From: Ben Pfaff <blp@nicira.com>
Date: Mon, 14 Jul 2008 20:43:17 +0000 (-0700)
Subject: rconn: Rewrite to use explicit state machine.
X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=b8769714c15d042e3db8e04af2713bf16a82faa2;p=openvswitch

rconn: Rewrite to use explicit state machine.
---

diff --git a/include/rconn.h b/include/rconn.h
index 8f3b0acf..1872aa9f 100644
--- a/include/rconn.h
+++ b/include/rconn.h
@@ -57,6 +57,12 @@ struct rconn *rconn_new(const char *name, int txq_limit,
                         int inactivity_probe_interval, int max_backoff);
 struct rconn *rconn_new_from_vconn(const char *name, int txq_limit,
                                    struct vconn *);
+struct rconn *rconn_create(int txq_limit, int inactivity_probe_interval,
+                           int max_backoff);
+void rconn_connect(struct rconn *, const char *name);
+void rconn_connect_unreliably(struct rconn *,
+                              const char *name, struct vconn *vconn);
+void rconn_disconnect(struct rconn *);
 void rconn_destroy(struct rconn *);
 
 void rconn_run(struct rconn *);
diff --git a/lib/rconn.c b/lib/rconn.c
index c410e736..bb7d3fe1 100644
--- a/lib/rconn.c
+++ b/lib/rconn.c
@@ -1,6 +1,6 @@
 /* Copyright (c) 2008 The Board of Trustees of The Leland Stanford
  * Junior University
- * 
+ *
  * We are making the OpenFlow specification and associated documentation
  * (Software) available for public use and benefit with the expectation
  * that others will use, modify and enhance the Software and contribute
@@ -13,10 +13,10 @@
  * distribute, sublicense, and/or sell copies of the Software, and to
  * permit persons to whom the Software is furnished to do so, subject to
  * the following conditions:
- * 
+ *
  * The above copyright notice and this permission notice shall be
  * included in all copies or substantial portions of the Software.
- * 
+ *
  * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
  * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
  * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
@@ -25,7 +25,7 @@
  * 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.
- * 
+ *
  * The name and trademarks of copyright holder(s) may NOT be used in
  * advertising or publicity pertaining to the Software or any
  * derivatives without specific, written prior permission.
@@ -34,6 +34,7 @@
 #include "rconn.h"
 #include <assert.h>
 #include <errno.h>
+#include <limits.h>
 #include <stdlib.h>
 #include <string.h>
 #include "buffer.h"
@@ -46,20 +47,50 @@
 #define THIS_MODULE VLM_rconn
 #include "vlog.h"
 
+#define STATES                                  \
+    STATE(VOID, 1 << 0)                         \
+    STATE(BACKOFF, 1 << 1)                      \
+    STATE(CONNECTING, 1 << 2)                   \
+    STATE(ACTIVE, 1 << 3)                       \
+    STATE(IDLE, 1 << 4)
+enum state {
+#define STATE(NAME, VALUE) S_##NAME = VALUE,
+    STATES
+#undef STATE
+};
+
+static const char *
+state_name(enum state state)
+{
+    switch (state) {
+#define STATE(NAME, VALUE) case S_##NAME: return #NAME;
+        STATES
+#undef STATE
+    }
+    return "***ERROR***";
+}
+
 /* A reliable connection to an OpenFlow switch or controller.
  *
  * See the large comment in rconn.h for more information. */
 struct rconn {
-    bool reliable;
-    char *name;
+    enum state state;
+    time_t state_entered;
+    unsigned int min_timeout;
+
     struct vconn *vconn;
-    bool connected;
+    char *name;
+    bool reliable;
+
     struct queue txq;
     int txq_limit;
-    time_t backoff_deadline;
+
     int backoff;
     int max_backoff;
+    time_t backoff_deadline;
+    time_t last_received;
     time_t last_connected;
+
     unsigned int packets_sent;
 
