[PATCH net-next 0/5] Involve rhashtable_lookup_insert routine

[PATCH net-next 0/5] Involve rhashtable_lookup_insert routine

[Ying Xue]

The series aims to involve rhashtable_lookup_insert() to guarantee
that the process of lookup and insertion of an object from/into hash
table is finished atomically, allowing rhashtable's users not to
introduce an extra lock during search and insertion. For example,
tipc socket is the first user benefiting from this enhancement. 

But before rhashtable_lookup_insert() is involved, the following
optimizations need to be first done:
- simplify rhashtable_lookup by reusing rhashtable_lookup_compare()
- introduce rhashtable_wakeup_worker() to further reduce duplicated
  code in patch #2
- fix an issue in patch #3
- involve rhashtable_lookup_insert(). But in this version, we firstly
  use rhashtable_lookup() to search duplicate key in both old and new
  bucket table; secondly introduce another __rhashtable_insert() helper
  function to reduce the duplicated code between rhashtable_insert()
  and rhashtable_lookup_insert().
- add patch #5 into the series as it depends on above patches. But in
  this version, no change is made comparing with its previous version.

Ying Xue (5):
  rhashtable: optimize rhashtable_lookup routine
  rhashtable: introduce rhashtable_wakeup_worker helper function
  rhashtable: use future table size to make expansion decision
  rhashtable: involve rhashtable_lookup_insert routine
  tipc: convert tipc reference table to use generic rhashtable

 include/linux/rhashtable.h |    1 +
 lib/rhashtable.c           |  162 ++++++++++-----
 net/tipc/Kconfig           |   12 --
 net/tipc/config.c          |   24 +--
 net/tipc/core.c            |   10 +-
 net/tipc/core.h            |    3 -
 net/tipc/socket.c          |  480 ++++++++++++++++----------------------------
 net/tipc/socket.h          |    4 +-
 8 files changed, 296 insertions(+), 400 deletions(-)

-- 
1.7.9.5


------------------------------------------------------------------------------
Dive into the World of Parallel Programming! The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net

[PATCH net-next 1/5] rhashtable: optimize rhashtable_lookup routine

[Ying Xue]

Define an internal compare function and relevant compare argument,
and then make use of rhashtable_lookup_compare() to lookup key in
hash table, reducing duplicated code between rhashtable_lookup()
and rhashtable_lookup_compare().

Signed-off-by: Ying Xue <ying.xue@windriver.com>
Cc: Thomas Graf <tgraf@suug.ch>
---
 lib/rhashtable.c |   41 ++++++++++++++++++-----------------------
 1 file changed, 18 insertions(+), 23 deletions(-)

diff --git a/lib/rhashtable.c b/lib/rhashtable.c
index cbad192..f2fdd7a 100644
--- a/lib/rhashtable.c
+++ b/lib/rhashtable.c
@@ -612,6 +612,19 @@ restart:
 }
 EXPORT_SYMBOL_GPL(rhashtable_remove);
 
+struct rhashtable_compare_arg {
+	struct rhashtable *ht;
+	const void *key;
+};
+
+static bool rhashtable_compare(void *ptr, void *arg)
+{
+	struct rhashtable_compare_arg *x = arg;
+	struct rhashtable *ht = x->ht;
+
+	return !memcmp(ptr + ht->p.key_offset, x->key, ht->p.key_len);
+}
+
 /**
  * rhashtable_lookup - lookup key in hash table
  * @ht:		hash table
@@ -627,32 +640,14 @@ EXPORT_SYMBOL_GPL(rhashtable_remove);
  */
 void *rhashtable_lookup(struct rhashtable *ht, const void *key)
 {
-	const struct bucket_table *tbl, *old_tbl;
-	struct rhash_head *he;
-	u32 hash;
+	struct rhashtable_compare_arg arg = {
+		.ht = ht,
+		.key = key,
+	};
 
 	BUG_ON(!ht->p.key_len);
 
-	rcu_read_lock();
-	old_tbl = rht_dereference_rcu(ht->tbl, ht);
-	tbl = rht_dereference_rcu(ht->future_tbl, ht);
-	hash = key_hashfn(ht, key, ht->p.key_len);
-restart:
-	rht_for_each_rcu(he, tbl, rht_bucket_index(tbl, hash)) {
-		if (memcmp(rht_obj(ht, he) + ht->p.key_offset, key,
-			   ht->p.key_len))
-			continue;
-		rcu_read_unlock();
-		return rht_obj(ht, he);
-	}
-
-	if (unlikely(tbl != old_tbl)) {
-		tbl = old_tbl;
-		goto restart;
-	}
-
-	rcu_read_unlock();
-	return NULL;
+	return rhashtable_lookup_compare(ht, key, &rhashtable_compare, &arg);
 }
 EXPORT_SYMBOL_GPL(rhashtable_lookup);
 
-- 
1.7.9.5


------------------------------------------------------------------------------
Dive into the World of Parallel Programming! The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net

[PATCH net-next 2/5] rhashtable: introduce rhashtable_wakeup_worker helper function

[Ying Xue]

Introduce rhashtable_wakeup_worker() helper function to reduce
duplicated code where to wake up worker.

Signed-off-by: Ying Xue <ying.xue@windriver.com>
Cc: Thomas Graf <tgraf@suug.ch>
---
 lib/rhashtable.c |   20 +++++++++++++-------
 1 file changed, 13 insertions(+), 7 deletions(-)

diff --git a/lib/rhashtable.c b/lib/rhashtable.c
index f2fdd7a..6eda22f 100644
--- a/lib/rhashtable.c
+++ b/lib/rhashtable.c
@@ -492,6 +492,17 @@ static void rht_deferred_worker(struct work_struct *work)
 	mutex_unlock(&ht->mutex);
 }
 
+static void rhashtable_wakeup_worker(struct rhashtable *ht)
+{
+	struct bucket_table *tbl = rht_dereference_rcu(ht->tbl, ht);
+	struct bucket_table *new_tbl = rht_dereference_rcu(ht->future_tbl, ht);
+
+	if (ht->tbl != ht->future_tbl &&
+	    ((ht->p.grow_decision && ht->p.grow_decision(ht, new_tbl->size)) ||
+	    (ht->p.shrink_decision && ht->p.shrink_decision(ht, tbl->size))))
+		schedule_delayed_work(&ht->run_work, 0);
+}
+
 /**
  * rhashtable_insert - insert object into hash hash table
  * @ht:		hash table
@@ -533,9 +544,7 @@ void rhashtable_insert(struct rhashtable *ht, struct rhash_head *obj)
 	atomic_inc(&ht->nelems);
 
 	/* Only grow the table if no resizing is currently in progress. */
-	if (ht->tbl != ht->future_tbl &&
-	    ht->p.grow_decision && ht->p.grow_decision(ht, tbl->size))
-		schedule_delayed_work(&ht->run_work, 0);
+	rhashtable_wakeup_worker(ht);
 
 	rcu_read_unlock();
 }
@@ -584,10 +593,7 @@ restart:
 
 		spin_unlock_bh(lock);
 
-		if (ht->tbl != ht->future_tbl &&
-		    ht->p.shrink_decision &&
-		    ht->p.shrink_decision(ht, tbl->size))
-			schedule_delayed_work(&ht->run_work, 0);
+		rhashtable_wakeup_worker(ht);
 
 		rcu_read_unlock();
 
-- 
1.7.9.5


------------------------------------------------------------------------------
Dive into the World of Parallel Programming! The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net

[PATCH net-next 3/5] rhashtable: use future table size to make expansion decision

[Ying Xue]

Should use future table size instead of old table size to decide
whether hash table is worth being expanded.

Signed-off-by: Ying Xue <ying.xue@windriver.com>
Cc: Thomas Graf <tgraf@suug.ch>
---
 lib/rhashtable.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/lib/rhashtable.c b/lib/rhashtable.c
index 6eda22f..f5288b1 100644
--- a/lib/rhashtable.c
+++ b/lib/rhashtable.c
@@ -478,13 +478,14 @@ EXPORT_SYMBOL_GPL(rhashtable_shrink);
 static void rht_deferred_worker(struct work_struct *work)
 {
 	struct rhashtable *ht;
-	struct bucket_table *tbl;
+	struct bucket_table *tbl, *new_tbl;
 
 	ht = container_of(work, struct rhashtable, run_work.work);
 	mutex_lock(&ht->mutex);
 	tbl = rht_dereference(ht->tbl, ht);
+	new_tbl = rht_dereference(ht->future_tbl, ht);
 
-	if (ht->p.grow_decision && ht->p.grow_decision(ht, tbl->size))
+	if (ht->p.grow_decision && ht->p.grow_decision(ht, new_tbl->size))
 		rhashtable_expand(ht);
 	else if (ht->p.shrink_decision && ht->p.shrink_decision(ht, tbl->size))
 		rhashtable_shrink(ht);
-- 
1.7.9.5


------------------------------------------------------------------------------
Dive into the World of Parallel Programming! The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net

[PATCH net-next 4/5] rhashtable: involve rhashtable_lookup_insert routine

[Ying Xue]

Involve a new function called rhashtable_lookup_insert() which makes
lookup and insertion atomic under bucket lock protection, helping us
avoid to introduce an extra lock when we search and insert an object
into hash table.

