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@xxxxxxxx>
> Cc: Ding Hui <dinghui@xxxxxxxxxxxxxx>
> Cc: Shifeng Li <lishifeng1992@xxxxxxx>
> Signed-off-by: Shifeng Li <lishifeng@xxxxxxxxxxxxxx>
> ---
> drivers/infiniband/core/device.c | 33 +++++++++++++++++---------------
> 1 file changed, 18 insertions(+), 15 deletions(-)
Applied to for-next, thanks
Jason
