[PATCH 0/5] i40iw fixes

[PATCH 0/5] i40iw fixes

[Shiraz Saleem]

This series includes a few fixes for i40iw driver to be
considered for 4.14-rc.

Mustafa Ismail (1):
  i40iw: Add missing VLAN priority

Shiraz Saleem (4):
  i40iw: Fail open if there are no available MSI-X vectors
  i40iw: Prevent multiple netdev event notifier registrations
  i40iw: Call i40iw_cm_disconn on modify QP to disconnect
  i40iw: Add support for port reuse on active side connections

 drivers/infiniband/hw/i40iw/i40iw.h       |   1 -
 drivers/infiniband/hw/i40iw/i40iw_cm.c    | 154 +++++++++++++++---------------
 drivers/infiniband/hw/i40iw/i40iw_cm.h    |   5 +
 drivers/infiniband/hw/i40iw/i40iw_main.c  |  39 ++++----
 drivers/infiniband/hw/i40iw/i40iw_utils.c |   6 +-
 drivers/infiniband/hw/i40iw/i40iw_verbs.c |  12 +++
 6 files changed, 120 insertions(+), 97 deletions(-)

-- 
2.8.3

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 1/5] i40iw: Fail open if there are no available MSI-X vectors

[Shiraz Saleem]

Check number of available MSI-X vectors for i40iw.
If there are no available vectors, fail the open.

Signed-off-by: Shiraz Saleem <shiraz.saleem-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
---
 drivers/infiniband/hw/i40iw/i40iw_main.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/i40iw/i40iw_main.c b/drivers/infiniband/hw/i40iw/i40iw_main.c
index cc742c3..5965b59 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_main.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_main.c
@@ -1400,6 +1400,11 @@ static enum i40iw_status_code i40iw_save_msix_info(struct i40iw_device *iwdev,
 	u32 i;
 	u32 size;
 
+	if (!ldev->msix_count) {
+		i40iw_pr_err("No MSI-X vectors\n");
+		return I40IW_ERR_CONFIG;
+	}
+
 	iwdev->msix_count = ldev->msix_count;
 
 	size = sizeof(struct i40iw_msix_vector) * iwdev->msix_count;
@@ -1550,7 +1555,7 @@ static enum i40iw_status_code i40iw_setup_init_state(struct i40iw_handler *hdl,
 
 	status = i40iw_save_msix_info(iwdev, ldev);
 	if (status)
-		goto exit;
+		return status;
 	iwdev->hw.dev_context = (void *)ldev->pcidev;
 	iwdev->hw.hw_addr = ldev->hw_addr;
 	status = i40iw_allocate_dma_mem(&iwdev->hw,
-- 
2.8.3

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 2/5] i40iw: Prevent multiple netdev event notifier registrations

[Shiraz Saleem]

Netdev event notifier registration/de-registration is not
synchronized with a lock and there is a possibility of a
duplicate registration of notifier before the unregister
completes.

Register netdev event notifiers during module init and
de-register them at module exit.

This avoids the need to tie the registration to first netdev
client interface open and de-registration to last client
interface close and the synchronization to achieve it.

This also fixes a crash due to duplicate registration.