Signed-off-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Thomas Graf <tgraf@suug.ch>
---
 include/linux/rhashtable.h |    1 +
 lib/rhashtable.c           |   98 ++++++++++++++++++++++++++++++++++++--------
 2 files changed, 83 insertions(+), 16 deletions(-)

diff --git a/include/linux/rhashtable.h b/include/linux/rhashtable.h
index de1459c7..73c913f 100644
--- a/include/linux/rhashtable.h
+++ b/include/linux/rhashtable.h
@@ -168,6 +168,7 @@ int rhashtable_shrink(struct rhashtable *ht);
 void *rhashtable_lookup(struct rhashtable *ht, const void *key);
 void *rhashtable_lookup_compare(struct rhashtable *ht, const void *key,
 				bool (*compare)(void *, void *), void *arg);
+bool rhashtable_lookup_insert(struct rhashtable *ht, struct rhash_head *obj);
 
 void rhashtable_destroy(struct rhashtable *ht);
 
diff --git a/lib/rhashtable.c b/lib/rhashtable.c
index f5288b1..c77b472 100644
--- a/lib/rhashtable.c
+++ b/lib/rhashtable.c
@@ -504,8 +504,27 @@ static void rhashtable_wakeup_worker(struct rhashtable *ht)
 		schedule_delayed_work(&ht->run_work, 0);
 }
 
+static void __rhashtable_insert(struct rhashtable *ht, struct rhash_head *obj,
+				struct bucket_table *tbl, u32 hash)
+{
+	struct rhash_head *head = rht_dereference_bucket(tbl->buckets[hash],
+							 tbl, hash);
+
+	if (rht_is_a_nulls(head))
+		INIT_RHT_NULLS_HEAD(obj->next, ht, hash);
+	else
+		RCU_INIT_POINTER(obj->next, head);
+
+	rcu_assign_pointer(tbl->buckets[hash], obj);
+
+	atomic_inc(&ht->nelems);
+
+	/* Only grow the table if no resizing is currently in progress. */
+	rhashtable_wakeup_worker(ht);
+}
+
 /**
- * rhashtable_insert - insert object into hash hash table
+ * rhashtable_insert - insert object into hash table
  * @ht:		hash table
  * @obj:	pointer to hash head inside object
  *
@@ -522,7 +541,6 @@ static void rhashtable_wakeup_worker(struct rhashtable *ht)
 void rhashtable_insert(struct rhashtable *ht, struct rhash_head *obj)
 {
 	struct bucket_table *tbl;
-	struct rhash_head *head;
 	spinlock_t *lock;
 	unsigned hash;
 
@@ -533,20 +551,9 @@ void rhashtable_insert(struct rhashtable *ht, struct rhash_head *obj)
 	lock = bucket_lock(tbl, hash);
 
 	spin_lock_bh(lock);
-	head = rht_dereference_bucket(tbl->buckets[hash], tbl, hash);
-	if (rht_is_a_nulls(head))
-		INIT_RHT_NULLS_HEAD(obj->next, ht, hash);
-	else
-		RCU_INIT_POINTER(obj->next, head);
-
-	rcu_assign_pointer(tbl->buckets[hash], obj);
+	__rhashtable_insert(ht, obj, tbl, hash);
 	spin_unlock_bh(lock);
 
-	atomic_inc(&ht->nelems);
-
-	/* Only grow the table if no resizing is currently in progress. */
-	rhashtable_wakeup_worker(ht);
-
 	rcu_read_unlock();
 }
 EXPORT_SYMBOL_GPL(rhashtable_insert);
@@ -560,7 +567,7 @@ EXPORT_SYMBOL_GPL(rhashtable_insert);
  * walk the bucket chain upon removal. The removal operation is thus
  * considerable slow if the hash table is not correctly sized.
  *
- * Will automatically shrink the table via rhashtable_expand() if the the
+ * Will automatically shrink the table via rhashtable_expand() if the
  * shrink_decision function specified at rhashtable_init() returns true.
  *
  * The caller must ensure that no concurrent table mutations occur. It is
@@ -641,7 +648,7 @@ static bool rhashtable_compare(void *ptr, void *arg)
  * for a entry with an identical key. The first matching entry is returned.
  *
  * This lookup function may only be used for fixed key hash table (key_len
- * paramter set). It will BUG() if used inappropriately.
+ * parameter set). It will BUG() if used inappropriately.
  *
  * Lookups may occur in parallel with hashtable mutations and resizing.
  */
@@ -702,6 +709,65 @@ restart:
 }
 EXPORT_SYMBOL_GPL(rhashtable_lookup_compare);
 
+/**
+ * rhashtable_lookup_insert - lookup and insert object into hash table
+ * @ht:		hash table
+ * @obj:	pointer to hash head inside object
+ *
+ * Locks down the bucket chain in both the old and new table if a resize
+ * is in progress to ensure that writers can't remove from the old table
+ * and can't insert to the new table during the atomic operation of search
+ * and insertion. Searches for duplicates in both the old and new table if
+ * a resize is in progress.
+ *
+ * This lookup function may only be used for fixed key hash table (key_len
+ * parameter set). It will BUG() if used inappropriately.
+ *
+ * It is safe to call this function from atomic context.
+ *
+ * Will trigger an automatic deferred table resizing if the size grows
+ * beyond the watermark indicated by grow_decision() which can be passed
+ * to rhashtable_init().
+ */
+bool rhashtable_lookup_insert(struct rhashtable *ht, struct rhash_head *obj)
+{
+	struct bucket_table *new_tbl, *old_tbl;
+	spinlock_t *new_bucket_lock, *old_bucket_lock;
+	u32 new_hash, old_hash;
+	bool success = true;
+
+	BUG_ON(!ht->p.key_len);
+
+	rcu_read_lock();
+
+	old_tbl = rht_dereference_rcu(ht->tbl, ht);
+	old_hash = head_hashfn(ht, old_tbl, obj);
+	old_bucket_lock = bucket_lock(old_tbl, old_hash);
+	spin_lock_bh(old_bucket_lock);
+
+	new_tbl = rht_dereference_rcu(ht->future_tbl, ht);
+	new_hash = head_hashfn(ht, new_tbl, obj);
+	new_bucket_lock = bucket_lock(new_tbl, new_hash);
+	if (unlikely(old_tbl != new_tbl))
+		spin_lock_bh_nested(new_bucket_lock, RHT_LOCK_NESTED);
+
+	if (rhashtable_lookup(ht, rht_obj(ht, obj) + ht->p.key_offset)) {
+		success = false;
+		goto exit;
+	}
+
+	__rhashtable_insert(ht, obj, new_tbl, new_hash);
+
+exit:
+	if (unlikely(old_tbl != new_tbl))
+		spin_unlock_bh(new_bucket_lock);
+	spin_unlock_bh(old_bucket_lock);
+	rcu_read_unlock();
+
+	return success;
+}
+EXPORT_SYMBOL_GPL(rhashtable_lookup_insert);
+
 static size_t rounded_hashtable_size(struct rhashtable_params *params)
 {
 	return max(roundup_pow_of_two(params->nelem_hint * 4 / 3),
-- 
1.7.9.5


------------------------------------------------------------------------------
Dive into the World of Parallel Programming! The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net

[PATCH net-next 5/5] tipc: convert tipc reference table to use generic rhashtable

[Ying Xue]

As tipc reference table is statically allocated, its memory size
requested on stack initialization stage is quite big even if the
maximum port number is just restricted to 8191 currently, however,
the number already becomes insufficient in practice. But if the
maximum ports is allowed to its theory value - 2^32, its consumed
memory size will reach a ridiculously unacceptable value. Apart from
this, heavy tipc users spend a considerable amount of time in
tipc_sk_get() due to the read-lock on ref_table_lock.

