Merge branch 'master' into next

Conflicts:
	COPYING
	datapath/datapath.h
	lib/automake.mk
	lib/dpif-provider.h
	lib/dpif.c
	lib/hmap.h
	lib/netdev-provider.h
	lib/netdev.c
	lib/stream-ssl.h
	ofproto/executer.c
	ofproto/ofproto.c
	ofproto/ofproto.h
	tests/automake.mk
	utilities/ovs-ofctl.c
	utilities/ovs-vsctl.in
	vswitchd/ovs-vswitchd.conf.5.in
	xenserver/etc_init.d_vswitch
	xenserver/etc_xensource_scripts_vif
	xenserver/opt_xensource_libexec_interface-reconfigure

COPYING

@@ -46,3 +46,7 @@ The files under ovsdb/simplejson are covered by the following license:
     WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
     OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
     DEALINGS IN THE SOFTWARE.
+
+Files lib/sflow*.[ch] are licensed under the terms of the InMon sFlow
+licence that is available at:
+        http://www.inmon.com/technology/sflowlicense.txt

ChangeLog

@@ -1,3 +1,14 @@
+v0.99.1 - 25 Jan 2010
+---------------------
+    - Add support for sFlow(R)
+    - Make headers compatible with C++
+    - Bug fixes
+
+v0.99.0 - 14 Jan 2010
+---------------------
+    - User-space forwarding engine
+    - Bug fixes
+
 v0.90.6 - 6 Oct 2009
 --------------------
     - Bug fixes

README

@@ -6,8 +6,8 @@ What is Open vSwitch?
 Open vSwitch is a multilayer software switch licensed under the open
 source Apache 2 license.  Our goal is to implement a production
 quality switch platform that supports standard management interfaces
-(e.g. NetFlow, RSPAN, ERSPAN, IOS-like CLI), and opens the forwarding
-functions to programmatic extension and control.
+(e.g. NetFlow, sFlow(R), RSPAN, ERSPAN, IOS-like CLI), and opens the
+forwarding functions to programmatic extension and control.
 
 Open vSwitch is well suited to function as a virtual switch in VM
 environments.  In addition to exposing standard control and visibility
@@ -20,7 +20,8 @@ The bulk of the code is written in platform-independent C and is
 easily ported to other environments.  The current release of Open
 vSwitch supports the following features:
 
-    * Visibility into inter-VM communication via NetFlow, SPAN, and RSPAN
+    * Visibility into inter-VM communication via NetFlow, sFlow, SPAN,
+      and RSPAN
     * Standard 802.1Q VLAN model with trunking
     * Per VM policing
     * NIC bonding with source-MAC load balancing

acinclude.m4

@@ -215,6 +215,7 @@ AC_DEFUN([OVS_CHECK_STRTOK_R],
                            char *token1, *token2;
                            token1 = strtok_r(string, ":", &save_ptr);
                            token2 = strtok_r(NULL, ":", &save_ptr);
+                           freopen ("/dev/null", "w", stdout);
                            printf ("%s %s\n", token1, token2);
                            return 0;
                           ]])],
@@ -259,4 +260,14 @@ dnl Example: OVS_ENABLE_OPTION([-Wdeclaration-after-statement])
 AC_DEFUN([OVS_ENABLE_OPTION], 
   [OVS_CHECK_CC_OPTION([$1], [WARNING_FLAGS="$WARNING_FLAGS $1"])
    AC_SUBST([WARNING_FLAGS])])
+
+dnl OVS_CONDITIONAL_CC_OPTION([OPTION], [CONDITIONAL])
+dnl Check whether the given C compiler OPTION is accepted.
+dnl If so, enable the given Automake CONDITIONAL.
+
+dnl Example: OVS_CONDITIONAL_CC_OPTION([-Wno-unused], [HAVE_WNO_UNUSED])
+AC_DEFUN([OVS_CONDITIONAL_CC_OPTION],
+  [OVS_CHECK_CC_OPTION(
+    [$1], [ovs_have_cc_option=yes], [ovs_have_cc_option=no])
+   AM_CONDITIONAL([$2], [test $ovs_have_cc_option = yes])])
 dnl ----------------------------------------------------------------------

configure.ac

@@ -13,7 +13,7 @@
 # limitations under the License.
 