BUG: unable to handle kernel paging request at ffffffffa0d60388
IP: [<ffffffff8160f75d>] notifier_call_chain+0x3d/0x70
PGD 190d067 PUD 190e063 PMD 76c840067 PTE 0
Oops: 0000 [#1] SMP
Modules linked in: i40e(OF-) fuse btrfs zlib_deflate raid6_pq xor vfat msdos
[..]
e1000e vxlan ip_tunnel ptp pps_core i2c_core video [last unloaded: i40iw]
CPU: 1 PID: 27101 Comm: modprobe Tainted: GF       W  O--------------   3.10.0-229.el7.x86_64 #1
Hardware name: Gigabyte Technology Co., Ltd. To be filled by O.E.M./Q87M-D2H, BIOS F7 01/17/2014
task: ffff88076e8a96c0 ti: ffff8806959c8000 task.ti: ffff8806959c8000
RIP: 0010:[<ffffffff8160f75d>]  [<ffffffff8160f75d>] notifier_call_chain+0x3d/0x70
RSP: 0018:ffff8806959cbb38  EFLAGS: 00010282
RAX: ffffffffa0d60380 RBX: 00000000fffffffd RCX: 0000000000000000
0708] RDX: 0000000000000000 RSI: ffff88081227a000 RDI: 0000000000000002
RBP: ffff8806959cbb60 R08: 0000000000000246 R09: 000000000000700c
R10: ffff88080e16ea40 R11: 00000000000ae8df R12: ffffffffa0d60380
R13: 0000000000000002 R14: ffff88076e738800 R15: 0000000000000000
FS:  00007f604ef4a740(0000) GS:ffff88083e240000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: ffffffffa0d60388 CR3: 0000000753cd2000 CR4: 00000000001407e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Stack:
ffffffff819e73a0 0000000000000000 0000000000000002 ffff88076e738800
00000000ffffffff ffff8806959cbba0 ffffffff8109d61d 0000000000000000
0000000000000000 ffff88076e738800 0000000000000000 ffff88076e738800
Call Trace:
[<ffffffff8109d61d>] __blocking_notifier_call_chain+0x4d/0x70
[<ffffffff8109d656>] blocking_notifier_call_chain+0x16/0x20
[<ffffffff8156b9e4>] __inet_del_ifa+0x154/0x2b0
[<ffffffff8156d102>] inetdev_event+0x182/0x530
[<ffffffff8160f76c>] notifier_call_chain+0x4c/0x70
[<ffffffff8109d446>] raw_notifier_call_chain+0x16/0x20
[<ffffffff814f71fd>] call_netdevice_notifiers+0x2d/0x60
[<ffffffff814f8845>] rollback_registered_many+0x105/0x220
[<ffffffff814f89a0>] rollback_registered+0x40/0x70
[<ffffffff814f9c88>] unregister_netdevice_queue+0x48/0x80
[<ffffffff814f9cdc>] unregister_netdev+0x1c/0x30
[<ffffffffa0067139>] i40e_vsi_release+0x2a9/0x2b0 [i40e]
[<ffffffffa00674e8>] i40e_remove+0x128/0x2b0 [i40e]
[<ffffffff813092db>] pci_device_remove+0x3b/0xb0
[<ffffffff813d26ef>] __device_release_driver+0x7f/0xf0
[<ffffffff813d3068>] driver_detach+0xb8/0xc0
[<ffffffff813d22db>] bus_remove_driver+0x9b/0x120
[<ffffffff813d36dc>] driver_unregister+0x2c/0x50
[<ffffffff81307d4c>] pci_unregister_driver+0x2c/0x90
[<ffffffffa008f9d0>] i40e_exit_module+0x10/0x23 [i40e]
[<ffffffff810dad0b>] SyS_delete_module+0x16b/0x2d0
[<ffffffff81013b0c>] ? do_notify_resume+0x9c/0xb0
[<ffffffff81613da9>] system_call_fastpath+0x16/0x1b
Code: e5 41 57 4d 89 c7 41 56 49 89 d6 41 55 49 89 f5 41 54 53 89 cb
75 14 eb 3d 0f 1f 44 00 00 83 eb 01 74 25 4d 85 e4 74 20 4c 89 e0 <4c>
8b 60 08 4c 89 f2 4c 89 ee 48 89 c7 ff 10 4d 85 ff 74 04 41
RIP  [<ffffffff8160f75d>] notifier_call_chain+0x3d/0x70

