ofproto-dpif-upcall: Reset ukey's last stats value if the datapath changed.

When the ukey's action set changes, it could cause the flow to use a
different datapath, for example, when it moves from tc to kernel.
This will cause the the cached previous datapath statistics to be used.

This change will reset the cached statistics when a change in
datapath is discovered.

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Eelco Chaudron <echaudro@redhat.com>
Signed-off-by: Ilya Maximets <i.maximets@ovn.org>

7 files changed, 116 insertions, 2 deletions

diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 9331f2cba..aa2e9292b 100644
--- a/lib/dpif-netdev.c
+++ b/lib/dpif-netdev.c
@@ -9531,6 +9531,7 @@ dpif_netdev_bond_stats_get(struct dpif *dpif, uint32_t bond_id,
 const struct dpif_class dpif_netdev_class = {
     "netdev",
     true,                       /* cleanup_required */
+    true,                       /* synced_dp_layers */
     dpif_netdev_init,
     dpif_netdev_enumerate,
     dpif_netdev_port_open_type,
diff --git a/lib/dpif-netlink.c b/lib/dpif-netlink.c
index a620a6ec5..620bf4674 100644
--- a/lib/dpif-netlink.c
+++ b/lib/dpif-netlink.c
@@ -4516,6 +4516,7 @@ dpif_netlink_cache_set_size(struct dpif *dpif_, uint32_t level, uint32_t size)
 const struct dpif_class dpif_netlink_class = {
     "system",
     false,                      /* cleanup_required */
+    false,                      /* synced_dp_layers */
     NULL,                       /* init */
     dpif_netlink_enumerate,
     NULL,
diff --git a/lib/dpif-provider.h b/lib/dpif-provider.h
index 12477a24f..b8ead8a02 100644
--- a/lib/dpif-provider.h
+++ b/lib/dpif-provider.h
@@ -127,6 +127,14 @@ struct dpif_class {
      * datapaths that can not exist without it (e.g. netdev datapath).  */
     bool cleanup_required;
 
+    /* If 'true' the specific dpif implementation synchronizes the various
+     * datapath implementation layers, i.e., the dpif's layer in combination
+     * with the underlying netdev offload layers. For example, dpif-netlink
+     * does not sync its kernel flows with the tc ones, i.e., only one gets
+     * installed. On the other hand, dpif-netdev installs both flows,
+     * internally keeps track of both, and represents them as one. */
+    bool synced_dp_layers;
+
     /* Called when the dpif provider is registered, typically at program
      * startup.  Returning an error from this function will prevent any
      * datapath with this class from being created.
diff --git a/lib/dpif.c b/lib/dpif.c
index fe4db83fb..3305401fe 100644
--- a/lib/dpif.c
+++ b/lib/dpif.c
@@ -2109,3 +2109,9 @@ dpif_cache_set_size(struct dpif *dpif, uint32_t level, uint32_t size)
            ? dpif->dpif_class->cache_set_size(dpif, level, size)
            : EOPNOTSUPP;
 }
+
+bool
+dpif_synced_dp_layers(struct dpif *dpif)
+{
+    return dpif->dpif_class->synced_dp_layers;
+}
diff --git a/lib/dpif.h b/lib/dpif.h
index 6cb4dae6d..129cbf6a1 100644
--- a/lib/dpif.h
+++ b/lib/dpif.h
@@ -939,6 +939,7 @@ int dpif_get_pmds_for_port(const struct dpif * dpif, odp_port_t port_no,
 char *dpif_get_dp_version(const struct dpif *);
 bool dpif_supports_tnl_push_pop(const struct dpif *);
 bool dpif_supports_explicit_drop_action(const struct dpif *);
+bool dpif_synced_dp_layers(struct dpif *);
 
 /* Log functions. */
 struct vlog_module;
diff --git a/ofproto/ofproto-dpif-upcall.c b/ofproto/ofproto-dpif-upcall.c
index 59126b901..9c8ab17f3 100644
--- a/ofproto/ofproto-dpif-upcall.c
+++ b/ofproto/ofproto-dpif-upcall.c
@@ -47,17 +47,20 @@
 
 #define UPCALL_MAX_BATCH 64
 #define REVALIDATE_MAX_BATCH 50
+#define UINT64_THREE_QUARTERS (UINT64_MAX / 4 * 3)
 
 VLOG_DEFINE_THIS_MODULE(ofproto_dpif_upcall);
 
