Merge branch 'topic/core-cleanup' into for-linus

[safe/jmp/linux-2.6] / fs / ceph / messenger.c

diff --git a/fs/ceph/messenger.c b/fs/ceph/messenger.c
index 0ddc2c7..cd4fadb 100644 (file)
--- a/fs/ceph/messenger.c
+++ b/fs/ceph/messenger.c
@@ -6,6 +6,7 @@
 #include <linux/inet.h>
 #include <linux/kthread.h>
 #include <linux/net.h>
+#include <linux/slab.h>
 #include <linux/socket.h>
 #include <linux/string.h>
 #include <net/tcp.h>
@@ -29,6 +30,10 @@ static char tag_msg = CEPH_MSGR_TAG_MSG;
 static char tag_ack = CEPH_MSGR_TAG_ACK;
 static char tag_keepalive = CEPH_MSGR_TAG_KEEPALIVE;
 
+#ifdef CONFIG_LOCKDEP
+static struct lock_class_key socket_class;
+#endif
+
 
 static void queue_con(struct ceph_connection *con);
 static void con_work(struct work_struct *);
@@ -227,6 +232,10 @@ static struct socket *ceph_tcp_connect(struct ceph_connection *con)
        con->sock = sock;
        sock->sk->sk_allocation = GFP_NOFS;
 
+#ifdef CONFIG_LOCKDEP
+       lockdep_set_class(&sock->sk->sk_lock, &socket_class);
+#endif
+
        set_sock_callbacks(sock, con);
 
        dout("connect %s\n", pr_addr(&con->peer_addr.in_addr));
@@ -332,6 +341,7 @@ static void reset_connection(struct ceph_connection *con)
                con->out_msg = NULL;
        }
        con->in_seq = 0;
+       con->in_seq_acked = 0;
 }
 
 /*
@@ -342,6 +352,9 @@ void ceph_con_close(struct ceph_connection *con)
        dout("con_close %p peer %s\n", con, pr_addr(&con->peer_addr.in_addr));
        set_bit(CLOSED, &con->state);  /* in case there's queued work */
        clear_bit(STANDBY, &con->state);  /* avoid connect_seq bump */
+       clear_bit(LOSSYTX, &con->state);  /* so we retry next connect */
+       clear_bit(KEEPALIVE_PENDING, &con->state);
+       clear_bit(WRITE_PENDING, &con->state);
        mutex_lock(&con->mutex);
        reset_connection(con);
        cancel_delayed_work(&con->work);
@@ -363,6 +376,14 @@ void ceph_con_open(struct ceph_connection *con, struct ceph_entity_addr *addr)
 }
 
 /*
+ * return true if this connection ever successfully opened
+ */
+bool ceph_con_opened(struct ceph_connection *con)
+{
+       return con->connect_seq > 0;
+}
+
+/*
  * generic get/put
  */
 struct ceph_connection *ceph_con_get(struct ceph_connection *con)
@@ -471,7 +492,14 @@ static void prepare_write_message(struct ceph_connection *con)
                list_move_tail(&m->list_head, &con->out_sent);
        }
 
-       m->hdr.seq = cpu_to_le64(++con->out_seq);
+       /*
+        * only assign outgoing seq # if we haven't sent this message
+        * yet.  if it is requeued, resend with it's original seq.
+        */
+       if (m->needs_out_seq) {
+               m->hdr.seq = cpu_to_le64(++con->out_seq);
+               m->needs_out_seq = false;
+       }
 
        dout("prepare_write_message %p seq %lld type %d len %d+%d+%d %d pgs\n",
             m, con->out_seq, le16_to_cpu(m->hdr.type),
@@ -827,13 +855,6 @@ static void prepare_read_connect(struct ceph_connection *con)
        con->in_base_pos = 0;
 }
 
-static void prepare_read_connect_retry(struct ceph_connection *con)
-{
-       dout("prepare_read_connect_retry %p\n", con);
-       con->in_base_pos = strlen(CEPH_BANNER) + sizeof(con->actual_peer_addr)
-               + sizeof(con->peer_addr_for_me);
-}
-
 static void prepare_read_ack(struct ceph_connection *con)
 {
        dout("prepare_read_ack %p\n", con);
@@ -1143,7 +1164,7 @@ static int process_connect(struct ceph_connection *con)
                }
                con->auth_retry = 1;
                prepare_write_connect(con->msgr, con, 0);
-               prepare_read_connect_retry(con);
+               prepare_read_connect(con);
                break;
 
        case CEPH_MSGR_TAG_RESETSESSION:
@@ -1320,6 +1341,7 @@ static int read_partial_message(struct ceph_connection *con)
        unsigned front_len, middle_len, data_len, data_off;
        int datacrc = con->msgr->nocrc;
        int skip;
+       u64 seq;
 
        dout("read_partial_message con %p msg %p\n", con, m);
 
