On 8/2/2020 10:44 PM, Michael Grzeschik wrote:
> To get the driver working with other chips using different port counts
> the dyn_mac_table should be flushed depending on the amount of physical
> ports.
> 
> Signed-off-by: Michael Grzeschik <m.grzesc...@pengutronix.de>
> 
> ---
> v1 -> v4: - extracted this change from bigger previous patch
> 
>  drivers/net/dsa/microchip/ksz8795.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/net/dsa/microchip/ksz8795.c 
> b/drivers/net/dsa/microchip/ksz8795.c
> index 947ea1e45f5b2c6..ba722f730bf0f7b 100644
> --- a/drivers/net/dsa/microchip/ksz8795.c
> +++ b/drivers/net/dsa/microchip/ksz8795.c
> @@ -750,11 +750,11 @@ static void ksz8795_port_stp_state_set(struct 
> dsa_switch *ds, int port,
>  
>  static void ksz8795_flush_dyn_mac_table(struct ksz_device *dev, int port)
>  {
> -     u8 learn[TOTAL_PORT_NUM];
> +     u8 *learn = kzalloc(dev->mib_port_cnt, GFP_KERNEL);

There is no error checking on this allocation, and you could consider
propagating an allocation failure back to the caller for graceful handling.
-- 
Florian

Reply via email to