* Copyright (c) 1999-2005, Mellanox Technologies, Inc. All rights reserved.
* Copyright (c) 2005-2006 Intel Corporation. All rights reserved.
*
- * This Software is licensed under one of the following licenses:
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
*
- * 1) under the terms of the "Common Public License 1.0" a copy of which is
- * available from the Open Source Initiative, see
- * http://www.opensource.org/licenses/cpl.php.
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
*
- * 2) under the terms of the "The BSD License" a copy of which is
- * available from the Open Source Initiative, see
- * http://www.opensource.org/licenses/bsd-license.php.
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
*
- * 3) under the terms of the "GNU General Public License (GPL) Version 2" a
- * copy of which is available from the Open Source Initiative, see
- * http://www.opensource.org/licenses/gpl-license.php.
- *
- * Licensee has the right to choose one of the above licenses.
- *
- * Redistributions of source code must retain the above copyright
- * notice and one of the license notices.
- *
- * Redistributions in binary form must reproduce both the above copyright
- * notice, one of the license notices in the documentation
- * and/or other materials provided with the distribution.
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
*
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
*/
#include <linux/completion.h>
#include <linux/inetdevice.h>
#include <net/tcp.h>
+#include <net/ipv6.h>
#include <rdma/rdma_cm.h>
#include <rdma/rdma_cm_ib.h>
struct rdma_bind_list *bind_list;
struct hlist_node node;
- struct list_head list;
- struct list_head listen_list;
+ struct list_head list; /* listen_any_list or cma_device.list */
+ struct list_head listen_list; /* per device listens */
struct cma_device *cma_dev;
struct list_head mc_list;
+ int internal_id;
enum cma_state state;
spinlock_t lock;
+ struct mutex qp_mutex;
+
struct completion comp;
atomic_t refcount;
- wait_queue_head_t wait_remove;
- atomic_t dev_remove;
+ struct mutex handler_mutex;
int backlog;
int timeout_ms;
} multicast;
struct list_head list;
void *context;
- struct sockaddr addr;
- u8 pad[sizeof(struct sockaddr_in6) -
- sizeof(struct sockaddr)];
+ struct sockaddr_storage addr;
};
struct cma_work {
struct rdma_cm_event event;
};
+struct cma_ndev_work {
+ struct work_struct work;
+ struct rdma_id_private *id;
+ struct rdma_cm_event event;
+};
+
union cma_ip_addr {
struct in6_addr ip6;
struct {
- __u32 pad[3];
- __u32 addr;
+ __be32 pad[3];
+ __be32 addr;
} ip4;
};
struct cma_hdr {
u8 cma_version;
u8 ip_version; /* IP version: 7:4 */
- __u16 port;
+ __be16 port;
union cma_ip_addr src_addr;
union cma_ip_addr dst_addr;
};
u8 sdp_version; /* Major version: 7:4 */
u8 ip_version; /* IP version: 7:4 */
u8 sdp_specific1[10];
- __u16 port;
- __u16 sdp_specific2;
+ __be16 port;
+ __be16 sdp_specific2;
union cma_ip_addr src_addr;
union cma_ip_addr dst_addr;
};
complete(&id_priv->comp);
}
-static int cma_disable_remove(struct rdma_id_private *id_priv,
+static int cma_disable_callback(struct rdma_id_private *id_priv,
enum cma_state state)
{
- unsigned long flags;
- int ret;
-
- spin_lock_irqsave(&id_priv->lock, flags);
- if (id_priv->state == state) {
- atomic_inc(&id_priv->dev_remove);
- ret = 0;
- } else
- ret = -EINVAL;
- spin_unlock_irqrestore(&id_priv->lock, flags);
- return ret;
-}
-
-static void cma_enable_remove(struct rdma_id_private *id_priv)
-{
- if (atomic_dec_and_test(&id_priv->dev_remove))
- wake_up(&id_priv->wait_remove);
+ mutex_lock(&id_priv->handler_mutex);
+ if (id_priv->state != state) {
+ mutex_unlock(&id_priv->handler_mutex);
+ return -EINVAL;
+ }
+ return 0;
}
static int cma_has_cm_dev(struct rdma_id_private *id_priv)
id_priv->id.event_handler = event_handler;
id_priv->id.ps = ps;
spin_lock_init(&id_priv->lock);
+ mutex_init(&id_priv->qp_mutex);
init_completion(&id_priv->comp);
atomic_set(&id_priv->refcount, 1);
- init_waitqueue_head(&id_priv->wait_remove);
- atomic_set(&id_priv->dev_remove, 0);
+ mutex_init(&id_priv->handler_mutex);
INIT_LIST_HEAD(&id_priv->listen_list);
INIT_LIST_HEAD(&id_priv->mc_list);
get_random_bytes(&id_priv->seq_num, sizeof id_priv->seq_num);
void rdma_destroy_qp(struct rdma_cm_id *id)
{
- ib_destroy_qp(id->qp);
+ struct rdma_id_private *id_priv;
+
+ id_priv = container_of(id, struct rdma_id_private, id);
+ mutex_lock(&id_priv->qp_mutex);
+ ib_destroy_qp(id_priv->id.qp);
+ id_priv->id.qp = NULL;
+ mutex_unlock(&id_priv->qp_mutex);
}
EXPORT_SYMBOL(rdma_destroy_qp);
