/* Resubmit statistics, via xlate_table_action(). */
int recurse; /* Current resubmit nesting depth. */
int resubmits; /* Total number of resubmits. */
+ bool in_group; /* Currently translating ofgroup, if true. */
uint32_t orig_skb_priority; /* Priority when packet arrived. */
uint8_t table_id; /* OpenFlow table ID where flow was found. */
* it did not arrive on a "real" port. 'ofpp_none_bundle' exists for
* when an input bundle is needed for validation (e.g., mirroring or
* OFPP_NORMAL processing). It is not connected to an 'ofproto' or have
- * any 'port' structs, so care must be taken when dealing with it.
- * The bundle's name and vlan mode are initialized in lookup_input_bundle() */
-static struct xbundle ofpp_none_bundle;
+ * any 'port' structs, so care must be taken when dealing with it. */
+static struct xbundle ofpp_none_bundle = {
+ .name = "OFPP_NONE",
+ .vlan_mode = PORT_VLAN_TRUNK
+};
/* Node in 'xport''s 'skb_priorities' map. Used to maintain a map from
* 'priority' (the datapath's term for QoS queue) to the dscp bits which all
OVS_REQ_RDLOCK(xlate_rwlock);
static void xlate_normal(struct xlate_ctx *);
static void xlate_report(struct xlate_ctx *, const char *);
- static void xlate_table_action(struct xlate_ctx *, ofp_port_t in_port,
- uint8_t table_id, bool may_packet_in);
+static void xlate_table_action(struct xlate_ctx *, ofp_port_t in_port,
+ uint8_t table_id, bool may_packet_in,
+ bool honor_table_miss);
static bool input_vid_is_valid(uint16_t vid, struct xbundle *, bool warn);
static uint16_t input_vid_to_vlan(const struct xbundle *, uint16_t vid);
static void output_normal(struct xlate_ctx *, const struct xbundle *,
: NULL;
}
-static enum stp_state
+static bool
xport_stp_learn_state(const struct xport *xport)
{
struct stp_port *sp = xport_get_stp_port(xport);
return stp_forward_in_state(sp ? stp_port_get_state(sp) : STP_DISABLED);
}
+static bool
+xport_stp_listen_state(const struct xport *xport)
+{
+ struct stp_port *sp = xport_get_stp_port(xport);
+ return stp_listen_in_state(sp ? stp_port_get_state(sp) : STP_DISABLED);
+}
+
/* Returns true if STP should process 'flow'. Sets fields in 'wc' that
* were used to make the determination.*/
static bool
/* Special-case OFPP_NONE, which a controller may use as the ingress
* port for traffic that it is sourcing. */
if (in_port == OFPP_NONE) {
- ofpp_none_bundle.name = "OFPP_NONE";
- ofpp_none_bundle.vlan_mode = PORT_VLAN_TRUNK;
return &ofpp_none_bundle;
}
actions_offset = nl_msg_start_nested(odp_actions, OVS_SAMPLE_ATTR_ACTIONS);
odp_port = ofp_port_to_odp_port(xbridge, flow->in_port.ofp_port);
- pid = dpif_port_get_pid(xbridge->dpif, odp_port);
+ pid = dpif_port_get_pid(xbridge->dpif, odp_port, 0);
cookie_offset = odp_put_userspace_action(pid, cookie, cookie_size, odp_actions);
nl_msg_end_nested(odp_actions, actions_offset);
} else if (xport->config & OFPUTIL_PC_NO_FWD) {
xlate_report(ctx, "OFPPC_NO_FWD set, skipping output");
return;
- } else if (check_stp && !xport_stp_forward_state(xport)) {
- xlate_report(ctx, "STP not in forwarding state, skipping output");
- return;
+ } else if (check_stp) {
+ if (eth_addr_equals(ctx->base_flow.dl_dst, eth_addr_stp)) {
+ if (!xport_stp_listen_state(xport)) {
+ xlate_report(ctx, "STP not in listening state, "
