From: <gregkh@linuxfoundation.org>
To: markb@mellanox.com, dledford@redhat.com, erezsh@mellanox.com,
gregkh@linuxfoundation.org, leon@kernel.org, maorg@mellanox.com
Cc: <stable@vger.kernel.org>, <stable-commits@vger.kernel.org>
Subject: Patch "IB/cm: Mark stale CM id's whenever the mad agent was unregistered" has been added to the 4.8-stable tree
Date: Mon, 21 Nov 2016 17:29:30 +0100 [thread overview]
Message-ID: <147974577014068@kroah.com> (raw)
This is a note to let you know that I've just added the patch titled
IB/cm: Mark stale CM id's whenever the mad agent was unregistered
to the 4.8-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
ib-cm-mark-stale-cm-id-s-whenever-the-mad-agent-was-unregistered.patch
and it can be found in the queue-4.8 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@vger.kernel.org> know about it.
>From 9db0ff53cb9b43ed75bacd42a89c1a0ab048b2b0 Mon Sep 17 00:00:00 2001
From: Mark Bloch <markb@mellanox.com>
Date: Thu, 27 Oct 2016 16:36:27 +0300
Subject: IB/cm: Mark stale CM id's whenever the mad agent was unregistered
From: Mark Bloch <markb@mellanox.com>
commit 9db0ff53cb9b43ed75bacd42a89c1a0ab048b2b0 upstream.
When there is a CM id object that has port assigned to it, it means that
the cm-id asked for the specific port that it should go by it, but if
that port was removed (hot-unplug event) the cm-id was not updated.
In order to fix that the port keeps a list of all the cm-id's that are
planning to go by it, whenever the port is removed it marks all of them
as invalid.
This commit fixes a kernel panic which happens when running traffic between
guests and we force reboot a guest mid traffic, it triggers a kernel panic:
Call Trace:
[<ffffffff815271fa>] ? panic+0xa7/0x16f
[<ffffffff8152b534>] ? oops_end+0xe4/0x100
[<ffffffff8104a00b>] ? no_context+0xfb/0x260
[<ffffffff81084db2>] ? del_timer_sync+0x22/0x30
[<ffffffff8104a295>] ? __bad_area_nosemaphore+0x125/0x1e0
[<ffffffff81084240>] ? process_timeout+0x0/0x10
[<ffffffff8104a363>] ? bad_area_nosemaphore+0x13/0x20
[<ffffffff8104aabf>] ? __do_page_fault+0x31f/0x480
[<ffffffff81065df0>] ? default_wake_function+0x0/0x20
[<ffffffffa0752675>] ? free_msg+0x55/0x70 [mlx5_core]
[<ffffffffa0753434>] ? cmd_exec+0x124/0x840 [mlx5_core]
[<ffffffff8105a924>] ? find_busiest_group+0x244/0x9f0
[<ffffffff8152d45e>] ? do_page_fault+0x3e/0xa0
[<ffffffff8152a815>] ? page_fault+0x25/0x30
[<ffffffffa024da25>] ? cm_alloc_msg+0x35/0xc0 [ib_cm]
[<ffffffffa024e821>] ? ib_send_cm_dreq+0xb1/0x1e0 [ib_cm]
[<ffffffffa024f836>] ? cm_destroy_id+0x176/0x320 [ib_cm]
[<ffffffffa024fb00>] ? ib_destroy_cm_id+0x10/0x20 [ib_cm]
[<ffffffffa034f527>] ? ipoib_cm_free_rx_reap_list+0xa7/0x110 [ib_ipoib]
[<ffffffffa034f590>] ? ipoib_cm_rx_reap+0x0/0x20 [ib_ipoib]
[<ffffffffa034f5a5>] ? ipoib_cm_rx_reap+0x15/0x20 [ib_ipoib]
[<ffffffff81094d20>] ? worker_thread+0x170/0x2a0
[<ffffffff8109b2a0>] ? autoremove_wake_function+0x0/0x40
[<ffffffff81094bb0>] ? worker_thread+0x0/0x2a0
[<ffffffff8109aef6>] ? kthread+0x96/0xa0
[<ffffffff8100c20a>] ? child_rip+0xa/0x20
[<ffffffff8109ae60>] ? kthread+0x0/0xa0
[<ffffffff8100c200>] ? child_rip+0x0/0x20
Fixes: a977049dacde ("[PATCH] IB: Add the kernel CM implementation")
Signed-off-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Erez Shitrit <erezsh@mellanox.com>