If tipc reference table is converted with generic rhashtable, above
mentioned both disadvantages would be resolved respectively: making
use of the new resizable hash table can avoid locking on the lookup;
smaller memory size is required at initial stage, for example, 256
hash bucket slots are requested at the beginning phase instead of
allocating the entire 8191 slots in old mode. The hash table will
grow if entries exceeds 75% of table size up to a total table size
of 1M, and it will automatically shrink if usage falls below 30%,
but the minimum table size is allowed down to 256.

Also converts ref_table_lock to a separate mutex to protect hash table
mutations on write side. Lastly defers the release of the socket
reference using call_rcu() to allow using an RCU read-side protected
call to rhashtable_lookup().

Signed-off-by: Ying Xue <ying.xue@windriver.com>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Acked-by: Erik Hugne <erik.hugne@ericsson.com>
Cc: Thomas Graf <tgraf@suug.ch>
---
 net/tipc/Kconfig  |   12 --
 net/tipc/config.c |   24 +--
 net/tipc/core.c   |   10 +-
 net/tipc/core.h   |    3 -
 net/tipc/socket.c |  480 +++++++++++++++++++----------------------------------
 net/tipc/socket.h |    4 +-
 6 files changed, 180 insertions(+), 353 deletions(-)

diff --git a/net/tipc/Kconfig b/net/tipc/Kconfig
index c890848..91c8a8e 100644
--- a/net/tipc/Kconfig
+++ b/net/tipc/Kconfig
@@ -20,18 +20,6 @@ menuconfig TIPC
 
 	  If in doubt, say N.
 
-config TIPC_PORTS
-	int "Maximum number of ports in a node"
-	depends on TIPC
-	range 127 65535
-	default "8191"
-	help
-	  Specifies how many ports can be supported by a node.
-	  Can range from 127 to 65535 ports; default is 8191.
-
-	  Setting this to a smaller value saves some memory,
-	  setting it to higher allows for more ports.
-
 config TIPC_MEDIA_IB
 	bool "InfiniBand media type support"
 	depends on TIPC && INFINIBAND_IPOIB
diff --git a/net/tipc/config.c b/net/tipc/config.c
index 876f4c6..0b3a90e 100644
--- a/net/tipc/config.c
+++ b/net/tipc/config.c
@@ -183,22 +183,6 @@ static struct sk_buff *cfg_set_own_addr(void)
 	return tipc_cfg_reply_error_string("cannot change to network mode");
 }
 
-static struct sk_buff *cfg_set_max_ports(void)
-{
-	u32 value;
-
-	if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED))
-		return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR);
-	value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area));
-	if (value == tipc_max_ports)
-		return tipc_cfg_reply_none();
-	if (value < 127 || value > 65535)
-		return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE
-						   " (max ports must be 127-65535)");
-	return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED
-		" (cannot change max ports while TIPC is active)");
-}
-
 static struct sk_buff *cfg_set_netid(void)
 {
 	u32 value;
@@ -285,15 +269,9 @@ struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area
 	case TIPC_CMD_SET_NODE_ADDR:
 		rep_tlv_buf = cfg_set_own_addr();
 		break;
-	case TIPC_CMD_SET_MAX_PORTS:
-		rep_tlv_buf = cfg_set_max_ports();
-		break;
 	case TIPC_CMD_SET_NETID:
 		rep_tlv_buf = cfg_set_netid();
 		break;
-	case TIPC_CMD_GET_MAX_PORTS:
-		rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_max_ports);
-		break;
 	case TIPC_CMD_GET_NETID:
 		rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_net_id);
 		break;
@@ -317,6 +295,8 @@ struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area
 	case TIPC_CMD_SET_REMOTE_MNG:
 	case TIPC_CMD_GET_REMOTE_MNG:
 	case TIPC_CMD_DUMP_LOG:
+	case TIPC_CMD_SET_MAX_PORTS:
+	case TIPC_CMD_GET_MAX_PORTS:
 		rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED
 							  " (obsolete command)");
 		break;
diff --git a/net/tipc/core.c b/net/tipc/core.c
index a5737b8..71b2ada 100644
--- a/net/tipc/core.c
+++ b/net/tipc/core.c
@@ -34,6 +34,8 @@
  * POSSIBILITY OF SUCH DAMAGE.
  */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include "core.h"
 #include "name_table.h"
 #include "subscr.h"
@@ -47,7 +49,6 @@ int tipc_random __read_mostly;
 
 /* configurable TIPC parameters */
 u32 tipc_own_addr __read_mostly;
-int tipc_max_ports __read_mostly;
 int tipc_net_id __read_mostly;
 int sysctl_tipc_rmem[3] __read_mostly;	/* min/default/max */
 
@@ -84,9 +85,9 @@ static void tipc_core_stop(void)
 	tipc_netlink_stop();
 	tipc_subscr_stop();
 	tipc_nametbl_stop();
-	tipc_sk_ref_table_stop();
 	tipc_socket_stop();
 	tipc_unregister_sysctl();
+	tipc_sk_rht_destroy();
 }
 
 /**
@@ -98,7 +99,7 @@ static int tipc_core_start(void)
 
 	get_random_bytes(&tipc_random, sizeof(tipc_random));
 
-	err = tipc_sk_ref_table_init(tipc_max_ports, tipc_random);
+	err = tipc_sk_rht_init();
 	if (err)
 		goto out_reftbl;
 
@@ -138,7 +139,7 @@ out_socket:
 out_netlink:
 	tipc_nametbl_stop();
 out_nametbl:
-	tipc_sk_ref_table_stop();
+	tipc_sk_rht_destroy();
 out_reftbl:
 	return err;
 }
@@ -150,7 +151,6 @@ static int __init tipc_init(void)
 	pr_info("Activated (version " TIPC_MOD_VER ")\n");
 
 	tipc_own_addr = 0;
-	tipc_max_ports = CONFIG_TIPC_PORTS;
 	tipc_net_id = 4711;
 
 	sysctl_tipc_rmem[0] = TIPC_CONN_OVERLOAD_LIMIT >> 4 <<
diff --git a/net/tipc/core.h b/net/tipc/core.h
index 8460213..56fe422 100644
--- a/net/tipc/core.h
+++ b/net/tipc/core.h
@@ -37,8 +37,6 @@
 #ifndef _TIPC_CORE_H
 #define _TIPC_CORE_H
 
-#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
-
 #include <linux/tipc.h>
 #include <linux/tipc_config.h>
 #include <linux/tipc_netlink.h>
@@ -79,7 +77,6 @@ int tipc_snprintf(char *buf, int len, const char *fmt, ...);
  * Global configuration variables
  */
 extern u32 tipc_own_addr __read_mostly;
-extern int tipc_max_ports __read_mostly;
 extern int tipc_net_id __read_mostly;
 extern int sysctl_tipc_rmem[3] __read_mostly;
 extern int sysctl_tipc_named_timeout __read_mostly;
diff --git a/net/tipc/socket.c b/net/tipc/socket.c
index 4731cad..3335ccf 100644
--- a/net/tipc/socket.c
+++ b/net/tipc/socket.c
@@ -34,22 +34,25 @@
  * POSSIBILITY OF SUCH DAMAGE.
  */
 
+#include <linux/rhashtable.h>
+#include <linux/jhash.h>
 #include "core.h"
 #include "name_table.h"
 #include "node.h"
 #include "link.h"
-#include <linux/export.h>
 #include "config.h"
 #include "socket.h"
 
-#define SS_LISTENING	-1	/* socket is listening */
-#define SS_READY	-2	/* socket is connectionless */
+#define SS_LISTENING		-1	/* socket is listening */
+#define SS_READY		-2	/* socket is connectionless */
 
-#define CONN_TIMEOUT_DEFAULT  8000	/* default connect timeout = 8s */
-#define CONN_PROBING_INTERVAL 3600000	/* [ms] => 1 h */
-#define TIPC_FWD_MSG	      1
-#define TIPC_CONN_OK          0
-#define TIPC_CONN_PROBING     1
+#define CONN_TIMEOUT_DEFAULT	8000	/* default connect timeout = 8s */
+#define CONN_PROBING_INTERVAL	3600000	/* [ms] => 1 h */
+#define TIPC_FWD_MSG		1
+#define TIPC_CONN_OK		0
+#define TIPC_CONN_PROBING	1
+#define TIPC_MAX_PORT		0xffffffff
+#define TIPC_MIN_PORT		1
 
