[PATCH net-next v2 2/5] net: xilinx: axienet: Fix dangling multicast addresses

[Sean Anderson]

On 8/19/24 21:33, Jakub Kicinski wrote:
> On Thu, 15 Aug 2024 15:36:11 -0400 Sean Anderson wrote:
>> If a multicast address is removed but there are still some multicast
>> addresses, that address would remain programmed into the frame filter.
>> Fix this by explicitly setting the enable bit for each filter.
>> 
>> Fixes: 8a3b7a252dca ("drivers/net/ethernet/xilinx: added Xilinx AXI Ethernet driver")
>> Signed-off-by: Sean Anderson <sean.anderson at linux.dev>
>> Reviewed-by: Simon Horman <horms at kernel.org>
> 
> Again, I'd go for net.
> 
>> diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet.h b/drivers/net/ethernet/xilinx/xilinx_axienet.h
>> index 0d5b300107e0..03fef656478e 100644
>> --- a/drivers/net/ethernet/xilinx/xilinx_axienet.h
>> +++ b/drivers/net/ethernet/xilinx/xilinx_axienet.h
>> @@ -170,6 +170,7 @@
>>  #define XAE_UAW0_OFFSET		0x00000700 /* Unicast address word 0 */
>>  #define XAE_UAW1_OFFSET		0x00000704 /* Unicast address word 1 */
>>  #define XAE_FMI_OFFSET		0x00000708 /* Filter Mask Index */
>> +#define XAE_FFE_OFFSET		0x0000070C /* Frame Filter Enable */
>>  #define XAE_AF0_OFFSET		0x00000710 /* Address Filter 0 */
>>  #define XAE_AF1_OFFSET		0x00000714 /* Address Filter 1 */
> 
> There is a conflict with current net / net-next here, because of
> 9ff2f816e2aa65ca9, you'll need to rebase / repost (which is why 
> I'm allowing myself the nit picks ;))
> 
>> diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
>> index e664611c29cf..1bcabb016ca9 100644
>> --- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
>> +++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
>> @@ -433,7 +433,7 @@ static int netdev_set_mac_address(struct net_device *ndev, void *p)
>>   */
>>  static void axienet_set_multicast_list(struct net_device *ndev)
>>  {
>> -	int i;
>> +	int i = 0;
> 
> Consider renaming i to addr_cnt ? or addr_num ?

Well, this doesn't really have anything to do with addresses. It selects
the "Filter Index" (as named by the datasheet) so I'd rename it `filter`
if anything.

This hardware is actually really unusual since the CAM acts on the
entire first 64 bytes of the packet. So you could theoretically filter
for the source address too, but I don't know why you'd want to since all
you can do is drop the packet. Seems like a big waste of resources to
me.

--Sean