diff mbox series

[for-rc] RDMA/siw: Remove NETDEV_GOING_DOWN event handler

Message ID 20240724085428.3813-1-showrya@chelsio.com (mailing list archive)
State Accepted
Headers show
Series [for-rc] RDMA/siw: Remove NETDEV_GOING_DOWN event handler | expand

Commit Message

Showrya M N July 24, 2024, 8:54 a.m. UTC 
Toggling link while running NVME-oF over siw hits a kernel panic
due to race condition within siw_handler and ib_destroy_qp().
The IB_EVENT_PORT_ERR event can alone handle destroying qps.
therefore remove unwanted processing in siw.

Suggested-by: Bernard Metzler <bmt@zurich.ibm.com>
Signed-off-by: Showrya M N <showrya@chelsio.com>
Signed-off-by: Potnuri Bharat Teja <bharat@chelsio.com>
---
 drivers/infiniband/sw/siw/siw.h      |  2 --
 drivers/infiniband/sw/siw/siw_main.c | 37 ----------------------------
 2 files changed, 39 deletions(-)

Comments

Zhu Yanjun July 25, 2024, 12:57 p.m. UTC | #1 
在 2024/7/24 10:54, Showrya M N 写道:
> Toggling link while running NVME-oF over siw hits a kernel panic
> due to race condition within siw_handler and ib_destroy_qp().
> The IB_EVENT_PORT_ERR event can alone handle destroying qps.
> therefore remove unwanted processing in siw.

In the link:
https://lore.kernel.org/all/000000000000fe34b1061e0ffa36@google.com/T/

The Call Trace is as below. Not sure if this call trace is the same with 
this commit.

Call Trace:
  <TASK>
  __debug_object_init+0x2a9/0x400 lib/debugobjects.c:654
  siw_device_goes_down drivers/infiniband/sw/siw/siw_main.c:395 [inline]
  siw_netdev_event+0x3bd/0x620 drivers/infiniband/sw/siw/siw_main.c:422
  notifier_call_chain+0x19f/0x3e0 kernel/notifier.c:93
  call_netdevice_notifiers_extack net/core/dev.c:2032 [inline]
  call_netdevice_notifiers net/core/dev.c:2046 [inline]
  __dev_close_many+0x146/0x300 net/core/dev.c:1532
  __dev_close net/core/dev.c:1570 [inline]
  __dev_change_flags+0x30e/0x6f0 net/core/dev.c:8835
  dev_change_flags+0x8b/0x1a0 net/core/dev.c:8909
  do_setlink+0xccd/0x41f0 net/core/rtnetlink.c:2900
  rtnl_setlink+0x40d/0x5a0 net/core/rtnetlink.c:3201
  rtnetlink_rcv_msg+0x73f/0xcf0 net/core/rtnetlink.c:6647
  netlink_rcv_skb+0x1e3/0x430 net/netlink/af_netlink.c:2550
  netlink_unicast_kernel net/netlink/af_netlink.c:1331 [inline]
  netlink_unicast+0x7f0/0x990 net/netlink/af_netlink.c:1357
  netlink_sendmsg+0x8e4/0xcb0 net/netlink/af_netlink.c:1901
  sock_sendmsg_nosec net/socket.c:730 [inline]
  __sock_sendmsg+0x221/0x270 net/socket.c:745
  sock_write_iter+0x2dd/0x400 net/socket.c:1160
  do_iter_readv_writev+0x60a/0x890
  vfs_writev+0x37c/0xbb0 fs/read_write.c:971
  do_writev+0x1b1/0x350 fs/read_write.c:1018
  do_syscall_x64 arch/x86/entry/common.c:52 [inline]
  do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83
  entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7fda35175f19
Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 
f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 
f0 ff ff 73 01 c3 48 c7 c1 a8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007fda35fff048 EFLAGS: 00000246 ORIG_RAX: 0000000000000014
RAX: ffffffffffffffda RBX: 00007fda35305f60 RCX: 00007fda35175f19
RDX: 0000000000000001 RSI: 00000000200003c0 RDI: 0000000000000006
RBP: 00007fda351e4e68 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 000000000000000b R14: 00007fda35305f60 R15: 00007ffc0c742898
  </TASK>