 COVERAGE_DEFINE(dumped_duplicate_flow);
 COVERAGE_DEFINE(dumped_new_flow);
 COVERAGE_DEFINE(handler_duplicate_upcall);
-COVERAGE_DEFINE(upcall_ukey_contention);
-COVERAGE_DEFINE(upcall_ukey_replace);
 COVERAGE_DEFINE(revalidate_missed_dp_flow);
+COVERAGE_DEFINE(ukey_dp_change);
+COVERAGE_DEFINE(ukey_invalid_stat_reset);
 COVERAGE_DEFINE(upcall_flow_limit_hit);
 COVERAGE_DEFINE(upcall_flow_limit_kill);
+COVERAGE_DEFINE(upcall_ukey_contention);
+COVERAGE_DEFINE(upcall_ukey_replace);
 
 /* A thread that reads upcalls from dpif, forwards each upcall's packet,
  * and possibly sets up a kernel flow as a cache. */
@@ -287,6 +290,7 @@ struct udpif_key {
 
     struct ovs_mutex mutex;                   /* Guards the following. */
     struct dpif_flow_stats stats OVS_GUARDED; /* Last known stats.*/
+    const char *dp_layer OVS_GUARDED;         /* Last known dp_layer. */
     long long int created OVS_GUARDED;        /* Estimate of creation time. */
     uint64_t dump_seq OVS_GUARDED;            /* Tracks udpif->dump_seq. */
     uint64_t reval_seq OVS_GUARDED;           /* Tracks udpif->reval_seq. */
@@ -1754,6 +1758,7 @@ ukey_create__(const struct nlattr *key, size_t key_len,
     ukey->created = ukey->flow_time = time_msec();
     memset(&ukey->stats, 0, sizeof ukey->stats);
     ukey->stats.used = used;
+    ukey->dp_layer = NULL;
     ukey->xcache = NULL;
 
     ukey->offloaded = false;
@@ -2330,6 +2335,13 @@ revalidate_ukey(struct udpif *udpif, struct udpif_key *ukey,
                     ? stats->n_bytes - ukey->stats.n_bytes
                     : 0);
 
+    if (stats->n_packets < ukey->stats.n_packets &&
+        ukey->stats.n_packets < UINT64_THREE_QUARTERS) {
+        /* Report cases where the packet counter is lower than the previous
+         * instance, but exclude the potential wrapping of an uint64_t. */
+        COVERAGE_INC(ukey_invalid_stat_reset);
+    }
+
     if (need_revalidate) {
         if (should_revalidate(udpif, push.n_packets, ukey->stats.used)) {
             if (!ukey->xcache) {
@@ -2443,6 +2455,15 @@ push_dp_ops(struct udpif *udpif, struct ukey_op *ops, size_t n_ops)
             push->tcp_flags = stats->tcp_flags | op->ukey->stats.tcp_flags;
             push->n_packets = stats->n_packets - op->ukey->stats.n_packets;
             push->n_bytes = stats->n_bytes - op->ukey->stats.n_bytes;
+
+            if (stats->n_packets < op->ukey->stats.n_packets &&
+                op->ukey->stats.n_packets < UINT64_THREE_QUARTERS) {
+                /* Report cases where the packet counter is lower than the
+                 * previous instance, but exclude the potential wrapping of an
+                 * uint64_t. */
+                COVERAGE_INC(ukey_invalid_stat_reset);
+            }
+
             ovs_mutex_unlock(&op->ukey->mutex);
         } else {
             push = stats;
@@ -2747,6 +2768,22 @@ revalidate(struct revalidator *revalidator)
                 continue;
             }
 
+            ukey->offloaded = f->attrs.offloaded;
+            if (!ukey->dp_layer
+                || (!dpif_synced_dp_layers(udpif->dpif)
+                    && strcmp(ukey->dp_layer, f->attrs.dp_layer))) {
+
+                if (ukey->dp_layer) {
+                    /* The dp_layer has changed this is probably due to an
+                     * earlier revalidate cycle moving it to/from hw offload.
+                     * In this case we should reset the ukey stored statistics,
+                     * as they are from the deleted DP flow. */
+                    COVERAGE_INC(ukey_dp_change);
+                    memset(&ukey->stats, 0, sizeof ukey->stats);
+                }
+                ukey->dp_layer = f->attrs.dp_layer;
+            }
+
             already_dumped = ukey->dump_seq == dump_seq;
             if (already_dumped) {
                 /* The flow has already been handled during this flow dump
diff --git a/tests/system-offloads-traffic.at b/tests/system-offloads-traffic.at
index d36da0580..0aba60f28 100644
--- a/tests/system-offloads-traffic.at
+++ b/tests/system-offloads-traffic.at
@@ -742,3 +742,63 @@ recirc_id(<recirc>),in_port(3),eth_type(0x0800),ipv4(frag=no), packets:29, bytes
 
