ndo_get_stats64() can race with ndo_stop(), which frees input and
output queue resources. Call synchronize_net() to avoid such races.

Fixes: 6a610a46bad1 ("octeon_ep: add support for ndo ops")
Signed-off-by: Shinas Rasheed <srasheed@marvell.com>
---
V3: 
  - No changes 

V2: https://lore.kernel.org/all/20241216075842.2394606-2-srasheed@marvell.com/
  - Changed sync mechanism to fix race conditions from using an atomic
    set_bit ops to a much simpler synchronize_net()

V1: https://lore.kernel.org/all/20241203072130.2316913-2-srasheed@marvell.com/

 drivers/net/ethernet/marvell/octeon_ep/octep_main.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
index 549436efc204..941bbaaa67b5 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
@@ -757,6 +757,7 @@ static int octep_stop(struct net_device *netdev)
 {
 	struct octep_device *oct = netdev_priv(netdev);
 
+	synchronize_net();
 	netdev_info(netdev, "Stopping the device ...\n");
 
 	octep_ctrl_net_set_link_status(oct, OCTEP_CTRL_NET_INVALID_VFID, false,
-- 
2.25.1

On Wed, 18 Dec 2024 03:51:08 -0800 Shinas Rasheed wrote:
> ndo_get_stats64() can race with ndo_stop(), which frees input and
> output queue resources. Call synchronize_net() to avoid such races.

synchronize_rcu() acts as a barrier.
What are the two operations you are separating with this barrier?

> diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
> index 549436efc204..941bbaaa67b5 100644
> --- a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
> +++ b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c
> @@ -757,6 +757,7 @@ static int octep_stop(struct net_device *netdev)
>  {
>  	struct octep_device *oct = netdev_priv(netdev);
>  
> +	synchronize_net();
>  	netdev_info(netdev, "Stopping the device ...\n");