1 files changed, 200 insertions, 116 deletions

diff --git a/drivers/staging/lustre/lustre/ldlm/ldlm_lock.c b/drivers/staging/lustre/lustre/ldlm/ldlm_lock.c
index 3c48b4fb96f1..a4a291acb659 100644
--- a/drivers/staging/lustre/lustre/ldlm/ldlm_lock.c
+++ b/drivers/staging/lustre/lustre/ldlm/ldlm_lock.c
@@ -39,6 +39,7 @@
 
 #include "../../include/linux/libcfs/libcfs.h"
 #include "../include/lustre_intent.h"
+#include "../include/lustre_swab.h"
 #include "../include/obd_class.h"
 #include "ldlm_internal.h"
 
@@ -63,17 +64,10 @@ static char *ldlm_typename[] = {
 	[LDLM_IBITS]	= "IBT",
 };
 
-static ldlm_policy_wire_to_local_t ldlm_policy_wire18_to_local[] = {
+static ldlm_policy_wire_to_local_t ldlm_policy_wire_to_local[] = {
 	[LDLM_PLAIN - LDLM_MIN_TYPE]	= ldlm_plain_policy_wire_to_local,
 	[LDLM_EXTENT - LDLM_MIN_TYPE]	= ldlm_extent_policy_wire_to_local,
-	[LDLM_FLOCK - LDLM_MIN_TYPE]	= ldlm_flock_policy_wire18_to_local,
-	[LDLM_IBITS - LDLM_MIN_TYPE]	= ldlm_ibits_policy_wire_to_local,
-};
-
-static ldlm_policy_wire_to_local_t ldlm_policy_wire21_to_local[] = {
-	[LDLM_PLAIN - LDLM_MIN_TYPE]	= ldlm_plain_policy_wire_to_local,
-	[LDLM_EXTENT - LDLM_MIN_TYPE]	= ldlm_extent_policy_wire_to_local,
-	[LDLM_FLOCK - LDLM_MIN_TYPE]	= ldlm_flock_policy_wire21_to_local,
+	[LDLM_FLOCK - LDLM_MIN_TYPE]	= ldlm_flock_policy_wire_to_local,
 	[LDLM_IBITS - LDLM_MIN_TYPE]	= ldlm_ibits_policy_wire_to_local,
 };
 
@@ -88,8 +82,8 @@ static ldlm_policy_local_to_wire_t ldlm_policy_local_to_wire[] = {
  * Converts lock policy from local format to on the wire lock_desc format
  */
 static void ldlm_convert_policy_to_wire(enum ldlm_type type,
-					const ldlm_policy_data_t *lpolicy,
-					ldlm_wire_policy_data_t *wpolicy)
+					const union ldlm_policy_data *lpolicy,
+					union ldlm_wire_policy_data *wpolicy)
 {
 	ldlm_policy_local_to_wire_t convert;
 
@@ -102,23 +96,17 @@ static void ldlm_convert_policy_to_wire(enum ldlm_type type,
  * Converts lock policy from on the wire lock_desc format to local format
  */
 void ldlm_convert_policy_to_local(struct obd_export *exp, enum ldlm_type type,
-				  const ldlm_wire_policy_data_t *wpolicy,
-				  ldlm_policy_data_t *lpolicy)
+				  const union ldlm_wire_policy_data *wpolicy,
+				  union ldlm_policy_data *lpolicy)
 {
 	ldlm_policy_wire_to_local_t convert;
-	int new_client;
 
