Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client

Pull Ceph updates from Sage Weil:
 "The two main changes are aio support in CephFS, and a series that
  fixes several issues in the authentication key timeout/renewal code.

  On top of that are a variety of cleanups and minor bug fixes"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client:
  libceph: remove outdated comment
  libceph: kill off ceph_x_ticket_handler::validity
  libceph: invalidate AUTH in addition to a service ticket
  libceph: fix authorizer invalidation, take 2
  libceph: clear messenger auth_retry flag if we fault
  libceph: fix ceph_msg_revoke()
  libceph: use list_for_each_entry_safe
  ceph: use i_size_{read,write} to get/set i_size
  ceph: re-send AIO write request when getting -EOLDSNAP error
  ceph: Asynchronous IO support
  ceph: Avoid to propagate the invalid page point
  ceph: fix double page_unlock() in page_mkwrite()
  rbd: delete an unnecessary check before rbd_dev_destroy()
  libceph: use list_next_entry instead of list_entry_next
  ceph: ceph_frag_contains_value can be boolean
  ceph: remove unused functions in ceph_frag.h

4 files changed, 109 insertions, 51 deletions

diff --git a/net/ceph/auth_x.c b/net/ceph/auth_x.c
index 10d87753ed87..9e43a315e662 100644
--- a/net/ceph/auth_x.c
+++ b/net/ceph/auth_x.c
@@ -152,7 +152,6 @@ static int process_one_ticket(struct ceph_auth_client *ac,
 	void *ticket_buf = NULL;
 	void *tp, *tpend;
 	void **ptp;
-	struct ceph_timespec new_validity;
 	struct ceph_crypto_key new_session_key;
 	struct ceph_buffer *new_ticket_blob;
 	unsigned long new_expires, new_renew_after;
@@ -193,8 +192,8 @@ static int process_one_ticket(struct ceph_auth_client *ac,
 	if (ret)
 		goto out;
 
-	ceph_decode_copy(&dp, &new_validity, sizeof(new_validity));
-	ceph_decode_timespec(&validity, &new_validity);
+	ceph_decode_timespec(&validity, dp);
+	dp += sizeof(struct ceph_timespec);
 	new_expires = get_seconds() + validity.tv_sec;
 	new_renew_after = new_expires - (validity.tv_sec / 4);
 	dout(" expires=%lu renew_after=%lu\n", new_expires,
@@ -233,10 +232,10 @@ static int process_one_ticket(struct ceph_auth_client *ac,
 		ceph_buffer_put(th->ticket_blob);
 	th->session_key = new_session_key;
 	th->ticket_blob = new_ticket_blob;
-	th->validity = new_validity;
 	th->secret_id = new_secret_id;
 	th->expires = new_expires;
 	th->renew_after = new_renew_after;
+	th->have_key = true;
 	dout(" got ticket service %d (%s) secret_id %lld len %d\n",
 	     type, ceph_entity_type_name(type), th->secret_id,
 	     (int)th->ticket_blob->vec.iov_len);
@@ -384,6 +383,24 @@ bad:
 	return -ERANGE;
 }
 
+static bool need_key(struct ceph_x_ticket_handler *th)
+{
+	if (!th->have_key)
+		return true;
+
+	return get_seconds() >= th->renew_after;
+}
+
+static bool have_key(struct ceph_x_ticket_handler *th)
+{
+	if (th->have_key) {
+		if (get_seconds() >= th->expires)
+			th->have_key = false;
+	}
+
+	return th->have_key;
+}
+
 static void ceph_x_validate_tickets(struct ceph_auth_client *ac, int *pneed)
 {
 	int want = ac->want_keys;
@@ -402,20 +419,18 @@ static void ceph_x_validate_tickets(struct ceph_auth_client *ac, int *pneed)
 			continue;
 
 		th = get_ticket_handler(ac, service);
-
 		if (IS_ERR(th)) {
 			*pneed |= service;
 			continue;
 		}
 
-		if (get_seconds() >= th->renew_after)
+		if (need_key(th))
 			*pneed |= service;
-		if (get_seconds() >= th->expires)
+		if (!have_key(th))
 			xi->have_keys &= ~service;
 	}
 }
 
-
 static int ceph_x_build_request(struct ceph_auth_client *ac,
 				void *buf, void *end)
 {
@@ -667,14 +682,26 @@ static void ceph_x_destroy(struct ceph_auth_client *ac)
 	ac->private = NULL;
 }
 