+ "skipping bpdu output");
+ return;
+ }
+ } else if (!xport_stp_forward_state(xport)) {
+ xlate_report(ctx, "STP not in forwarding state, "
+ "skipping output");
+ return;
+ }
}
if (mbridge_has_mirrors(ctx->xbridge->mbridge) && xport->xbundle) {
ctx->xout->slow |= special;
} else if (may_receive(peer, ctx)) {
if (xport_stp_forward_state(peer)) {
- xlate_table_action(ctx, flow->in_port.ofp_port, 0, true);
+ xlate_table_action(ctx, flow->in_port.ofp_port, 0, true, true);
} else {
/* Forwarding is disabled by STP. Let OFPP_NORMAL and the
* learning action look at the packet, then drop it. */
struct flow old_base_flow = ctx->base_flow;
size_t old_size = ctx->xout->odp_actions.size;
mirror_mask_t old_mirrors = ctx->xout->mirrors;
- xlate_table_action(ctx, flow->in_port.ofp_port, 0, true);
+ xlate_table_action(ctx, flow->in_port.ofp_port, 0, true, true);
ctx->xout->mirrors = old_mirrors;
ctx->base_flow = old_base_flow;
ctx->xout->odp_actions.size = old_size;
&ctx->xout->odp_actions);
flow->tunnel = flow_tnl; /* Restore tunnel metadata */
} else {
- ofp_port_t vlandev_port;
-
odp_port = xport->odp_port;
+ out_port = odp_port;
if (ofproto_has_vlan_splinters(ctx->xbridge->ofproto)) {
+ ofp_port_t vlandev_port;
+
wc->masks.vlan_tci |= htons(VLAN_VID_MASK | VLAN_CFI);
- }
- vlandev_port = vsp_realdev_to_vlandev(ctx->xbridge->ofproto, ofp_port,
- flow->vlan_tci);
- if (vlandev_port == ofp_port) {
- out_port = odp_port;
- } else {
- out_port = ofp_port_to_odp_port(ctx->xbridge, vlandev_port);
- flow->vlan_tci = htons(0);
+ vlandev_port = vsp_realdev_to_vlandev(ctx->xbridge->ofproto,
+ ofp_port, flow->vlan_tci);
+ if (vlandev_port != ofp_port) {
+ out_port = ofp_port_to_odp_port(ctx->xbridge, vlandev_port);
+ flow->vlan_tci = htons(0);
+ }
}
}
ctx->rule = rule;
actions = rule_dpif_get_actions(rule);
do_xlate_actions(actions->ofpacts, actions->ofpacts_len, ctx);
- rule_actions_unref(actions);
ctx->rule = old_rule;
ctx->recurse--;
}
}
static void
-xlate_table_action(struct xlate_ctx *ctx,
- ofp_port_t in_port, uint8_t table_id, bool may_packet_in)
+xlate_table_action(struct xlate_ctx *ctx, ofp_port_t in_port, uint8_t table_id,
+ bool may_packet_in, bool honor_table_miss)
{
if (xlate_resubmit_resource_check(ctx)) {
ofp_port_t old_in_port = ctx->xin->flow.in_port.ofp_port;
bool skip_wildcards = ctx->xin->skip_wildcards;
uint8_t old_table_id = ctx->table_id;
struct rule_dpif *rule;
+ enum rule_dpif_lookup_verdict verdict;
+ enum ofputil_port_config config = 0;
ctx->table_id = table_id;
* original input port (otherwise OFPP_NORMAL and OFPP_IN_PORT will
* have surprising behavior). */
ctx->xin->flow.in_port.ofp_port = in_port;
- rule_dpif_lookup_in_table(ctx->xbridge->ofproto, &ctx->xin->flow,
- !skip_wildcards ? &ctx->xout->wc : NULL,
- table_id, &rule);
+ verdict = rule_dpif_lookup_from_table(ctx->xbridge->ofproto,
+ &ctx->xin->flow,
+ !skip_wildcards
+ ? &ctx->xout->wc : NULL,
+ honor_table_miss,
+ &ctx->table_id, &rule);
ctx->xin->flow.in_port.ofp_port = old_in_port;
if (ctx->xin->resubmit_hook) {
ctx->xin->resubmit_hook(ctx->xin, rule, ctx->recurse);
}
- if (!rule && may_packet_in) {
- struct xport *xport;
-
- /* XXX
- * check if table configuration flags
- * OFPTC11_TABLE_MISS_CONTROLLER, default.