     /* Throughout this file, "probe" is shorthand for "inactivity probe".
@@ -67,19 +98,38 @@ struct rconn {
      * an echo request as an inactivity probe packet.  We should receive back
      * a response. */
     int probe_interval;         /* Secs of inactivity before sending probe. */
-    time_t probe_sent;          /* Time at which last probe sent, or 0 if none
-                                 * has been sent since 'last_connected'. */
 };
 
-static struct rconn *create_rconn(const char *name, int txq_limit,
-                                  int probe_interval, int max_backoff,
-                                  struct vconn *);
+static unsigned int sat_add(unsigned int x, unsigned int y);
+static unsigned int sat_mul(unsigned int x, unsigned int y);
+static unsigned int elapsed_in_this_state(const struct rconn *);
+static bool timeout(struct rconn *, unsigned int secs);
+static void state_transition(struct rconn *, enum state);
 static int try_send(struct rconn *);
+static void reconnect(struct rconn *);
 static void disconnect(struct rconn *, int error);
-static time_t probe_deadline(const struct rconn *);
 
-/* Creates and returns a new rconn that connects (and re-connects as necessary)
- * to the vconn named 'name'.
+/* Creates a new rconn, connects it (reliably) to 'name', and returns it. */
+struct rconn *
+rconn_new(const char *name, int txq_limit, int inactivity_probe_interval,
+          int max_backoff) 
+{
+    struct rconn *rc = rconn_create(txq_limit, inactivity_probe_interval,
+                                    max_backoff);
+    rconn_connect(rc, name);
+    return rc;
+}
+
+/* Creates a new rconn, connects it (unreliably) to 'vconn', and returns it. */
+struct rconn *
+rconn_new_from_vconn(const char *name, int txq_limit, struct vconn *vconn) 
+{
+    struct rconn *rc = rconn_create(txq_limit, 60, 0);
+    rconn_connect_unreliably(rc, name, vconn);
+    return rc;
+}
+
+/* Creates and returns a new rconn.
  *
  * 'txq_limit' is the maximum length of the send queue, in packets.
  *
@@ -94,21 +144,74 @@ static time_t probe_deadline(const struct rconn *);
  * failure until it reaches 'max_backoff'.  If 0 is specified, the default of
  * 60 seconds is used. */
 struct rconn *
-rconn_new(const char *name, int txq_limit, int probe_interval, int max_backoff)
+rconn_create(int txq_limit, int probe_interval, int max_backoff)
 {
-    return create_rconn(name, txq_limit, probe_interval, max_backoff, NULL);
+    struct rconn *rc = xcalloc(1, sizeof *rc);
+
+    rc->state = S_VOID;
+    rc->state_entered = time(0);
+    rc->min_timeout = 0;
+
+    rc->vconn = NULL;
+    rc->name = xstrdup("void");
+    rc->reliable = false;
+
+    queue_init(&rc->txq);
+    assert(txq_limit > 0);
+    rc->txq_limit = txq_limit;
+
+    rc->backoff = 0;
+    rc->max_backoff = max_backoff ? max_backoff : 60;
+    rc->backoff_deadline = TIME_MIN;
+    rc->last_received = time(0);
+    rc->last_connected = time(0);
+
+    rc->packets_sent = 0;
+
+    rc->probe_interval = probe_interval ? MAX(5, probe_interval) : 0;
+
+    return rc;
 }
 
-/* Creates and returns a new rconn that is initially connected to 'vconn' and
- * has the given 'name'.  The rconn will not re-connect after the connection
- * drops.
- *
- * 'txq_limit' is the maximum length of the send queue, in packets. */
-struct rconn *
-rconn_new_from_vconn(const char *name, int txq_limit, struct vconn *vconn)
+void
+rconn_connect(struct rconn *rc, const char *name)
+{
+    rconn_disconnect(rc);
+    free(rc->name);
+    rc->name = xstrdup(name);
+    rc->reliable = true;
+    reconnect(rc);
+}
+
+void
+rconn_connect_unreliably(struct rconn *rc,
+                         const char *name, struct vconn *vconn)
 {
     assert(vconn != NULL);
-    return create_rconn(name, txq_limit, 0, 0, vconn);
+    rconn_disconnect(rc);
+    free(rc->name);
+    rc->name = xstrdup(name);
+    rc->reliable = false;
+    rc->vconn = vconn;
+    rc->last_connected = time(0);
+    state_transition(rc, S_ACTIVE);
+}
+
+void
+rconn_disconnect(struct rconn *rc)
+{
+    if (rc->vconn) {
+        vconn_close(rc->vconn);
+        rc->vconn = NULL;
+    }
+    free(rc->name);
+    rc->name = xstrdup("void");
+    rc->reliable = false;
+
+    rc->backoff = 0;
+    rc->backoff_deadline = TIME_MIN;
+
+    state_transition(rc, S_VOID);
 }
 