 AC_PREREQ(2.63)
-AC_INIT(openvswitch, 0.90.6, [email protected])
+AC_INIT(openvswitch, 0.99.1, [email protected])
 NX_BUILDNR
 AC_CONFIG_SRCDIR([datapath/datapath.c])
 AC_CONFIG_MACRO_DIR([m4])
@@ -76,6 +76,7 @@ OVS_ENABLE_OPTION([-Wold-style-definition])
 OVS_ENABLE_OPTION([-Wmissing-prototypes])
 OVS_ENABLE_OPTION([-Wmissing-field-initializers])
 OVS_ENABLE_OPTION([-Wno-override-init])
+OVS_CONDITIONAL_CC_OPTION([-Wno-unused], [HAVE_WNO_UNUSED])
 
 AC_ARG_VAR(KARCH, [Kernel Architecture String])
 AC_SUBST(KARCH)

datapath/actions.c

@@ -400,6 +400,28 @@ output_control(struct datapath *dp, struct sk_buff *skb, u32 arg, gfp_t gfp)
 	return dp_output_control(dp, skb, _ODPL_ACTION_NR, arg);
 }
 
+/* Send a copy of this packet up to the sFlow agent, along with extra
+ * information about what happened to it. */
+static void sflow_sample(struct datapath *dp, struct sk_buff *skb,
+			 const union odp_action *a, int n_actions,
+			 gfp_t gfp, struct net_bridge_port *nbp)
+{
+	struct odp_sflow_sample_header *hdr;
+	unsigned int actlen = n_actions * sizeof(union odp_action);
+	unsigned int hdrlen = sizeof(struct odp_sflow_sample_header);
+	struct sk_buff *nskb;
+
+	nskb = skb_copy_expand(skb, actlen + hdrlen, 0, gfp);
+	if (!nskb)
+		return;
+
+	memcpy(__skb_push(nskb, actlen), a, actlen);
+	hdr = (struct odp_sflow_sample_header*)__skb_push(nskb, hdrlen);
+	hdr->n_actions = n_actions;
+	hdr->sample_pool = atomic_read(&nbp->sflow_pool);
+	dp_output_control(dp, nskb, _ODPL_SFLOW_NR, 0);
+}
+
 /* Execute a list of actions against 'skb'. */
 int execute_actions(struct datapath *dp, struct sk_buff *skb,
 		    struct odp_flow_key *key,
@@ -412,6 +434,17 @@ int execute_actions(struct datapath *dp, struct sk_buff *skb,
 	 * is slightly obscure just to avoid that. */
 	int prev_port = -1;
 	int err;
+
+	if (dp->sflow_probability) {
+		struct net_bridge_port *p = skb->dev->br_port;
+		if (p) {
+			atomic_inc(&p->sflow_pool);
+			if (dp->sflow_probability == UINT_MAX ||
+			    net_random() < dp->sflow_probability)
+				sflow_sample(dp, skb, a, n_actions, gfp, p);
+		}
+	}
+
 	for (; n_actions > 0; a++, n_actions--) {
 		WARN_ON_ONCE(skb_shared(skb));
 		if (prev_port != -1) {

datapath/datapath.c

@@ -349,6 +349,7 @@ static int new_nbp(struct datapath *dp, struct net_device *dev, int port_no)
 	p->port_no = port_no;
 	p->dp = dp;
 	p->dev = dev;
+	atomic_set(&p->sflow_pool, 0);
 	if (!is_dp_dev(dev))
 		rcu_assign_pointer(dev->br_port, p);
 	else {
@@ -646,9 +647,7 @@ int vswitch_skb_checksum_setup(struct sk_buff *skb)
 out:
 	return err;
 }
-#else
-int vswitch_skb_checksum_setup(struct sk_buff *skb) { return 0; }
-#endif /* CONFIG_XEN && linux == 2.6.18 */
+#endif /* CONFIG_XEN && HAVE_PROTO_DATA_VALID */
 