-static int cma_modify_qp_rtr(struct rdma_cm_id *id)
+static int cma_modify_qp_rtr(struct rdma_id_private *id_priv,
+ struct rdma_conn_param *conn_param)
{
struct ib_qp_attr qp_attr;
int qp_attr_mask, ret;
- if (!id->qp)
- return 0;
+ mutex_lock(&id_priv->qp_mutex);
+ if (!id_priv->id.qp) {
+ ret = 0;
+ goto out;
+ }
/* Need to update QP attributes from default values. */
qp_attr.qp_state = IB_QPS_INIT;
- ret = rdma_init_qp_attr(id, &qp_attr, &qp_attr_mask);
+ ret = rdma_init_qp_attr(&id_priv->id, &qp_attr, &qp_attr_mask);
if (ret)
- return ret;
+ goto out;
- ret = ib_modify_qp(id->qp, &qp_attr, qp_attr_mask);
+ ret = ib_modify_qp(id_priv->id.qp, &qp_attr, qp_attr_mask);
if (ret)
- return ret;
+ goto out;
qp_attr.qp_state = IB_QPS_RTR;
- ret = rdma_init_qp_attr(id, &qp_attr, &qp_attr_mask);
+ ret = rdma_init_qp_attr(&id_priv->id, &qp_attr, &qp_attr_mask);
if (ret)
- return ret;
+ goto out;
- return ib_modify_qp(id->qp, &qp_attr, qp_attr_mask);
+ if (conn_param)
+ qp_attr.max_dest_rd_atomic = conn_param->responder_resources;
+ ret = ib_modify_qp(id_priv->id.qp, &qp_attr, qp_attr_mask);
+out:
+ mutex_unlock(&id_priv->qp_mutex);
+ return ret;
}
-static int cma_modify_qp_rts(struct rdma_cm_id *id)
+static int cma_modify_qp_rts(struct rdma_id_private *id_priv,
+ struct rdma_conn_param *conn_param)
{
struct ib_qp_attr qp_attr;
int qp_attr_mask, ret;
- if (!id->qp)
- return 0;
+ mutex_lock(&id_priv->qp_mutex);
+ if (!id_priv->id.qp) {
+ ret = 0;
+ goto out;
+ }
qp_attr.qp_state = IB_QPS_RTS;
- ret = rdma_init_qp_attr(id, &qp_attr, &qp_attr_mask);
+ ret = rdma_init_qp_attr(&id_priv->id, &qp_attr, &qp_attr_mask);
if (ret)
- return ret;
+ goto out;
- return ib_modify_qp(id->qp, &qp_attr, qp_attr_mask);
+ if (conn_param)
+ qp_attr.max_rd_atomic = conn_param->initiator_depth;
+ ret = ib_modify_qp(id_priv->id.qp, &qp_attr, qp_attr_mask);
+out:
+ mutex_unlock(&id_priv->qp_mutex);
+ return ret;
}
-static int cma_modify_qp_err(struct rdma_cm_id *id)
+static int cma_modify_qp_err(struct rdma_id_private *id_priv)
{
struct ib_qp_attr qp_attr;
+ int ret;
- if (!id->qp)
- return 0;
+ mutex_lock(&id_priv->qp_mutex);
+ if (!id_priv->id.qp) {
+ ret = 0;
+ goto out;
+ }
qp_attr.qp_state = IB_QPS_ERR;
- return ib_modify_qp(id->qp, &qp_attr, IB_QP_STATE);
+ ret = ib_modify_qp(id_priv->id.qp, &qp_attr, IB_QP_STATE);
+out:
+ mutex_unlock(&id_priv->qp_mutex);
+ return ret;
}
static int cma_ib_init_qp_attr(struct rdma_id_private *id_priv,
struct in6_addr *ip6;
if (addr->sa_family == AF_INET)
- return ZERONET(((struct sockaddr_in *) addr)->sin_addr.s_addr);
+ return ipv4_is_zeronet(
+ ((struct sockaddr_in *)addr)->sin_addr.s_addr);
else {
ip6 = &((struct sockaddr_in6 *) addr)->sin6_addr;
return (ip6->s6_addr32[0] | ip6->s6_addr32[1] |
static inline int cma_loopback_addr(struct sockaddr *addr)
{
- return LOOPBACK(((struct sockaddr_in *) addr)->sin_addr.s_addr);
+ if (addr->sa_family == AF_INET)
+ return ipv4_is_loopback(
+ ((struct sockaddr_in *) addr)->sin_addr.s_addr);
+ else
+ return ipv6_addr_loopback(
+ &((struct sockaddr_in6 *) addr)->sin6_addr);
}
static inline int cma_any_addr(struct sockaddr *addr)
}
static int cma_get_net_info(void *hdr, enum rdma_port_space ps,
- u8 *ip_ver, __u16 *port,
+ u8 *ip_ver, __be16 *port,
union cma_ip_addr **src, union cma_ip_addr **dst)
{
switch (ps) {
static void cma_save_net_info(struct rdma_addr *addr,
struct rdma_addr *listen_addr,
- u8 ip_ver, __u16 port,
+ u8 ip_ver, __be16 port,
union cma_ip_addr *src, union cma_ip_addr *dst)
{
struct sockaddr_in *listen4, *ip4;
}
}
-static inline int cma_internal_listen(struct rdma_id_private *id_priv)
-{
- return (id_priv->state == CMA_LISTEN) && id_priv->cma_dev &&
- cma_any_addr(&id_priv->id.route.addr.src_addr);
-}
-
-static void cma_destroy_listen(struct rdma_id_private *id_priv)
-{
- cma_exch(id_priv, CMA_DESTROYING);
-
- if (id_priv->cma_dev) {
- switch (rdma_node_get_transport(id_priv->id.device->node_type)) {
- case RDMA_TRANSPORT_IB:
- if (id_priv->cm_id.ib && !IS_ERR(id_priv->cm_id.ib))
- ib_destroy_cm_id(id_priv->cm_id.ib);
- break;
- case RDMA_TRANSPORT_IWARP:
- if (id_priv->cm_id.iw && !IS_ERR(id_priv->cm_id.iw))
- iw_destroy_cm_id(id_priv->cm_id.iw);
- break;
- default:
- break;
- }
- cma_detach_from_dev(id_priv);
- }
- list_del(&id_priv->listen_list);
-
- cma_deref_id(id_priv);
- wait_for_completion(&id_priv->comp);
-
- kfree(id_priv);
-}
-
static void cma_cancel_listens(struct rdma_id_private *id_priv)
{
struct rdma_id_private *dev_id_priv;
+ /*
+ * Remove from listen_any_list to prevent added devices from spawning
+ * additional listen requests.