 /* Disconnects 'rc' and frees the underlying storage. */
@@ -123,68 +226,118 @@ rconn_destroy(struct rconn *rc)
     }
 }
 
+static void
+run_VOID(struct rconn *rc)
+{
+    /* Nothing to do. */
+}
+
+static void
+reconnect(struct rconn *rc)
+{
+    int retval;
+
+    VLOG_WARN("%s: connecting...", rc->name);
+    retval = vconn_open(rc->name, &rc->vconn);
+    if (!retval) {
+        rc->backoff_deadline = time(0) + rc->backoff;
+        state_transition(rc, S_CONNECTING);
+    } else {
+        VLOG_WARN("%s: connection failed (%s)", rc->name, strerror(retval));
+        disconnect(rc, 0);
+    }
+}
+
+static void
+run_BACKOFF(struct rconn *rc)
+{
+    if (timeout(rc, rc->backoff)) {
+        reconnect(rc);
+    }
+}
+
+static void
+run_CONNECTING(struct rconn *rc)
+{
+    int error = vconn_connect(rc->vconn);
+    if (!error) {
+        VLOG_WARN("%s: connected", rc->name);
+        if (vconn_is_passive(rc->vconn)) {
+            fatal(0, "%s: passive vconn not supported in switch",
+                  rc->name);
+        }
+        state_transition(rc, S_ACTIVE);
+        rc->last_connected = rc->state_entered;
+    } else if (error != EAGAIN) {
+        VLOG_WARN("%s: connection failed (%s)", rc->name, strerror(error));
+        disconnect(rc, error);
+    } else if (timeout(rc, MAX(1, rc->backoff))) {
+        VLOG_WARN("%s: connection timed out", rc->name);
+        rc->backoff_deadline = TIME_MAX; /* Prevent resetting backoff. */
+        disconnect(rc, 0);
+    }
+}
+
+static void
+do_tx_work(struct rconn *rc)
+{
+    while (rc->txq.n > 0) {
+        int error = try_send(rc);
+        if (error) {
+            break;
+        }
+    }
+}
+
+static void
+run_ACTIVE(struct rconn *rc)
+{
+    if (rc->probe_interval) {
+        unsigned int base = MAX(rc->last_received, rc->state_entered);
+        unsigned int arg = base + rc->probe_interval - rc->state_entered;
+        if (timeout(rc, arg)) {
+            queue_push_tail(&rc->txq, make_echo_request());
+            VLOG_DBG("%s: idle %u seconds, sending inactivity probe",
+                     rc->name, (unsigned int) (time(0) - base));
+            state_transition(rc, S_IDLE);
+            return;
+        }
+    }
+
+    do_tx_work(rc);
+}
+
+static void
+run_IDLE(struct rconn *rc)
+{
+    if (timeout(rc, rc->probe_interval)) {
+        VLOG_ERR("%s: no response to inactivity probe after %u "
+                 "seconds, disconnecting",
+                 rc->name, elapsed_in_this_state(rc));
+        disconnect(rc, 0);
+    } else {
+        do_tx_work(rc);
+    }
+}
+
 /* Performs whatever activities are necessary to maintain 'rc': if 'rc' is
  * disconnected, attempts to (re)connect, backing off as necessary; if 'rc' is
  * connected, attempts to send packets in the send queue, if any. */
 void
 rconn_run(struct rconn *rc)
 {
-    if (!rc->vconn) {
-        if (rc->reliable && time(0) >= rc->backoff_deadline) {
-            int retval;
-
-            VLOG_WARN("%s: connecting...", rc->name);
-            retval = vconn_open(rc->name, &rc->vconn);
-            if (!retval) {
-                rc->backoff_deadline = time(0) + rc->backoff;
-                rc->connected = false;
-            } else {
-                VLOG_WARN("%s: connection failed (%s)",
-                          rc->name, strerror(retval)); 
-                disconnect(rc, 0);
-            }
+    int old_state;
+    do {
+        old_state = rc->state;
+        rc->min_timeout = UINT_MAX;
+        switch (rc->state) {
+#define STATE(NAME, VALUE) case S_##NAME: run_##NAME(rc); break;
+            STATES
+#undef STATE
+        default:
+            NOT_REACHED();
         }
-    } else if (!rc->connected) {
-        int error = vconn_connect(rc->vconn);
-        if (!error) {
-            VLOG_WARN("%s: connected", rc->name);
-            if (vconn_is_passive(rc->vconn)) {
-                fatal(0, "%s: passive vconn not supported in switch",
-                      rc->name);
-            }
-            rc->connected = true;
-        } else if (error != EAGAIN) {
-            VLOG_WARN("%s: connection failed (%s)", rc->name, strerror(error));
-            disconnect(rc, 0);
-        } else if (time(0) >= rc->backoff_deadline) {
-            VLOG_WARN("%s: connection timed out", rc->name);
-            rc->backoff_deadline = TIME_MAX; /* Prevent resetting backoff. */
-            disconnect(rc, 0);
-        }
-    } else {
-        if (rc->probe_interval) {
-            time_t now = time(0);
-            if (now >= probe_deadline(rc)) {
-                if (!rc->probe_sent) {
-                    queue_push_tail(&rc->txq, make_echo_request());
-                    rc->probe_sent = now;
-                    VLOG_DBG("%s: idle %d seconds, sending inactivity probe",
-                             rc->name, (int) (now - rc->last_connected)); 
-                } else {
-                    VLOG_ERR("%s: no response to inactivity probe after %d "
-                             "seconds, disconnecting",
-                             rc->name, (int) (now - rc->probe_sent));
-                    disconnect(rc, 0);
-                }
-            }
-        }
-        while (rc->txq.n > 0) {
-            int error = try_send(rc);
-            if (error) {
-                break;
-            }
-        }
-    }
+    } while (rc->state != old_state);
 }
 