 OVS_TRAFFIC_VSWITCHD_STOP
 AT_CLEANUP
+
+
+AT_SETUP([offloads - offload flow to none-offload])
+OVS_TRAFFIC_VSWITCHD_START([], [], [-- set Open_vSwitch . other_config:hw-offload=true])
+
+ADD_NAMESPACES(at_ns0, at_ns1)
+
+ADD_VETH(p0, at_ns0, br0, "10.1.1.1/24")
+ADD_VETH(p1, at_ns1, br0, "10.1.1.2/24")
+
+AT_DATA([flows.txt], [dnl
+add in_port=ovs-p0,actions=ovs-p1
+add in_port=ovs-p1,actions=ovs-p0
+])
+AT_CHECK([ovs-ofctl add-flows br0 flows.txt])
+
+NS_CHECK_EXEC([at_ns0], [ping -q -c 10 -i 0.1 -w 2 10.1.1.2 | FORMAT_PING], [0], [dnl
+10 packets transmitted, 10 received, 0% packet loss, time 0ms
+])
+
+AT_CHECK([ovs-appctl dpctl/dump-flows type=tc | grep "eth_type(0x0800)" | sort | strip_recirc | strip_used], [0], [dnl
+recirc_id(<recirc>),in_port(2),eth(),eth_type(0x0800),ipv4(frag=no), packets:9, bytes:756, used:0.0s, actions:3
+recirc_id(<recirc>),in_port(3),eth(),eth_type(0x0800),ipv4(frag=no), packets:9, bytes:756, used:0.0s, actions:2
+])
+
+dnl Here we use an output action with truncate, which will force a kernel flow.
+AT_DATA([flows2.txt], [dnl
+modify in_port=ovs-p0,actions=output(port=ovs-p1, max_len=128)
+modify in_port=ovs-p1,actions=output(port=ovs-p0, max_len=128)
+])
+AT_CHECK([ovs-ofctl add-flows br0 flows2.txt])
+AT_CHECK([ovs-appctl revalidator/wait], [0])
+
+NS_CHECK_EXEC([at_ns0], [ping -q -c 10 -i 0.1 -w 2 10.1.1.2 | FORMAT_PING], [0], [dnl
+10 packets transmitted, 10 received, 0% packet loss, time 0ms
+])
+
+AT_CHECK([ovs-appctl dpctl/dump-flows type=ovs | grep "eth_type(0x0800)" | sort | strip_recirc | strip_used], [0], [dnl
+recirc_id(<recirc>),in_port(2),eth(),eth_type(0x0800),ipv4(frag=no), packets:10, bytes:980, used:0.0s, actions:trunc(128),3
+recirc_id(<recirc>),in_port(3),eth(),eth_type(0x0800),ipv4(frag=no), packets:10, bytes:980, used:0.0s, actions:trunc(128),2
+])
+
+AT_CHECK([ovs-ofctl add-flows br0 flows.txt])
+AT_CHECK([ovs-appctl revalidator/wait], [0])
+
+NS_CHECK_EXEC([at_ns0], [ping -q -c 10 -i 0.1 -w 2 10.1.1.2 | FORMAT_PING], [0], [dnl
+10 packets transmitted, 10 received, 0% packet loss, time 0ms
+])
+
+AT_CHECK([ovs-appctl dpctl/dump-flows type=tc | grep "eth_type(0x0800)" | sort | strip_recirc | strip_used], [0], [dnl
+recirc_id(<recirc>),in_port(2),eth(),eth_type(0x0800),ipv4(frag=no), packets:10, bytes:840, used:0.0s, actions:3
+recirc_id(<recirc>),in_port(3),eth(),eth_type(0x0800),ipv4(frag=no), packets:10, bytes:840, used:0.0s, actions:2
+])
+
+AT_CHECK([ovs-appctl coverage/read-counter ukey_invalid_stat_reset], [0], [dnl
+0
+])
+
+OVS_TRAFFIC_VSWITCHD_STOP
+AT_CLEANUP