+ */
mutex_lock(&lock);
list_del(&id_priv->list);
while (!list_empty(&id_priv->listen_list)) {
dev_id_priv = list_entry(id_priv->listen_list.next,
struct rdma_id_private, listen_list);
- cma_destroy_listen(dev_id_priv);
+ /* sync with device removal to avoid duplicate destruction */
+ list_del_init(&dev_id_priv->list);
+ list_del(&dev_id_priv->listen_list);
+ mutex_unlock(&lock);
+
+ rdma_destroy_id(&dev_id_priv->id);
+ mutex_lock(&lock);
}
mutex_unlock(&lock);
}
cma_cancel_route(id_priv);
break;
case CMA_LISTEN:
- if (cma_any_addr(&id_priv->id.route.addr.src_addr) &&
- !id_priv->cma_dev)
+ if (cma_any_addr((struct sockaddr *) &id_priv->id.route.addr.src_addr)
+ && !id_priv->cma_dev)
cma_cancel_listens(id_priv);
break;
default:
cma_deref_id(id_priv);
wait_for_completion(&id_priv->comp);
+ if (id_priv->internal_id)
+ cma_deref_id(id_priv->id.context);
+
kfree(id_priv->id.route.path_rec);
kfree(id_priv);
}
{
int ret;
- ret = cma_modify_qp_rtr(&id_priv->id);
+ ret = cma_modify_qp_rtr(id_priv, NULL);
if (ret)
goto reject;
- ret = cma_modify_qp_rts(&id_priv->id);
+ ret = cma_modify_qp_rts(id_priv, NULL);
if (ret)
goto reject;
return 0;
reject:
- cma_modify_qp_err(&id_priv->id);
+ cma_modify_qp_err(id_priv);
ib_send_cm_rej(id_priv->cm_id.ib, IB_CM_REJ_CONSUMER_DEFINED,
NULL, 0, NULL, 0);
return ret;
struct rdma_cm_event event;
int ret = 0;
- if (cma_disable_remove(id_priv, CMA_CONNECT))
+ if ((ib_event->event != IB_CM_TIMEWAIT_EXIT &&
+ cma_disable_callback(id_priv, CMA_CONNECT)) ||
+ (ib_event->event == IB_CM_TIMEWAIT_EXIT &&
+ cma_disable_callback(id_priv, CMA_DISCONNECT)))
return 0;
memset(&event, 0, sizeof event);
event.event = RDMA_CM_EVENT_DISCONNECTED;
break;
case IB_CM_TIMEWAIT_EXIT:
+ event.event = RDMA_CM_EVENT_TIMEWAIT_EXIT;
+ break;
case IB_CM_MRA_RECEIVED:
/* ignore event */
goto out;
case IB_CM_REJ_RECEIVED:
- cma_modify_qp_err(&id_priv->id);
+ cma_modify_qp_err(id_priv);
event.status = ib_event->param.rej_rcvd.reason;
event.event = RDMA_CM_EVENT_REJECTED;
event.param.conn.private_data = ib_event->private_data;
event.param.conn.private_data_len = IB_CM_REJ_PRIVATE_DATA_SIZE;
break;
default:
- printk(KERN_ERR "RDMA CMA: unexpected IB CM event: %d",
+ printk(KERN_ERR "RDMA CMA: unexpected IB CM event: %d\n",
ib_event->event);
goto out;
}
/* Destroy the CM ID by returning a non-zero value. */
id_priv->cm_id.ib = NULL;
cma_exch(id_priv, CMA_DESTROYING);
- cma_enable_remove(id_priv);
+ mutex_unlock(&id_priv->handler_mutex);
rdma_destroy_id(&id_priv->id);
return ret;
}
out:
- cma_enable_remove(id_priv);
+ mutex_unlock(&id_priv->handler_mutex);
return ret;
}
struct rdma_cm_id *id;
struct rdma_route *rt;
union cma_ip_addr *src, *dst;
- __u16 port;
+ __be16 port;
u8 ip_ver;
+ int ret;
if (cma_get_net_info(ib_event->private_data, listen_id->ps,
&ip_ver, &port, &src, &dst))
if (rt->num_paths == 2)
rt->path_rec[1] = *ib_event->param.req_rcvd.alternate_path;
- ib_addr_set_sgid(&rt->addr.dev_addr, &rt->path_rec[0].sgid);
ib_addr_set_dgid(&rt->addr.dev_addr, &rt->path_rec[0].dgid);
- ib_addr_set_pkey(&rt->addr.dev_addr, be16_to_cpu(rt->path_rec[0].pkey));
- rt->addr.dev_addr.dev_type = RDMA_NODE_IB_CA;
+ ret = rdma_translate_ip((struct sockaddr *) &id->route.addr.src_addr,
+ &id->route.addr.dev_addr);
+ if (ret)
+ goto destroy_id;
id_priv = container_of(id, struct rdma_id_private, id);
id_priv->state = CMA_CONNECT;
struct rdma_id_private *id_priv;
struct rdma_cm_id *id;
union cma_ip_addr *src, *dst;
- __u16 port;
+ __be16 port;
u8 ip_ver;
int ret;
cma_save_net_info(&id->route.addr, &listen_id->route.addr,
ip_ver, port, src, dst);
- ret = rdma_translate_ip(&id->route.addr.src_addr,
+ ret = rdma_translate_ip((struct sockaddr *) &id->route.addr.src_addr,
&id->route.addr.dev_addr);
if (ret)
goto err;
int offset, ret;
listen_id = cm_id->context;
- if (cma_disable_remove(listen_id, CMA_LISTEN))
+ if (cma_disable_callback(listen_id, CMA_LISTEN))
return -ECONNABORTED;
memset(&event, 0, sizeof event);
event.param.ud.private_data_len =
IB_CM_SIDR_REQ_PRIVATE_DATA_SIZE - offset;
} else {
- ib_send_cm_mra(cm_id, CMA_CM_MRA_SETTING, NULL, 0);
conn_id = cma_new_conn_id(&listen_id->id, ib_event);
cma_set_req_event_data(&event, &ib_event->param.req_rcvd,
ib_event->private_data, offset);
goto out;
}
- atomic_inc(&conn_id->dev_remove);
+ mutex_lock_nested(&conn_id->handler_mutex, SINGLE_DEPTH_NESTING);
mutex_lock(&lock);
ret = cma_acquire_dev(conn_id);
mutex_unlock(&lock);
cm_id->cm_handler = cma_ib_handler;
ret = conn_id->id.event_handler(&conn_id->id, &event);
- if (!ret)
+ if (!ret) {
+ /*
+ * Acquire mutex to prevent user executing rdma_destroy_id()
+ * while we're accessing the cm_id.
