Hi All,
here's a potential locking cycle in
2.5.62/net/sctp/input.c.
Confirmation/correction appreciated.
Thanks,
Dawson
ERROR: 1 thread local-local deadlock.
<sctp_hashbucket_t.lock>-><sctp_endpoint_common_t.addr_lock> occurred 2 times
<sctp_endpoint_common_t.addr_lock>-><sctp_hashbucket_t.lock> occurred 2 times
depth = 2:
net/sctp/input.c:__sctp_lookup_association:634
->net/sctp/input.c:__sctp_lookup_association:634
hash = sctp_assoc_hashfn(laddr->v4.sin_port, paddr->v4.sin_port);
head = &sctp_proto.assoc_hashbucket[hash];
read_lock(&head->lock);
for (epb = head->chain; epb; epb = epb->next) {
asoc = sctp_assoc(epb);
transport = sctp_assoc_is_match(asoc, laddr, paddr);
->__sctp_lookup_association:637
->end=net/sctp/associola.c:sctp_assoc_is_match:754:read_lock
->net/sctp/associola.c:sctp_assoc_is_match:754
sctp_read_lock(&asoc->base.addr_lock);
depth = 2:
net/sctp/input.c:__sctp_rcv_lookup_endpoint:536
->net/sctp/input.c:__sctp_rcv_lookup_endpoint:536
->__sctp_rcv_lookup_endpoint:539
->end=net/sctp/endpointola.c:sctp_endpoint_is_match:241:read_lock
While the readlocks are recursive, can be called in a place without
*->base.addr_lock held (e.g., sctp_v4_err).
<sctp_endpoint_common_t.addr_lock>-><sctp_hashbucket_t.lock> =
depth = 4:
net/sctp/endpointola.c:sctp_endpoint_is_peeled_off:313
->net/sctp/endpointola.c:sctp_endpoint_is_peeled_off:313
->sctp_endpoint_is_peeled_off:317
sctp_read_lock(&ep->base.addr_lock);
bp = &ep->base.bind_addr;
list_for_each(pos, &bp->address_list) {
addr = list_entry(pos, struct sockaddr_storage_list, list);
if (sctp_has_association(&addr->a, paddr)) {
->net/sctp/input.c:sctp_has_association:675
->sctp_lookup_association:662
->end=__sctp_lookup_association:634:read_lock
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun Mar 23 2003 - 22:00:40 EST