Commit ec302645 authored by Sage Weil's avatar Sage Weil

ceph: use connection mutex to protect read and write stages

Use a single mutex (previously out_mutex) to protect both read and write
activity from concurrent ceph_con_* calls.  Drop the mutex when doing
callbacks to avoid nested locking (the callback may need to call something
like ceph_con_close).
Signed-off-by: default avatarSage Weil <sage@newdream.net>
parent 529cfcc4
...@@ -316,7 +316,6 @@ static void reset_connection(struct ceph_connection *con) ...@@ -316,7 +316,6 @@ static void reset_connection(struct ceph_connection *con)
{ {
/* reset connection, out_queue, msg_ and connect_seq */ /* reset connection, out_queue, msg_ and connect_seq */
/* discard existing out_queue and msg_seq */ /* discard existing out_queue and msg_seq */
mutex_lock(&con->out_mutex);
ceph_msg_remove_list(&con->out_queue); ceph_msg_remove_list(&con->out_queue);
ceph_msg_remove_list(&con->out_sent); ceph_msg_remove_list(&con->out_sent);
...@@ -332,7 +331,6 @@ static void reset_connection(struct ceph_connection *con) ...@@ -332,7 +331,6 @@ static void reset_connection(struct ceph_connection *con)
con->out_msg = NULL; con->out_msg = NULL;
} }
con->in_seq = 0; con->in_seq = 0;
mutex_unlock(&con->out_mutex);
} }
/* /*
...@@ -343,7 +341,9 @@ void ceph_con_close(struct ceph_connection *con) ...@@ -343,7 +341,9 @@ void ceph_con_close(struct ceph_connection *con)
dout("con_close %p peer %s\n", con, pr_addr(&con->peer_addr.in_addr)); 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 */ set_bit(CLOSED, &con->state); /* in case there's queued work */
clear_bit(STANDBY, &con->state); /* avoid connect_seq bump */ clear_bit(STANDBY, &con->state); /* avoid connect_seq bump */
mutex_lock(&con->mutex);
reset_connection(con); reset_connection(con);
mutex_unlock(&con->mutex);
queue_con(con); queue_con(con);
} }
...@@ -392,7 +392,7 @@ void ceph_con_init(struct ceph_messenger *msgr, struct ceph_connection *con) ...@@ -392,7 +392,7 @@ void ceph_con_init(struct ceph_messenger *msgr, struct ceph_connection *con)
memset(con, 0, sizeof(*con)); memset(con, 0, sizeof(*con));
atomic_set(&con->nref, 1); atomic_set(&con->nref, 1);
con->msgr = msgr; con->msgr = msgr;
mutex_init(&con->out_mutex); mutex_init(&con->mutex);
INIT_LIST_HEAD(&con->out_queue); INIT_LIST_HEAD(&con->out_queue);
INIT_LIST_HEAD(&con->out_sent); INIT_LIST_HEAD(&con->out_sent);
INIT_DELAYED_WORK(&con->work, con_work); INIT_DELAYED_WORK(&con->work, con_work);
...@@ -571,11 +571,13 @@ static void prepare_connect_authorizer(struct ceph_connection *con) ...@@ -571,11 +571,13 @@ static void prepare_connect_authorizer(struct ceph_connection *con)
int auth_len = 0; int auth_len = 0;
int auth_protocol = 0; int auth_protocol = 0;
mutex_unlock(&con->mutex);
if (con->ops->get_authorizer) if (con->ops->get_authorizer)
con->ops->get_authorizer(con, &auth_buf, &auth_len, con->ops->get_authorizer(con, &auth_buf, &auth_len,
&auth_protocol, &con->auth_reply_buf, &auth_protocol, &con->auth_reply_buf,
&con->auth_reply_buf_len, &con->auth_reply_buf_len,
con->auth_retry); con->auth_retry);
mutex_lock(&con->mutex);
con->out_connect.authorizer_protocol = cpu_to_le32(auth_protocol); con->out_connect.authorizer_protocol = cpu_to_le32(auth_protocol);
con->out_connect.authorizer_len = cpu_to_le32(auth_len); con->out_connect.authorizer_len = cpu_to_le32(auth_len);
...@@ -1094,10 +1096,13 @@ static int process_connect(struct ceph_connection *con) ...@@ -1094,10 +1096,13 @@ static int process_connect(struct ceph_connection *con)