 /* Causes the next call to poll_block() to wake up when rconn_run() should be
@@ -192,27 +345,18 @@ rconn_run(struct rconn *rc)
 void
 rconn_run_wait(struct rconn *rc)
 {
-    if (rc->vconn) {
-        if (rc->txq.n) {
-            vconn_wait(rc->vconn, WAIT_SEND);
-        }
-        if (rc->probe_interval) {
-            poll_timer_wait((probe_deadline(rc) - time(0)) * 1000);
-        }
-    } else {
-        poll_timer_wait((rc->backoff_deadline - time(0)) * 1000);
+    if (rc->min_timeout != UINT_MAX) {
+        poll_timer_wait(sat_mul(rc->min_timeout, 1000));
     }
-}
+    /* Reset timeout to 1 second.  This will have no effect ordinarily, because
+     * rconn_run() will typically set it back to a higher value.  If, however,
+     * the caller fails to call rconn_run() before its next call to
+     * rconn_wait() we won't potentially block forever. */
+    rc->min_timeout = 1;
 
-/* Returns the time at which, should nothing be received, we should send out an
- * inactivity probe (if none has yet been sent) or conclude that the connection
- * is dead (if a probe has already been sent). */
-static time_t
-probe_deadline(const struct rconn *rc) 
-{
-    assert(rc->probe_interval);
-    return (rc->probe_interval
-            + (rc->probe_sent ? rc->probe_sent : rc->last_connected));
+    if ((rc->state & (S_ACTIVE | S_IDLE)) && rc->txq.n) {
+        vconn_wait(rc->vconn, WAIT_SEND);
+    }
 }
 