+ */
+ mutex_lock(&lock);
+ if (cma_comp(conn_id, CMA_CONNECT) &&
+ !cma_is_ud_ps(conn_id->id.ps))
+ ib_send_cm_mra(cm_id, CMA_CM_MRA_SETTING, NULL, 0);
+ mutex_unlock(&lock);
+ mutex_unlock(&conn_id->handler_mutex);
goto out;
+ }
/* Destroy the CM ID by returning a non-zero value. */
conn_id->cm_id.ib = NULL;
release_conn_id:
cma_exch(conn_id, CMA_DESTROYING);
- cma_enable_remove(conn_id);
+ mutex_unlock(&conn_id->handler_mutex);
rdma_destroy_id(&conn_id->id);
out:
- cma_enable_remove(listen_id);
+ mutex_unlock(&listen_id->handler_mutex);
return ret;
}
{
struct cma_hdr *cma_data, *cma_mask;
struct sdp_hh *sdp_data, *sdp_mask;
- __u32 ip4_addr;
+ __be32 ip4_addr;
struct in6_addr ip6_addr;
memset(compare, 0, sizeof *compare);
sdp_set_ip_ver(sdp_data, 4);
sdp_set_ip_ver(sdp_mask, 0xF);
sdp_data->dst_addr.ip4.addr = ip4_addr;
- sdp_mask->dst_addr.ip4.addr = ~0;
+ sdp_mask->dst_addr.ip4.addr = htonl(~0);
} else {
cma_set_ip_ver(cma_data, 4);
cma_set_ip_ver(cma_mask, 0xF);
cma_data->dst_addr.ip4.addr = ip4_addr;
- cma_mask->dst_addr.ip4.addr = ~0;
+ cma_mask->dst_addr.ip4.addr = htonl(~0);
}
break;
case AF_INET6:
struct sockaddr_in *sin;
int ret = 0;
- if (cma_disable_remove(id_priv, CMA_CONNECT))
+ if (cma_disable_callback(id_priv, CMA_CONNECT))
return 0;
memset(&event, 0, sizeof event);
/* Destroy the CM ID by returning a non-zero value. */
id_priv->cm_id.iw = NULL;
cma_exch(id_priv, CMA_DESTROYING);
- cma_enable_remove(id_priv);
+ mutex_unlock(&id_priv->handler_mutex);
rdma_destroy_id(&id_priv->id);
return ret;
}
- cma_enable_remove(id_priv);
+ mutex_unlock(&id_priv->handler_mutex);
return ret;
}
struct net_device *dev = NULL;
struct rdma_cm_event event;
int ret;
+ struct ib_device_attr attr;
listen_id = cm_id->context;
- if (cma_disable_remove(listen_id, CMA_LISTEN))
+ if (cma_disable_callback(listen_id, CMA_LISTEN))
return -ECONNABORTED;
/* Create a new RDMA id for the new IW CM ID */
new_cm_id = rdma_create_id(listen_id->id.event_handler,
listen_id->id.context,
RDMA_PS_TCP);
- if (!new_cm_id) {
+ if (IS_ERR(new_cm_id)) {
ret = -ENOMEM;
goto out;
}
conn_id = container_of(new_cm_id, struct rdma_id_private, id);
- atomic_inc(&conn_id->dev_remove);
+ mutex_lock_nested(&conn_id->handler_mutex, SINGLE_DEPTH_NESTING);
conn_id->state = CMA_CONNECT;
- dev = ip_dev_find(iw_event->local_addr.sin_addr.s_addr);
+ dev = ip_dev_find(&init_net, iw_event->local_addr.sin_addr.s_addr);
if (!dev) {
ret = -EADDRNOTAVAIL;
- cma_enable_remove(conn_id);
+ mutex_unlock(&conn_id->handler_mutex);
rdma_destroy_id(new_cm_id);
goto out;
}
ret = rdma_copy_addr(&conn_id->id.route.addr.dev_addr, dev, NULL);
if (ret) {
- cma_enable_remove(conn_id);
+ mutex_unlock(&conn_id->handler_mutex);
rdma_destroy_id(new_cm_id);
goto out;
}
ret = cma_acquire_dev(conn_id);
mutex_unlock(&lock);
if (ret) {
- cma_enable_remove(conn_id);
+ mutex_unlock(&conn_id->handler_mutex);
rdma_destroy_id(new_cm_id);
goto out;
}
sin = (struct sockaddr_in *) &new_cm_id->route.addr.dst_addr;
*sin = iw_event->remote_addr;
+ ret = ib_query_device(conn_id->id.device, &attr);
+ if (ret) {
+ mutex_unlock(&conn_id->handler_mutex);
+ rdma_destroy_id(new_cm_id);
+ goto out;
+ }
+
memset(&event, 0, sizeof event);
event.event = RDMA_CM_EVENT_CONNECT_REQUEST;
event.param.conn.private_data = iw_event->private_data;
event.param.conn.private_data_len = iw_event->private_data_len;
+ event.param.conn.initiator_depth = attr.max_qp_init_rd_atom;
+ event.param.conn.responder_resources = attr.max_qp_rd_atom;
ret = conn_id->id.event_handler(&conn_id->id, &event);
if (ret) {
/* User wants to destroy the CM ID */
conn_id->cm_id.iw = NULL;
cma_exch(conn_id, CMA_DESTROYING);
- cma_enable_remove(conn_id);
+ mutex_unlock(&conn_id->handler_mutex);
rdma_destroy_id(&conn_id->id);
+ goto out;
}
+ mutex_unlock(&conn_id->handler_mutex);
+
out:
if (dev)
dev_put(dev);
- cma_enable_remove(listen_id);
+ mutex_unlock(&listen_id->handler_mutex);
return ret;
}
if (IS_ERR(id_priv->cm_id.ib))
return PTR_ERR(id_priv->cm_id.ib);
- addr = &id_priv->id.route.addr.src_addr;
+ addr = (struct sockaddr *) &id_priv->id.route.addr.src_addr;
svc_id = cma_get_service_id(id_priv->id.ps, addr);
if (cma_any_addr(addr))
ret = ib_cm_listen(id_priv->cm_id.ib, svc_id, 0, NULL);
dev_id_priv->state = CMA_ADDR_BOUND;
memcpy(&id->route.addr.src_addr, &id_priv->id.route.addr.src_addr,
- ip_addr_size(&id_priv->id.route.addr.src_addr));
+ ip_addr_size((struct sockaddr *) &id_priv->id.route.addr.src_addr));
cma_attach_to_dev(dev_id_priv, cma_dev);
list_add_tail(&dev_id_priv->listen_list, &id_priv->listen_list);