le32_to_cpu(con->out_connect.protocol_version), le32_to_cpu(con->out_connect.protocol_version),
le32_to_cpu(con->in_reply.protocol_version)); le32_to_cpu(con->in_reply.protocol_version));
con->error_msg = "protocol version mismatch"; con->error_msg = "protocol version mismatch";
if (con->ops->bad_proto)
con->ops->bad_proto(con);
reset_connection(con); reset_connection(con);
set_bit(CLOSED, &con->state); /* in case there's queued work */ set_bit(CLOSED, &con->state); /* in case there's queued work */
mutex_unlock(&con->mutex);
if (con->ops->bad_proto)
con->ops->bad_proto(con);
mutex_lock(&con->mutex);
return -1; return -1;
case CEPH_MSGR_TAG_BADAUTHORIZER: case CEPH_MSGR_TAG_BADAUTHORIZER:
...@@ -1133,9 +1138,11 @@ static int process_connect(struct ceph_connection *con) ...@@ -1133,9 +1138,11 @@ static int process_connect(struct ceph_connection *con)
prepare_read_connect(con); prepare_read_connect(con);
/* Tell ceph about it. */ /* Tell ceph about it. */
mutex_unlock(&con->mutex);
pr_info("reset on %s%lld\n", ENTITY_NAME(con->peer_name)); pr_info("reset on %s%lld\n", ENTITY_NAME(con->peer_name));
if (con->ops->peer_reset) if (con->ops->peer_reset)
con->ops->peer_reset(con); con->ops->peer_reset(con);
mutex_lock(&con->mutex);
break; break;
case CEPH_MSGR_TAG_RETRY_SESSION: case CEPH_MSGR_TAG_RETRY_SESSION:
...@@ -1221,7 +1228,6 @@ static void process_ack(struct ceph_connection *con) ...@@ -1221,7 +1228,6 @@ static void process_ack(struct ceph_connection *con)
u64 ack = le64_to_cpu(con->in_temp_ack); u64 ack = le64_to_cpu(con->in_temp_ack);
u64 seq; u64 seq;
mutex_lock(&con->out_mutex);
while (!list_empty(&con->out_sent)) { while (!list_empty(&con->out_sent)) {
m = list_first_entry(&con->out_sent, struct ceph_msg, m = list_first_entry(&con->out_sent, struct ceph_msg,
list_head); list_head);
...@@ -1232,7 +1238,6 @@ static void process_ack(struct ceph_connection *con) ...@@ -1232,7 +1238,6 @@ static void process_ack(struct ceph_connection *con)
le16_to_cpu(m->hdr.type), m); le16_to_cpu(m->hdr.type), m);
ceph_msg_remove(m); ceph_msg_remove(m);
} }
mutex_unlock(&con->out_mutex);
prepare_read_tag(con); prepare_read_tag(con);
} }
...@@ -1366,8 +1371,10 @@ static int read_partial_message(struct ceph_connection *con) ...@@ -1366,8 +1371,10 @@ static int read_partial_message(struct ceph_connection *con)
/* find pages for data payload */ /* find pages for data payload */
want = calc_pages_for(data_off & ~PAGE_MASK, data_len); want = calc_pages_for(data_off & ~PAGE_MASK, data_len);
ret = -1; ret = -1;
mutex_unlock(&con->mutex);
if (con->ops->prepare_pages) if (con->ops->prepare_pages)
ret = con->ops->prepare_pages(con, m, want); ret = con->ops->prepare_pages(con, m, want);
mutex_lock(&con->mutex);
if (ret < 0) { if (ret < 0) {
dout("%p prepare_pages failed, skipping payload\n", m); dout("%p prepare_pages failed, skipping payload\n", m);
con->in_base_pos = -data_len - sizeof(m->footer); con->in_base_pos = -data_len - sizeof(m->footer);
...@@ -1454,9 +1461,8 @@ static void process_message(struct ceph_connection *con) ...@@ -1454,9 +1461,8 @@ static void process_message(struct ceph_connection *con)
if (con->peer_name.type == 0) if (con->peer_name.type == 0)
con->peer_name = msg->hdr.src.name; con->peer_name = msg->hdr.src.name;
mutex_lock(&con->out_mutex);
con->in_seq++; con->in_seq++;
mutex_unlock(&con->out_mutex); mutex_unlock(&con->mutex);
dout("===== %p %llu from %s%lld %d=%s len %d+%d (%u %u %u) =====\n", dout("===== %p %llu from %s%lld %d=%s len %d+%d (%u %u %u) =====\n",
