@@ -203,7 +203,9 @@ static int rxe_init(struct rxe_dev *rxe)
spin_lock_init(&rxe->pending_lock);
INIT_LIST_HEAD(&rxe->pending_mmaps);
+ /* init multicast support */
spin_lock_init(&rxe->mcg_lock);
+ rxe->mcg_tree = RB_ROOT;
mutex_init(&rxe->usdev_lock);
@@ -40,9 +40,10 @@ void rxe_cq_disable(struct rxe_cq *cq);
void rxe_cq_cleanup(struct rxe_pool_elem *arg);
/* rxe_mcast.c */
-void rxe_mc_cleanup(struct rxe_pool_elem *arg);
+struct rxe_mcg *rxe_lookup_mcg(struct rxe_dev *rxe, union ib_gid *mgid);
int rxe_attach_mcast(struct ib_qp *ibqp, union ib_gid *mgid, u16 mlid);
int rxe_detach_mcast(struct ib_qp *ibqp, union ib_gid *mgid, u16 mlid);
+void rxe_mc_cleanup(struct rxe_pool_elem *elem);
/* rxe_mmap.c */
struct rxe_mmap_info {
@@ -25,56 +25,225 @@ static int rxe_mcast_delete(struct rxe_dev *rxe, union ib_gid *mgid)
return dev_mc_del(rxe->ndev, ll_addr);
}
-/* caller should hold rxe->mcg_lock */
-static struct rxe_mcg *__rxe_create_mcg(struct rxe_dev *rxe,
- struct rxe_pool *pool,
+/**
+ * __rxe_insert_mcg - insert an mcg into red-black tree (rxe->mcg_tree)
+ * @mcg: mcg object with an embedded red-black tree node
+ *
+ * Context: caller must hold a reference to mcg and rxe->mcg_lock and
+ * is responsible to avoid adding the same mcg twice to the tree.
+ */
+static void __rxe_insert_mcg(struct rxe_mcg *mcg)
+{
+ struct rb_root *tree = &mcg->rxe->mcg_tree;
+ struct rb_node **link = &tree->rb_node;
+ struct rb_node *node = NULL;
+ struct rxe_mcg *tmp;
+ int cmp;
+
+ while (*link) {
+ node = *link;
+ tmp = rb_entry(node, struct rxe_mcg, node);
+
+ cmp = memcmp(&tmp->mgid, &mcg->mgid, sizeof(mcg->mgid));
+ if (cmp > 0)
+ link = &(*link)->rb_left;
+ else
+ link = &(*link)->rb_right;
+ }
+
+ rb_link_node(&mcg->node, node, link);
+ rb_insert_color(&mcg->node, tree);
+}
+
+/**
+ * __rxe_remove_mcg - remove an mcg from red-black tree holding lock
+ * @mcg: mcast group object with an embedded red-black tree node
+ *
+ * Context: caller must hold a reference to mcg and rxe->mcg_lock
+ */
+static void __rxe_remove_mcg(struct rxe_mcg *mcg)
+{
+ rb_erase(&mcg->node, &mcg->rxe->mcg_tree);
+}
+
+/**
+ * __rxe_lookup_mcg - lookup mcg in rxe->mcg_tree while holding lock
+ * @rxe: rxe device object
+ * @mgid: multicast IP address
+ *
+ * Context: caller must hold rxe->mcg_lock
+ * Returns: mcg on success and takes a ref to mcg else NULL
+ */
+static struct rxe_mcg *__rxe_lookup_mcg(struct rxe_dev *rxe,
union ib_gid *mgid)
{
+ struct rb_root *tree = &rxe->mcg_tree;
struct rxe_mcg *mcg;
- int err;
+ struct rb_node *node;
+ int cmp;
- mcg = rxe_alloc_locked(pool);
- if (!mcg)
- return ERR_PTR(-ENOMEM);
+ node = tree->rb_node;
- err = rxe_mcast_add(rxe, mgid);
- if (unlikely(err)) {
- rxe_drop_ref(mcg);
- return ERR_PTR(err);
+ while (node) {
+ mcg = rb_entry(node, struct rxe_mcg, node);
+
+ cmp = memcmp(&mcg->mgid, mgid, sizeof(*mgid));
+
+ if (cmp > 0)
+ node = node->rb_left;
+ else if (cmp < 0)
+ node = node->rb_right;
+ else
+ break;
}
+ if (node) {
+ rxe_add_ref(mcg);
+ return mcg;
+ }
+
+ return NULL;
+}
+
+/**
+ * rxe_lookup_mcg - lookup up mcg in red-back tree
+ * @rxe: rxe device object
+ * @mgid: multicast IP address
+ *
+ * Returns: mcg if found else NULL
+ */
+struct rxe_mcg *rxe_lookup_mcg(struct rxe_dev *rxe, union ib_gid *mgid)
+{
+ struct rxe_mcg *mcg;
+
+ spin_lock_bh(&rxe->mcg_lock);
+ mcg = __rxe_lookup_mcg(rxe, mgid);
+ spin_unlock_bh(&rxe->mcg_lock);
+
+ return mcg;
+}
+
+/**
+ * __rxe_init_mcg - initialize a new mcg
+ * @rxe: rxe device
+ * @mgid: multicast address as a gid
+ * @mcg: new mcg object
+ *
+ * Context: caller should hold rxe->mcg lock
+ * Returns: 0 on success else an error
+ */
+static int __rxe_init_mcg(struct rxe_dev *rxe, union ib_gid *mgid,
+ struct rxe_mcg *mcg)
+{
+ int err;
+
+ err = rxe_mcast_add(rxe, mgid);
+ if (unlikely(err))
+ return err;
+
+ memcpy(&mcg->mgid, mgid, sizeof(mcg->mgid));
INIT_LIST_HEAD(&mcg->qp_list);
mcg->rxe = rxe;
- /* rxe_alloc_locked takes a ref on mcg but that will be
+ /* caller holds a ref on mcg but that will be
* dropped when mcg goes out of scope. We need to take a ref
* on the pointer that will be saved in the red-black tree
- * by rxe_add_key and used to lookup mcg from mgid later.