+ atomic_inc(&id_priv->refcount);
+ dev_id_priv->internal_id = 1;
ret = rdma_listen(id, id_priv->backlog);
if (ret)
- goto err;
-
- return;
-err:
- cma_destroy_listen(dev_id_priv);
+ printk(KERN_WARNING "RDMA CMA: cma_listen_on_dev, error %d, "
+ "listening on device %s\n", ret, cma_dev->device->name);
}
static void cma_listen_on_all(struct rdma_id_private *id_priv)
static int cma_bind_any(struct rdma_cm_id *id, sa_family_t af)
{
- struct sockaddr_in addr_in;
+ struct sockaddr_storage addr_in;
memset(&addr_in, 0, sizeof addr_in);
- addr_in.sin_family = af;
+ addr_in.ss_family = af;
return rdma_bind_addr(id, (struct sockaddr *) &addr_in);
}
path_rec.pkey = cpu_to_be16(ib_addr_get_pkey(&addr->dev_addr));
path_rec.numb_path = 1;
path_rec.reversible = 1;
- path_rec.service_id = cma_get_service_id(id_priv->id.ps, &addr->dst_addr);
+ path_rec.service_id = cma_get_service_id(id_priv->id.ps,
+ (struct sockaddr *) &addr->dst_addr);
comp_mask = IB_SA_PATH_REC_DGID | IB_SA_PATH_REC_SGID |
IB_SA_PATH_REC_PKEY | IB_SA_PATH_REC_NUMB_PATH |
IB_SA_PATH_REC_REVERSIBLE | IB_SA_PATH_REC_SERVICE_ID;
- if (addr->src_addr.sa_family == AF_INET) {
+ if (addr->src_addr.ss_family == AF_INET) {
path_rec.qos_class = cpu_to_be16((u16) id_priv->tos);
comp_mask |= IB_SA_PATH_REC_QOS_CLASS;
} else {
struct rdma_id_private *id_priv = work->id;
int destroy = 0;
- atomic_inc(&id_priv->dev_remove);
+ mutex_lock(&id_priv->handler_mutex);
if (!cma_comp_exch(id_priv, work->old_state, work->new_state))
goto out;
destroy = 1;
}
out:
- cma_enable_remove(id_priv);
+ mutex_unlock(&id_priv->handler_mutex);
+ cma_deref_id(id_priv);
+ if (destroy)
+ rdma_destroy_id(&id_priv->id);
+ kfree(work);
+}
+
+static void cma_ndev_work_handler(struct work_struct *_work)
+{
+ struct cma_ndev_work *work = container_of(_work, struct cma_ndev_work, work);
+ struct rdma_id_private *id_priv = work->id;
+ int destroy = 0;
+
+ mutex_lock(&id_priv->handler_mutex);
+ if (id_priv->state == CMA_DESTROYING ||
+ id_priv->state == CMA_DEVICE_REMOVAL)
+ goto out;
+
+ if (id_priv->id.event_handler(&id_priv->id, &work->event)) {
+ cma_exch(id_priv, CMA_DESTROYING);
+ destroy = 1;
+ }
+
+out:
+ mutex_unlock(&id_priv->handler_mutex);
cma_deref_id(id_priv);
if (destroy)
rdma_destroy_id(&id_priv->id);
struct rdma_cm_event event;
memset(&event, 0, sizeof event);
- atomic_inc(&id_priv->dev_remove);
+ mutex_lock(&id_priv->handler_mutex);
/*
* Grab mutex to block rdma_destroy_id() from removing the device while
if (id_priv->id.event_handler(&id_priv->id, &event)) {
cma_exch(id_priv, CMA_DESTROYING);
- cma_enable_remove(id_priv);
+ mutex_unlock(&id_priv->handler_mutex);
cma_deref_id(id_priv);
rdma_destroy_id(&id_priv->id);
return;
}
out:
- cma_enable_remove(id_priv);
+ mutex_unlock(&id_priv->handler_mutex);
cma_deref_id(id_priv);
}
ib_addr_get_sgid(&id_priv->id.route.addr.dev_addr, &gid);
ib_addr_set_dgid(&id_priv->id.route.addr.dev_addr, &gid);
- if (cma_zero_addr(&id_priv->id.route.addr.src_addr)) {
+ if (cma_zero_addr((struct sockaddr *) &id_priv->id.route.addr.src_addr)) {
src_in = (struct sockaddr_in *)&id_priv->id.route.addr.src_addr;
dst_in = (struct sockaddr_in *)&id_priv->id.route.addr.dst_addr;
src_in->sin_family = dst_in->sin_family;
if (cma_any_addr(dst_addr))
ret = cma_resolve_loopback(id_priv);
else
- ret = rdma_resolve_ip(&addr_client, &id->route.addr.src_addr,
+ ret = rdma_resolve_ip(&addr_client, (struct sockaddr *) &id->route.addr.src_addr,
dst_addr, &id->route.addr.dev_addr,
timeout_ms, addr_handler, id_priv);
if (ret)
static int cma_alloc_any_port(struct idr *ps, struct rdma_id_private *id_priv)
{
struct rdma_bind_list *bind_list;
- int port, ret;
+ int port, ret, low, high;
bind_list = kzalloc(sizeof *bind_list, GFP_KERNEL);
if (!bind_list)
return -ENOMEM;
retry:
+ /* FIXME: add proper port randomization per like inet_csk_get_port */
do {
ret = idr_get_new_above(ps, bind_list, next_port, &port);
} while ((ret == -EAGAIN) && idr_pre_get(ps, GFP_KERNEL));
if (ret)
goto err1;
- if (port > sysctl_local_port_range[1]) {
- if (next_port != sysctl_local_port_range[0]) {
+ inet_get_local_port_range(&low, &high);
+ if (port > high) {
+ if (next_port != low) {
idr_remove(ps, port);
- next_port = sysctl_local_port_range[0];
+ next_port = low;
goto retry;
}
ret = -EADDRNOTAVAIL;
goto err2;
}
- if (port == sysctl_local_port_range[1])
- next_port = sysctl_local_port_range[0];
+ if (port == high)
+ next_port = low;
else
next_port = port + 1;
* We don't support binding to any address if anyone is bound to
* a specific address on the same port.
