sctp: fully initialize sctp_outq in sctp_outq_init
authorNeil Horman <nhorman@tuxdriver.com>
Wed, 12 Jun 2013 18:26:44 +0000 (14:26 -0400)
committerBen Hutchings <ben@decadent.org.uk>
Tue, 10 Sep 2013 00:57:06 +0000 (01:57 +0100)
commit681c76440d4d1a0f1956f0f655931c12cfabd5bd
treef8f22c752e20c346680cae561c78b7f1a233a9e3
parent2dec485f87d4bcfe9808e379f002a8444c09321c
sctp: fully initialize sctp_outq in sctp_outq_init

[ Upstream commit c5c7774d7eb4397891edca9ebdf750ba90977a69 ]

In commit 2f94aabd9f6c925d77aecb3ff020f1cc12ed8f86
(refactor sctp_outq_teardown to insure proper re-initalization)
we modified sctp_outq_teardown to use sctp_outq_init to fully re-initalize the
outq structure.  Steve West recently asked me why I removed the q->error = 0
initalization from sctp_outq_teardown.  I did so because I was operating under
the impression that sctp_outq_init would properly initalize that value for us,
but it doesn't.  sctp_outq_init operates under the assumption that the outq
struct is all 0's (as it is when called from sctp_association_init), but using
it in __sctp_outq_teardown violates that assumption. We should do a memset in
sctp_outq_init to ensure that the entire structure is in a known state there
instead.

Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Reported-by: "West, Steve (NSN - US/Fort Worth)" <steve.west@nsn.com>
CC: Vlad Yasevich <vyasevich@gmail.com>
CC: netdev@vger.kernel.org
CC: davem@davemloft.net
Acked-by: Vlad Yasevich <vyasevich@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
net/sctp/outqueue.c