Zhu Yanjun

> 
> Suggested-by: Bernard Metzler <bmt@zurich.ibm.com>
> Signed-off-by: Showrya M N <showrya@chelsio.com>
> Signed-off-by: Potnuri Bharat Teja <bharat@chelsio.com>
> ---
>   drivers/infiniband/sw/siw/siw.h      |  2 --
>   drivers/infiniband/sw/siw/siw_main.c | 37 ----------------------------
>   2 files changed, 39 deletions(-)
> 
> diff --git a/drivers/infiniband/sw/siw/siw.h b/drivers/infiniband/sw/siw/siw.h
> index 75253f2b3e3d..86d4d6a2170e 100644
> --- a/drivers/infiniband/sw/siw/siw.h
> +++ b/drivers/infiniband/sw/siw/siw.h
> @@ -94,8 +94,6 @@ struct siw_device {
>   	atomic_t num_mr;
>   	atomic_t num_srq;
>   	atomic_t num_ctx;
> -
> -	struct work_struct netdev_down;
>   };
>   
>   struct siw_ucontext {
> diff --git a/drivers/infiniband/sw/siw/siw_main.c b/drivers/infiniband/sw/siw/siw_main.c
> index 61ad8ca3d1a2..9a50a9dcce39 100644
> --- a/drivers/infiniband/sw/siw/siw_main.c
> +++ b/drivers/infiniband/sw/siw/siw_main.c
> @@ -364,39 +364,6 @@ static struct siw_device *siw_device_create(struct net_device *netdev)
>   	return NULL;
>   }
>   
> -/*
> - * Network link becomes unavailable. Mark all
> - * affected QP's accordingly.
> - */
> -static void siw_netdev_down(struct work_struct *work)
> -{
> -	struct siw_device *sdev =
> -		container_of(work, struct siw_device, netdev_down);
> -
> -	struct siw_qp_attrs qp_attrs;
> -	struct list_head *pos, *tmp;
> -
> -	memset(&qp_attrs, 0, sizeof(qp_attrs));
> -	qp_attrs.state = SIW_QP_STATE_ERROR;
> -
> -	list_for_each_safe(pos, tmp, &sdev->qp_list) {
> -		struct siw_qp *qp = list_entry(pos, struct siw_qp, devq);
> -
> -		down_write(&qp->state_lock);
> -		WARN_ON(siw_qp_modify(qp, &qp_attrs, SIW_QP_ATTR_STATE));
> -		up_write(&qp->state_lock);
> -	}
> -	ib_device_put(&sdev->base_dev);
> -}
> -
> -static void siw_device_goes_down(struct siw_device *sdev)
> -{
> -	if (ib_device_try_get(&sdev->base_dev)) {
> -		INIT_WORK(&sdev->netdev_down, siw_netdev_down);
> -		schedule_work(&sdev->netdev_down);
> -	}
> -}
> -
>   static int siw_netdev_event(struct notifier_block *nb, unsigned long event,
>   			    void *arg)
>   {
> @@ -418,10 +385,6 @@ static int siw_netdev_event(struct notifier_block *nb, unsigned long event,
>   		siw_port_event(sdev, 1, IB_EVENT_PORT_ACTIVE);
>   		break;
>   
> -	case NETDEV_GOING_DOWN:
> -		siw_device_goes_down(sdev);
> -		break;
> -
>   	case NETDEV_DOWN:
>   		sdev->state = IB_PORT_DOWN;
>   		siw_port_event(sdev, 1, IB_EVENT_PORT_ERR);
Leon Romanovsky July 29, 2024, 8:47 a.m. UTC | #2 
On Wed, 24 Jul 2024 14:24:28 +0530, Showrya M N wrote:
> Toggling link while running NVME-oF over siw hits a kernel panic
> due to race condition within siw_handler and ib_destroy_qp().
> The IB_EVENT_PORT_ERR event can alone handle destroying qps.
> therefore remove unwanted processing in siw.
> 
> 

Applied, thanks!

[1/1] RDMA/siw: Remove NETDEV_GOING_DOWN event handler
      https://git.kernel.org/rdma/rdma/c/60dc7fcafea817

Best regards,
diff mbox series

Patch

diff --git a/drivers/infiniband/sw/siw/siw.h b/drivers/infiniband/sw/siw/siw.h
index 75253f2b3e3d..86d4d6a2170e 100644
--- a/drivers/infiniband/sw/siw/siw.h
+++ b/drivers/infiniband/sw/siw/siw.h
@@ -94,8 +94,6 @@  struct siw_device {
 	atomic_t num_mr;
 	atomic_t num_srq;
 	atomic_t num_ctx;
-
-	struct work_struct netdev_down;
 };
 
 struct siw_ucontext {
diff --git a/drivers/infiniband/sw/siw/siw_main.c b/drivers/infiniband/sw/siw/siw_main.c
index 61ad8ca3d1a2..9a50a9dcce39 100644
--- a/drivers/infiniband/sw/siw/siw_main.c
+++ b/drivers/infiniband/sw/siw/siw_main.c
@@ -364,39 +364,6 @@  static struct siw_device *siw_device_create(struct net_device *netdev)
 	return NULL;
 }
 
-/*
- * Network link becomes unavailable. Mark all
- * affected QP's accordingly.
- */
-static void siw_netdev_down(struct work_struct *work)
-{
-	struct siw_device *sdev =
-		container_of(work, struct siw_device, netdev_down);
-
-	struct siw_qp_attrs qp_attrs;
-	struct list_head *pos, *tmp;
-
-	memset(&qp_attrs, 0, sizeof(qp_attrs));
-	qp_attrs.state = SIW_QP_STATE_ERROR;
-
-	list_for_each_safe(pos, tmp, &sdev->qp_list) {
-		struct siw_qp *qp = list_entry(pos, struct siw_qp, devq);
-
-		down_write(&qp->state_lock);
-		WARN_ON(siw_qp_modify(qp, &qp_attrs, SIW_QP_ATTR_STATE));
-		up_write(&qp->state_lock);
-	}
-	ib_device_put(&sdev->base_dev);
-}
-
-static void siw_device_goes_down(struct siw_device *sdev)
-{
-	if (ib_device_try_get(&sdev->base_dev)) {
-		INIT_WORK(&sdev->netdev_down, siw_netdev_down);
-		schedule_work(&sdev->netdev_down);
-	}
-}
-
 static int siw_netdev_event(struct notifier_block *nb, unsigned long event,
 			    void *arg)
 {
@@ -418,10 +385,6 @@  static int siw_netdev_event(struct notifier_block *nb, unsigned long event,
 		siw_port_event(sdev, 1, IB_EVENT_PORT_ACTIVE);
 		break;
 
-	case NETDEV_GOING_DOWN:
-		siw_device_goes_down(sdev);
-		break;
-
 	case NETDEV_DOWN:
 		sdev->state = IB_PORT_DOWN;
 		siw_port_event(sdev, 1, IB_EVENT_PORT_ERR);