Signed-off-by: Shiraz Saleem <shiraz.saleem-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
---
 drivers/infiniband/hw/i40iw/i40iw.h       |  1 -
 drivers/infiniband/hw/i40iw/i40iw_main.c  | 32 ++++++++++++++++---------------
 drivers/infiniband/hw/i40iw/i40iw_utils.c |  6 +++---
 3 files changed, 20 insertions(+), 19 deletions(-)

diff --git a/drivers/infiniband/hw/i40iw/i40iw.h b/drivers/infiniband/hw/i40iw/i40iw.h
index 9b15664..a65e4cbd 100644
--- a/drivers/infiniband/hw/i40iw/i40iw.h
+++ b/drivers/infiniband/hw/i40iw/i40iw.h
@@ -201,7 +201,6 @@ enum init_completion_state {
 	CEQ_CREATED,
 	ILQ_CREATED,
 	IEQ_CREATED,
-	INET_NOTIFIER,
 	IP_ADDR_REGISTERED,
 	RDMA_DEV_REGISTERED
 };
diff --git a/drivers/infiniband/hw/i40iw/i40iw_main.c b/drivers/infiniband/hw/i40iw/i40iw_main.c
index 5965b59..27590ae 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_main.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_main.c
@@ -99,8 +99,6 @@ static struct notifier_block i40iw_net_notifier = {
 	.notifier_call = i40iw_net_event
 };
 
-static atomic_t i40iw_notifiers_registered;
-
 /**
  * i40iw_find_i40e_handler - find a handler given a client info
  * @ldev: pointer to a client info
@@ -1376,11 +1374,20 @@ static enum i40iw_status_code i40iw_initialize_dev(struct i40iw_device *iwdev,
  */
 static void i40iw_register_notifiers(void)
 {
-	if (atomic_inc_return(&i40iw_notifiers_registered) == 1) {
-		register_inetaddr_notifier(&i40iw_inetaddr_notifier);
-		register_inet6addr_notifier(&i40iw_inetaddr6_notifier);
-		register_netevent_notifier(&i40iw_net_notifier);
-	}
+	register_inetaddr_notifier(&i40iw_inetaddr_notifier);
+	register_inet6addr_notifier(&i40iw_inetaddr6_notifier);
+	register_netevent_notifier(&i40iw_net_notifier);
+}
+
+/**
+ * i40iw_unregister_notifiers - unregister tcp ip notifiers
+ */
+
+static void i40iw_unregister_notifiers(void)
+{
+	unregister_netevent_notifier(&i40iw_net_notifier);
+	unregister_inetaddr_notifier(&i40iw_inetaddr_notifier);
+	unregister_inet6addr_notifier(&i40iw_inetaddr6_notifier);
 }
 
 /**
@@ -1467,12 +1474,6 @@ static void i40iw_deinit_device(struct i40iw_device *iwdev)
 		if (!iwdev->reset)
 			i40iw_del_macip_entry(iwdev, (u8)iwdev->mac_ip_table_idx);
 		/* fallthrough */
-	case INET_NOTIFIER:
-		if (!atomic_dec_return(&i40iw_notifiers_registered)) {
-			unregister_netevent_notifier(&i40iw_net_notifier);
-			unregister_inetaddr_notifier(&i40iw_inetaddr_notifier);
-			unregister_inet6addr_notifier(&i40iw_inetaddr6_notifier);
-		}
 		/* fallthrough */
 	case PBLE_CHUNK_MEM:
 		i40iw_destroy_pble_pool(dev, iwdev->pble_rsrc);
@@ -1672,8 +1673,6 @@ static int i40iw_open(struct i40e_info *ldev, struct i40e_client *client)
 			break;
 		iwdev->init_state = PBLE_CHUNK_MEM;
 		iwdev->virtchnl_wq = alloc_ordered_workqueue("iwvch", WQ_MEM_RECLAIM);
-		i40iw_register_notifiers();
-		iwdev->init_state = INET_NOTIFIER;
 		status = i40iw_add_mac_ip(iwdev);
 		if (status)
 			break;
@@ -2023,6 +2022,8 @@ static int __init i40iw_init_module(void)
 	i40iw_client.type = I40E_CLIENT_IWARP;
 	spin_lock_init(&i40iw_handler_lock);
 	ret = i40e_register_client(&i40iw_client);
+	i40iw_register_notifiers();
+
 	return ret;
 }
 