 /* Attempts to receive a packet from 'rc'.  If successful, returns the packet;
@@ -221,15 +365,17 @@ probe_deadline(const struct rconn *rc)
 struct buffer *
 rconn_recv(struct rconn *rc)
 {
-    if (rc->vconn && rc->connected) {
+    if (rc->state & (S_ACTIVE | S_IDLE)) {
         struct buffer *buffer;
         int error = vconn_recv(rc->vconn, &buffer);
         if (!error) {
-            rc->last_connected = time(0);
-            rc->probe_sent = 0;
+            rc->last_received = time(0);
+            if (rc->state == S_IDLE) {
+                state_transition(rc, S_ACTIVE);
+            }
             return buffer;
         } else if (error != EAGAIN) {
-            disconnect(rc, error); 
+            disconnect(rc, error);
         }
     }
     return NULL;
@@ -238,7 +384,7 @@ rconn_recv(struct rconn *rc)
 /* Causes the next call to poll_block() to wake up when a packet may be ready
  * to be received by vconn_recv() on 'rc'.  */
 void
-rconn_recv_wait(struct rconn *rc) 
+rconn_recv_wait(struct rconn *rc)
 {
     if (rc->vconn) {
         vconn_wait(rc->vconn, WAIT_RECV);
@@ -291,7 +437,7 @@ rconn_force_send(struct rconn *rc, struct buffer *b)
 /* Returns true if 'rc''s send buffer is full,
  * false if it has room for at least one more packet. */
 bool
-rconn_is_full(const struct rconn *rc) 
+rconn_is_full(const struct rconn *rc)
 {
     return rc->txq.n >= rc->txq_limit;
 }
@@ -300,22 +446,22 @@ rconn_is_full(const struct rconn *rc)
  * vconn.  A packet is not counted as sent while it is still queued in the
  * rconn, only when it has been successfuly passed to the vconn.  */
 unsigned int
-rconn_packets_sent(const struct rconn *rc) 
+rconn_packets_sent(const struct rconn *rc)
 {
     return rc->packets_sent;
 }
 
 /* Returns 'rc''s name (the 'name' argument passed to rconn_new()). */
 const char *
-rconn_get_name(const struct rconn *rc) 
+rconn_get_name(const struct rconn *rc)
 {
     return rc->name;
 }
 
 /* Returns true if 'rconn' is connected or in the process of reconnecting,
- * false if 'rconn' is disconnected and will not be reconnected. */
+ * false if 'rconn' is disconnected and will not reconnect on its own. */
 bool
-rconn_is_alive(const struct rconn *rconn) 
+rconn_is_alive(const struct rconn *rconn)
 {
     return rconn->reliable || rconn->vconn;
 }
@@ -324,13 +470,13 @@ rconn_is_alive(const struct rconn *rconn)
 bool
 rconn_is_connected(const struct rconn *rconn)
 {
-    return rconn->vconn && !vconn_connect(rconn->vconn);
+    return rconn->state & (S_ACTIVE | S_IDLE);
 }
 
 /* Returns 0 if 'rconn' is connected, otherwise the number of seconds that it
  * has been disconnected. */
 int
-rconn_disconnected_duration(const struct rconn *rconn) 
+rconn_disconnected_duration(const struct rconn *rconn)
 {
     return rconn_is_connected(rconn) ? 0 : time(0) - rconn->last_connected;
 }
@@ -343,29 +489,6 @@ rconn_get_ip(const struct rconn *rconn)
     return rconn->vconn ? vconn_get_ip(rconn->vconn) : 0;
 }
 
