mirror of
https://github.com/torvalds/linux.git
synced 2024-09-21 23:43:02 +00:00
ed2e923945
When creating a timewait socket, we need to arm the timer before
allowing other cpus to find it. The signal allowing cpus to find
the socket is setting tw_refcnt to non zero value.
As we set tw_refcnt in __inet_twsk_hashdance(), we therefore need to
call inet_twsk_schedule() first.
This also means we need to remove tw_refcnt changes from
inet_twsk_schedule() and let the caller handle it.
Note that because we use mod_timer_pinned(), we have the guarantee
the timer wont expire before we set tw_refcnt as we run in BH context.
To make things more readable I introduced inet_twsk_reschedule() helper.
When rearming the timer, we can use mod_timer_pending() to make sure
we do not rearm a canceled timer.
Note: This bug can possibly trigger if packets of a flow can hit
multiple cpus. This does not normally happen, unless flow steering
is broken somehow. This explains this bug was spotted ~5 months after
its introduction.
A similar fix is needed for SYN_RECV sockets in reqsk_queue_hash_req(),
but will be provided in a separate patch for proper tracking.
Fixes:
|
||
---|---|---|
.. | ||
6lowpan | ||
9p | ||
802 | ||
8021q | ||
appletalk | ||
atm | ||
ax25 | ||
batman-adv | ||
bluetooth | ||
bridge | ||
caif | ||
can | ||
ceph | ||
core | ||
dcb | ||
dccp | ||
decnet | ||
dns_resolver | ||
dsa | ||
ethernet | ||
hsr | ||
ieee802154 | ||
ipv4 | ||
ipv6 | ||
ipx | ||
irda | ||
iucv | ||
key | ||
l2tp | ||
lapb | ||
llc | ||
mac80211 | ||
mac802154 | ||
mpls | ||
netfilter | ||
netlabel | ||
netlink | ||
netrom | ||
nfc | ||
openvswitch | ||
packet | ||
phonet | ||
rds | ||
rfkill | ||
rose | ||
rxrpc | ||
sched | ||
sctp | ||
sunrpc | ||
switchdev | ||
tipc | ||
unix | ||
vmw_vsock | ||
wimax | ||
wireless | ||
x25 | ||
xfrm | ||
compat.c | ||
Kconfig | ||
Makefile | ||
socket.c | ||
sysctl_net.c |