@@ -2034,6 +2035,7 @@ static int __init i40iw_init_module(void)
  */
 static void __exit i40iw_exit_module(void)
 {
+	i40iw_unregister_notifiers();
 	i40e_unregister_client(&i40iw_client);
 }
 
diff --git a/drivers/infiniband/hw/i40iw/i40iw_utils.c b/drivers/infiniband/hw/i40iw/i40iw_utils.c
index 62f1f45..e52dbbb 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_utils.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_utils.c
@@ -160,7 +160,7 @@ int i40iw_inetaddr_event(struct notifier_block *notifier,
 		return NOTIFY_DONE;
 
 	iwdev = &hdl->device;
-	if (iwdev->init_state < INET_NOTIFIER)
+	if (iwdev->init_state < IP_ADDR_REGISTERED || iwdev->closing)
 		return NOTIFY_DONE;
 
 	netdev = iwdev->ldev->netdev;
@@ -217,7 +217,7 @@ int i40iw_inet6addr_event(struct notifier_block *notifier,
 		return NOTIFY_DONE;
 
 	iwdev = &hdl->device;
-	if (iwdev->init_state < INET_NOTIFIER)
+	if (iwdev->init_state < IP_ADDR_REGISTERED || iwdev->closing)
 		return NOTIFY_DONE;
 
 	netdev = iwdev->ldev->netdev;
@@ -266,7 +266,7 @@ int i40iw_net_event(struct notifier_block *notifier, unsigned long event, void *
 		if (!iwhdl)
 			return NOTIFY_DONE;
 		iwdev = &iwhdl->device;
-		if (iwdev->init_state < INET_NOTIFIER)
+		if (iwdev->init_state < IP_ADDR_REGISTERED || iwdev->closing)
 			return NOTIFY_DONE;
 		p = (__be32 *)neigh->primary_key;
 		i40iw_copy_ip_ntohl(local_ipaddr, p);
-- 
2.8.3

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 3/5] i40iw: Call i40iw_cm_disconn on modify QP to disconnect

[Shiraz Saleem]

If QP modify to closing/terminate/error fails, connection is
not torn down as there is no corresponding asynchronous
event that will initiate the teardown.

Add explicit call to i40iw_cm_disconn if not waiting in
modify QP, otherwise schedule it in CM timer.

Signed-off-by: Shiraz Saleem <shiraz.saleem-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
---
 drivers/infiniband/hw/i40iw/i40iw_verbs.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/infiniband/hw/i40iw/i40iw_verbs.c b/drivers/infiniband/hw/i40iw/i40iw_verbs.c
index 1aa4110..28b3d02 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_verbs.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_verbs.c
@@ -1027,7 +1027,19 @@ int i40iw_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 				iwqp->hw_tcp_state = I40IW_TCP_STATE_CLOSED;
 				iwqp->last_aeq = I40IW_AE_RESET_SENT;
 				spin_unlock_irqrestore(&iwqp->lock, flags);
+				i40iw_cm_disconn(iwqp);
 			}
+		} else {
+			spin_lock_irqsave(&iwqp->lock, flags);
+			if (iwqp->cm_id) {
+				if (atomic_inc_return(&iwqp->close_timer_started) == 1) {
+					iwqp->cm_id->add_ref(iwqp->cm_id);
+					i40iw_schedule_cm_timer(iwqp->cm_node,
+								(struct i40iw_puda_buf *)iwqp,
+								 I40IW_TIMER_TYPE_CLOSE, 1, 0);
+				}
+			}
+			spin_unlock_irqrestore(&iwqp->lock, flags);
 		}
 	}
 	return 0;
-- 
2.8.3

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 4/5] i40iw: Add missing VLAN priority

[Shiraz Saleem]

From: Mustafa Ismail <mustafa.ismail-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>

Set the VLAN priority which is in the upper 3 bits of the VLAN
tag field in the QP context.