 /**
  * struct tipc_sock - TIPC socket structure
@@ -59,7 +62,7 @@
  * @conn_instance: TIPC instance used when connection was established
  * @published: non-zero if port has one or more associated names
  * @max_pkt: maximum packet size "hint" used when building messages sent by port
- * @ref: unique reference to port in TIPC object registry
+ * @portid: unique port identity in TIPC socket hash table
  * @phdr: preformatted message header used when sending messages
  * @port_list: adjacent ports in TIPC's global list of ports
  * @publications: list of publications for port
@@ -74,6 +77,8 @@
  * @link_cong: non-zero if owner must sleep because of link congestion
  * @sent_unacked: # messages sent by socket, and not yet acked by peer
  * @rcv_unacked: # messages read by user, but not yet acked back to peer
+ * @node: hash table node
+ * @rcu: rcu struct for tipc_sock
  */
 struct tipc_sock {
 	struct sock sk;
@@ -82,7 +87,7 @@ struct tipc_sock {
 	u32 conn_instance;
 	int published;
 	u32 max_pkt;
-	u32 ref;
+	u32 portid;
 	struct tipc_msg phdr;
 	struct list_head sock_list;
 	struct list_head publications;
@@ -95,6 +100,8 @@ struct tipc_sock {
 	bool link_cong;
 	uint sent_unacked;
 	uint rcv_unacked;
+	struct rhash_head node;
+	struct rcu_head rcu;
 };
 
 static int tipc_backlog_rcv(struct sock *sk, struct sk_buff *skb);
@@ -103,16 +110,14 @@ static void tipc_write_space(struct sock *sk);
 static int tipc_release(struct socket *sock);
 static int tipc_accept(struct socket *sock, struct socket *new_sock, int flags);
 static int tipc_wait_for_sndmsg(struct socket *sock, long *timeo_p);
-static void tipc_sk_timeout(unsigned long ref);
+static void tipc_sk_timeout(unsigned long portid);
 static int tipc_sk_publish(struct tipc_sock *tsk, uint scope,
 			   struct tipc_name_seq const *seq);
 static int tipc_sk_withdraw(struct tipc_sock *tsk, uint scope,
 			    struct tipc_name_seq const *seq);
-static u32 tipc_sk_ref_acquire(struct tipc_sock *tsk);
-static void tipc_sk_ref_discard(u32 ref);
-static struct tipc_sock *tipc_sk_get(u32 ref);
-static struct tipc_sock *tipc_sk_get_next(u32 *ref);
-static void tipc_sk_put(struct tipc_sock *tsk);
+static struct tipc_sock *tipc_sk_lookup(u32 portid);
+static int tipc_sk_insert(struct tipc_sock *tsk);
+static void tipc_sk_remove(struct tipc_sock *tsk);
 
 static const struct proto_ops packet_ops;
 static const struct proto_ops stream_ops;
@@ -174,6 +179,9 @@ static const struct nla_policy tipc_nl_sock_policy[TIPC_NLA_SOCK_MAX + 1] = {
  *   - port reference
  */
 
+/* Protects tipc socket hash table mutations */
+static struct rhashtable tipc_sk_rht;
+
 static u32 tsk_peer_node(struct tipc_sock *tsk)
 {
 	return msg_destnode(&tsk->phdr);
@@ -305,7 +313,6 @@ static int tipc_sk_create(struct net *net, struct socket *sock,
 	struct sock *sk;
 	struct tipc_sock *tsk;
 	struct tipc_msg *msg;
-	u32 ref;
 
 	/* Validate arguments */
 	if (unlikely(protocol != 0))
@@ -339,24 +346,22 @@ static int tipc_sk_create(struct net *net, struct socket *sock,
 		return -ENOMEM;
 
 	tsk = tipc_sk(sk);
-	ref = tipc_sk_ref_acquire(tsk);
-	if (!ref) {
-		pr_warn("Socket create failed; reference table exhausted\n");
-		return -ENOMEM;
-	}
 	tsk->max_pkt = MAX_PKT_DEFAULT;
-	tsk->ref = ref;
 	INIT_LIST_HEAD(&tsk->publications);
 	msg = &tsk->phdr;
 	tipc_msg_init(msg, TIPC_LOW_IMPORTANCE, TIPC_NAMED_MSG,
 		      NAMED_H_SIZE, 0);
-	msg_set_origport(msg, ref);
 
 	/* Finish initializing socket data structures */
 	sock->ops = ops;
 	sock->state = state;
 	sock_init_data(sock, sk);
-	k_init_timer(&tsk->timer, (Handler)tipc_sk_timeout, ref);
+	if (tipc_sk_insert(tsk)) {
+		pr_warn("Socket create failed; port numbrer exhausted\n");
+		return -EINVAL;
+	}
+	msg_set_origport(msg, tsk->portid);
+	k_init_timer(&tsk->timer, (Handler)tipc_sk_timeout, tsk->portid);
 	sk->sk_backlog_rcv = tipc_backlog_rcv;
 	sk->sk_rcvbuf = sysctl_tipc_rmem[1];
 	sk->sk_data_ready = tipc_data_ready;
@@ -442,6 +447,13 @@ int tipc_sock_accept_local(struct socket *sock, struct socket **newsock,
 	return ret;
 }
 
+static void tipc_sk_callback(struct rcu_head *head)
+{
+	struct tipc_sock *tsk = container_of(head, struct tipc_sock, rcu);
+
+	sock_put(&tsk->sk);
+}
+
 /**
  * tipc_release - destroy a TIPC socket
  * @sock: socket to destroy
@@ -491,7 +503,7 @@ static int tipc_release(struct socket *sock)
 			    (sock->state == SS_CONNECTED)) {
 				sock->state = SS_DISCONNECTING;
 				tsk->connected = 0;
-				tipc_node_remove_conn(dnode, tsk->ref);
+				tipc_node_remove_conn(dnode, tsk->portid);
 			}
 			if (tipc_msg_reverse(skb, &dnode, TIPC_ERR_NO_PORT))
 				tipc_link_xmit_skb(skb, dnode, 0);
@@ -499,16 +511,16 @@ static int tipc_release(struct socket *sock)
 	}
 
 	tipc_sk_withdraw(tsk, 0, NULL);
-	tipc_sk_ref_discard(tsk->ref);
 	k_cancel_timer(&tsk->timer);
+	tipc_sk_remove(tsk);
 	if (tsk->connected) {
 		skb = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE, TIPC_CONN_MSG,
 				      SHORT_H_SIZE, 0, dnode, tipc_own_addr,
 				      tsk_peer_port(tsk),
-				      tsk->ref, TIPC_ERR_NO_PORT);
+				      tsk->portid, TIPC_ERR_NO_PORT);
 		if (skb)
-			tipc_link_xmit_skb(skb, dnode, tsk->ref);
-		tipc_node_remove_conn(dnode, tsk->ref);
+			tipc_link_xmit_skb(skb, dnode, tsk->portid);
+		tipc_node_remove_conn(dnode, tsk->portid);
 	}
 	k_term_timer(&tsk->timer);
 
@@ -518,7 +530,8 @@ static int tipc_release(struct socket *sock)
 	/* Reject any messages that accumulated in backlog queue */
 	sock->state = SS_DISCONNECTING;
 	release_sock(sk);
-	sock_put(sk);
+
+	call_rcu(&tsk->rcu, tipc_sk_callback);
 	sock->sk = NULL;
 
 	return 0;
@@ -611,7 +624,7 @@ static int tipc_getname(struct socket *sock, struct sockaddr *uaddr,
 		addr->addr.id.ref = tsk_peer_port(tsk);
 		addr->addr.id.node = tsk_peer_node(tsk);
 	} else {
-		addr->addr.id.ref = tsk->ref;
+		addr->addr.id.ref = tsk->portid;
 		addr->addr.id.node = tipc_own_addr;
 	}
 
