consistent with cgroup configs 2014-01-03 23:41:42 +01:00
neighbour.c neigh: recompute reachabletime before returning from neigh_periodic_work() 2014-02-27 18:21:17 -05:00
net_namespace.c
net-procfs.c
net-sysfs.c net: fix "queues" uevent between network namespaces 2014-01-19 20:02:02 -08:00
net-sysfs.h net: netdev_kobject_init: annotate with __init 2014-01-05 20:27:54 -05:00
net-traces.c
netclassid_cgroup.c net: net_cls: move cgroupfs classid handling into core 2014-01-03 23:41:41 +01:00
netevent.c
netpoll.c netpoll: fix netconsole IPv6 setup 2014-02-06 21:28:06 -08:00
netprio_cgroup.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next 2014-01-25 11:17:34 -08:00
pktgen.c pktgen: Use ether_addr_copy 2014-01-21 18:13:05 -08:00
request_sock.c
rtnetlink.c net: correct error path in rtnl_newlink() 2014-02-13 17:08:29 -05:00
scm.c
secure_seq.c
skbuff.c skbuff: skb_segment: orphan frags before copying 2014-03-11 16:26:38 -04:00
sock_diag.c
sock.c tcp: tcp_release_cb() should release socket ownership 2014-03-11 16:45:59 -04:00
stream.c net: replace macros net_random and net_srandom with direct calls to prandom 2014-01-14 15:15:25 -08:00
sysctl_net_core.c
timestamping.c
user_dma.c
utils.c