Signed-off-by: Mustafa Ismail <mustafa.ismail-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
Signed-off-by: Shiraz Saleem <shiraz.saleem-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
---
 drivers/infiniband/hw/i40iw/i40iw_cm.c | 3 ++-
 drivers/infiniband/hw/i40iw/i40iw_cm.h | 2 ++
 2 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/i40iw/i40iw_cm.c b/drivers/infiniband/hw/i40iw/i40iw_cm.c
index 14f36ba..b721544 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_cm.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_cm.c
@@ -3255,7 +3255,8 @@ static void i40iw_init_tcp_ctx(struct i40iw_cm_node *cm_node,
 	tcp_info->snd_mss = cpu_to_le32(((u32)cm_node->tcp_cntxt.mss));
 	if (cm_node->vlan_id < VLAN_TAG_PRESENT) {
 		tcp_info->insert_vlan_tag = true;
-		tcp_info->vlan_tag = cpu_to_le16(cm_node->vlan_id);
+		tcp_info->vlan_tag = cpu_to_le16(((u16)cm_node->user_pri << I40IW_VLAN_PRIO_SHIFT) |
+						  cm_node->vlan_id);
 	}
 	if (cm_node->ipv4) {
 		tcp_info->src_port = cpu_to_le16(cm_node->loc_port);
diff --git a/drivers/infiniband/hw/i40iw/i40iw_cm.h b/drivers/infiniband/hw/i40iw/i40iw_cm.h
index 2e52e38..8626e7f 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_cm.h
+++ b/drivers/infiniband/hw/i40iw/i40iw_cm.h
@@ -71,6 +71,8 @@
 #define	I40IW_HW_IRD_SETTING_32	32
 #define	I40IW_HW_IRD_SETTING_64	64
 
+#define I40IW_VLAN_PRIO_SHIFT   13
+
 enum ietf_mpa_flags {
 	IETF_MPA_FLAGS_MARKERS = 0x80,	/* receive Markers */
 	IETF_MPA_FLAGS_CRC = 0x40,	/* receive Markers */
-- 
2.8.3

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 5/5] i40iw: Add support for port reuse on active side connections

[Shiraz Saleem]

During OpenMPI scale up testing, we observe rdma_connect
failures if ports are reused on multiple connections.
This is because the Control Queue-Pair (CQP) command to add
the reused port to Accelerated Port Bit VectorTable (APBVT)
fails as there already exists an entry.

Check for duplicate port before invoking the CQP command
to add APBVT entry and delete the entry only if the port
is not in use.

Signed-off-by: Shiraz Saleem <shiraz.saleem-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
---
 drivers/infiniband/hw/i40iw/i40iw_cm.c | 151 ++++++++++++++++-----------------
 drivers/infiniband/hw/i40iw/i40iw_cm.h |   3 +
 2 files changed, 78 insertions(+), 76 deletions(-)

diff --git a/drivers/infiniband/hw/i40iw/i40iw_cm.c b/drivers/infiniband/hw/i40iw/i40iw_cm.c
index b721544..5230dd3 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_cm.c
+++ b/drivers/infiniband/hw/i40iw/i40iw_cm.c
@@ -1504,23 +1504,40 @@ static void i40iw_add_hte_node(struct i40iw_cm_core *cm_core,
 }
 
 /**
- * listen_port_in_use - determine if port is in use
- * @port: Listen port number
+ * i40iw_port_in_use - determine if port is in use
+ * @port: port number
+ * @active_side: flag for listener side vs active side
  */
-static bool i40iw_listen_port_in_use(struct i40iw_cm_core *cm_core, u16 port)
+static bool i40iw_port_in_use(struct i40iw_cm_core *cm_core, u16 port, bool active_side)
 {
 	struct i40iw_cm_listener *listen_node;
+	struct i40iw_cm_node *cm_node;
 	unsigned long flags;
 	bool ret = false;
 