@@ -946,7 +959,7 @@ static int tipc_sendmsg(struct kiocb *iocb, struct socket *sock,
 	}
 
 new_mtu:
-	mtu = tipc_node_get_mtu(dnode, tsk->ref);
+	mtu = tipc_node_get_mtu(dnode, tsk->portid);
 	__skb_queue_head_init(&head);
 	rc = tipc_msg_build(mhdr, m, 0, dsz, mtu, &head);
 	if (rc < 0)
@@ -955,7 +968,7 @@ new_mtu:
 	do {
 		skb = skb_peek(&head);
 		TIPC_SKB_CB(skb)->wakeup_pending = tsk->link_cong;
-		rc = tipc_link_xmit(&head, dnode, tsk->ref);
+		rc = tipc_link_xmit(&head, dnode, tsk->portid);
 		if (likely(rc >= 0)) {
 			if (sock->state != SS_READY)
 				sock->state = SS_CONNECTING;
@@ -1028,7 +1041,7 @@ static int tipc_send_stream(struct kiocb *iocb, struct socket *sock,
 	struct tipc_msg *mhdr = &tsk->phdr;
 	struct sk_buff_head head;
 	DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name);
-	u32 ref = tsk->ref;
+	u32 portid = tsk->portid;
 	int rc = -EINVAL;
 	long timeo;
 	u32 dnode;
@@ -1067,7 +1080,7 @@ next:
 		goto exit;
 	do {
 		if (likely(!tsk_conn_cong(tsk))) {
-			rc = tipc_link_xmit(&head, dnode, ref);
+			rc = tipc_link_xmit(&head, dnode, portid);
 			if (likely(!rc)) {
 				tsk->sent_unacked++;
 				sent += send;
@@ -1076,7 +1089,7 @@ next:
 				goto next;
 			}
 			if (rc == -EMSGSIZE) {
-				tsk->max_pkt = tipc_node_get_mtu(dnode, ref);
+				tsk->max_pkt = tipc_node_get_mtu(dnode, portid);
 				goto next;
 			}
 			if (rc != -ELINKCONG)
@@ -1130,8 +1143,8 @@ static void tipc_sk_finish_conn(struct tipc_sock *tsk, u32 peer_port,
 	tsk->probing_state = TIPC_CONN_OK;
 	tsk->connected = 1;
 	k_start_timer(&tsk->timer, tsk->probing_interval);
-	tipc_node_add_conn(peer_node, tsk->ref, peer_port);
-	tsk->max_pkt = tipc_node_get_mtu(peer_node, tsk->ref);
+	tipc_node_add_conn(peer_node, tsk->portid, peer_port);
+	tsk->max_pkt = tipc_node_get_mtu(peer_node, tsk->portid);
 }
 
 /**
@@ -1238,7 +1251,7 @@ static void tipc_sk_send_ack(struct tipc_sock *tsk, uint ack)
 	if (!tsk->connected)
 		return;
 	skb = tipc_msg_create(CONN_MANAGER, CONN_ACK, INT_H_SIZE, 0, dnode,
-			      tipc_own_addr, peer_port, tsk->ref, TIPC_OK);
+			      tipc_own_addr, peer_port, tsk->portid, TIPC_OK);
 	if (!skb)
 		return;
 	msg = buf_msg(skb);
@@ -1552,7 +1565,7 @@ static int filter_connect(struct tipc_sock *tsk, struct sk_buff **buf)
 				tsk->connected = 0;
 				/* let timer expire on it's own */
 				tipc_node_remove_conn(tsk_peer_node(tsk),
-						      tsk->ref);
+						      tsk->portid);
 			}
 			retval = TIPC_OK;
 		}
@@ -1743,7 +1756,7 @@ int tipc_sk_rcv(struct sk_buff *skb)
 	u32 dnode;
 
 	/* Validate destination and message */
-	tsk = tipc_sk_get(dport);
+	tsk = tipc_sk_lookup(dport);
 	if (unlikely(!tsk)) {
 		rc = tipc_msg_eval(skb, &dnode);
 		goto exit;
@@ -1763,7 +1776,7 @@ int tipc_sk_rcv(struct sk_buff *skb)
 			rc = -TIPC_ERR_OVERLOAD;
 	}
 	spin_unlock_bh(&sk->sk_lock.slock);
-	tipc_sk_put(tsk);
+	sock_put(sk);
 	if (likely(!rc))
 		return 0;
 exit:
@@ -2050,20 +2063,20 @@ restart:
 				goto restart;
 			}
 			if (tipc_msg_reverse(skb, &dnode, TIPC_CONN_SHUTDOWN))
-				tipc_link_xmit_skb(skb, dnode, tsk->ref);
-			tipc_node_remove_conn(dnode, tsk->ref);
+				tipc_link_xmit_skb(skb, dnode, tsk->portid);
+			tipc_node_remove_conn(dnode, tsk->portid);
 		} else {
 			dnode = tsk_peer_node(tsk);
 			skb = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE,
 					      TIPC_CONN_MSG, SHORT_H_SIZE,
 					      0, dnode, tipc_own_addr,
 					      tsk_peer_port(tsk),
-					      tsk->ref, TIPC_CONN_SHUTDOWN);
-			tipc_link_xmit_skb(skb, dnode, tsk->ref);
+					      tsk->portid, TIPC_CONN_SHUTDOWN);
+			tipc_link_xmit_skb(skb, dnode, tsk->portid);
 		}
 		tsk->connected = 0;
 		sock->state = SS_DISCONNECTING;
-		tipc_node_remove_conn(dnode, tsk->ref);
+		tipc_node_remove_conn(dnode, tsk->portid);
 		/* fall through */
 
 	case SS_DISCONNECTING:
@@ -2084,14 +2097,14 @@ restart:
 	return res;
 }
 
-static void tipc_sk_timeout(unsigned long ref)
+static void tipc_sk_timeout(unsigned long portid)
 {
 	struct tipc_sock *tsk;
 	struct sock *sk;
 	struct sk_buff *skb = NULL;
 	u32 peer_port, peer_node;
 
-	tsk = tipc_sk_get(ref);
+	tsk = tipc_sk_lookup(portid);
 	if (!tsk)
 		return;
 
@@ -2108,20 +2121,20 @@ static void tipc_sk_timeout(unsigned long ref)
 		/* Previous probe not answered -> self abort */
 		skb = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE, TIPC_CONN_MSG,
 				      SHORT_H_SIZE, 0, tipc_own_addr,
-				      peer_node, ref, peer_port,
+				      peer_node, portid, peer_port,
 				      TIPC_ERR_NO_PORT);
 	} else {
 		skb = tipc_msg_create(CONN_MANAGER, CONN_PROBE, INT_H_SIZE,
 				      0, peer_node, tipc_own_addr,
-				      peer_port, ref, TIPC_OK);
+				      peer_port, portid, TIPC_OK);
 		tsk->probing_state = TIPC_CONN_PROBING;
 		k_start_timer(&tsk->timer, tsk->probing_interval);
 	}
 	bh_unlock_sock(sk);
 	if (skb)
-		tipc_link_xmit_skb(skb, peer_node, ref);
+		tipc_link_xmit_skb(skb, peer_node, portid);
 exit:
-	tipc_sk_put(tsk);
+	sock_put(sk);
 }
 
 static int tipc_sk_publish(struct tipc_sock *tsk, uint scope,
@@ -2132,12 +2145,12 @@ static int tipc_sk_publish(struct tipc_sock *tsk, uint scope,
 
 	if (tsk->connected)
 		return -EINVAL;
-	key = tsk->ref + tsk->pub_count + 1;
-	if (key == tsk->ref)
+	key = tsk->portid + tsk->pub_count + 1;
+	if (key == tsk->portid)
 		return -EADDRINUSE;
 
 	publ = tipc_nametbl_publish(seq->type, seq->lower, seq->upper,
-				    scope, tsk->ref, key);
+				    scope, tsk->portid, key);
 	if (unlikely(!publ))
 		return -EINVAL;
 
@@ -2188,9 +2201,9 @@ static int tipc_sk_show(struct tipc_sock *tsk, char *buf,
 		ret = tipc_snprintf(buf, len, "<%u.%u.%u:%u>:",
 				    tipc_zone(tipc_own_addr),
 				    tipc_cluster(tipc_own_addr),
-				    tipc_node(tipc_own_addr), tsk->ref);
+				    tipc_node(tipc_own_addr), tsk->portid);
 	else
-		ret = tipc_snprintf(buf, len, "%-10u:", tsk->ref);
+		ret = tipc_snprintf(buf, len, "%-10u:", tsk->portid);
 
 	if (tsk->connected) {
 		u32 dport = tsk_peer_port(tsk);
@@ -2224,13 +2237,15 @@ static int tipc_sk_show(struct tipc_sock *tsk, char *buf,
 
 struct sk_buff *tipc_sk_socks_show(void)
 {
+	const struct bucket_table *tbl;
+	struct rhash_head *pos;
 	struct sk_buff *buf;
 	struct tlv_desc *rep_tlv;
 	char *pb;
 	int pb_len;
 	struct tipc_sock *tsk;
 	int str_len = 0;
-	u32 ref = 0;
+	int i;
 
 	buf = tipc_cfg_reply_alloc(TLV_SPACE(ULTRA_STRING_MAX_LEN));
 	if (!buf)
@@ -2239,14 +2254,18 @@ struct sk_buff *tipc_sk_socks_show(void)
 	pb = TLV_DATA(rep_tlv);
 	pb_len = ULTRA_STRING_MAX_LEN;
 