Reviewed-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
drivers/infiniband/core/cm.c | 126 +++++++++++++++++++++++++++++++++++++------
1 file changed, 110 insertions(+), 16 deletions(-)
--- a/drivers/infiniband/core/cm.c
+++ b/drivers/infiniband/core/cm.c
@@ -80,6 +80,8 @@ static struct ib_cm {
__be32 random_id_operand;
struct list_head timewait_list;
struct workqueue_struct *wq;
+ /* Sync on cm change port state */
+ spinlock_t state_lock;
} cm;
/* Counter indexes ordered by attribute ID */
@@ -161,6 +163,8 @@ struct cm_port {
struct ib_mad_agent *mad_agent;
struct kobject port_obj;
u8 port_num;
+ struct list_head cm_priv_prim_list;
+ struct list_head cm_priv_altr_list;
struct cm_counter_group counter_group[CM_COUNTER_GROUPS];
};
@@ -241,6 +245,12 @@ struct cm_id_private {
u8 service_timeout;
u8 target_ack_delay;
+ struct list_head prim_list;
+ struct list_head altr_list;
+ /* Indicates that the send port mad is registered and av is set */
+ int prim_send_port_not_ready;
+ int altr_send_port_not_ready;
+
struct list_head work_list;
atomic_t work_count;
};
@@ -259,20 +269,47 @@ static int cm_alloc_msg(struct cm_id_pri
struct ib_mad_agent *mad_agent;
struct ib_mad_send_buf *m;
struct ib_ah *ah;
+ struct cm_av *av;
+ unsigned long flags, flags2;
+ int ret = 0;
+ /* don't let the port to be released till the agent is down */
+ spin_lock_irqsave(&cm.state_lock, flags2);
+ spin_lock_irqsave(&cm.lock, flags);
+ if (!cm_id_priv->prim_send_port_not_ready)
+ av = &cm_id_priv->av;
+ else if (!cm_id_priv->altr_send_port_not_ready &&
+ (cm_id_priv->alt_av.port))
+ av = &cm_id_priv->alt_av;
+ else {
+ pr_info("%s: not valid CM id\n", __func__);
+ ret = -ENODEV;
+ spin_unlock_irqrestore(&cm.lock, flags);
+ goto out;
+ }
+ spin_unlock_irqrestore(&cm.lock, flags);
+ /* Make sure the port haven't released the mad yet */
mad_agent = cm_id_priv->av.port->mad_agent;
- ah = ib_create_ah(mad_agent->qp->pd, &cm_id_priv->av.ah_attr);
- if (IS_ERR(ah))
- return PTR_ERR(ah);
+ if (!mad_agent) {
+ pr_info("%s: not a valid MAD agent\n", __func__);
+ ret = -ENODEV;
+ goto out;
+ }
+ ah = ib_create_ah(mad_agent->qp->pd, &av->ah_attr);
+ if (IS_ERR(ah)) {
+ ret = PTR_ERR(ah);
+ goto out;
+ }
m = ib_create_send_mad(mad_agent, cm_id_priv->id.remote_cm_qpn,
- cm_id_priv->av.pkey_index,
+ av->pkey_index,
0, IB_MGMT_MAD_HDR, IB_MGMT_MAD_DATA,
GFP_ATOMIC,
IB_MGMT_BASE_VERSION);
if (IS_ERR(m)) {
ib_destroy_ah(ah);
- return PTR_ERR(m);
+ ret = PTR_ERR(m);
+ goto out;
}
/* Timeout set by caller if response is expected. */
@@ -282,7 +319,10 @@ static int cm_alloc_msg(struct cm_id_pri
atomic_inc(&cm_id_priv->refcount);
m->context[0] = cm_id_priv;
*msg = m;
- return 0;
+
+out:
+ spin_unlock_irqrestore(&cm.state_lock, flags2);
+ return ret;
}
static int cm_alloc_response_msg(struct cm_port *port,
@@ -352,7 +392,8 @@ static void cm_init_av_for_response(stru
grh, &av->ah_attr);
}
-static int cm_init_av_by_path(struct ib_sa_path_rec *path, struct cm_av *av)
+static int cm_init_av_by_path(struct ib_sa_path_rec *path, struct cm_av *av,
+ struct cm_id_private *cm_id_priv)
{
struct cm_device *cm_dev;
struct cm_port *port = NULL;
@@ -387,7 +428,17 @@ static int cm_init_av_by_path(struct ib_
&av->ah_attr);
av->timeout = path->packet_life_time + 1;
- return 0;
+ spin_lock_irqsave(&cm.lock, flags);
+ if (&cm_id_priv->av == av)