*/
- if (cma_any_addr(&id_priv->id.route.addr.src_addr))
+ if (cma_any_addr((struct sockaddr *) &id_priv->id.route.addr.src_addr))
return -EADDRNOTAVAIL;
hlist_for_each_entry(cur_id, node, &bind_list->owners, node) {
- if (cma_any_addr(&cur_id->id.route.addr.src_addr))
+ if (cma_any_addr((struct sockaddr *) &cur_id->id.route.addr.src_addr))
return -EADDRNOTAVAIL;
cur_sin = (struct sockaddr_in *) &cur_id->id.route.addr.src_addr;
}
mutex_lock(&lock);
- if (cma_any_port(&id_priv->id.route.addr.src_addr))
+ if (cma_any_port((struct sockaddr *) &id_priv->id.route.addr.src_addr))
ret = cma_alloc_any_port(ps, id_priv);
else
ret = cma_use_port(ps, id_priv);
struct rdma_id_private *id_priv;
int ret;
- if (addr->sa_family != AF_INET)
+ if (addr->sa_family != AF_INET && addr->sa_family != AF_INET6)
return -EAFNOSUPPORT;
id_priv = container_of(id, struct rdma_id_private, id);
static int cma_format_hdr(void *hdr, enum rdma_port_space ps,
struct rdma_route *route)
{
- struct sockaddr_in *src4, *dst4;
struct cma_hdr *cma_hdr;
struct sdp_hh *sdp_hdr;
- src4 = (struct sockaddr_in *) &route->addr.src_addr;
- dst4 = (struct sockaddr_in *) &route->addr.dst_addr;
-
- switch (ps) {
- case RDMA_PS_SDP:
- sdp_hdr = hdr;
- if (sdp_get_majv(sdp_hdr->sdp_version) != SDP_MAJ_VERSION)
- return -EINVAL;
- sdp_set_ip_ver(sdp_hdr, 4);
- sdp_hdr->src_addr.ip4.addr = src4->sin_addr.s_addr;
- sdp_hdr->dst_addr.ip4.addr = dst4->sin_addr.s_addr;
- sdp_hdr->port = src4->sin_port;
- break;
- default:
- cma_hdr = hdr;
- cma_hdr->cma_version = CMA_VERSION;
- cma_set_ip_ver(cma_hdr, 4);
- cma_hdr->src_addr.ip4.addr = src4->sin_addr.s_addr;
- cma_hdr->dst_addr.ip4.addr = dst4->sin_addr.s_addr;
- cma_hdr->port = src4->sin_port;
- break;
+ if (route->addr.src_addr.ss_family == AF_INET) {
+ struct sockaddr_in *src4, *dst4;
+
+ src4 = (struct sockaddr_in *) &route->addr.src_addr;
+ dst4 = (struct sockaddr_in *) &route->addr.dst_addr;
+
+ switch (ps) {
+ case RDMA_PS_SDP:
+ sdp_hdr = hdr;
+ if (sdp_get_majv(sdp_hdr->sdp_version) != SDP_MAJ_VERSION)
+ return -EINVAL;
+ sdp_set_ip_ver(sdp_hdr, 4);
+ sdp_hdr->src_addr.ip4.addr = src4->sin_addr.s_addr;
+ sdp_hdr->dst_addr.ip4.addr = dst4->sin_addr.s_addr;
+ sdp_hdr->port = src4->sin_port;
+ break;
+ default:
+ cma_hdr = hdr;
+ cma_hdr->cma_version = CMA_VERSION;
+ cma_set_ip_ver(cma_hdr, 4);
+ cma_hdr->src_addr.ip4.addr = src4->sin_addr.s_addr;
+ cma_hdr->dst_addr.ip4.addr = dst4->sin_addr.s_addr;
+ cma_hdr->port = src4->sin_port;
+ break;
+ }
+ } else {
+ struct sockaddr_in6 *src6, *dst6;
+
+ src6 = (struct sockaddr_in6 *) &route->addr.src_addr;
+ dst6 = (struct sockaddr_in6 *) &route->addr.dst_addr;
+
+ switch (ps) {
+ case RDMA_PS_SDP:
+ sdp_hdr = hdr;
+ if (sdp_get_majv(sdp_hdr->sdp_version) != SDP_MAJ_VERSION)
+ return -EINVAL;
+ sdp_set_ip_ver(sdp_hdr, 6);
+ sdp_hdr->src_addr.ip6 = src6->sin6_addr;
+ sdp_hdr->dst_addr.ip6 = dst6->sin6_addr;
+ sdp_hdr->port = src6->sin6_port;
+ break;
+ default:
+ cma_hdr = hdr;
+ cma_hdr->cma_version = CMA_VERSION;
+ cma_set_ip_ver(cma_hdr, 6);
+ cma_hdr->src_addr.ip6 = src6->sin6_addr;
+ cma_hdr->dst_addr.ip6 = dst6->sin6_addr;
+ cma_hdr->port = src6->sin6_port;
+ break;
+ }
}
return 0;
}
struct ib_cm_sidr_rep_event_param *rep = &ib_event->param.sidr_rep_rcvd;
int ret = 0;
- if (cma_disable_remove(id_priv, CMA_CONNECT))
+ if (cma_disable_callback(id_priv, CMA_CONNECT))
return 0;
memset(&event, 0, sizeof event);
event.status = 0;
break;
default:
- printk(KERN_ERR "RDMA CMA: unexpected IB CM event: %d",
+ printk(KERN_ERR "RDMA CMA: unexpected IB CM event: %d\n",
ib_event->event);
goto out;
}
/* Destroy the CM ID by returning a non-zero value. */
id_priv->cm_id.ib = NULL;
cma_exch(id_priv, CMA_DESTROYING);