-	tsk = tipc_sk_get_next(&ref);
-	for (; tsk; tsk = tipc_sk_get_next(&ref)) {
-		lock_sock(&tsk->sk);
-		str_len += tipc_sk_show(tsk, pb + str_len,
-					pb_len - str_len, 0);
-		release_sock(&tsk->sk);
-		tipc_sk_put(tsk);
+	rcu_read_lock();
+	tbl = rht_dereference_rcu((&tipc_sk_rht)->tbl, &tipc_sk_rht);
+	for (i = 0; i < tbl->size; i++) {
+		rht_for_each_entry_rcu(tsk, pos, tbl, i, node) {
+			spin_lock_bh(&tsk->sk.sk_lock.slock);
+			str_len += tipc_sk_show(tsk, pb + str_len,
+						pb_len - str_len, 0);
+			spin_unlock_bh(&tsk->sk.sk_lock.slock);
+		}
 	}
+	rcu_read_unlock();
+
 	str_len += 1;	/* for "\0" */
 	skb_put(buf, TLV_SPACE(str_len));
 	TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len);
@@ -2259,255 +2278,91 @@ struct sk_buff *tipc_sk_socks_show(void)
  */
 void tipc_sk_reinit(void)
 {
+	const struct bucket_table *tbl;
+	struct rhash_head *pos;
+	struct tipc_sock *tsk;
 	struct tipc_msg *msg;
-	u32 ref = 0;
-	struct tipc_sock *tsk = tipc_sk_get_next(&ref);
+	int i;
 
-	for (; tsk; tsk = tipc_sk_get_next(&ref)) {
-		lock_sock(&tsk->sk);
-		msg = &tsk->phdr;
-		msg_set_prevnode(msg, tipc_own_addr);
-		msg_set_orignode(msg, tipc_own_addr);
-		release_sock(&tsk->sk);
-		tipc_sk_put(tsk);
+	rcu_read_lock();
+	tbl = rht_dereference_rcu((&tipc_sk_rht)->tbl, &tipc_sk_rht);
+	for (i = 0; i < tbl->size; i++) {
+		rht_for_each_entry_rcu(tsk, pos, tbl, i, node) {
+			spin_lock_bh(&tsk->sk.sk_lock.slock);
+			msg = &tsk->phdr;
+			msg_set_prevnode(msg, tipc_own_addr);
+			msg_set_orignode(msg, tipc_own_addr);
+			spin_unlock_bh(&tsk->sk.sk_lock.slock);
+		}
 	}
+	rcu_read_unlock();
 }
 
-/**
- * struct reference - TIPC socket reference entry
- * @tsk: pointer to socket associated with reference entry
- * @ref: reference value for socket (combines instance & array index info)
- */
-struct reference {
-	struct tipc_sock *tsk;
-	u32 ref;
-};
-
-/**
- * struct tipc_ref_table - table of TIPC socket reference entries
- * @entries: pointer to array of reference entries
- * @capacity: array index of first unusable entry
- * @init_point: array index of first uninitialized entry
- * @first_free: array index of first unused socket reference entry
- * @last_free: array index of last unused socket reference entry
- * @index_mask: bitmask for array index portion of reference values
- * @start_mask: initial value for instance value portion of reference values
- */
-struct ref_table {
-	struct reference *entries;
-	u32 capacity;
-	u32 init_point;
-	u32 first_free;
-	u32 last_free;
-	u32 index_mask;
-	u32 start_mask;
-};
-
-/* Socket reference table consists of 2**N entries.
- *
- * State	Socket ptr	Reference
- * -----        ----------      ---------
- * In use        non-NULL       XXXX|own index
- *				(XXXX changes each time entry is acquired)
- * Free            NULL         YYYY|next free index
- *				(YYYY is one more than last used XXXX)
- * Uninitialized   NULL         0
- *
- * Entry 0 is not used; this allows index 0 to denote the end of the free list.
- *
- * Note that a reference value of 0 does not necessarily indicate that an
- * entry is uninitialized, since the last entry in the free list could also
- * have a reference value of 0 (although this is unlikely).
- */
-
-static struct ref_table tipc_ref_table;
-
-static DEFINE_RWLOCK(ref_table_lock);
-
-/**
- * tipc_ref_table_init - create reference table for sockets
- */
-int tipc_sk_ref_table_init(u32 req_sz, u32 start)
+static struct tipc_sock *tipc_sk_lookup(u32 portid)
 {
-	struct reference *table;
-	u32 actual_sz;
-
-	/* account for unused entry, then round up size to a power of 2 */
-
-	req_sz++;
-	for (actual_sz = 16; actual_sz < req_sz; actual_sz <<= 1) {
-		/* do nothing */
-	};
-
-	/* allocate table & mark all entries as uninitialized */
-	table = vzalloc(actual_sz * sizeof(struct reference));
-	if (table == NULL)
-		return -ENOMEM;
-
-	tipc_ref_table.entries = table;
-	tipc_ref_table.capacity = req_sz;
-	tipc_ref_table.init_point = 1;
-	tipc_ref_table.first_free = 0;
-	tipc_ref_table.last_free = 0;
-	tipc_ref_table.index_mask = actual_sz - 1;
-	tipc_ref_table.start_mask = start & ~tipc_ref_table.index_mask;
+	struct tipc_sock *tsk;
 
-	return 0;
-}
+	rcu_read_lock();
+	tsk = rhashtable_lookup(&tipc_sk_rht, &portid);
+	if (tsk)
+		sock_hold(&tsk->sk);
+	rcu_read_unlock();
 
-/**
- * tipc_ref_table_stop - destroy reference table for sockets
- */
-void tipc_sk_ref_table_stop(void)
-{
-	if (!tipc_ref_table.entries)
-		return;
-	vfree(tipc_ref_table.entries);
-	tipc_ref_table.entries = NULL;
+	return tsk;
 }
 
-/* tipc_ref_acquire - create reference to a socket
- *
- * Register an socket pointer in the reference table.
- * Returns a unique reference value that is used from then on to retrieve the
- * socket pointer, or to determine if the socket has been deregistered.
- */
-u32 tipc_sk_ref_acquire(struct tipc_sock *tsk)
+static int tipc_sk_insert(struct tipc_sock *tsk)
 {
-	u32 index;
-	u32 index_mask;
-	u32 next_plus_upper;
-	u32 ref = 0;
-	struct reference *entry;
-
-	if (unlikely(!tsk)) {
-		pr_err("Attempt to acquire ref. to non-existent obj\n");
-		return 0;
-	}
-	if (unlikely(!tipc_ref_table.entries)) {
-		pr_err("Ref. table not found in acquisition attempt\n");
-		return 0;
-	}
-
-	/* Take a free entry, if available; otherwise initialize a new one */
-	write_lock_bh(&ref_table_lock);
-	index = tipc_ref_table.first_free;
-	entry = &tipc_ref_table.entries[index];
+	u32 remaining = (TIPC_MAX_PORT - TIPC_MIN_PORT) + 1;
+	u32 portid = prandom_u32() % remaining + TIPC_MIN_PORT;
 
-	if (likely(index)) {
-		index = tipc_ref_table.first_free;
-		entry = &tipc_ref_table.entries[index];
-		index_mask = tipc_ref_table.index_mask;
-		next_plus_upper = entry->ref;
-		tipc_ref_table.first_free = next_plus_upper & index_mask;
-		ref = (next_plus_upper & ~index_mask) + index;
-		entry->tsk = tsk;
-	} else if (tipc_ref_table.init_point < tipc_ref_table.capacity) {
-		index = tipc_ref_table.init_point++;
-		entry = &tipc_ref_table.entries[index];
-		ref = tipc_ref_table.start_mask + index;
+	while (remaining--) {
+		portid++;
+		if ((portid < TIPC_MIN_PORT) || (portid > TIPC_MAX_PORT))
+			portid = TIPC_MIN_PORT;
+		tsk->portid = portid;
+		sock_hold(&tsk->sk);
+		if (rhashtable_lookup_insert(&tipc_sk_rht, &tsk->node))
+			return 0;
+		sock_put(&tsk->sk);
 	}
 
-	if (ref) {
-		entry->ref = ref;
-		entry->tsk = tsk;
-	}
-	write_unlock_bh(&ref_table_lock);
-	return ref;
+	return -1;
 }
 
-/* tipc_sk_ref_discard - invalidate reference to an socket
- *
- * Disallow future references to an socket and free up the entry for re-use.
- */
-void tipc_sk_ref_discard(u32 ref)
+static void tipc_sk_remove(struct tipc_sock *tsk)
 {
-	struct reference *entry;
-	u32 index;
-	u32 index_mask;
-
-	if (unlikely(!tipc_ref_table.entries)) {
-		pr_err("Ref. table not found during discard attempt\n");
-		return;
-	}
-
-	index_mask = tipc_ref_table.index_mask;
-	index = ref & index_mask;
-	entry = &tipc_ref_table.entries[index];
-
-	write_lock_bh(&ref_table_lock);
+	struct sock *sk = &tsk->sk;
 
