net/sctp: terminate rhashtable walk correctly
authorVegard Nossum <vegard.nossum@oracle.com>
Sat, 23 Jul 2016 07:42:35 +0000 (09:42 +0200)
committerDavid S. Miller <davem@davemloft.net>
Tue, 26 Jul 2016 00:43:57 +0000 (17:43 -0700)
I was seeing a lot of these:

    BUG: sleeping function called from invalid context at mm/slab.h:388
    in_atomic(): 0, irqs_disabled(): 0, pid: 14971, name: trinity-c2
    Preemption disabled at:[<ffffffff819bcd46>] rhashtable_walk_start+0x46/0x150

     [<ffffffff81149abb>] preempt_count_add+0x1fb/0x280
     [<ffffffff83295722>] _raw_spin_lock+0x12/0x40
     [<ffffffff811aac87>] console_unlock+0x2f7/0x930
     [<ffffffff811ab5bb>] vprintk_emit+0x2fb/0x520
     [<ffffffff811aba6a>] vprintk_default+0x1a/0x20
     [<ffffffff812c171a>] printk+0x94/0xb0
     [<ffffffff811d6ed0>] print_stack_trace+0xe0/0x170
     [<ffffffff8115835e>] ___might_sleep+0x3be/0x460
     [<ffffffff81158490>] __might_sleep+0x90/0x1a0
     [<ffffffff8139b823>] kmem_cache_alloc+0x153/0x1e0
     [<ffffffff819bca1e>] rhashtable_walk_init+0xfe/0x2d0
     [<ffffffff82ec64de>] sctp_transport_walk_start+0x1e/0x60
     [<ffffffff82edd8ad>] sctp_transport_seq_start+0x4d/0x150
     [<ffffffff8143a82b>] seq_read+0x27b/0x1180
     [<ffffffff814f97fc>] proc_reg_read+0xbc/0x180
     [<ffffffff813d471b>] __vfs_read+0xdb/0x610
     [<ffffffff813d4d3a>] vfs_read+0xea/0x2d0
     [<ffffffff813d615b>] SyS_pread64+0x11b/0x150
     [<ffffffff8100334c>] do_syscall_64+0x19c/0x410
     [<ffffffff832960a5>] return_from_SYSCALL_64+0x0/0x6a
     [<ffffffffffffffff>] 0xffffffffffffffff

Apparently we always need to call rhashtable_walk_stop(), even when
rhashtable_walk_start() fails:

 * rhashtable_walk_start - Start a hash table walk
 * @iter:       Hash table iterator
 *
 * Start a hash table walk.  Note that we take the RCU lock in all
 * cases including when we return an error.  So you must always call
 * rhashtable_walk_stop to clean up.

otherwise we never call rcu_read_unlock() and we get the splat above.

Fixes: 53fa1036 ("sctp: fix some rhashtable functions using in sctp proc/diag")
See-also: 53fa1036 ("sctp: fix some rhashtable functions using in sctp proc/diag")
See-also: f2dba9c6 ("rhashtable: Introduce rhashtable_walk_*")
Cc: Xin Long <lucien.xin@gmail.com>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: stable@vger.kernel.org
Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/sctp/socket.c

index d2681cb1dd30044d62b443311923a94659ce9395..8812e1bf6c1c18fa7e20fdc478633615630162b6 100644 (file)
@@ -4393,6 +4393,7 @@ int sctp_transport_walk_start(struct rhashtable_iter *iter)
 
        err = rhashtable_walk_start(iter);
        if (err && err != -EAGAIN) {
+               rhashtable_walk_stop(iter);
                rhashtable_walk_exit(iter);
                return err;
        }