-static void ceph_x_invalidate_authorizer(struct ceph_auth_client *ac,
-				   int peer_type)
+static void invalidate_ticket(struct ceph_auth_client *ac, int peer_type)
 {
 	struct ceph_x_ticket_handler *th;
 
 	th = get_ticket_handler(ac, peer_type);
 	if (!IS_ERR(th))
-		memset(&th->validity, 0, sizeof(th->validity));
+		th->have_key = false;
+}
+
+static void ceph_x_invalidate_authorizer(struct ceph_auth_client *ac,
+					 int peer_type)
+{
+	/*
+	 * We are to invalidate a service ticket in the hopes of
+	 * getting a new, hopefully more valid, one.  But, we won't get
+	 * it unless our AUTH ticket is good, so invalidate AUTH ticket
+	 * as well, just in case.
+	 */
+	invalidate_ticket(ac, peer_type);
+	invalidate_ticket(ac, CEPH_ENTITY_TYPE_AUTH);
 }
 
 static int calcu_signature(struct ceph_x_authorizer *au,
diff --git a/net/ceph/auth_x.h b/net/ceph/auth_x.h
index e8b7c6917d47..40b1a3cf7397 100644
--- a/net/ceph/auth_x.h
+++ b/net/ceph/auth_x.h
@@ -16,7 +16,7 @@ struct ceph_x_ticket_handler {
 	unsigned int service;
 
 	struct ceph_crypto_key session_key;
-	struct ceph_timespec validity;
+	bool have_key;
 
 	u64 secret_id;
 	struct ceph_buffer *ticket_blob;
diff --git a/net/ceph/messenger.c b/net/ceph/messenger.c
index 9981039ef4ff..9cfedf565f5b 100644
--- a/net/ceph/messenger.c
+++ b/net/ceph/messenger.c
@@ -23,9 +23,6 @@
 #include <linux/ceph/pagelist.h>
 #include <linux/export.h>
 
-#define list_entry_next(pos, member)					\
-	list_entry(pos->member.next, typeof(*pos), member)
-
 /*
  * Ceph uses the messenger to exchange ceph_msg messages with other
  * hosts in the system.  The messenger provides ordered and reliable
@@ -672,6 +669,8 @@ static void reset_connection(struct ceph_connection *con)
 	}
 	con->in_seq = 0;
 	con->in_seq_acked = 0;
+
+	con->out_skip = 0;
 }
 
 /*
@@ -771,6 +770,8 @@ static u32 get_global_seq(struct ceph_messenger *msgr, u32 gt)
 
 static void con_out_kvec_reset(struct ceph_connection *con)
 {
+	BUG_ON(con->out_skip);
+
 	con->out_kvec_left = 0;
 	con->out_kvec_bytes = 0;
 	con->out_kvec_cur = &con->out_kvec[0];
@@ -779,9 +780,9 @@ static void con_out_kvec_reset(struct ceph_connection *con)
 static void con_out_kvec_add(struct ceph_connection *con,
 				size_t size, void *data)
 {
-	int index;
+	int index = con->out_kvec_left;
 
-	index = con->out_kvec_left;
+	BUG_ON(con->out_skip);
 	BUG_ON(index >= ARRAY_SIZE(con->out_kvec));
 
 	con->out_kvec[index].iov_len = size;
@@ -790,6 +791,27 @@ static void con_out_kvec_add(struct ceph_connection *con,
 	con->out_kvec_bytes += size;
 }
 
+/*
+ * Chop off a kvec from the end.  Return residual number of bytes for
+ * that kvec, i.e. how many bytes would have been written if the kvec
+ * hadn't been nuked.
+ */
+static int con_out_kvec_skip(struct ceph_connection *con)
+{
+	int off = con->out_kvec_cur - con->out_kvec;
+	int skip = 0;
+
+	if (con->out_kvec_bytes > 0) {
+		skip = con->out_kvec[off + con->out_kvec_left - 1].iov_len;
+		BUG_ON(con->out_kvec_bytes < skip);
+		BUG_ON(!con->out_kvec_left);
+		con->out_kvec_bytes -= skip;
+		con->out_kvec_left--;
+	}
+
+	return skip;
+}
+
 #ifdef CONFIG_BLOCK
 