-	if (unlikely(!entry->tsk)) {
-		pr_err("Attempt to discard ref. to non-existent socket\n");
-		goto exit;
-	}
-	if (unlikely(entry->ref != ref)) {
-		pr_err("Attempt to discard non-existent reference\n");
-		goto exit;
+	if (rhashtable_remove(&tipc_sk_rht, &tsk->node)) {
+		WARN_ON(atomic_read(&sk->sk_refcnt) == 1);
+		__sock_put(sk);
 	}
-
-	/* Mark entry as unused; increment instance part of entry's
-	 *   reference to invalidate any subsequent references
-	 */
-
-	entry->tsk = NULL;
-	entry->ref = (ref & ~index_mask) + (index_mask + 1);
-
-	/* Append entry to free entry list */
-	if (unlikely(tipc_ref_table.first_free == 0))
-		tipc_ref_table.first_free = index;
-	else
-		tipc_ref_table.entries[tipc_ref_table.last_free].ref |= index;
-	tipc_ref_table.last_free = index;
-exit:
-	write_unlock_bh(&ref_table_lock);
 }
 
-/* tipc_sk_get - find referenced socket and return pointer to it
- */
-struct tipc_sock *tipc_sk_get(u32 ref)
+int tipc_sk_rht_init(void)
 {
-	struct reference *entry;
-	struct tipc_sock *tsk;
+	struct rhashtable_params rht_params = {
+		.nelem_hint = 256,
+		.head_offset = offsetof(struct tipc_sock, node),
+		.key_offset = offsetof(struct tipc_sock, portid),
+		.key_len = sizeof(u32), /* portid */
+		.hashfn = jhash,
+		.max_shift = 20, /* 1M */
+		.min_shift = 8,  /* 256 */
+		.grow_decision = rht_grow_above_75,
+		.shrink_decision = rht_shrink_below_30,
+	};
 
-	if (unlikely(!tipc_ref_table.entries))
-		return NULL;
-	read_lock_bh(&ref_table_lock);
-	entry = &tipc_ref_table.entries[ref & tipc_ref_table.index_mask];
-	tsk = entry->tsk;
-	if (likely(tsk && (entry->ref == ref)))
-		sock_hold(&tsk->sk);
-	else
-		tsk = NULL;
-	read_unlock_bh(&ref_table_lock);
-	return tsk;
+	return rhashtable_init(&tipc_sk_rht, &rht_params);
 }
 
-/* tipc_sk_get_next - lock & return next socket after referenced one
-*/
-struct tipc_sock *tipc_sk_get_next(u32 *ref)
+void tipc_sk_rht_destroy(void)
 {
-	struct reference *entry;
-	struct tipc_sock *tsk = NULL;
-	uint index = *ref & tipc_ref_table.index_mask;
+	/* Wait for socket readers to complete */
+	synchronize_net();
 
-	read_lock_bh(&ref_table_lock);
-	while (++index < tipc_ref_table.capacity) {
-		entry = &tipc_ref_table.entries[index];
-		if (!entry->tsk)
-			continue;
-		tsk = entry->tsk;
-		sock_hold(&tsk->sk);
-		*ref = entry->ref;
-		break;
-	}
-	read_unlock_bh(&ref_table_lock);
-	return tsk;
-}
-
-static void tipc_sk_put(struct tipc_sock *tsk)
-{
-	sock_put(&tsk->sk);
+	rhashtable_destroy(&tipc_sk_rht);
 }
 
 /**
@@ -2829,7 +2684,7 @@ static int __tipc_nl_add_sk(struct sk_buff *skb, struct netlink_callback *cb,
 	attrs = nla_nest_start(skb, TIPC_NLA_SOCK);
 	if (!attrs)
 		goto genlmsg_cancel;
-	if (nla_put_u32(skb, TIPC_NLA_SOCK_REF, tsk->ref))
+	if (nla_put_u32(skb, TIPC_NLA_SOCK_REF, tsk->portid))
 		goto attr_msg_cancel;
 	if (nla_put_u32(skb, TIPC_NLA_SOCK_ADDR, tipc_own_addr))
 		goto attr_msg_cancel;
@@ -2859,22 +2714,29 @@ int tipc_nl_sk_dump(struct sk_buff *skb, struct netlink_callback *cb)
 {
 	int err;
 	struct tipc_sock *tsk;
-	u32 prev_ref = cb->args[0];
-	u32 ref = prev_ref;
-
-	tsk = tipc_sk_get_next(&ref);
-	for (; tsk; tsk = tipc_sk_get_next(&ref)) {
-		lock_sock(&tsk->sk);
-		err = __tipc_nl_add_sk(skb, cb, tsk);
-		release_sock(&tsk->sk);
-		tipc_sk_put(tsk);
-		if (err)
-			break;
+	const struct bucket_table *tbl;
+	struct rhash_head *pos;
+	u32 prev_portid = cb->args[0];
+	u32 portid = prev_portid;
+	int i;
 
-		prev_ref = ref;
+	rcu_read_lock();
+	tbl = rht_dereference_rcu((&tipc_sk_rht)->tbl, &tipc_sk_rht);
+	for (i = 0; i < tbl->size; i++) {
+		rht_for_each_entry_rcu(tsk, pos, tbl, i, node) {
+			spin_lock_bh(&tsk->sk.sk_lock.slock);
+			portid = tsk->portid;
+			err = __tipc_nl_add_sk(skb, cb, tsk);
+			spin_unlock_bh(&tsk->sk.sk_lock.slock);
+			if (err)
+				break;
+
+			prev_portid = portid;
+		}
 	}
+	rcu_read_unlock();
 
-	cb->args[0] = prev_ref;
+	cb->args[0] = prev_portid;
 
 	return skb->len;
 }
@@ -2962,12 +2824,12 @@ static int __tipc_nl_list_sk_publ(struct sk_buff *skb,
 int tipc_nl_publ_dump(struct sk_buff *skb, struct netlink_callback *cb)
 {
 	int err;
-	u32 tsk_ref = cb->args[0];
+	u32 tsk_portid = cb->args[0];
 	u32 last_publ = cb->args[1];
 	u32 done = cb->args[2];
 	struct tipc_sock *tsk;
 
-	if (!tsk_ref) {
+	if (!tsk_portid) {
 		struct nlattr **attrs;
 		struct nlattr *sock[TIPC_NLA_SOCK_MAX + 1];
 
@@ -2984,13 +2846,13 @@ int tipc_nl_publ_dump(struct sk_buff *skb, struct netlink_callback *cb)
 		if (!sock[TIPC_NLA_SOCK_REF])
 			return -EINVAL;
 
-		tsk_ref = nla_get_u32(sock[TIPC_NLA_SOCK_REF]);
+		tsk_portid = nla_get_u32(sock[TIPC_NLA_SOCK_REF]);
 	}
 
 	if (done)
 		return 0;
 
-	tsk = tipc_sk_get(tsk_ref);
+	tsk = tipc_sk_lookup(tsk_portid);
 	if (!tsk)
 		return -EINVAL;
 
@@ -2999,9 +2861,9 @@ int tipc_nl_publ_dump(struct sk_buff *skb, struct netlink_callback *cb)
 	if (!err)
 		done = 1;
 	release_sock(&tsk->sk);
-	tipc_sk_put(tsk);
+	sock_put(&tsk->sk);
 
-	cb->args[0] = tsk_ref;
+	cb->args[0] = tsk_portid;
 	cb->args[1] = last_publ;
 	cb->args[2] = done;
 
diff --git a/net/tipc/socket.h b/net/tipc/socket.h
index d340893..c7d46d06 100644
--- a/net/tipc/socket.h
+++ b/net/tipc/socket.h
@@ -46,8 +46,8 @@ int tipc_sk_rcv(struct sk_buff *buf);
 struct sk_buff *tipc_sk_socks_show(void);
 void tipc_sk_mcast_rcv(struct sk_buff *buf);
 void tipc_sk_reinit(void);
-int tipc_sk_ref_table_init(u32 requested_size, u32 start);
-void tipc_sk_ref_table_stop(void);
+int tipc_sk_rht_init(void);
+void tipc_sk_rht_destroy(void);
 int tipc_nl_sk_dump(struct sk_buff *skb, struct netlink_callback *cb);
 int tipc_nl_publ_dump(struct sk_buff *skb, struct netlink_callback *cb);
 
-- 
1.7.9.5


------------------------------------------------------------------------------
Dive into the World of Parallel Programming! The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net