- * OFPTC11_TABLE_MISS_CONTINUE,
- * OFPTC11_TABLE_MISS_DROP
- * When OF1.0, OFPTC11_TABLE_MISS_CONTINUE is used. What to do? */
- xport = get_ofp_port(ctx->xbridge, ctx->xin->flow.in_port.ofp_port);
- choose_miss_rule(xport ? xport->config : 0,
- ctx->xbridge->miss_rule,
- ctx->xbridge->no_packet_in_rule, &rule);
+ switch (verdict) {
+ case RULE_DPIF_LOOKUP_VERDICT_MATCH:
+ goto match;
+ case RULE_DPIF_LOOKUP_VERDICT_CONTROLLER:
+ if (may_packet_in) {
+ struct xport *xport;
+
+ xport = get_ofp_port(ctx->xbridge,
+ ctx->xin->flow.in_port.ofp_port);
+ config = xport ? xport->config : 0;
+ break;
+ }
+ /* Fall through to drop */
+ case RULE_DPIF_LOOKUP_VERDICT_DROP:
+ config = OFPUTIL_PC_NO_PACKET_IN;
+ break;
+ default:
+ OVS_NOT_REACHED();
}
+
+ choose_miss_rule(config, ctx->xbridge->miss_rule,
+ ctx->xbridge->no_packet_in_rule, &rule);
+
+match:
if (rule) {
xlate_recursively(ctx, rule);
rule_dpif_unref(rule);
static void
xlate_group_action__(struct xlate_ctx *ctx, struct group_dpif *group)
{
+ ctx->in_group = true;
+
switch (group_dpif_get_type(group)) {
case OFPGT11_ALL:
case OFPGT11_INDIRECT:
OVS_NOT_REACHED();
}
group_dpif_release(group);
+
+ ctx->in_group = false;
+}
+
+static bool
+xlate_group_resource_check(struct xlate_ctx *ctx)
+{
+ if (!xlate_resubmit_resource_check(ctx)) {
+ return false;
+ } else if (ctx->in_group) {
+ /* Prevent nested translation of OpenFlow groups.
+ *
+ * OpenFlow allows this restriction. We enforce this restriction only
+ * because, with the current architecture, we would otherwise have to
+ * take a possibly recursive read lock on the ofgroup rwlock, which is
+ * unsafe given that POSIX allows taking a read lock to block if there
+ * is a thread blocked on taking the write lock. Other solutions
+ * without this restriction are also possible, but seem unwarranted
+ * given the current limited use of groups. */
+ static struct vlog_rate_limit rl = VLOG_RATE_LIMIT_INIT(1, 1);
+
+ VLOG_ERR_RL(&rl, "cannot recursively translate OpenFlow group");
+ return false;
+ } else {
+ return true;
+ }
}
static bool
xlate_group_action(struct xlate_ctx *ctx, uint32_t group_id)
{
- if (xlate_resubmit_resource_check(ctx)) {
+ if (xlate_group_resource_check(ctx)) {
struct group_dpif *group;
bool got_group;
table_id = ctx->table_id;
}
- xlate_table_action(ctx, in_port, table_id, false);
+ xlate_table_action(ctx, in_port, table_id, false, false);
}
static void
&ctx->xout->odp_actions,
&ctx->xout->wc);
- odp_execute_actions(NULL, packet, &md, ctx->xout->odp_actions.data,
+ odp_execute_actions(NULL, packet, false, &md, ctx->xout->odp_actions.data,
ctx->xout->odp_actions.size, NULL);
pin = xmalloc(sizeof *pin);
pin->controller_id = controller_id;
pin->send_len = len;
- pin->generated_by_table_miss = (ctx->rule
- && rule_dpif_is_table_miss(ctx->rule));
+ /* If a rule is a table-miss rule then this is
+ * a table-miss handled by a table-miss rule.
+ *
+ * Else, if rule is internal and has a controller action,
+ * the later being implied by the rule being processed here,
+ * then this is a table-miss handled without a table-miss rule.
+ *
+ * Otherwise this is not a table-miss. */
+ pin->miss_type = OFPROTO_PACKET_IN_NO_MISS;
+ if (ctx->rule) {
+ if (rule_dpif_is_table_miss(ctx->rule)) {
+ pin->miss_type = OFPROTO_PACKET_IN_MISS_FLOW;
+ } else if (rule_dpif_is_internal(ctx->rule)) {
+ pin->miss_type = OFPROTO_PACKET_IN_MISS_WITHOUT_FLOW;
+ }
+ }
ofproto_dpif_send_packet_in(ctx->xbridge->ofproto, pin);
ofpbuf_delete(packet);
}
{
struct flow_wildcards *wc = &ctx->xout->wc;
struct flow *flow = &ctx->xin->flow;
- ovs_be16 vlan_tci = flow->vlan_tci;
int n;
ovs_assert(eth_type_mpls(mpls->ethertype));
n = flow_count_mpls_labels(flow, wc);
if (!n) {
- if (mpls->position == OFPACT_MPLS_BEFORE_VLAN) {
- vlan_tci = 0;
- } else {
- flow->vlan_tci = 0;
- }
ctx->xout->slow |= commit_odp_actions(flow, &ctx->base_flow,
&ctx->xout->odp_actions,
&ctx->xout->wc);
}
flow_push_mpls(flow, n, mpls->ethertype, wc);
- flow->vlan_tci = vlan_tci;
}
static void
break;
case OFPP_TABLE:
xlate_table_action(ctx, ctx->xin->flow.in_port.ofp_port,
- 0, may_packet_in);
+ 0, may_packet_in, true);
break;
case OFPP_NORMAL:
xlate_normal(ctx);
ovs_assert(ctx->table_id < ogt->table_id);
xlate_table_action(ctx, ctx->xin->flow.in_port.ofp_port,
- ogt->table_id, true);
+ ogt->table_id, true, true);
break;
}
struct xlate_ctx ctx;
size_t ofpacts_len;
bool tnl_may_send;
+ bool is_icmp;
COVERAGE_INC(xlate_actions);
if (is_ip_any(flow)) {
wc->masks.nw_frag |= FLOW_NW_FRAG_MASK;
}
+ is_icmp = is_icmpv4(flow) || is_icmpv6(flow);
tnl_may_send = tnl_xlate_init(&ctx.base_flow, flow, wc);
if (ctx.xbridge->netflow) {
ctx.recurse = 0;
ctx.resubmits = 0;
+ ctx.in_group = false;
ctx.orig_skb_priority = flow->skb_priority;
ctx.table_id = 0;
ctx.exit = false;
if (!xin->ofpacts && !ctx.rule) {
- rule_dpif_lookup(ctx.xbridge->ofproto, flow,
- !xin->skip_wildcards ? wc : NULL, &rule);
+ ctx.table_id = rule_dpif_lookup(ctx.xbridge->ofproto, flow,
+ !xin->skip_wildcards ? wc : NULL,
+ &rule);
if (ctx.xin->resubmit_stats) {
rule_dpif_credit_stats(rule, ctx.xin->resubmit_stats);
}
* use non-header fields as part of the cache. */
flow_wildcards_clear_non_packet_fields(wc);
+ /* ICMPv4 and ICMPv6 have 8-bit "type" and "code" fields. struct flow uses
+ * the low 8 bits of the 16-bit tp_src and tp_dst members to represent
+ * these fields. The datapath interface, on the other hand, represents
+ * them with just 8 bits each. This means that if the high 8 bits of the
+ * masks for these fields somehow become set, then they will get chopped
+ * off by a round trip through the datapath, and revalidation will spot
+ * that as an inconsistency and delete the flow. Avoid the problem here by
+ * making sure that only the low 8 bits of either field can be unwildcarded
+ * for ICMP.
+ */
+ if (is_icmp) {
+ wc->masks.tp_src &= htons(UINT8_MAX);
+ wc->masks.tp_dst &= htons(UINT8_MAX);
+ }
+
out:
- rule_actions_unref(actions);
rule_dpif_unref(rule);
}
struct xport *xport;
struct ofpact_output output;
struct flow flow;
- union flow_in_port in_port_;
ofpact_init(&output.ofpact, OFPACT_OUTPUT, sizeof output);
/* Use OFPP_NONE as the in_port to avoid special packet processing. */
- in_port_.ofp_port = OFPP_NONE;
- flow_extract(packet, 0, 0, NULL, &in_port_, &flow);
+ flow_extract(packet, NULL, &flow);
+ flow.in_port.ofp_port = OFPP_NONE;
ovs_rwlock_rdlock(&xlate_rwlock);
xport = xport_lookup(ofport);