dpif-netdev: Avoid copying netdev_flow_key in emc_processing().

Before this commit, emc_processing() copied a netdev_flow_key if there was
no exact-match cache (EMC) hit.  This commit eliminates the copy by
constructing the netdev_flow_key in the place it would be copied.

Found by inspection.

Shahbaz (CCed) reports that this reduces the cost of an EMC miss by 72
cycles in his test case in which the EMC is disabled.  Presumably this
is similarly valuable in cases where the EMC merely has few hits.

For the original version of this patch, which was against a slightly
earlier version of OVS, Daniele reported that:

    - With EMC disabled, this increases throughput from 4.8 Mpps to 5.4
      Mpps.

    - With EMC enabled, this decreases throughput from 12.4 to 12.0 Mpps.

CC: Muhammad Shahbaz <mshahbaz@cs.princeton.edu>
Signed-off-by: Ben Pfaff <blp@ovn.org>
Acked-by: Andy Zhou <azhou@ovn.org>
Acked-by: Daniele Di Proietto <diproiettod@vmware.com>

diff --git a/AUTHORS b/AUTHORS
index 7ae64d3..420aa2d 100644 (file)
--- a/AUTHORS
+++ b/AUTHORS
@@ -352,6 +352,7 @@ Mike Kruze              mkruze@nicira.com
 Min Chen                ustcer.tonychan@gmail.com
 Mikael Doverhag         mdoverhag@nicira.com
 Mrinmoy Das             mrdas@ixiacom.com
+Muhammad Shahbaz        mshahbaz@cs.princeton.edu
 Murali R                muralirdev@gmail.com
 Nagi Reddy Jonnala      njonnala@Brocade.com
 Niels van Adrichem      N.L.M.vanAdrichem@tudelft.nl

diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 833174a..9abb948 100644 (file)
--- a/lib/dpif-netdev.c
+++ b/lib/dpif-netdev.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2009, 2010, 2011, 2012, 2013, 2014 Nicira, Inc.
+ * Copyright (c) 2009, 2010, 2011, 2012, 2013, 2014, 2016 Nicira, Inc.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -3297,7 +3297,6 @@ emc_processing(struct dp_netdev_pmd_thread *pmd, struct dp_packet **packets,
                struct packet_batch batches[], size_t *n_batches)
 {
     struct emc_cache *flow_cache = &pmd->flow_cache;
-    struct netdev_flow_key key;
     size_t i, n_missed = 0, n_dropped = 0;
 
     for (i = 0; i < cnt; i++) {
@@ -3314,19 +3313,19 @@ emc_processing(struct dp_netdev_pmd_thread *pmd, struct dp_packet **packets,
             OVS_PREFETCH(dp_packet_data(packets[i+1]));
         }
 
-        miniflow_extract(packets[i], &key.mf);
-        key.len = 0; /* Not computed yet. */
-        key.hash = dpif_netdev_packet_get_rss_hash(packets[i], &key.mf);
+        struct netdev_flow_key *key = &keys[n_missed];
+        miniflow_extract(packets[i], &key->mf);
+        key->len = 0; /* Not computed yet. */
+        key->hash = dpif_netdev_packet_get_rss_hash(packets[i], &key->mf);
 
-        flow = emc_lookup(flow_cache, &key);
+        flow = emc_lookup(flow_cache, key);
         if (OVS_LIKELY(flow)) {
-            dp_netdev_queue_batches(packets[i], flow, &key.mf, batches,
+            dp_netdev_queue_batches(packets[i], flow, &key->mf, batches,
                                     n_batches);
         } else {
             /* Exact match cache missed. Group missed packets together at
              * the beginning of the 'packets' array.  */
-            packets[n_missed] = packets[i];
-            keys[n_missed++] = key;
+            packets[n_missed++] = packets[i];
         }
     }