Skip to content

Commit

Permalink
sctp: clear the transport of some out_chunk_list chunks in sctp_assoc…
Browse files Browse the repository at this point in the history
…_rm_peer

If a transport is removed by asconf but there still are some chunks with
this transport queuing on out_chunk_list, later an use-after-free issue
will be caused when accessing this transport from these chunks in
sctp_outq_flush().

This is an old bug, we fix it by clearing the transport of these chunks
in out_chunk_list when removing a transport in sctp_assoc_rm_peer().

Reported-by: syzbot+56a40ceee5fb35932f4d@syzkaller.appspotmail.com
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
  • Loading branch information
lxin authored and davem330 committed Oct 30, 2018
1 parent 2b0ab72 commit df132ef
Showing 1 changed file with 7 additions and 3 deletions.
10 changes: 7 additions & 3 deletions net/sctp/associola.c
Original file line number Diff line number Diff line change
Expand Up @@ -499,8 +499,9 @@ void sctp_assoc_set_primary(struct sctp_association *asoc,
void sctp_assoc_rm_peer(struct sctp_association *asoc,
struct sctp_transport *peer)
{
struct list_head *pos;
struct sctp_transport *transport;
struct sctp_transport *transport;
struct list_head *pos;
struct sctp_chunk *ch;

pr_debug("%s: association:%p addr:%pISpc\n",
__func__, asoc, &peer->ipaddr.sa);
Expand Down Expand Up @@ -564,7 +565,6 @@ void sctp_assoc_rm_peer(struct sctp_association *asoc,
*/
if (!list_empty(&peer->transmitted)) {
struct sctp_transport *active = asoc->peer.active_path;
struct sctp_chunk *ch;

/* Reset the transport of each chunk on this list */
list_for_each_entry(ch, &peer->transmitted,
Expand All @@ -586,6 +586,10 @@ void sctp_assoc_rm_peer(struct sctp_association *asoc,
sctp_transport_hold(active);
}

list_for_each_entry(ch, &asoc->outqueue.out_chunk_list, list)
if (ch->transport == peer)
ch->transport = NULL;

asoc->peer.transport_count--;

sctp_transport_free(peer);
Expand Down

0 comments on commit df132ef

Please sign in to comment.