+ list_add_tail(&cm_id_priv->prim_list, &port->cm_priv_prim_list);
+ else if (&cm_id_priv->alt_av == av)
+ list_add_tail(&cm_id_priv->altr_list, &port->cm_priv_altr_list);
+ else
+ ret = -EINVAL;
+
+ spin_unlock_irqrestore(&cm.lock, flags);
+
+ return ret;
}
static int cm_alloc_id(struct cm_id_private *cm_id_priv)
@@ -677,6 +728,8 @@ struct ib_cm_id *ib_create_cm_id(struct
spin_lock_init(&cm_id_priv->lock);
init_completion(&cm_id_priv->comp);
INIT_LIST_HEAD(&cm_id_priv->work_list);
+ INIT_LIST_HEAD(&cm_id_priv->prim_list);
+ INIT_LIST_HEAD(&cm_id_priv->altr_list);
atomic_set(&cm_id_priv->work_count, -1);
atomic_set(&cm_id_priv->refcount, 1);
return &cm_id_priv->id;
@@ -892,6 +945,15 @@ retest:
break;
}
+ spin_lock_irq(&cm.lock);
+ if (!list_empty(&cm_id_priv->altr_list) &&
+ (!cm_id_priv->altr_send_port_not_ready))
+ list_del(&cm_id_priv->altr_list);
+ if (!list_empty(&cm_id_priv->prim_list) &&
+ (!cm_id_priv->prim_send_port_not_ready))
+ list_del(&cm_id_priv->prim_list);
+ spin_unlock_irq(&cm.lock);
+
cm_free_id(cm_id->local_id);
cm_deref_id(cm_id_priv);
wait_for_completion(&cm_id_priv->comp);
@@ -1192,12 +1254,13 @@ int ib_send_cm_req(struct ib_cm_id *cm_i
goto out;
}
- ret = cm_init_av_by_path(param->primary_path, &cm_id_priv->av);
+ ret = cm_init_av_by_path(param->primary_path, &cm_id_priv->av,
+ cm_id_priv);
if (ret)
goto error1;
if (param->alternate_path) {
ret = cm_init_av_by_path(param->alternate_path,
- &cm_id_priv->alt_av);
+ &cm_id_priv->alt_av, cm_id_priv);
if (ret)
goto error1;
}
@@ -1653,7 +1716,8 @@ static int cm_req_handler(struct cm_work
dev_put(gid_attr.ndev);
}
work->path[0].gid_type = gid_attr.gid_type;
- ret = cm_init_av_by_path(&work->path[0], &cm_id_priv->av);
+ ret = cm_init_av_by_path(&work->path[0], &cm_id_priv->av,
+ cm_id_priv);
}
if (ret) {
int err = ib_get_cached_gid(work->port->cm_dev->ib_device,
@@ -1672,7 +1736,8 @@ static int cm_req_handler(struct cm_work
goto rejected;
}
if (req_msg->alt_local_lid) {
- ret = cm_init_av_by_path(&work->path[1], &cm_id_priv->alt_av);
+ ret = cm_init_av_by_path(&work->path[1], &cm_id_priv->alt_av,
+ cm_id_priv);
if (ret) {
ib_send_cm_rej(cm_id, IB_CM_REJ_INVALID_ALT_GID,
&work->path[0].sgid,
@@ -2727,7 +2792,8 @@ int ib_send_cm_lap(struct ib_cm_id *cm_i
goto out;
}
- ret = cm_init_av_by_path(alternate_path, &cm_id_priv->alt_av);
+ ret = cm_init_av_by_path(alternate_path, &cm_id_priv->alt_av,
+ cm_id_priv);
if (ret)
goto out;
cm_id_priv->alt_av.timeout =
@@ -2839,7 +2905,8 @@ static int cm_lap_handler(struct cm_work
cm_init_av_for_response(work->port, work->mad_recv_wc->wc,
work->mad_recv_wc->recv_buf.grh,
&cm_id_priv->av);
- cm_init_av_by_path(param->alternate_path, &cm_id_priv->alt_av);
+ cm_init_av_by_path(param->alternate_path, &cm_id_priv->alt_av,
+ cm_id_priv);
ret = atomic_inc_and_test(&cm_id_priv->work_count);
if (!ret)
list_add_tail(&work->list, &cm_id_priv->work_list);
@@ -3031,7 +3098,7 @@ int ib_send_cm_sidr_req(struct ib_cm_id
return -EINVAL;
cm_id_priv = container_of(cm_id, struct cm_id_private, id);
- ret = cm_init_av_by_path(param->path, &cm_id_priv->av);
+ ret = cm_init_av_by_path(param->path, &cm_id_priv->av, cm_id_priv);
if (ret)
goto out;
@@ -3468,7 +3535,9 @@ out:
static int cm_migrate(struct ib_cm_id *cm_id)
{
struct cm_id_private *cm_id_priv;
+ struct cm_av tmp_av;