-	/** some badness for 2.0.0 clients, but 2.0.0 isn't supported */
-	new_client = (exp_connect_flags(exp) & OBD_CONNECT_FULL20) != 0;
-	if (new_client)
-		convert = ldlm_policy_wire21_to_local[type - LDLM_MIN_TYPE];
-	else
-		convert = ldlm_policy_wire18_to_local[type - LDLM_MIN_TYPE];
+	convert = ldlm_policy_wire_to_local[type - LDLM_MIN_TYPE];
 
 	convert(wpolicy, lpolicy);
 }
 
-char *ldlm_it2str(int it)
+const char *ldlm_it2str(enum ldlm_intent_flags it)
 {
 	switch (it) {
 	case IT_OPEN:
@@ -140,7 +128,7 @@ char *ldlm_it2str(int it)
 	case IT_LAYOUT:
 		return "layout";
 	default:
-		CERROR("Unknown intent %d\n", it);
+		CERROR("Unknown intent 0x%08x\n", it);
 		return "UNKNOWN";
 	}
 }
@@ -512,7 +500,6 @@ int ldlm_lock_change_resource(struct ldlm_namespace *ns, struct ldlm_lock *lock,
 
 	return 0;
 }
-EXPORT_SYMBOL(ldlm_lock_change_resource);
 
 /** \defgroup ldlm_handles LDLM HANDLES
  * Ways to get hold of locks without any addresses.
@@ -595,7 +582,6 @@ void ldlm_lock2desc(struct ldlm_lock *lock, struct ldlm_lock_desc *desc)
 				    &lock->l_policy_data,
 				    &desc->l_policy_data);
 }
-EXPORT_SYMBOL(ldlm_lock2desc);
 
 /**
  * Add a lock to list of conflicting locks to send AST to.
@@ -658,7 +644,7 @@ static void ldlm_add_ast_work_item(struct ldlm_lock *lock,
  * r/w reference type is determined by \a mode
  * Calls ldlm_lock_addref_internal.
  */
-void ldlm_lock_addref(const struct lustre_handle *lockh, __u32 mode)
+void ldlm_lock_addref(const struct lustre_handle *lockh, enum ldlm_mode mode)
 {
 	struct ldlm_lock *lock;
 
@@ -676,7 +662,8 @@ EXPORT_SYMBOL(ldlm_lock_addref);
  * Removes lock from LRU if it is there.
  * Assumes the LDLM lock is already locked.
  */
-void ldlm_lock_addref_internal_nolock(struct ldlm_lock *lock, __u32 mode)
+void ldlm_lock_addref_internal_nolock(struct ldlm_lock *lock,
+				      enum ldlm_mode mode)
 {
 	ldlm_lock_remove_from_lru(lock);
 	if (mode & (LCK_NL | LCK_CR | LCK_PR)) {
@@ -700,7 +687,7 @@ void ldlm_lock_addref_internal_nolock(struct ldlm_lock *lock, __u32 mode)
  *
  * \retval -EAGAIN lock is being canceled.
  */
-int ldlm_lock_addref_try(const struct lustre_handle *lockh, __u32 mode)
+int ldlm_lock_addref_try(const struct lustre_handle *lockh, enum ldlm_mode mode)
 {
 	struct ldlm_lock *lock;
 	int	       result;
@@ -726,7 +713,7 @@ EXPORT_SYMBOL(ldlm_lock_addref_try);
  * Locks LDLM lock and calls ldlm_lock_addref_internal_nolock to do the work.
  * Only called for local locks.
  */
-void ldlm_lock_addref_internal(struct ldlm_lock *lock, __u32 mode)
+void ldlm_lock_addref_internal(struct ldlm_lock *lock, enum ldlm_mode mode)
 {
 	lock_res_and_lock(lock);
 	ldlm_lock_addref_internal_nolock(lock, mode);
@@ -740,7 +727,8 @@ void ldlm_lock_addref_internal(struct ldlm_lock *lock, __u32 mode)
  * Does NOT add lock to LRU if no r/w references left to accommodate flock locks
  * that cannot be placed in LRU.
  */
-void ldlm_lock_decref_internal_nolock(struct ldlm_lock *lock, __u32 mode)
+void ldlm_lock_decref_internal_nolock(struct ldlm_lock *lock,
+				      enum ldlm_mode mode)
 {
 	LDLM_DEBUG(lock, "ldlm_lock_decref(%s)", ldlm_lockname[mode]);
 	if (mode & (LCK_NL | LCK_CR | LCK_PR)) {
@@ -766,7 +754,7 @@ void ldlm_lock_decref_internal_nolock(struct ldlm_lock *lock, __u32 mode)
  * on the namespace.
  * For blocked LDLM locks if r/w count drops to zero, blocking_ast is called.
  */
-void ldlm_lock_decref_internal(struct ldlm_lock *lock, __u32 mode)
+void ldlm_lock_decref_internal(struct ldlm_lock *lock, enum ldlm_mode mode)
 {
 	struct ldlm_namespace *ns;
 
