Received-SPF: pass (google.com: domain of linux-kernel+bounces-75180-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) client-ip=2604:1380:45d1:ec00::1;
Date: Wed, 21 Feb 2024 13:17:05 -0400
From: Jason Gunthorpe <jgg@ziepe.ca>
To: Shifeng Li <lishifeng@sangfor.com.cn>
Cc: leon@kernel.org, wenglianfa@huawei.com, gustavoars@kernel.org,
	linux-rdma@vger.kernel.org, linux-kernel@vger.kernel.org,
	Ding Hui <dinghui@sangfor.com.cn>,
	Shifeng Li <lishifeng1992@126.com>
Subject: Re: [PATCH v2] RDMA/device: Fix a race between mad_client and
 cm_client init
Message-ID: <20240221171705.GE13491@ziepe.ca>
References: <20240203035313.98991-1-lishifeng@sangfor.com.cn>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20240203035313.98991-1-lishifeng@sangfor.com.cn>


On Fri, Feb 02, 2024 at 07:53:13PM -0800, Shifeng Li wrote:
> The mad_client will be initialized in enable_device_and_get(), while the
> devices_rwsem will be downgraded to a read semaphore. There is a window
> that leads to the failed initialization for cm_client, since it can not
> get matched mad port from ib_mad_port_list, and the matched mad port will
> be added to the list after that.
> 
>     mad_client    |                       cm_client
> ------------------|--------------------------------------------------------
> ib_register_device|
> enable_device_and_get
> down_write(&devices_rwsem)
> xa_set_mark(&devices, DEVICE_REGISTERED)
> downgrade_write(&devices_rwsem)
>                   |
>                   |ib_cm_init
>                   |ib_register_client(&cm_client)
>                   |down_read(&devices_rwsem)
>                   |xa_for_each_marked (&devices, DEVICE_REGISTERED)
>                   |add_client_context
>                   |cm_add_one
>                   |ib_register_mad_agent
>                   |ib_get_mad_port
>                   |__ib_get_mad_port
>                   |list_for_each_entry(entry, &ib_mad_port_list, port_list)
>                   |return NULL
>                   |up_read(&devices_rwsem)
>                   |
> add_client_context|
> ib_mad_init_device|
> ib_mad_port_open  |
> list_add_tail(&port_priv->port_list, &ib_mad_port_list)
> up_read(&devices_rwsem)
>                   |
> 
> Fix it by using down_write(&devices_rwsem) in ib_register_client().
> 
> Fixes: d0899892edd0 ("RDMA/device: Provide APIs from the core code to help unregistration")
> Suggested-by: Jason Gunthorpe <jgg@ziepe.ca>
> Cc: Ding Hui <dinghui@sangfor.com.cn>
> Cc: Shifeng Li <lishifeng1992@126.com>
> Signed-off-by: Shifeng Li <lishifeng@sangfor.com.cn>
> ---
>  drivers/infiniband/core/device.c | 33 +++++++++++++++++---------------
>  1 file changed, 18 insertions(+), 15 deletions(-)

Applied to for-next, thanks

Jason