dccp ccid-2: Implementation of circular Ack Vector buffer with overflow handling

This completes the implementation of a circular buffer for Ack Vectors, by 
extending the current (linear array-based) implementation.  The changes are:

 (a) An `overflow' flag to deal with the case of overflow. As before, dynamic
     growth of the buffer will not be supported; but code will be added to deal
     robustly with overflowing Ack Vector buffers.

 (b) A `tail_seqno' field. When naively implementing the algorithm of Appendix A
     in RFC 4340, problems arise whenever subsequent Ack Vector records overlap,
     which can bring the entire run length calculation completely out of synch.
     (This is documented on http://www.erg.abdn.ac.uk/users/gerrit/dccp/notes/\
                                             ack_vectors/tracking_tail_ackno/ .)
 (c) The buffer lengthi is now computed dynamically (i.e. current fill level),
     as the span between head to tail.

As a result, dccp_ackvec_pending() is now simpler - the #ifdef is no longer 
necessary since buf_empty is always true when IP_DCCP_ACKVEC is not configured.

Note on overflow handling: 
-------------------------
 The Ack Vector code previously simply started to drop packets when the
 Ack Vector buffer overflowed. This means that the userspace application
 will not be able to receive, only because of an Ack Vector storage problem.
 
 Furthermore, overflow may be transient, so that applications may later
 recover from the overflow. Recovering from dropped packets is more difficult
 (e.g. video key frames).
 
 Hence the patch uses a different policy: when the buffer overflows, the oldest
 entries are subsequently overwritten. This has a higher chance of recovery.
 Details are on http://www.erg.abdn.ac.uk/users/gerrit/dccp/notes/ack_vectors/

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>

1 files changed, 30 insertions, 1 deletions

diff --git a/net/dccp/ackvec.c b/net/dccp/ackvec.c
index de84dd34c93f..1184d5e5dc96 100644
--- a/net/dccp/ackvec.c
+++ b/net/dccp/ackvec.c
@@ -29,7 +29,7 @@ struct dccp_ackvec *dccp_ackvec_alloc(const gfp_t priority)
         struct dccp_ackvec *av = kmem_cache_zalloc(dccp_ackvec_slab, priority);
 
         if (av != NULL) {
-                av->av_buf_head = DCCPAV_MAX_ACKVEC_LEN - 1;
+                av->av_buf_head = av->av_buf_tail = DCCPAV_MAX_ACKVEC_LEN - 1;
                 INIT_LIST_HEAD(&av->av_records);
         }
         return av;
@@ -72,6 +72,14 @@ int dccp_ackvec_update_records(struct dccp_ackvec *av, u64 seqno, u8 nonce_sum)
         avr->avr_ack_nonce  = nonce_sum;
         avr->avr_ack_runlen = dccp_ackvec_runlen(av->av_buf + av->av_buf_head);
         /*
+         * When the buffer overflows, we keep no more than one record. This is
+         * the simplest way of disambiguating sender-Acks dating from before the
+         * overflow from sender-Acks which refer to after the overflow; a simple
+         * solution is preferable here since we are handling an exception.
+         */
+        if (av->av_overflow)
+                dccp_ackvec_purge_records(av);
+        /*
          * Since GSS is incremented for each packet, the list is automatically
          * arranged in descending order of @ack_seqno.
          */
@@ -85,6 +93,27 @@ int dccp_ackvec_update_records(struct dccp_ackvec *av, u64 seqno, u8 nonce_sum)
 }
 
 /*
+ * Buffer index and length computation using modulo-buffersize arithmetic.
+ * Note that, as pointers move from right to left, head is `before' tail.
+ */
+static inline u16 __ackvec_idx_add(const u16 a, const u16 b)
+{
+        return (a + b) % DCCPAV_MAX_ACKVEC_LEN;
+}
+
+static inline u16 __ackvec_idx_sub(const u16 a, const u16 b)
+{
+        return __ackvec_idx_add(a, DCCPAV_MAX_ACKVEC_LEN - b);
+}
+
+u16 dccp_ackvec_buflen(const struct dccp_ackvec *av)
+{
+        if (unlikely(av->av_overflow))
+                return DCCPAV_MAX_ACKVEC_LEN;
+        return __ackvec_idx_sub(av->av_buf_tail, av->av_buf_head);
+}
+
+/*
  * If several packets are missing, the HC-Receiver may prefer to enter multiple
  * bytes with run length 0, rather than a single byte with a larger run length;
  * this simplifies table updates if one of the missing packets arrives.