- * Adding key makes object visible to outside so this should
+ * by __rxe_insert_mcg and used to lookup mcg from mgid later.
+ * Inserting mcg makes it visible to outside so this should
* be done last after the object is ready.
*/
rxe_add_ref(mcg);
- rxe_add_key_locked(mcg, mgid);
+ __rxe_insert_mcg(mcg);
- return mcg;
+ return 0;
}
-static struct rxe_mcg *rxe_get_mcg(struct rxe_dev *rxe,
- union ib_gid *mgid)
+/**
+ * rxe_get_mcg - lookup or allocate a mcg
+ * @rxe: rxe device object
+ * @mgid: multicast IP address as a gid
+ *
+ * Returns: mcg on success else ERR_PTR(error)
+ */
+static struct rxe_mcg *rxe_get_mcg(struct rxe_dev *rxe, union ib_gid *mgid)
{
- struct rxe_mcg *mcg;
struct rxe_pool *pool = &rxe->mc_grp_pool;
+ struct rxe_mcg *mcg, *tmp;
+ int err;
- if (rxe->attr.max_mcast_qp_attach == 0)
+ if (rxe->attr.max_mcast_grp == 0)
return ERR_PTR(-EINVAL);
- spin_lock_bh(&rxe->mcg_lock);
- mcg = rxe_pool_get_key_locked(pool, mgid);
+ /* check to see if mcg already exists */
+ mcg = rxe_lookup_mcg(rxe, mgid);
+ if (mcg)
+ return mcg;
+
+ /* speculative alloc of new mcg */
+ mcg = rxe_alloc(pool);
if (!mcg)
- mcg = __rxe_create_mcg(rxe, pool, mgid);
- spin_unlock_bh(&rxe->mcg_lock);
+ return ERR_PTR(-ENOMEM);
+ spin_lock_bh(&rxe->mcg_lock);
+ /* re-check to see if someone else just added it */
+ tmp = __rxe_lookup_mcg(rxe, mgid);
+ if (tmp) {
+ rxe_drop_ref(mcg);
+ mcg = tmp;
+ goto out;
+ }
+
+ if (atomic_inc_return(&rxe->mcg_num) > rxe->attr.max_mcast_grp) {
+ err = -ENOMEM;
+ goto err_dec;
+ }
+
+ err = __rxe_init_mcg(rxe, mgid, mcg);
+ if (err)
+ goto err_dec;
+out:
+ spin_unlock_bh(&rxe->mcg_lock);
return mcg;
+
+err_dec:
+ atomic_dec(&rxe->mcg_num);
+ spin_unlock_bh(&rxe->mcg_lock);
+ rxe_drop_ref(mcg);
+ return ERR_PTR(err);
+}
+
+/**
+ * __rxe_destroy_mcg - destroy mcg object holding rxe->mcg_lock
+ * @mcg: the mcg object
+ *
+ * Context: caller is holding rxe->mcg_lock
+ * no qp's are attached to mcg
+ */
+static void __rxe_destroy_mcg(struct rxe_mcg *mcg)
+{
+ /* remove mcg from red-black tree then drop ref */
+ __rxe_remove_mcg(mcg);
+ rxe_drop_ref(mcg);
+
+ rxe_mcast_delete(mcg->rxe, &mcg->mgid);
+}
+
+/**
+ * rxe_destroy_mcg - destroy mcg object
+ * @mcg: the mcg object
+ *
+ * Context: no qp's are attached to mcg
+ */
+static void rxe_destroy_mcg(struct rxe_mcg *mcg)
+{
+ spin_lock_bh(&mcg->rxe->mcg_lock);
+ __rxe_destroy_mcg(mcg);
+ spin_unlock_bh(&mcg->rxe->mcg_lock);
+}
+
+void rxe_mc_cleanup(struct rxe_pool_elem *elem)
+{
+ /* nothing left to do for now */
}
static int rxe_attach_mcg(struct rxe_dev *rxe, struct rxe_qp *qp,
@@ -129,30 +298,6 @@ static int rxe_attach_mcg(struct rxe_dev *rxe, struct rxe_qp *qp,
return err;
}
-/* caller should be holding rxe->mcg_lock */
-static void __rxe_destroy_mcg(struct rxe_mcg *mcg)
-{
- /* first remove mcg from red-black tree then drop ref */
- rxe_drop_key_locked(mcg);
- rxe_drop_ref(mcg);
-
- rxe_mcast_delete(mcg->rxe, &mcg->mgid);
-}
-
-static void rxe_destroy_mcg(struct rxe_mcg *mcg)
-{
- struct rxe_dev *rxe = mcg->rxe;