 /*
@@ -1042,7 +1064,7 @@ static bool ceph_msg_data_pagelist_advance(struct ceph_msg_data_cursor *cursor,
 	/* Move on to the next page */
 
 	BUG_ON(list_is_last(&cursor->page->lru, &pagelist->head));
-	cursor->page = list_entry_next(cursor->page, lru);
+	cursor->page = list_next_entry(cursor->page, lru);
 	cursor->last_piece = cursor->resid <= PAGE_SIZE;
 
 	return true;
@@ -1166,7 +1188,7 @@ static bool ceph_msg_data_advance(struct ceph_msg_data_cursor *cursor,
 	if (!cursor->resid && cursor->total_resid) {
 		WARN_ON(!cursor->last_piece);
 		BUG_ON(list_is_last(&cursor->data->links, cursor->data_head));
-		cursor->data = list_entry_next(cursor->data, links);
+		cursor->data = list_next_entry(cursor->data, links);
 		__ceph_msg_data_cursor_init(cursor);
 		new_piece = true;
 	}
@@ -1197,7 +1219,6 @@ static void prepare_write_message_footer(struct ceph_connection *con)
 	m->footer.flags |= CEPH_MSG_FOOTER_COMPLETE;
 
 	dout("prepare_write_message_footer %p\n", con);
-	con->out_kvec_is_msg = true;
 	con->out_kvec[v].iov_base = &m->footer;
 	if (con->peer_features & CEPH_FEATURE_MSG_AUTH) {
 		if (con->ops->sign_message)
@@ -1225,7 +1246,6 @@ static void prepare_write_message(struct ceph_connection *con)
 	u32 crc;
 
 	con_out_kvec_reset(con);
-	con->out_kvec_is_msg = true;
 	con->out_msg_done = false;
 
 	/* Sneak an ack in there first?  If we can get it into the same
@@ -1265,18 +1285,19 @@ static void prepare_write_message(struct ceph_connection *con)
 
 	/* tag + hdr + front + middle */
 	con_out_kvec_add(con, sizeof (tag_msg), &tag_msg);
-	con_out_kvec_add(con, sizeof (m->hdr), &m->hdr);
+	con_out_kvec_add(con, sizeof(con->out_hdr), &con->out_hdr);
 	con_out_kvec_add(con, m->front.iov_len, m->front.iov_base);
 
 	if (m->middle)
 		con_out_kvec_add(con, m->middle->vec.iov_len,
 			m->middle->vec.iov_base);
 
-	/* fill in crc (except data pages), footer */
+	/* fill in hdr crc and finalize hdr */
 	crc = crc32c(0, &m->hdr, offsetof(struct ceph_msg_header, crc));
 	con->out_msg->hdr.crc = cpu_to_le32(crc);
-	con->out_msg->footer.flags = 0;
+	memcpy(&con->out_hdr, &con->out_msg->hdr, sizeof(con->out_hdr));
 
+	/* fill in front and middle crc, footer */
 	crc = crc32c(0, m->front.iov_base, m->front.iov_len);
 	con->out_msg->footer.front_crc = cpu_to_le32(crc);
 	if (m->middle) {
@@ -1288,6 +1309,7 @@ static void prepare_write_message(struct ceph_connection *con)
 	dout("%s front_crc %u middle_crc %u\n", __func__,
 	     le32_to_cpu(con->out_msg->footer.front_crc),
 	     le32_to_cpu(con->out_msg->footer.middle_crc));
+	con->out_msg->footer.flags = 0;
 
 	/* is there a data payload? */
 	con->out_msg->footer.data_crc = 0;
@@ -1492,7 +1514,6 @@ static int write_partial_kvec(struct ceph_connection *con)
 		}
 	}
 	con->out_kvec_left = 0;
-	con->out_kvec_is_msg = false;
 	ret = 1;
 out:
 	dout("write_partial_kvec %p %d left in %d kvecs ret = %d\n", con,
@@ -1584,6 +1605,7 @@ static int write_partial_skip(struct ceph_connection *con)
 {
 	int ret;
 
+	dout("%s %p %d left\n", __func__, con, con->out_skip);
 	while (con->out_skip > 0) {
 		size_t size = min(con->out_skip, (int) PAGE_CACHE_SIZE);
 