-static struct rconn *
-create_rconn(const char *name, int txq_limit, int probe_interval,
-             int max_backoff, struct vconn *vconn)
-{
-    struct rconn *rc = xmalloc(sizeof *rc);
-    assert(txq_limit > 0);
-    rc->reliable = vconn == NULL;
-    rc->connected = vconn != NULL;
-    rc->name = xstrdup(name);
-    rc->vconn = vconn;
-    queue_init(&rc->txq);
-    rc->txq_limit = txq_limit;
-    rc->backoff_deadline = time(0);
-    rc->backoff = 0;
-    rc->max_backoff = max_backoff ? max_backoff : 60;
-    rc->last_connected = time(0);
-    rc->probe_interval = (probe_interval
-                              ? MAX(5, probe_interval) : 0);
-    rc->probe_sent = 0;
-    rc->packets_sent = 0;
-    return rc;
-}
-
 /* Tries to send a packet from 'rc''s send buffer.  Returns 0 if successful,
  * otherwise a positive errno value. */
 static int
@@ -389,33 +512,71 @@ try_send(struct rconn *rc)
  * nonzero, then it should be EOF to indicate the connection was closed by the
  * peer in a normal fashion or a positive errno value. */
 static void
-disconnect(struct rconn *rc, int error) 
+disconnect(struct rconn *rc, int error)
 {
-    time_t now = time(0);
-    
-    if (rc->vconn) {
-        if (error > 0) {
-            VLOG_WARN("%s: connection dropped (%s)",
-                      rc->name, strerror(error)); 
-        } else if (error == EOF) {
-            if (rc->reliable) {
-                VLOG_WARN("%s: connection closed", rc->name);
+    if (rc->reliable) {
+        time_t now = time(0);
+
+        if (rc->state & (S_CONNECTING | S_ACTIVE | S_IDLE)) {
+            if (error > 0) {
+                VLOG_WARN("%s: connection dropped (%s)",
+                          rc->name, strerror(error));
+            } else if (error == EOF) {
+                if (rc->reliable) {
+                    VLOG_WARN("%s: connection closed", rc->name);
+                }
+            } else {
+                VLOG_WARN("%s: connection dropped", rc->name);
             }
-        } else {
-            VLOG_WARN("%s: connection dropped", rc->name); 
+            vconn_close(rc->vconn);
+            rc->vconn = NULL;
+            queue_clear(&rc->txq);
         }
-        vconn_close(rc->vconn);
-        rc->vconn = NULL;
-        queue_clear(&rc->txq);
-    }
 
-    if (now >= rc->backoff_deadline) {
-        rc->backoff = 1;
+        if (now >= rc->backoff_deadline) {
+            rc->backoff = 1;
+        } else {
+            rc->backoff = MIN(rc->max_backoff, MAX(1, 2 * rc->backoff));
+            VLOG_WARN("%s: waiting %d seconds before reconnect\n",
+                      rc->name, rc->backoff);
+        }
+        rc->backoff_deadline = now + rc->backoff;
+        state_transition(rc, S_BACKOFF);
     } else {
-        rc->backoff = MIN(rc->max_backoff, MAX(1, 2 * rc->backoff));
-        VLOG_WARN("%s: waiting %d seconds before reconnect\n",
-                  rc->name, rc->backoff);
+        rconn_disconnect(rc);
     }
-    rc->backoff_deadline = now + rc->backoff;
-    rc->probe_sent = 0;
+}
+
+static unsigned int
+elapsed_in_this_state(const struct rconn *rc)
+{
+    return time(0) - rc->state_entered;
+}
+
+static bool
+timeout(struct rconn *rc, unsigned int secs)
+{
+    rc->min_timeout = MIN(rc->min_timeout, secs);
+    return time(0) >= sat_add(rc->state_entered, secs);
+}
+
+static void
+state_transition(struct rconn *rc, enum state state)
+{
+    VLOG_DBG("%s: entering %s", rc->name, state_name(state));
+    rc->state = state;
+    rc->state_entered = time(0);
+}
+
+static unsigned int
+sat_add(unsigned int x, unsigned int y)
+{
+    return x + y >= x ? x + y : UINT_MAX;
+}
+
+static unsigned int
+sat_mul(unsigned int x, unsigned int y)
+{
+    assert(y);
+    return x <= UINT_MAX / y ? x * y : UINT_MAX;
 }