- cma_enable_remove(id_priv);
+ mutex_unlock(&id_priv->handler_mutex);
rdma_destroy_id(&id_priv->id);
return ret;
}
out:
- cma_enable_remove(id_priv);
+ mutex_unlock(&id_priv->handler_mutex);
return ret;
}
req.path = route->path_rec;
req.service_id = cma_get_service_id(id_priv->id.ps,
- &route->addr.dst_addr);
+ (struct sockaddr *) &route->addr.dst_addr);
req.timeout_ms = 1 << (CMA_CM_RESPONSE_TIMEOUT - 8);
req.max_cm_retries = CMA_MAX_CM_RETRIES;
req.alternate_path = &route->path_rec[1];
req.service_id = cma_get_service_id(id_priv->id.ps,
- &route->addr.dst_addr);
+ (struct sockaddr *) &route->addr.dst_addr);
req.qp_num = id_priv->qp_num;
req.qp_type = IB_QPT_RC;
req.starting_psn = id_priv->seq_num;
sin = (struct sockaddr_in*) &id_priv->id.route.addr.dst_addr;
cm_id->remote_addr = *sin;
- ret = cma_modify_qp_rtr(&id_priv->id);
+ ret = cma_modify_qp_rtr(id_priv, conn_param);
if (ret)
goto out;
struct rdma_conn_param *conn_param)
{
struct ib_cm_rep_param rep;
- struct ib_qp_attr qp_attr;
- int qp_attr_mask, ret;
-
- if (id_priv->id.qp) {
- ret = cma_modify_qp_rtr(&id_priv->id);
- if (ret)
- goto out;
+ int ret;
- qp_attr.qp_state = IB_QPS_RTS;
- ret = ib_cm_init_qp_attr(id_priv->cm_id.ib, &qp_attr,
- &qp_attr_mask);
- if (ret)
- goto out;
+ ret = cma_modify_qp_rtr(id_priv, conn_param);
+ if (ret)
+ goto out;
- qp_attr.max_rd_atomic = conn_param->initiator_depth;
- ret = ib_modify_qp(id_priv->id.qp, &qp_attr, qp_attr_mask);
- if (ret)
- goto out;
- }
+ ret = cma_modify_qp_rts(id_priv, conn_param);
+ if (ret)
+ goto out;
memset(&rep, 0, sizeof rep);
rep.qp_num = id_priv->qp_num;
struct iw_cm_conn_param iw_param;
int ret;
- ret = cma_modify_qp_rtr(&id_priv->id);
+ ret = cma_modify_qp_rtr(id_priv, conn_param);
if (ret)
return ret;
return 0;
reject:
- cma_modify_qp_err(id);
+ cma_modify_qp_err(id_priv);
rdma_reject(id, NULL, 0);
return ret;
}
switch (rdma_node_get_transport(id->device->node_type)) {
case RDMA_TRANSPORT_IB:
- ret = cma_modify_qp_err(id);
+ ret = cma_modify_qp_err(id_priv);
if (ret)
goto out;
/* Initiate or respond to a disconnect. */
int ret;
id_priv = mc->id_priv;
- if (cma_disable_remove(id_priv, CMA_ADDR_BOUND) &&
- cma_disable_remove(id_priv, CMA_ADDR_RESOLVED))
+ if (cma_disable_callback(id_priv, CMA_ADDR_BOUND) &&
+ cma_disable_callback(id_priv, CMA_ADDR_RESOLVED))
return 0;
+ mutex_lock(&id_priv->qp_mutex);
if (!status && id_priv->id.qp)
status = ib_attach_mcast(id_priv->id.qp, &multicast->rec.mgid,
multicast->rec.mlid);
+ mutex_unlock(&id_priv->qp_mutex);
memset(&event, 0, sizeof event);
event.status = status;
ret = id_priv->id.event_handler(&id_priv->id, &event);
if (ret) {
cma_exch(id_priv, CMA_DESTROYING);
- cma_enable_remove(id_priv);
+ mutex_unlock(&id_priv->handler_mutex);
rdma_destroy_id(&id_priv->id);
return 0;
}
- cma_enable_remove(id_priv);
+ mutex_unlock(&id_priv->handler_mutex);
return 0;
}
/* IPv6 address is an SA assigned MGID. */
memcpy(mgid, &sin6->sin6_addr, sizeof *mgid);
} else {
- ip_ib_mc_map(sin->sin_addr.s_addr, mc_map);
+ ip_ib_mc_map(sin->sin_addr.s_addr, dev_addr->broadcast, mc_map);
if (id_priv->id.ps == RDMA_PS_UDP)
mc_map[7] = 0x01; /* Use RDMA CM signature */
- mc_map[8] = ib_addr_get_pkey(dev_addr) >> 8;
- mc_map[9] = (unsigned char) ib_addr_get_pkey(dev_addr);
*mgid = *(union ib_gid *) (mc_map + 4);
}
}
if (ret)
return ret;
- cma_set_mgid(id_priv, &mc->addr, &rec.mgid);
+ cma_set_mgid(id_priv, (struct sockaddr *) &mc->addr, &rec.mgid);
if (id_priv->id.ps == RDMA_PS_UDP)
rec.qkey = cpu_to_be32(RDMA_UDP_QKEY);
ib_addr_get_sgid(dev_addr, &rec.port_gid);
}
EXPORT_SYMBOL(rdma_leave_multicast);
+static int cma_netdev_change(struct net_device *ndev, struct rdma_id_private *id_priv)