-	spin_lock_irqsave(&cm_core->listen_list_lock, flags);
-	list_for_each_entry(listen_node, &cm_core->listen_nodes, list) {
-		if (listen_node->loc_port == port) {
-			ret = true;
-			break;
+	if (active_side) {
+		/* search connected node list */
+		spin_lock_irqsave(&cm_core->ht_lock, flags);
+		list_for_each_entry(cm_node, &cm_core->connected_nodes, list) {
+			if (cm_node->loc_port == port) {
+				ret = true;
+				break;
+			}
+		}
+			if (!ret)
+				clear_bit(port, cm_core->active_side_ports);
+		spin_unlock_irqrestore(&cm_core->ht_lock, flags);
+	} else {
+		spin_lock_irqsave(&cm_core->listen_list_lock, flags);
+		list_for_each_entry(listen_node, &cm_core->listen_nodes, list) {
+			if (listen_node->loc_port == port) {
+				ret = true;
+				break;
+			}
 		}
+		spin_unlock_irqrestore(&cm_core->listen_list_lock, flags);
 	}
-	spin_unlock_irqrestore(&cm_core->listen_list_lock, flags);
+
 	return ret;
 }
 
@@ -1868,7 +1885,7 @@ static int i40iw_dec_refcnt_listen(struct i40iw_cm_core *cm_core,
 		spin_unlock_irqrestore(&cm_core->listen_list_lock, flags);
 
 		if (listener->iwdev) {
-			if (apbvt_del && !i40iw_listen_port_in_use(cm_core, listener->loc_port))
+			if (apbvt_del && !i40iw_port_in_use(cm_core, listener->loc_port, false))
 				i40iw_manage_apbvt(listener->iwdev,
 						   listener->loc_port,
 						   I40IW_MANAGE_APBVT_DEL);
@@ -2247,21 +2264,21 @@ static void i40iw_rem_ref_cm_node(struct i40iw_cm_node *cm_node)
 	if (cm_node->listener) {
 		i40iw_dec_refcnt_listen(cm_core, cm_node->listener, 0, true);
 	} else {
-		if (!i40iw_listen_port_in_use(cm_core, cm_node->loc_port) &&
-		    cm_node->apbvt_set) {
+		if (!i40iw_port_in_use(cm_core, cm_node->loc_port, true) && cm_node->apbvt_set) {
 			i40iw_manage_apbvt(cm_node->iwdev,
 					   cm_node->loc_port,
 					   I40IW_MANAGE_APBVT_DEL);
-			i40iw_get_addr_info(cm_node, &nfo);
-			if (cm_node->qhash_set) {
-				i40iw_manage_qhash(cm_node->iwdev,
-						   &nfo,
-						   I40IW_QHASH_TYPE_TCP_ESTABLISHED,
-						   I40IW_QHASH_MANAGE_TYPE_DELETE,
-						   NULL,
-						   false);
-				cm_node->qhash_set = 0;
-			}
+			cm_node->apbvt_set = 0;
+		}
+		i40iw_get_addr_info(cm_node, &nfo);
+		if (cm_node->qhash_set) {
+			i40iw_manage_qhash(cm_node->iwdev,
+					   &nfo,
+					   I40IW_QHASH_TYPE_TCP_ESTABLISHED,
+					   I40IW_QHASH_MANAGE_TYPE_DELETE,
+					   NULL,
+					   false);
+			cm_node->qhash_set = 0;
 		}
 	}
 
@@ -3738,10 +3755,8 @@ int i40iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	struct sockaddr_in *raddr;
 	struct sockaddr_in6 *laddr6;
 	struct sockaddr_in6 *raddr6;
-	bool qhash_set = false;
-	int apbvt_set = 0;
-	int err = 0;
-	enum i40iw_status_code status;
+	int ret = 0;
+	unsigned long flags;
 