@@ -786,11 +774,16 @@ void ldlm_lock_decref_internal(struct ldlm_lock *lock, __u32 mode)
 	}
 
 	if (!lock->l_readers && !lock->l_writers &&
-	    ldlm_is_cbpending(lock)) {
+	    (ldlm_is_cbpending(lock) || lock->l_req_mode == LCK_GROUP)) {
 		/* If we received a blocked AST and this was the last reference,
 		 * run the callback.
+		 * Group locks are special:
+		 * They must not go in LRU, but they are not called back
+		 * like non-group locks, instead they are manually released.
+		 * They have an l_writers reference which they keep until
+		 * they are manually released, so we remove them when they have
+		 * no more reader or writer references. - LU-6368
 		 */
-
 		LDLM_DEBUG(lock, "final decref done on cbpending lock");
 
 		LDLM_LOCK_GET(lock); /* dropped by bl thread */
@@ -832,7 +825,7 @@ void ldlm_lock_decref_internal(struct ldlm_lock *lock, __u32 mode)
 /**
  * Decrease reader/writer refcount for LDLM lock with handle \a lockh
  */
-void ldlm_lock_decref(const struct lustre_handle *lockh, __u32 mode)
+void ldlm_lock_decref(const struct lustre_handle *lockh, enum ldlm_mode mode)
 {
 	struct ldlm_lock *lock = __ldlm_handle2lock(lockh, 0);
 
@@ -846,10 +839,9 @@ EXPORT_SYMBOL(ldlm_lock_decref);
  * Decrease reader/writer refcount for LDLM lock with handle
  * \a lockh and mark it for subsequent cancellation once r/w refcount
  * drops to zero instead of putting into LRU.
- *
- * Typical usage is for GROUP locks which we cannot allow to be cached.
  */
-void ldlm_lock_decref_and_cancel(const struct lustre_handle *lockh, __u32 mode)
+void ldlm_lock_decref_and_cancel(const struct lustre_handle *lockh,
+				 enum ldlm_mode mode)
 {
 	struct ldlm_lock *lock = __ldlm_handle2lock(lockh, 0);
 
@@ -1055,88 +1047,173 @@ void ldlm_grant_lock(struct ldlm_lock *lock, struct list_head *work_list)
 }
 
 /**
- * Search for a lock with given properties in a queue.
+ * Describe the overlap between two locks.  itree_overlap_cb data.
+ */
+struct lock_match_data {
+	struct ldlm_lock	*lmd_old;
+	struct ldlm_lock	*lmd_lock;
+	enum ldlm_mode		*lmd_mode;
+	union ldlm_policy_data	*lmd_policy;
+	__u64			 lmd_flags;
+	int			 lmd_unref;
+};
+
+/**
+ * Check if the given @lock meets the criteria for a match.
+ * A reference on the lock is taken if matched.
  *
- * \retval a referenced lock or NULL.  See the flag descriptions below, in the
- * comment above ldlm_lock_match
+ * \param lock	test-against this lock
+ * \param data	parameters
  */
-static struct ldlm_lock *search_queue(struct list_head *queue,
-				      enum ldlm_mode *mode,
-				      ldlm_policy_data_t *policy,
-				      struct ldlm_lock *old_lock,
-				      __u64 flags, int unref)
+static int lock_matches(struct ldlm_lock *lock, struct lock_match_data *data)
 {
-	struct ldlm_lock *lock;
-	struct list_head       *tmp;
+	union ldlm_policy_data *lpol = &lock->l_policy_data;
+	enum ldlm_mode match;
 