Re: [PATCH net-next 1/5] rhashtable: optimize rhashtable_lookup routine

[Thomas Graf]

On 01/06/15 at 03:23pm, Ying Xue wrote:
> Define an internal compare function and relevant compare argument,
> and then make use of rhashtable_lookup_compare() to lookup key in
> hash table, reducing duplicated code between rhashtable_lookup()
> and rhashtable_lookup_compare().
> 
> Signed-off-by: Ying Xue <ying.xue@windriver.com>
> Cc: Thomas Graf <tgraf@suug.ch>

Acked-by: Thomas Graf <tgraf@suug.ch>

Re: [PATCH net-next 2/5] rhashtable: introduce rhashtable_wakeup_worker helper function

[Thomas Graf]

On 01/06/15 at 03:23pm, Ying Xue wrote:
> Introduce rhashtable_wakeup_worker() helper function to reduce
> duplicated code where to wake up worker.
> 
> Signed-off-by: Ying Xue <ying.xue@windriver.com>
> Cc: Thomas Graf <tgraf@suug.ch>
> ---
>  lib/rhashtable.c |   20 +++++++++++++-------
>  1 file changed, 13 insertions(+), 7 deletions(-)
> 
> diff --git a/lib/rhashtable.c b/lib/rhashtable.c
> index f2fdd7a..6eda22f 100644
> --- a/lib/rhashtable.c
> +++ b/lib/rhashtable.c
> @@ -492,6 +492,17 @@ static void rht_deferred_worker(struct work_struct *work)
>  	mutex_unlock(&ht->mutex);
>  }
>  
> +static void rhashtable_wakeup_worker(struct rhashtable *ht)
> +{
> +	struct bucket_table *tbl = rht_dereference_rcu(ht->tbl, ht);
> +	struct bucket_table *new_tbl = rht_dereference_rcu(ht->future_tbl, ht);
> +
> +	if (ht->tbl != ht->future_tbl &&

I just noticed that this is a bug in the original code. It should be

	if (ht->tbl == ht->future_tbl &&

Do you want to fix this in this series?


> +	    ((ht->p.grow_decision && ht->p.grow_decision(ht, new_tbl->size)) ||
> +	    (ht->p.shrink_decision && ht->p.shrink_decision(ht, tbl->size))))

Indent to inner brakcet

> +		schedule_delayed_work(&ht->run_work, 0);
> +}
> +
>  /**
>   * rhashtable_insert - insert object into hash hash table
>   * @ht:		hash table
> @@ -533,9 +544,7 @@ void rhashtable_insert(struct rhashtable *ht, struct rhash_head *obj)
>  	atomic_inc(&ht->nelems);
>  
>  	/* Only grow the table if no resizing is currently in progress. */

This comment should be moved to the function as well.

Re: [PATCH net-next 3/5] rhashtable: use future table size to make expansion decision

[Thomas Graf]

On 01/06/15 at 03:23pm, Ying Xue wrote:
> Should use future table size instead of old table size to decide
> whether hash table is worth being expanded.
> 
> Signed-off-by: Ying Xue <ying.xue@windriver.com>
> Cc: Thomas Graf <tgraf@suug.ch>
> ---
>  lib/rhashtable.c |    5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)

Apologies as you were probably mislead by the bug as commented
on in the previous patch.

I don't think we need this. future_tbl only points to a different
table until the old table entries are linked from the new table.
The condition in the resize check where meant to exclude this
phase so we would newer get to the deferred worker while relinking
is happening.

Re: [PATCH net-next 4/5] rhashtable: involve rhashtable_lookup_insert routine

[Thomas Graf]

On 01/06/15 at 03:23pm, Ying Xue wrote:
> Involve a new function called rhashtable_lookup_insert() which makes
> lookup and insertion atomic under bucket lock protection, helping us
> avoid to introduce an extra lock when we search and insert an object
> into hash table.
> 
> Signed-off-by: Ying Xue <ying.xue@windriver.com>
> Signed-off-by: Thomas Graf <tgraf@suug.ch>

LGTM now, thanks!
Acked-by: Thomas Graf <tgraf@suug.ch>

Re: [PATCH net-next 5/5] tipc: convert tipc reference table to use generic rhashtable

[Thomas Graf]

On 01/06/15 at 03:23pm, Ying Xue wrote:
> As tipc reference table is statically allocated, its memory size
> requested on stack initialization stage is quite big even if the
> maximum port number is just restricted to 8191 currently, however,
> the number already becomes insufficient in practice. But if the
> maximum ports is allowed to its theory value - 2^32, its consumed
> memory size will reach a ridiculously unacceptable value. Apart from
> this, heavy tipc users spend a considerable amount of time in
> tipc_sk_get() due to the read-lock on ref_table_lock.
> 
> If tipc reference table is converted with generic rhashtable, above
> mentioned both disadvantages would be resolved respectively: making
> use of the new resizable hash table can avoid locking on the lookup;
> smaller memory size is required at initial stage, for example, 256
> hash bucket slots are requested at the beginning phase instead of
> allocating the entire 8191 slots in old mode. The hash table will
> grow if entries exceeds 75% of table size up to a total table size
> of 1M, and it will automatically shrink if usage falls below 30%,
> but the minimum table size is allowed down to 256.
> 
> Also converts ref_table_lock to a separate mutex to protect hash table
> mutations on write side. Lastly defers the release of the socket
> reference using call_rcu() to allow using an RCU read-side protected
> call to rhashtable_lookup().
> 
> Signed-off-by: Ying Xue <ying.xue@windriver.com>
> Acked-by: Jon Maloy <jon.maloy@ericsson.com>
> Acked-by: Erik Hugne <erik.hugne@ericsson.com>
> Cc: Thomas Graf <tgraf@suug.ch>

The rhashtable usage looks good to me. I assume the TIPC maintainers
have reviewed the other pieces.

Acked-by: Thomas Graf <tgraf@suug.ch>

Re: [PATCH net-next 3/5] rhashtable: use future table size to make expansion decision

[Ying Xue]

On 01/06/2015 05:35 PM, Thomas Graf wrote:
> On 01/06/15 at 03:23pm, Ying Xue wrote:
>> Should use future table size instead of old table size to decide
>> whether hash table is worth being expanded.
>>
>> Signed-off-by: Ying Xue <ying.xue@windriver.com>
>> Cc: Thomas Graf <tgraf@suug.ch>
>> ---
>>  lib/rhashtable.c |    5 +++--
>>  1 file changed, 3 insertions(+), 2 deletions(-)
> 
> Apologies as you were probably mislead by the bug as commented
> on in the previous patch.
> 
> I don't think we need this. future_tbl only points to a different
> table until the old table entries are linked from the new table.
> The condition in the resize check where meant to exclude this
> phase so we would newer get to the deferred worker while relinking
> is happening.
> 
> 

Thank you for above nice explanation. Regarding my understanding, as
rhashtable_expand() and rhashtable_shrink() are always under the
protection of "ht->mutex", the "future_tbl" and "tbl" absolutely point
to the same bucket array once rhashtable_expand() or rhashtable_shrink()
returns. Therefore, if rht_deferred_worker() takes the "ht->mutex" lock,
the both "future_tbl" and "tbl" should point to the same bucket array.
So the change made in the patch is useless for us, right?

But as you pointed in above patch, there is a bug in
rhashtable_wakeup_worker(). As long as ht->tbl == ht->future_tbl, we
should wake up the work. OK, I will drop the patch and fix the error in
patch #2.

In all, thank you for quickly reviewing the series, and I would deliver
the next version soon in which your all comments would be resolved.

Please wait a moment.

Regards,
Ying



------------------------------------------------------------------------------
Dive into the World of Parallel Programming! The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net

Re: [PATCH net-next 3/5] rhashtable: use future table size to make expansion decision

[Thomas Graf]

On 01/06/15 at 05:56pm, Ying Xue wrote:
> Thank you for above nice explanation. Regarding my understanding, as
> rhashtable_expand() and rhashtable_shrink() are always under the
> protection of "ht->mutex", the "future_tbl" and "tbl" absolutely point
> to the same bucket array once rhashtable_expand() or rhashtable_shrink()
> returns. Therefore, if rht_deferred_worker() takes the "ht->mutex" lock,
> the both "future_tbl" and "tbl" should point to the same bucket array.
> So the change made in the patch is useless for us, right?

Correct.

> But as you pointed in above patch, there is a bug in
> rhashtable_wakeup_worker(). As long as ht->tbl == ht->future_tbl, we
> should wake up the work. OK, I will drop the patch and fix the error in
> patch #2.

Awesome, thanks!