+{
+ struct rdma_dev_addr *dev_addr;
+ struct cma_ndev_work *work;
+
+ dev_addr = &id_priv->id.route.addr.dev_addr;
+
+ if ((dev_addr->src_dev == ndev) &&
+ memcmp(dev_addr->src_dev_addr, ndev->dev_addr, ndev->addr_len)) {
+ printk(KERN_INFO "RDMA CM addr change for ndev %s used by id %p\n",
+ ndev->name, &id_priv->id);
+ work = kzalloc(sizeof *work, GFP_KERNEL);
+ if (!work)
+ return -ENOMEM;
+
+ INIT_WORK(&work->work, cma_ndev_work_handler);
+ work->id = id_priv;
+ work->event.event = RDMA_CM_EVENT_ADDR_CHANGE;
+ atomic_inc(&id_priv->refcount);
+ queue_work(cma_wq, &work->work);
+ }
+
+ return 0;
+}
+
+static int cma_netdev_callback(struct notifier_block *self, unsigned long event,
+ void *ctx)
+{
+ struct net_device *ndev = (struct net_device *)ctx;
+ struct cma_device *cma_dev;
+ struct rdma_id_private *id_priv;
+ int ret = NOTIFY_DONE;
+
+ if (dev_net(ndev) != &init_net)
+ return NOTIFY_DONE;
+
+ if (event != NETDEV_BONDING_FAILOVER)
+ return NOTIFY_DONE;
+
+ if (!(ndev->flags & IFF_MASTER) || !(ndev->priv_flags & IFF_BONDING))
+ return NOTIFY_DONE;
+
+ mutex_lock(&lock);
+ list_for_each_entry(cma_dev, &dev_list, list)
+ list_for_each_entry(id_priv, &cma_dev->id_list, list) {
+ ret = cma_netdev_change(ndev, id_priv);
+ if (ret)
+ goto out;
+ }
+
+out:
+ mutex_unlock(&lock);
+ return ret;
+}
+
+static struct notifier_block cma_nb = {
+ .notifier_call = cma_netdev_callback
+};
+
static void cma_add_one(struct ib_device *device)
{
struct cma_device *cma_dev;
{
struct rdma_cm_event event;
enum cma_state state;
+ int ret = 0;
/* Record that we want to remove the device */
state = cma_exch(id_priv, CMA_DEVICE_REMOVAL);
return 0;
cma_cancel_operation(id_priv, state);
- wait_event(id_priv->wait_remove, !atomic_read(&id_priv->dev_remove));
+ mutex_lock(&id_priv->handler_mutex);
/* Check for destruction from another callback. */
if (!cma_comp(id_priv, CMA_DEVICE_REMOVAL))
- return 0;
+ goto out;
memset(&event, 0, sizeof event);
event.event = RDMA_CM_EVENT_DEVICE_REMOVAL;
- return id_priv->id.event_handler(&id_priv->id, &event);
+ ret = id_priv->id.event_handler(&id_priv->id, &event);
+out:
+ mutex_unlock(&id_priv->handler_mutex);
+ return ret;
}
static void cma_process_remove(struct cma_device *cma_dev)
id_priv = list_entry(cma_dev->id_list.next,
struct rdma_id_private, list);
- if (cma_internal_listen(id_priv)) {
- cma_destroy_listen(id_priv);
- continue;
- }
-
+ list_del(&id_priv->listen_list);
list_del_init(&id_priv->list);
atomic_inc(&id_priv->refcount);
mutex_unlock(&lock);
- ret = cma_remove_id_dev(id_priv);
+ ret = id_priv->internal_id ? 1 : cma_remove_id_dev(id_priv);
cma_deref_id(id_priv);
if (ret)
rdma_destroy_id(&id_priv->id);
static int cma_init(void)
{
- int ret;
+ int ret, low, high, remaining;
get_random_bytes(&next_port, sizeof next_port);
- next_port = ((unsigned int) next_port %
- (sysctl_local_port_range[1] - sysctl_local_port_range[0])) +
- sysctl_local_port_range[0];
+ inet_get_local_port_range(&low, &high);
+ remaining = (high - low) + 1;
+ next_port = ((unsigned int) next_port % remaining) + low;
+
cma_wq = create_singlethread_workqueue("rdma_cm");
if (!cma_wq)
return -ENOMEM;
ib_sa_register_client(&sa_client);
rdma_addr_register_client(&addr_client);
+ register_netdevice_notifier(&cma_nb);
ret = ib_register_client(&cma_client);
if (ret)
return 0;
err:
+ unregister_netdevice_notifier(&cma_nb);
rdma_addr_unregister_client(&addr_client);
ib_sa_unregister_client(&sa_client);
destroy_workqueue(cma_wq);
static void cma_cleanup(void)
{
ib_unregister_client(&cma_client);
+ unregister_netdevice_notifier(&cma_nb);
rdma_addr_unregister_client(&addr_client);
ib_sa_unregister_client(&sa_client);
destroy_workqueue(cma_wq);