 	ibqp = i40iw_get_qp(cm_id->device, conn_param->qpn);
 	if (!ibqp)
@@ -3790,32 +3805,6 @@ int i40iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	cm_info.user_pri = rt_tos2priority(cm_id->tos);
 	i40iw_debug(&iwdev->sc_dev, I40IW_DEBUG_DCB, "%s TOS:[%d] UP:[%d]\n",
 		    __func__, cm_id->tos, cm_info.user_pri);
-	if ((cm_info.ipv4 && (laddr->sin_addr.s_addr != raddr->sin_addr.s_addr)) ||
-	    (!cm_info.ipv4 && memcmp(laddr6->sin6_addr.in6_u.u6_addr32,
-				     raddr6->sin6_addr.in6_u.u6_addr32,
-				     sizeof(laddr6->sin6_addr.in6_u.u6_addr32)))) {
-		status = i40iw_manage_qhash(iwdev,
-					    &cm_info,
-					    I40IW_QHASH_TYPE_TCP_ESTABLISHED,
-					    I40IW_QHASH_MANAGE_TYPE_ADD,
-					    NULL,
-					    true);
-		if (status)
-			return -EINVAL;
-		qhash_set = true;
-	}
-	status = i40iw_manage_apbvt(iwdev, cm_info.loc_port, I40IW_MANAGE_APBVT_ADD);
-	if (status) {
-		i40iw_manage_qhash(iwdev,
-				   &cm_info,
-				   I40IW_QHASH_TYPE_TCP_ESTABLISHED,
-				   I40IW_QHASH_MANAGE_TYPE_DELETE,
-				   NULL,
-				   false);
-		return -EINVAL;
-	}
-
-	apbvt_set = 1;
 	cm_id->add_ref(cm_id);
 	cm_node = i40iw_create_cm_node(&iwdev->cm_core, iwdev,
 				       conn_param->private_data_len,
@@ -3823,17 +3812,40 @@ int i40iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 				       &cm_info);
 
 	if (IS_ERR(cm_node)) {
-		err = PTR_ERR(cm_node);
-		goto err_out;
+		ret = PTR_ERR(cm_node);
+		cm_id->rem_ref(cm_id);
+		return ret;
+	}
+
+	if ((cm_info.ipv4 && (laddr->sin_addr.s_addr != raddr->sin_addr.s_addr)) ||
+	    (!cm_info.ipv4 && memcmp(laddr6->sin6_addr.in6_u.u6_addr32,
+				     raddr6->sin6_addr.in6_u.u6_addr32,
+				     sizeof(laddr6->sin6_addr.in6_u.u6_addr32)))) {
+		if (i40iw_manage_qhash(iwdev, &cm_info, I40IW_QHASH_TYPE_TCP_ESTABLISHED,
+				       I40IW_QHASH_MANAGE_TYPE_ADD, NULL, true)) {
+			ret = -EINVAL;
+			goto err;
+		}
+		cm_node->qhash_set = true;
 	}
 
+	spin_lock_irqsave(&iwdev->cm_core.ht_lock, flags);
+	if (!test_and_set_bit(cm_info.loc_port, iwdev->cm_core.active_side_ports)) {
+		spin_unlock_irqrestore(&iwdev->cm_core.ht_lock, flags);
+		if (i40iw_manage_apbvt(iwdev, cm_info.loc_port, I40IW_MANAGE_APBVT_ADD)) {
+			ret =  -EINVAL;
+			goto err;
+		}
+	} else {
+		spin_unlock_irqrestore(&iwdev->cm_core.ht_lock, flags);
+	}
+
+	cm_node->apbvt_set = true;
 	i40iw_record_ird_ord(cm_node, (u16)conn_param->ird, (u16)conn_param->ord);
 	if (cm_node->send_rdma0_op == SEND_RDMA_READ_ZERO &&
 	    !cm_node->ord_size)
 		cm_node->ord_size = 1;
 
-	cm_node->apbvt_set = apbvt_set;
-	cm_node->qhash_set = qhash_set;
 	iwqp->cm_node = cm_node;
 	cm_node->iwqp = iwqp;
 	iwqp->cm_id = cm_id;
@@ -3841,11 +3853,9 @@ int i40iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 
 	if (cm_node->state != I40IW_CM_STATE_OFFLOADED) {
 		cm_node->state = I40IW_CM_STATE_SYN_SENT;
-		err = i40iw_send_syn(cm_node, 0);
-		if (err) {
-			i40iw_rem_ref_cm_node(cm_node);
-			goto err_out;
-		}
+		ret = i40iw_send_syn(cm_node, 0);
+		if (ret)
+			goto err;
 	}
 
 	i40iw_debug(cm_node->dev,
@@ -3854,9 +3864,10 @@ int i40iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 		    cm_node->rem_port,
 		    cm_node,
 		    cm_node->cm_id);
+
 	return 0;
 