-
- spin_lock_bh(&rxe->mcg_lock);
- __rxe_destroy_mcg(mcg);
- spin_unlock_bh(&rxe->mcg_lock);
-}
-
-void rxe_mc_cleanup(struct rxe_pool_elem *elem)
-{
- /* nothing left to do for now */
-}
-
static int rxe_detach_mcg(struct rxe_dev *rxe, struct rxe_qp *qp,
union ib_gid *mgid)
{
@@ -160,17 +305,16 @@ static int rxe_detach_mcg(struct rxe_dev *rxe, struct rxe_qp *qp,
struct rxe_mca *mca, *tmp;
int err;
- spin_lock_bh(&rxe->mcg_lock);
- mcg = rxe_pool_get_key_locked(&rxe->mc_grp_pool, mgid);
- if (!mcg) {
- /* we didn't find the mcast group for mgid */
- err = -EINVAL;
- goto out_unlock;
- }
+ mcg = rxe_lookup_mcg(rxe, mgid);
+ if (!mcg)
+ return -EINVAL;
+ spin_lock_bh(&rxe->mcg_lock);
list_for_each_entry_safe(mca, tmp, &mcg->qp_list, qp_list) {
if (mca->qp == qp) {
list_del(&mca->qp_list);
+ atomic_dec(&qp->mcg_num);
+ rxe_drop_ref(qp);
/* if the number of qp's attached to the
* mcast group falls to zero go ahead and
@@ -181,10 +325,8 @@ static int rxe_detach_mcg(struct rxe_dev *rxe, struct rxe_qp *qp,
if (atomic_dec_return(&mcg->qp_num) <= 0)
__rxe_destroy_mcg(mcg);
- atomic_dec(&qp->mcg_num);
-
/* drop the ref from get key. This will free the
- * object if num_qp is zero.
+ * object if qp_num is zero.
*/
rxe_drop_ref(mcg);
kfree(mca);
@@ -246,7 +246,7 @@ static void rxe_rcv_mcast_pkt(struct rxe_dev *rxe, struct sk_buff *skb)
memcpy(&dgid, &ipv6_hdr(skb)->daddr, sizeof(dgid));
/* lookup mcast group corresponding to mgid, takes a ref */
- mcg = rxe_pool_get_key(&rxe->mc_grp_pool, &dgid);
+ mcg = rxe_lookup_mcg(rxe, &dgid);
if (!mcg)
goto drop; /* mcast group not registered */
@@ -300,7 +300,7 @@ static void rxe_rcv_mcast_pkt(struct rxe_dev *rxe, struct sk_buff *skb)
spin_unlock_bh(&rxe->mcg_lock);
- rxe_drop_ref(mcg); /* drop ref from rxe_pool_get_key. */
+ rxe_drop_ref(mcg);
if (likely(!skb))
return;
@@ -353,6 +353,7 @@ struct rxe_mw {
struct rxe_mcg {
struct rxe_pool_elem elem;
+ struct rb_node node;
struct rxe_dev *rxe;
struct list_head qp_list;
union ib_gid mgid;
@@ -396,7 +397,10 @@ struct rxe_dev {
struct rxe_pool mw_pool;
struct rxe_pool mc_grp_pool;
+ /* multicast support */
spinlock_t mcg_lock;
+ struct rb_root mcg_tree;
+ atomic_t mcg_num;
spinlock_t pending_lock; /* guard pending_mmaps */
struct list_head pending_mmaps;
@@ -477,6 +481,4 @@ static inline struct rxe_pd *rxe_mw_pd(struct rxe_mw *mw)
int rxe_register_device(struct rxe_dev *rxe, const char *ibdev_name);
-void rxe_mc_cleanup(struct rxe_pool_elem *elem);
-
#endif /* RXE_VERBS_H */
Continuing to decouple mcg from rxe pools. Create red-black tree code in rxe_mcast.c to hold mcg index. Replace pool key calls by calls to local red-black routines. Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> --- drivers/infiniband/sw/rxe/rxe.c | 2 + drivers/infiniband/sw/rxe/rxe_loc.h | 3 +- drivers/infiniband/sw/rxe/rxe_mcast.c | 258 ++++++++++++++++++++------ drivers/infiniband/sw/rxe/rxe_recv.c | 4 +- drivers/infiniband/sw/rxe/rxe_verbs.h | 6 +- 5 files changed, 210 insertions(+), 63 deletions(-)