net/smc: fix refcount non-blocking connect() -part 2
PATCH 1/1 net/smc: fix refcount non-blocking connect() -part 2
From: Ursula Braun <[email protected]>
If an SMC socket is immediately terminated after a non-blocking connect() has been called, a memory leak is possible.
Due to the sock_hold move in
commit 301428ea3708 ("net/smc: fix refcounting for non-blocking connect()") an extra sock_put() is needed in smc_connect_work(), if the internal TCP socket is aborted and cancels the sk_stream_wait_connect() of the connect worker.
--- net/smc/af_smc.c | 1 + 1 file changed, 1 insertion(+)
net/smc/af_smc.c +1 -0
@@ -799,6 +799,7 @@ static void smc_connect_work(struct work_struct *work) smc->sk.sk_err =EPIPE ; else if (signal_pending(current)) smc->sk.sk_err = -sock_intr_errno(timeo); +sock_put (&smc->sk);/* passive closing */ goto out; }
Signature · 1 line
2.17.1
From: Karsten Graul <[email protected]>
Date: Tue, 12 Nov 2019 16:03:41 +0100
Quoted 10 lines
From: Ursula Braun <[email protected]>
If an SMC socket is immediately terminated after a non-blocking
connect()has been called, a memory leak is possible.
Due to thesock_holdmove in
commit301428ea3708("net/smc: fix refcounting for non-blockingconnect()") an extrasock_put()is needed insmc_connect_work(), if the internal TCP socket is aborted and cancels thesk_stream_wait_connect()of the connect worker.
Quoted 4 lines
Reported-by [email protected]Fixes 301428ea3708 ("net/smc: fix refcounting for non-blocking connect()")Signed-off-by Ursula Braun <[email protected]>Signed-off-by Karsten Graul <[email protected]>
Applied.
And since 301428ea3708 went to -stable, I'll queue this up too.
Thanks.