-	list_for_each(tmp, queue) {
-		enum ldlm_mode match;
+	if (lock == data->lmd_old)
+		return INTERVAL_ITER_STOP;
 
-		lock = list_entry(tmp, struct ldlm_lock, l_res_link);
-
-		if (lock == old_lock)
-			break;
+	/*
+	 * Check if this lock can be matched.
+	 * Used by LU-2919(exclusive open) for open lease lock
+	 */
+	if (ldlm_is_excl(lock))
+		return INTERVAL_ITER_CONT;
 
-		/* Check if this lock can be matched.
-		 * Used by LU-2919(exclusive open) for open lease lock
-		 */
-		if (ldlm_is_excl(lock))
-			continue;
+	/*
+	 * llite sometimes wants to match locks that will be
+	 * canceled when their users drop, but we allow it to match
+	 * if it passes in CBPENDING and the lock still has users.
+	 * this is generally only going to be used by children
+	 * whose parents already hold a lock so forward progress
+	 * can still happen.
+	 */
+	if (ldlm_is_cbpending(lock) &&
+	    !(data->lmd_flags & LDLM_FL_CBPENDING))
+		return INTERVAL_ITER_CONT;
 
-		/* llite sometimes wants to match locks that will be
-		 * canceled when their users drop, but we allow it to match
-		 * if it passes in CBPENDING and the lock still has users.
-		 * this is generally only going to be used by children
-		 * whose parents already hold a lock so forward progress
-		 * can still happen.
-		 */
-		if (ldlm_is_cbpending(lock) && !(flags & LDLM_FL_CBPENDING))
-			continue;
-		if (!unref && ldlm_is_cbpending(lock) &&
-		    lock->l_readers == 0 && lock->l_writers == 0)
-			continue;
+	if (!data->lmd_unref && ldlm_is_cbpending(lock) &&
+	    !lock->l_readers && !lock->l_writers)
+		return INTERVAL_ITER_CONT;
 
-		if (!(lock->l_req_mode & *mode))
-			continue;
-		match = lock->l_req_mode;
+	if (!(lock->l_req_mode & *data->lmd_mode))
+		return INTERVAL_ITER_CONT;
+	match = lock->l_req_mode;
 
-		if (lock->l_resource->lr_type == LDLM_EXTENT &&
-		    (lock->l_policy_data.l_extent.start >
-		     policy->l_extent.start ||
-		     lock->l_policy_data.l_extent.end < policy->l_extent.end))
-			continue;
+	switch (lock->l_resource->lr_type) {
+	case LDLM_EXTENT:
+		if (lpol->l_extent.start > data->lmd_policy->l_extent.start ||
+		    lpol->l_extent.end < data->lmd_policy->l_extent.end)
+			return INTERVAL_ITER_CONT;
 