msg, le64_to_cpu(msg->hdr.seq), msg, le64_to_cpu(msg->hdr.seq),
...@@ -1467,6 +1473,8 @@ static void process_message(struct ceph_connection *con) ...@@ -1467,6 +1473,8 @@ static void process_message(struct ceph_connection *con)
le32_to_cpu(msg->hdr.data_len), le32_to_cpu(msg->hdr.data_len),
con->in_front_crc, con->in_middle_crc, con->in_data_crc); con->in_front_crc, con->in_middle_crc, con->in_data_crc);
con->ops->dispatch(con, msg); con->ops->dispatch(con, msg);
mutex_lock(&con->mutex);
prepare_read_tag(con); prepare_read_tag(con);
} }
...@@ -1483,7 +1491,7 @@ static int try_write(struct ceph_connection *con) ...@@ -1483,7 +1491,7 @@ static int try_write(struct ceph_connection *con)
dout("try_write start %p state %lu nref %d\n", con, con->state, dout("try_write start %p state %lu nref %d\n", con, con->state,
atomic_read(&con->nref)); atomic_read(&con->nref));
mutex_lock(&con->out_mutex); mutex_lock(&con->mutex);
more: more:
dout("try_write out_kvec_bytes %d\n", con->out_kvec_bytes); dout("try_write out_kvec_bytes %d\n", con->out_kvec_bytes);
...@@ -1576,7 +1584,7 @@ do_next: ...@@ -1576,7 +1584,7 @@ do_next:
done: done:
ret = 0; ret = 0;
out: out:
mutex_unlock(&con->out_mutex); mutex_unlock(&con->mutex);
dout("try_write done on %p\n", con); dout("try_write done on %p\n", con);
return ret; return ret;
} }
...@@ -1600,6 +1608,8 @@ static int try_read(struct ceph_connection *con) ...@@ -1600,6 +1608,8 @@ static int try_read(struct ceph_connection *con)
dout("try_read start on %p\n", con); dout("try_read start on %p\n", con);
msgr = con->msgr; msgr = con->msgr;
mutex_lock(&con->mutex);
more: more:
dout("try_read tag %d in_base_pos %d\n", (int)con->in_tag, dout("try_read tag %d in_base_pos %d\n", (int)con->in_tag,
con->in_base_pos); con->in_base_pos);
...@@ -1693,6 +1703,7 @@ more: ...@@ -1693,6 +1703,7 @@ more:
done: done:
ret = 0; ret = 0;
out: out:
mutex_unlock(&con->mutex);
dout("try_read done on %p\n", con); dout("try_read done on %p\n", con);
return ret; return ret;
...@@ -1818,6 +1829,8 @@ static void ceph_fault(struct ceph_connection *con) ...@@ -1818,6 +1829,8 @@ static void ceph_fault(struct ceph_connection *con)
clear_bit(BUSY, &con->state); /* to avoid an improbable race */ clear_bit(BUSY, &con->state); /* to avoid an improbable race */
mutex_lock(&con->mutex);
con_close_socket(con); con_close_socket(con);
if (con->in_msg) { if (con->in_msg) {
...@@ -1827,24 +1840,24 @@ static void ceph_fault(struct ceph_connection *con) ...@@ -1827,24 +1840,24 @@ static void ceph_fault(struct ceph_connection *con)
/* If there are no messages in the queue, place the connection /* If there are no messages in the queue, place the connection
* in a STANDBY state (i.e., don't try to reconnect just yet). */ * in a STANDBY state (i.e., don't try to reconnect just yet). */
mutex_lock(&con->out_mutex);
if (list_empty(&con->out_queue) && !con->out_keepalive_pending) { if (list_empty(&con->out_queue) && !con->out_keepalive_pending) {
dout("fault setting STANDBY\n"); dout("fault setting STANDBY\n");
set_bit(STANDBY, &con->state); set_bit(STANDBY, &con->state);
mutex_unlock(&con->out_mutex); mutex_unlock(&con->mutex);
goto out; goto out;
} }
/* Requeue anything that hasn't been acked, and retry after a /* Requeue anything that hasn't been acked, and retry after a
* delay. */ * delay. */
list_splice_init(&con->out_sent, &con->out_queue); list_splice_init(&con->out_sent, &con->out_queue);