  /* Types of checksums that we can receive (these all refer to L4 checksums):
  * 1. CHECKSUM_NONE: Device that did not compute checksum, contains full
@@ -796,8 +795,7 @@ dp_output_control(struct datapath *dp, struct sk_buff *skb, int queue_no,
 	int err;
 
 	WARN_ON_ONCE(skb_shared(skb));
-	BUG_ON(queue_no != _ODPL_MISS_NR && queue_no != _ODPL_ACTION_NR);
-
+	BUG_ON(queue_no != _ODPL_MISS_NR && queue_no != _ODPL_ACTION_NR && queue_no != _ODPL_SFLOW_NR);
 	queue = &dp->queues[queue_no];
 	err = -ENOBUFS;
 	if (skb_queue_len(queue) >= DP_MAX_QUEUE_LEN)
@@ -1499,6 +1497,7 @@ static long openvswitch_ioctl(struct file *f, unsigned int cmd,
 	int dp_idx = iminor(f->f_dentry->d_inode);
 	struct datapath *dp;
 	int drop_frags, listeners, port_no;
+	unsigned int sflow_probability;
 	int err;
 
 	/* Handle commands with special locking requirements up front. */
@@ -1562,6 +1561,16 @@ static long openvswitch_ioctl(struct file *f, unsigned int cmd,
 		set_listen_mask(f, listeners);
 		break;
 
+	case ODP_GET_SFLOW_PROBABILITY:
+		err = put_user(dp->sflow_probability, (unsigned int __user *)argp);
+		break;
+
+	case ODP_SET_SFLOW_PROBABILITY:
+		err = get_user(sflow_probability, (unsigned int __user *)argp);
+		if (!err)
+			dp->sflow_probability = sflow_probability;
+		break;
+
 	case ODP_PORT_QUERY:
 		err = query_port(dp, (struct odp_port __user *)argp);
 		break;

datapath/datapath.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2009 Nicira Networks.
+ * Copyright (c) 2009, 2010 Nicira Networks.
  * Distributed under the terms of the GNU GPL version 2.
  *
  * Significant portions of this file may be copied from parts of the Linux
@@ -79,9 +79,22 @@ struct dp_bucket {
 	struct sw_flow *flows[];
 };
 
-#define DP_N_QUEUES 2
+#define DP_N_QUEUES 3
 #define DP_MAX_QUEUE_LEN 100
 
+/**
+ * struct dp_stats_percpu - per-cpu packet processing statistics for a given
+ * datapath.
+ * @n_frags: Number of IP fragments processed by datapath.
+ * @n_hit: Number of received packets for which a matching flow was found in
+ * the flow table.
+ * @n_miss: Number of received packets that had no matching flow in the flow
+ * table.  The sum of @n_hit and @n_miss is the number of packets that have
+ * been received by the datapath.
+ * @n_lost: Number of received packets that had no matching flow in the flow
+ * table that could not be sent to userspace (normally due to an overflow in
+ * one of the datapath's queues).
+ */
 struct dp_stats_percpu {
 	u64 n_frags;
 	u64 n_hit;
@@ -95,10 +108,29 @@ struct dp_port_group {
 	u16 ports[];
 };
 
+/**
+ * struct datapath - datapath for flow-based packet switching
+ * @mutex: Mutual exclusion for ioctls.
+ * @dp_idx: Datapath number (index into the dps[] array in datapath.c).
+ * @ifobj: Represents /sys/class/net/<devname>/brif.
+ * @drop_frags: Drop all IP fragments if nonzero.
+ * @queues: %DP_N_QUEUES sets of queued packets for userspace to handle.
+ * @waitqueue: Waitqueue, for waiting for new packets in @queues.
+ * @n_flows: Number of flows currently in flow table.
+ * @table: Current flow table (RCU protected).
+ * @groups: Port groups, used by ODPAT_OUTPUT_GROUP action (RCU protected).
+ * @n_ports: Number of ports currently in @ports.
+ * @ports: Map from port number to &struct net_bridge_port.  %ODPP_LOCAL port
+ * always exists, other ports may be %NULL.
+ * @port_list: List of all ports in @ports in arbitrary order.
+ * @stats_percpu: Per-CPU datapath statistics.
+ * @sflow_probability: Number of packets out of UINT_MAX to sample to the
+ * %ODPL_SFLOW queue, e.g. (@sflow_probability/UINT_MAX) is the probability of
+ * sampling a given packet.
+ */
 struct datapath {
 	struct mutex mutex;
 	int dp_idx;
-
 	struct kobject ifobj;
 
 	int drop_frags;
@@ -117,19 +149,37 @@ struct datapath {
 	/* Switch ports. */
 	unsigned int n_ports;
 	struct net_bridge_port *ports[DP_MAX_PORTS];
-	struct list_head port_list; /* All ports, including local_port. */
+	struct list_head port_list;
 
 	/* Stats. */
 	struct dp_stats_percpu *stats_percpu;
+
+	/* sFlow Sampling */
+	unsigned int sflow_probability;
 };
 
+/**
+ * struct net_bridge_port - one port within a datapath
+ * @port_no: Index into @dp's @ports array.
+ * @dp: Datapath to which this port belongs.
+ * @dev: The network device attached to this port.  The @br_port member in @dev
+ * points back to this &struct net_bridge_port.
+ * @kobj: Represents /sys/class/net/<devname>/brport.
+ * @linkname: The name of the link from /sys/class/net/<datapath>/brif to this
+ * &struct net_bridge_port.  (We keep this around so that we can delete it
+ * if @dev gets renamed.)  Set to the null string when no link exists.
+ * @node: Element in @dp's @port_list.
+ * @sflow_pool: Number of packets that were candidates for sFlow sampling,
+ * regardless of whether they were actually chosen and sent down to userspace.
+ */
 struct net_bridge_port {
 	u16 port_no;
 	struct datapath	*dp;
 	struct net_device *dev;
 	struct kobject kobj;
 	char linkname[IFNAMSIZ];
-	struct list_head node;   /* Element in datapath.ports. */
+	struct list_head node;
+	atomic_t sflow_pool;
 };
 
 extern struct notifier_block dp_device_notifier;
@@ -160,16 +210,15 @@ static inline const char *dp_name(const struct datapath *dp)
 	return dp->ports[ODPP_LOCAL]->dev->name;
 }
 
-#ifdef CONFIG_XEN
-int skb_checksum_setup(struct sk_buff *skb);
+#if defined(CONFIG_XEN) && defined(HAVE_PROTO_DATA_VALID)
+int vswitch_skb_checksum_setup(struct sk_buff *skb);
 #else
-static inline int skb_checksum_setup(struct sk_buff *skb)
+static inline int vswitch_skb_checksum_setup(struct sk_buff *skb)
 {
 	return 0;
 }
 #endif
 
-int vswitch_skb_checksum_setup(struct sk_buff *skb);
 void forward_ip_summed(struct sk_buff *skb);
 
 #endif /* datapath.h */

datapath/dp_dev.c

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2009 Nicira Networks.
+ * Copyright (c) 2009, 2010 Nicira Networks.
  * Distributed under the terms of the GNU GPL version 2.
  *
  * Significant portions of this file may be copied from parts of the Linux
@@ -10,6 +10,7 @@
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
 #include <linux/ethtool.h>
+#include <linux/preempt.h>
 #include <linux/rcupdate.h>
 #include <linux/skbuff.h>
 #include <linux/workqueue.h>
@@ -62,9 +63,13 @@ int dp_dev_recv(struct net_device *netdev, struct sk_buff *skb)
 	else
 		netif_rx_ni(skb);
 	netdev->last_rx = jiffies;
+
+	preempt_disable();
 	lb_stats = per_cpu_ptr(dp_dev->lstats, smp_processor_id());
 	lb_stats->rx_packets++;
 	lb_stats->rx_bytes += len;
+	preempt_enable();
+
 	return len;
 }