 		if (unlikely(match == LCK_GROUP) &&
-		    lock->l_resource->lr_type == LDLM_EXTENT &&
-		    policy->l_extent.gid != LDLM_GID_ANY &&
-		    lock->l_policy_data.l_extent.gid != policy->l_extent.gid)
-			continue;
-
-		/* We match if we have existing lock with same or wider set
+		    data->lmd_policy->l_extent.gid != LDLM_GID_ANY &&
+		    lpol->l_extent.gid != data->lmd_policy->l_extent.gid)
+			return INTERVAL_ITER_CONT;
+		break;
+	case LDLM_IBITS:
+		/*
+		 * We match if we have existing lock with same or wider set
 		 * of bits.
 		 */
-		if (lock->l_resource->lr_type == LDLM_IBITS &&
-		    ((lock->l_policy_data.l_inodebits.bits &
-		      policy->l_inodebits.bits) !=
-		      policy->l_inodebits.bits))
-			continue;
+		if ((lpol->l_inodebits.bits &
+		     data->lmd_policy->l_inodebits.bits) !=
+		    data->lmd_policy->l_inodebits.bits)
+			return INTERVAL_ITER_CONT;
+		break;
+	default:
+		break;
+	}
+	/*
+	 * We match if we have existing lock with same or wider set
+	 * of bits.
+	 */
+	if (!data->lmd_unref && LDLM_HAVE_MASK(lock, GONE))
+		return INTERVAL_ITER_CONT;
+
+	if ((data->lmd_flags & LDLM_FL_LOCAL_ONLY) &&
+	    !ldlm_is_local(lock))
+		return INTERVAL_ITER_CONT;
+
+	if (data->lmd_flags & LDLM_FL_TEST_LOCK) {
+		LDLM_LOCK_GET(lock);
+		ldlm_lock_touch_in_lru(lock);
+	} else {
+		ldlm_lock_addref_internal_nolock(lock, match);
+	}
+
+	*data->lmd_mode = match;
+	data->lmd_lock = lock;
+
+	return INTERVAL_ITER_STOP;
+}
+
+static unsigned int itree_overlap_cb(struct interval_node *in, void *args)
+{
+	struct ldlm_interval *node = to_ldlm_interval(in);
+	struct lock_match_data *data = args;
+	struct ldlm_lock *lock;
+	int rc;
+
+	list_for_each_entry(lock, &node->li_group, l_sl_policy) {
+		rc = lock_matches(lock, data);
+		if (rc == INTERVAL_ITER_STOP)
+			return INTERVAL_ITER_STOP;
+	}
+	return INTERVAL_ITER_CONT;
+}
+
+/**
+ * Search for a lock with given parameters in interval trees.
+ *
+ * \param res	search for a lock in this resource
+ * \param data	parameters
+ *
+ * \retval	a referenced lock or NULL.
+ */
+static struct ldlm_lock *search_itree(struct ldlm_resource *res,
+				      struct lock_match_data *data)
+{
+	struct interval_node_extent ext = {
+		.start	= data->lmd_policy->l_extent.start,
+		.end	= data->lmd_policy->l_extent.end
+	};
+	int idx;
+
+	for (idx = 0; idx < LCK_MODE_NUM; idx++) {
+		struct ldlm_interval_tree *tree = &res->lr_itree[idx];
 
-		if (!unref && LDLM_HAVE_MASK(lock, GONE))
+		if (!tree->lit_root)
 			continue;
 
-		if ((flags & LDLM_FL_LOCAL_ONLY) && !ldlm_is_local(lock))
+		if (!(tree->lit_mode & *data->lmd_mode))
 			continue;
 
-		if (flags & LDLM_FL_TEST_LOCK) {
-			LDLM_LOCK_GET(lock);
-			ldlm_lock_touch_in_lru(lock);
-		} else {
-			ldlm_lock_addref_internal_nolock(lock, match);
-		}
-		*mode = match;
-		return lock;
+		interval_search(tree->lit_root, &ext,
+				itree_overlap_cb, data);
 	}
+	return data->lmd_lock;
+}
 
+/**
+ * Search for a lock with given properties in a queue.
+ *
+ * \param queue	search for a lock in this queue
+ * \param data	parameters
+ *
+ * \retval	a referenced lock or NULL.
+ */
+static struct ldlm_lock *search_queue(struct list_head *queue,
+				      struct lock_match_data *data)
+{
+	struct ldlm_lock *lock;
+	int rc;
+
+	list_for_each_entry(lock, queue, l_res_link) {
+		rc = lock_matches(lock, data);
+		if (rc == INTERVAL_ITER_STOP)
+			return data->lmd_lock;
+	}
 	return NULL;
 }
 
@@ -1147,7 +1224,6 @@ void ldlm_lock_fail_match_locked(struct ldlm_lock *lock)
 		wake_up_all(&lock->l_waitq);
 	}
 }
-EXPORT_SYMBOL(ldlm_lock_fail_match_locked);
 
 /**
  * Mark lock as "matchable" by OST.
@@ -1208,35 +1284,45 @@ EXPORT_SYMBOL(ldlm_lock_allow_match);
 enum ldlm_mode ldlm_lock_match(struct ldlm_namespace *ns, __u64 flags,
 			       const struct ldlm_res_id *res_id,
 			       enum ldlm_type type,
-			       ldlm_policy_data_t *policy,
+			       union ldlm_policy_data *policy,
 			       enum ldlm_mode mode,
 			       struct lustre_handle *lockh, int unref)
 {
+	struct lock_match_data data = {
+		.lmd_old	= NULL,
+		.lmd_lock	= NULL,
+		.lmd_mode	= &mode,
+		.lmd_policy	= policy,
+		.lmd_flags	= flags,
+		.lmd_unref	= unref,
+	};
 	struct ldlm_resource *res;
-	struct ldlm_lock *lock, *old_lock = NULL;
+	struct ldlm_lock *lock;
 	int rc = 0;
 
 	if (!ns) {
-		old_lock = ldlm_handle2lock(lockh);
-		LASSERT(old_lock);
+		data.lmd_old = ldlm_handle2lock(lockh);
+		LASSERT(data.lmd_old);
 
-		ns = ldlm_lock_to_ns(old_lock);
-		res_id = &old_lock->l_resource->lr_name;
-		type = old_lock->l_resource->lr_type;
-		mode = old_lock->l_req_mode;
+		ns = ldlm_lock_to_ns(data.lmd_old);
+		res_id = &data.lmd_old->l_resource->lr_name;
+		type = data.lmd_old->l_resource->lr_type;
+		*data.lmd_mode = data.lmd_old->l_req_mode;
 	}
 
 	res = ldlm_resource_get(ns, NULL, res_id, type, 0);
 	if (IS_ERR(res)) {
-		LASSERT(!old_lock);
+		LASSERT(!data.lmd_old);
 		return 0;
 	}
 
 	LDLM_RESOURCE_ADDREF(res);
 	lock_res(res);
 
-	lock = search_queue(&res->lr_granted, &mode, policy, old_lock,
-			    flags, unref);
+	if (res->lr_type == LDLM_EXTENT)
+		lock = search_itree(res, &data);
+	else
+		lock = search_queue(&res->lr_granted, &data);
 	if (lock) {
 		rc = 1;
 		goto out;
@@ -1245,14 +1331,12 @@ enum ldlm_mode ldlm_lock_match(struct ldlm_namespace *ns, __u64 flags,
 		rc = 0;
 		goto out;
 	}
-	lock = search_queue(&res->lr_waiting, &mode, policy, old_lock,
-			    flags, unref);
+	lock = search_queue(&res->lr_waiting, &data);
 	if (lock) {
 		rc = 1;
 		goto out;
 	}
-
- out:
+out:
 	unlock_res(res);
 	LDLM_RESOURCE_DELREF(res);
 	ldlm_resource_putref(res);
@@ -1324,8 +1408,8 @@ enum ldlm_mode ldlm_lock_match(struct ldlm_namespace *ns, __u64 flags,
 				  (type == LDLM_PLAIN || type == LDLM_IBITS) ?
 					res_id->name[3] : policy->l_extent.end);
 	}
-	if (old_lock)
-		LDLM_LOCK_PUT(old_lock);
+	if (data.lmd_old)
+		LDLM_LOCK_PUT(data.lmd_old);
 
 	return rc ? mode : 0;
 }