@@ -1354,6 +1376,25 @@ static int read_partial_message(struct ceph_connection *con)
                return -EIO;
        data_off = le16_to_cpu(con->in_hdr.data_off);
 
+       /* verify seq# */
+       seq = le64_to_cpu(con->in_hdr.seq);
+       if ((s64)seq - (s64)con->in_seq < 1) {
+               pr_info("skipping %s%lld %s seq %lld, expected %lld\n",
+                       ENTITY_NAME(con->peer_name),
+                       pr_addr(&con->peer_addr.in_addr),
+                       seq, con->in_seq + 1);
+               con->in_base_pos = -front_len - middle_len - data_len -
+                       sizeof(m->footer);
+               con->in_tag = CEPH_MSGR_TAG_READY;
+               con->in_seq++;
+               return 0;
+       } else if ((s64)seq - (s64)con->in_seq > 1) {
+               pr_err("read_partial_message bad seq %lld expected %lld\n",
+                      seq, con->in_seq + 1);
+               con->error_msg = "bad message sequence # for incoming message";
+               return -EBADMSG;
+       }
+
        /* allocate message? */
        if (!con->in_msg) {
                dout("got hdr type %d front %d data %d\n", con->in_hdr.type,
@@ -1365,6 +1406,7 @@ static int read_partial_message(struct ceph_connection *con)
                        con->in_base_pos = -front_len - middle_len - data_len -
                                sizeof(m->footer);
                        con->in_tag = CEPH_MSGR_TAG_READY;
+                       con->in_seq++;
                        return 0;
                }
                if (IS_ERR(con->in_msg)) {
@@ -1840,8 +1882,6 @@ static void ceph_fault(struct ceph_connection *con)
                goto out;
        }
 
-       clear_bit(BUSY, &con->state);  /* to avoid an improbable race */
-
        mutex_lock(&con->mutex);
        if (test_bit(CLOSED, &con->state))
                goto out_unlock;
@@ -1951,6 +1991,10 @@ void ceph_con_send(struct ceph_connection *con, struct ceph_msg *msg)
        msg->hdr.src.addr = con->msgr->my_enc_addr;
        msg->hdr.orig_src = msg->hdr.src;
 
+       BUG_ON(msg->front.iov_len != le32_to_cpu(msg->hdr.front_len));
+
+       msg->needs_out_seq = true;
+
        /* queue */
        mutex_lock(&con->mutex);
        BUG_ON(!list_empty(&msg->list_head));
@@ -2016,6 +2060,7 @@ void ceph_con_revoke_message(struct ceph_connection *con, struct ceph_msg *msg)
                ceph_msg_put(con->in_msg);
                con->in_msg = NULL;
                con->in_tag = CEPH_MSGR_TAG_READY;
+               con->in_seq++;
        } else {
                dout("con_revoke_pages %p msg %p pages %p no-op\n",
                     con, con->in_msg, msg);
@@ -2049,15 +2094,19 @@ struct ceph_msg *ceph_msg_new(int type, int front_len,
        kref_init(&m->kref);
        INIT_LIST_HEAD(&m->list_head);
 
+       m->hdr.tid = 0;
        m->hdr.type = cpu_to_le16(type);
+       m->hdr.priority = cpu_to_le16(CEPH_MSG_PRIO_DEFAULT);
+       m->hdr.version = 0;
        m->hdr.front_len = cpu_to_le32(front_len);
        m->hdr.middle_len = 0;
        m->hdr.data_len = cpu_to_le32(page_len);
        m->hdr.data_off = cpu_to_le16(page_off);
-       m->hdr.priority = cpu_to_le16(CEPH_MSG_PRIO_DEFAULT);
+       m->hdr.reserved = 0;
        m->footer.front_crc = 0;
        m->footer.middle_crc = 0;
        m->footer.data_crc = 0;
+       m->footer.flags = 0;
        m->front_max = front_len;
        m->front_is_vmalloc = false;
        m->more_to_follow = false;