@@ -2506,13 +2528,13 @@ more:
 
 more_kvec:
 	/* kvec data queued? */
-	if (con->out_skip) {
-		ret = write_partial_skip(con);
+	if (con->out_kvec_left) {
+		ret = write_partial_kvec(con);
 		if (ret <= 0)
 			goto out;
 	}
-	if (con->out_kvec_left) {
-		ret = write_partial_kvec(con);
+	if (con->out_skip) {
+		ret = write_partial_skip(con);
 		if (ret <= 0)
 			goto out;
 	}
@@ -2805,13 +2827,17 @@ static bool con_backoff(struct ceph_connection *con)
 
 static void con_fault_finish(struct ceph_connection *con)
 {
+	dout("%s %p\n", __func__, con);
+
 	/*
 	 * in case we faulted due to authentication, invalidate our
 	 * current tickets so that we can get new ones.
 	 */
-	if (con->auth_retry && con->ops->invalidate_authorizer) {
-		dout("calling invalidate_authorizer()\n");
-		con->ops->invalidate_authorizer(con);
+	if (con->auth_retry) {
+		dout("auth_retry %d, invalidating\n", con->auth_retry);
+		if (con->ops->invalidate_authorizer)
+			con->ops->invalidate_authorizer(con);
+		con->auth_retry = 0;
 	}
 
 	if (con->ops->fault)
@@ -3050,16 +3076,31 @@ void ceph_msg_revoke(struct ceph_msg *msg)
 		ceph_msg_put(msg);
 	}
 	if (con->out_msg == msg) {
-		dout("%s %p msg %p - was sending\n", __func__, con, msg);
-		con->out_msg = NULL;
-		if (con->out_kvec_is_msg) {
-			con->out_skip = con->out_kvec_bytes;
-			con->out_kvec_is_msg = false;
+		BUG_ON(con->out_skip);
+		/* footer */
+		if (con->out_msg_done) {
+			con->out_skip += con_out_kvec_skip(con);
+		} else {
+			BUG_ON(!msg->data_length);
+			if (con->peer_features & CEPH_FEATURE_MSG_AUTH)
+				con->out_skip += sizeof(msg->footer);
+			else
+				con->out_skip += sizeof(msg->old_footer);
 		}
+		/* data, middle, front */
+		if (msg->data_length)
+			con->out_skip += msg->cursor.total_resid;
+		if (msg->middle)
+			con->out_skip += con_out_kvec_skip(con);
+		con->out_skip += con_out_kvec_skip(con);
+
+		dout("%s %p msg %p - was sending, will write %d skip %d\n",
+		     __func__, con, msg, con->out_kvec_bytes, con->out_skip);
 		msg->hdr.seq = 0;
-
+		con->out_msg = NULL;
 		ceph_msg_put(msg);
 	}
+
 	mutex_unlock(&con->mutex);
 }
 
@@ -3361,9 +3402,7 @@ static void ceph_msg_free(struct ceph_msg *m)
 static void ceph_msg_release(struct kref *kref)
 {
 	struct ceph_msg *m = container_of(kref, struct ceph_msg, kref);
-	LIST_HEAD(data);
-	struct list_head *links;
-	struct list_head *next;
+	struct ceph_msg_data *data, *next;
 
 	dout("%s %p\n", __func__, m);
 	WARN_ON(!list_empty(&m->list_head));
@@ -3376,12 +3415,8 @@ static void ceph_msg_release(struct kref *kref)
 		m->middle = NULL;
 	}
 
-	list_splice_init(&m->data, &data);
-	list_for_each_safe(links, next, &data) {
-		struct ceph_msg_data *data;
-
-		data = list_entry(links, struct ceph_msg_data, links);
-		list_del_init(links);
+	list_for_each_entry_safe(data, next, &m->data, links) {
+		list_del_init(&data->links);
 		ceph_msg_data_destroy(data);
 	}
 	m->data_length = 0;
diff --git a/net/ceph/mon_client.c b/net/ceph/mon_client.c
index edda01626a45..de85dddc3dc0 100644
--- a/net/ceph/mon_client.c
+++ b/net/ceph/mon_client.c
@@ -364,10 +364,6 @@ static bool have_debugfs_info(struct ceph_mon_client *monc)
 	return monc->client->have_fsid && monc->auth->global_id > 0;
 }
 
-/*
- * The monitor responds with mount ack indicate mount success.  The
- * included client ticket allows the client to talk to MDSs and OSDs.
- */
 static void ceph_monc_handle_map(struct ceph_mon_client *monc,
 				 struct ceph_msg *msg)
 {