mutex_unlock(&con->out_mutex);
if (con->delay == 0) if (con->delay == 0)
con->delay = BASE_DELAY_INTERVAL; con->delay = BASE_DELAY_INTERVAL;
else if (con->delay < MAX_DELAY_INTERVAL) else if (con->delay < MAX_DELAY_INTERVAL)
con->delay *= 2; con->delay *= 2;
mutex_unlock(&con->mutex);
/* explicitly schedule work to try to reconnect again later. */ /* explicitly schedule work to try to reconnect again later. */
dout("fault queueing %p delay %lu\n", con, con->delay); dout("fault queueing %p delay %lu\n", con, con->delay);
con->ops->get(con); con->ops->get(con);
...@@ -1920,7 +1933,7 @@ void ceph_con_send(struct ceph_connection *con, struct ceph_msg *msg) ...@@ -1920,7 +1933,7 @@ void ceph_con_send(struct ceph_connection *con, struct ceph_msg *msg)
msg->hdr.dst_erank = con->peer_addr.erank; msg->hdr.dst_erank = con->peer_addr.erank;
/* queue */ /* queue */
mutex_lock(&con->out_mutex); mutex_lock(&con->mutex);
BUG_ON(!list_empty(&msg->list_head)); BUG_ON(!list_empty(&msg->list_head));
list_add_tail(&msg->list_head, &con->out_queue); list_add_tail(&msg->list_head, &con->out_queue);
dout("----- %p to %s%lld %d=%s len %d+%d+%d -----\n", msg, dout("----- %p to %s%lld %d=%s len %d+%d+%d -----\n", msg,
...@@ -1929,7 +1942,7 @@ void ceph_con_send(struct ceph_connection *con, struct ceph_msg *msg) ...@@ -1929,7 +1942,7 @@ void ceph_con_send(struct ceph_connection *con, struct ceph_msg *msg)
le32_to_cpu(msg->hdr.front_len), le32_to_cpu(msg->hdr.front_len),
le32_to_cpu(msg->hdr.middle_len), le32_to_cpu(msg->hdr.middle_len),
le32_to_cpu(msg->hdr.data_len)); le32_to_cpu(msg->hdr.data_len));
mutex_unlock(&con->out_mutex); mutex_unlock(&con->mutex);
/* if there wasn't anything waiting to send before, queue /* if there wasn't anything waiting to send before, queue
* new work */ * new work */
...@@ -1942,7 +1955,7 @@ void ceph_con_send(struct ceph_connection *con, struct ceph_msg *msg) ...@@ -1942,7 +1955,7 @@ void ceph_con_send(struct ceph_connection *con, struct ceph_msg *msg)
*/ */
void ceph_con_revoke(struct ceph_connection *con, struct ceph_msg *msg) void ceph_con_revoke(struct ceph_connection *con, struct ceph_msg *msg)
{ {
mutex_lock(&con->out_mutex); mutex_lock(&con->mutex);
if (!list_empty(&msg->list_head)) { if (!list_empty(&msg->list_head)) {
dout("con_revoke %p msg %p\n", con, msg); dout("con_revoke %p msg %p\n", con, msg);
list_del_init(&msg->list_head); list_del_init(&msg->list_head);
...@@ -1959,7 +1972,7 @@ void ceph_con_revoke(struct ceph_connection *con, struct ceph_msg *msg) ...@@ -1959,7 +1972,7 @@ void ceph_con_revoke(struct ceph_connection *con, struct ceph_msg *msg)
} else { } else {
dout("con_revoke %p msg %p - not queued (sent?)\n", con, msg); dout("con_revoke %p msg %p - not queued (sent?)\n", con, msg);
} }
mutex_unlock(&con->out_mutex); mutex_unlock(&con->mutex);
} }
/* /*
......
...@@ -155,8 +155,9 @@ struct ceph_connection { ...@@ -155,8 +155,9 @@ struct ceph_connection {
void *auth_reply_buf; /* where to put the authorizer reply */ void *auth_reply_buf; /* where to put the authorizer reply */
int auth_reply_buf_len; int auth_reply_buf_len;
struct mutex mutex;
/* out queue */ /* out queue */
struct mutex out_mutex;
struct list_head out_queue; struct list_head out_queue;
struct list_head out_sent; /* sending or sent but unacked */ struct list_head out_sent; /* sending or sent but unacked */
u64 out_seq; /* last message queued for send */ u64 out_seq; /* last message queued for send */
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment