APPLIED: [PATCH] net/mlx5: Avoid dereferencing uninitialized pointer

[Tim Gardner]

On 03/29/2017 04:49 AM, Talat Batheesh wrote:
> BugLink: http://bugs.launchpad.net/bugs/1676786
> 
> In NETDEV_CHANGEUPPER event the upper_info field is valid
> only when linking is true. Otherwise it should be ignored.
> 
> Fixes: 7907f23adc18 (net/mlx5: Implement RoCE LAG feature)
> Signed-off-by: Talat Batheesh <talatb at mellanox.com>
> Reviewed-by: Aviv Heller <avivh at mellanox.com>
> Reviewed-by: Moni Shoua <monis at mellanox.com>
> Signed-off-by: Saeed Mahameed <saeedm at mellanox.com>
> Signed-off-by: David S. Miller <davem at davemloft.net>
> (cherry picked from commit e497ec680c4cd51e76bfcdd49363d9ab8d32a757)
> ---
>  drivers/net/ethernet/mellanox/mlx5/core/lag.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag.c b/drivers/net/ethernet/mellanox/mlx5/core/lag.c
> index 5595724..b5d5519 100644
> --- a/drivers/net/ethernet/mellanox/mlx5/core/lag.c
> +++ b/drivers/net/ethernet/mellanox/mlx5/core/lag.c
> @@ -294,7 +294,7 @@ static int mlx5_handle_changeupper_event(struct mlx5_lag *ldev,
>  					 struct netdev_notifier_changeupper_info *info)
>  {
>  	struct net_device *upper = info->upper_dev, *ndev_tmp;
> -	struct netdev_lag_upper_info *lag_upper_info;
> +	struct netdev_lag_upper_info *lag_upper_info = NULL;
>  	bool is_bonded;
>  	int bond_status = 0;
>  	int num_slaves = 0;
> @@ -303,7 +303,8 @@ static int mlx5_handle_changeupper_event(struct mlx5_lag *ldev,
>  	if (!netif_is_lag_master(upper))
>  		return 0;
>  
> -	lag_upper_info = info->upper_info;
> +	if (info->linking)
> +		lag_upper_info = info->upper_info;
>  
>  	/* The event may still be of interest if the slave does not belong to
>  	 * us, but is enslaved to a master which has one or more of our netdevs
> 

I can't find this patch in linux or linux-next so I'm marking it SAUCE.
A patch is 'cherry picked' only when it comes from Linus' repository, or
sometimes linux-next.

-- 
Tim Gardner tim.gardner at canonical.com