-err_out:
+err:
 	if (cm_info.ipv4)
 		i40iw_debug(&iwdev->sc_dev,
 			    I40IW_DEBUG_CM,
@@ -3868,22 +3879,10 @@ int i40iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 			    "Api - connect() FAILED: dest addr=%pI6",
 			    cm_info.rem_addr);
 
-	if (qhash_set)
-		i40iw_manage_qhash(iwdev,
-				   &cm_info,
-				   I40IW_QHASH_TYPE_TCP_ESTABLISHED,
-				   I40IW_QHASH_MANAGE_TYPE_DELETE,
-				   NULL,
-				   false);
-
-	if (apbvt_set && !i40iw_listen_port_in_use(&iwdev->cm_core,
-						   cm_info.loc_port))
-		i40iw_manage_apbvt(iwdev,
-				   cm_info.loc_port,
-				   I40IW_MANAGE_APBVT_DEL);
+	i40iw_rem_ref_cm_node(cm_node);
 	cm_id->rem_ref(cm_id);
 	iwdev->cm_core.stats_connect_errs++;
-	return err;
+	return ret;
 }
 
 /**
diff --git a/drivers/infiniband/hw/i40iw/i40iw_cm.h b/drivers/infiniband/hw/i40iw/i40iw_cm.h
index 8626e7f..45abef7 100644
--- a/drivers/infiniband/hw/i40iw/i40iw_cm.h
+++ b/drivers/infiniband/hw/i40iw/i40iw_cm.h
@@ -71,6 +71,7 @@
 #define	I40IW_HW_IRD_SETTING_32	32
 #define	I40IW_HW_IRD_SETTING_64	64
 
+#define MAX_PORTS		65536
 #define I40IW_VLAN_PRIO_SHIFT   13
 
 enum ietf_mpa_flags {
@@ -413,6 +414,8 @@ struct i40iw_cm_core {
 	spinlock_t ht_lock; /* manage hash table */
 	spinlock_t listen_list_lock; /* listen list */
 
+	unsigned long active_side_ports[BITS_TO_LONGS(MAX_PORTS)];
+
 	u64	stats_nodes_created;
 	u64	stats_nodes_destroyed;
 	u64	stats_listen_created;
-- 
2.8.3

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 0/5] i40iw fixes

[Doug Ledford]

On Tue, 2017-09-19 at 09:19 -0500, Shiraz Saleem wrote:
> This series includes a few fixes for i40iw driver to be
> considered for 4.14-rc.
> 
> Mustafa Ismail (1):
>   i40iw: Add missing VLAN priority
> 
> Shiraz Saleem (4):
>   i40iw: Fail open if there are no available MSI-X vectors
>   i40iw: Prevent multiple netdev event notifier registrations
>   i40iw: Call i40iw_cm_disconn on modify QP to disconnect
>   i40iw: Add support for port reuse on active side connections
> 
>  drivers/infiniband/hw/i40iw/i40iw.h       |   1 -
>  drivers/infiniband/hw/i40iw/i40iw_cm.c    | 154 +++++++++++++++-----
> ----------
>  drivers/infiniband/hw/i40iw/i40iw_cm.h    |   5 +
>  drivers/infiniband/hw/i40iw/i40iw_main.c  |  39 ++++----
>  drivers/infiniband/hw/i40iw/i40iw_utils.c |   6 +-
>  drivers/infiniband/hw/i40iw/i40iw_verbs.c |  12 +++
>  6 files changed, 120 insertions(+), 97 deletions(-)

The fifth patch was a bit large...if it weren't for the fact that it
fixes failed connects under load I would have skipped it.  But, I took
the entire series to -rc because it actually fixes an issue.  Thanks.


-- 
Doug Ledford <dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
    GPG KeyID: B826A3330E572FDD
    Key fingerprint = AE6B 1BDA 122B 23B4 265B  1274 B826 A333 0E57 2FDD

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html