Merge master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6

* master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6: (109 commits)
  [ETHTOOL]: Fix UFO typo
  [SCTP]: Fix persistent slowdown in sctp when a gap ack consumes rx buffer.
  [SCTP]: Send only 1 window update SACK per message.
  [SCTP]: Don't do CRC32C checksum over loopback.
  [SCTP] Reset rtt_in_progress for the chunk when processing its sack.
  [SCTP]: Reject sctp packets with broadcast addresses.
  [SCTP]: Limit association max_retrans setting in setsockopt.
  [PFKEYV2]: Fix inconsistent typing in struct sadb_x_kmprivate.
  [IPV6]: Sum real space for RTAs.
  [IRDA]: Use put_unaligned() in irlmp_do_discovery().
  [BRIDGE]: Add support for NETIF_F_HW_CSUM devices
  [NET]: Add NETIF_F_GEN_CSUM and NETIF_F_ALL_CSUM
  [TG3]: Convert to non-LLTX
  [TG3]: Remove unnecessary tx_lock
  [TCP]: Add tcp_slow_start_after_idle sysctl.
  [BNX2]: Update version and reldate
  [BNX2]: Use CPU native page size
  [BNX2]: Use compressed firmware
  [BNX2]: Add firmware decompression
  [BNX2]: Allow WoL settings on new 5708 chips
  ...

Manual fixup for conflict in drivers/net/tulip/winbond-840.c

1 files changed, 9 insertions, 1 deletions

diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index 8bc279219a72..9e58144f4851 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -5293,10 +5293,18 @@ static int sctp_eat_data(const struct sctp_association *asoc,
          * seems a bit troublesome in that frag_point varies based on
          * PMTU.  In cases, such as loopback, this might be a rather
          * large spill over.
+         * NOTE: If we have a full receive buffer here, we only renege if
+         * our receiver can still make progress without the tsn being
+         * received. We do this because in the event that the associations
+         * receive queue is empty we are filling a leading gap, and since
+         * reneging moves the gap to the end of the tsn stream, we are likely
+         * to stall again very shortly. Avoiding the renege when we fill a
+         * leading gap is a good heuristic for avoiding such steady state
+         * stalls.
          */
         if (!asoc->rwnd || asoc->rwnd_over ||
             (datalen > asoc->rwnd + asoc->frag_point) ||
-            rcvbuf_over) {
+            (rcvbuf_over && (!skb_queue_len(&sk->sk_receive_queue)))) {
 
                 /* If this is the next TSN, consider reneging to make
                  * room.   Note: Playing nice with a confused sender.  A