unsigned long flags;
+ int tmp_send_port_not_ready;
int ret = 0;
cm_id_priv = container_of(cm_id, struct cm_id_private, id);
@@ -3477,7 +3546,14 @@ static int cm_migrate(struct ib_cm_id *c
(cm_id->lap_state == IB_CM_LAP_UNINIT ||
cm_id->lap_state == IB_CM_LAP_IDLE)) {
cm_id->lap_state = IB_CM_LAP_IDLE;
+ /* Swap address vector */
+ tmp_av = cm_id_priv->av;
cm_id_priv->av = cm_id_priv->alt_av;
+ cm_id_priv->alt_av = tmp_av;
+ /* Swap port send ready state */
+ tmp_send_port_not_ready = cm_id_priv->prim_send_port_not_ready;
+ cm_id_priv->prim_send_port_not_ready = cm_id_priv->altr_send_port_not_ready;
+ cm_id_priv->altr_send_port_not_ready = tmp_send_port_not_ready;
} else
ret = -EINVAL;
spin_unlock_irqrestore(&cm_id_priv->lock, flags);
@@ -3888,6 +3964,9 @@ static void cm_add_one(struct ib_device
port->cm_dev = cm_dev;
port->port_num = i;
+ INIT_LIST_HEAD(&port->cm_priv_prim_list);
+ INIT_LIST_HEAD(&port->cm_priv_altr_list);
+
ret = cm_create_port_fs(port);
if (ret)
goto error1;
@@ -3945,6 +4024,8 @@ static void cm_remove_one(struct ib_devi
{
struct cm_device *cm_dev = client_data;
struct cm_port *port;
+ struct cm_id_private *cm_id_priv;
+ struct ib_mad_agent *cur_mad_agent;
struct ib_port_modify port_modify = {
.clr_port_cap_mask = IB_PORT_CM_SUP
};
@@ -3968,15 +4049,27 @@ static void cm_remove_one(struct ib_devi
port = cm_dev->port[i-1];
ib_modify_port(ib_device, port->port_num, 0, &port_modify);
+ /* Mark all the cm_id's as not valid */
+ spin_lock_irq(&cm.lock);
+ list_for_each_entry(cm_id_priv, &port->cm_priv_altr_list, altr_list)
+ cm_id_priv->altr_send_port_not_ready = 1;
+ list_for_each_entry(cm_id_priv, &port->cm_priv_prim_list, prim_list)
+ cm_id_priv->prim_send_port_not_ready = 1;
+ spin_unlock_irq(&cm.lock);
/*
* We flush the queue here after the going_down set, this
* verify that no new works will be queued in the recv handler,
* after that we can call the unregister_mad_agent
*/
flush_workqueue(cm.wq);
- ib_unregister_mad_agent(port->mad_agent);
+ spin_lock_irq(&cm.state_lock);
+ cur_mad_agent = port->mad_agent;
+ port->mad_agent = NULL;
+ spin_unlock_irq(&cm.state_lock);
+ ib_unregister_mad_agent(cur_mad_agent);
cm_remove_port_fs(port);
}
+
device_unregister(cm_dev->device);
kfree(cm_dev);
}
@@ -3989,6 +4082,7 @@ static int __init ib_cm_init(void)
INIT_LIST_HEAD(&cm.device_list);
rwlock_init(&cm.device_lock);
spin_lock_init(&cm.lock);
+ spin_lock_init(&cm.state_lock);
cm.listen_service_table = RB_ROOT;
cm.listen_service_id = be64_to_cpu(IB_CM_ASSIGN_SERVICE_ID);
cm.remote_id_table = RB_ROOT;
Patches currently in stable-queue which might be from markb@mellanox.com are
queue-4.8/ib-core-avoid-unsigned-int-overflow-in-sg_alloc_table.patch
queue-4.8/ib-mlx4-check-gid_index-return-value.patch
queue-4.8/ib-mlx5-fix-memory-leak-in-query-device.patch
queue-4.8/ib-cm-mark-stale-cm-id-s-whenever-the-mad-agent-was-unregistered.patch
queue-4.8/ib-mlx4-fix-create-cq-error-flow.patch
reply other threads:[~2016-11-21 16:31 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=147974577014068@kroah.com \
--to=gregkh@linuxfoundation.org \
--cc=dledford@redhat.com \
--cc=erezsh@mellanox.com \
--cc=leon@kernel.org \
--cc=maorg@mellanox.com \
--cc=markb@mellanox.com \
--cc=stable-commits@vger.kernel.org \
--cc=stable@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.