[dpdk-dev] [PATCH] ethdev: fix setting of MAC address
Igor Ryzhov
iryzhov at nfware.com
Mon Dec 18 11:53:47 CET 2017
Hello Andrew,
Don't you think that it's not correct that net/sfc works that way?
If we go further, dev->dev_ops->mac_addr_set not only should be called
before ether_addr_copy.
It should return status code, and in case of error ether_addr_copy
shouldn't be called at all.
Am I wrong?
Best regards,
Igor
On Mon, Dec 18, 2017 at 1:35 PM, Andrew Rybchenko <arybchenko at solarflare.com
> wrote:
> On 12/14/2017 08:15 PM, Olivier Matz wrote:
>
>> From: Laurent Hardy <laurent.hardy at 6wind.com>
>>
>> When a new mac address is set, it is saved in dev->data->mac_addrs
>> before the ethdev handler is called.
>>
>> First, it is inconsistent with the other ethdev functions
>> rte_eth_dev_mac_addr_remove() and rte_eth_dev_mac_addr_add().
>>
>> Moreover, it prevents the drivers from wrongly comparing the old address
>> and the new one, like it's done in i40evf driver:
>>
>> if (is_same_ether_addr(mac_addr, dev->data->mac_addrs))
>> return;
>>
>> Fixes: 943c2d899a0c ("net/i40e: set VF MAC from VF")
>> Fixes: 854d8ad4ef68 ("ethdev: add default mac address modifier")
>> Cc: stable at dpdk.org
>>
>> Signed-off-by: Laurent Hardy <laurent.hardy at 6wind.com>
>> ---
>> lib/librte_ether/rte_ethdev.c | 4 ++--
>> 1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.
>> c
>> index 4f492e3db..297c02a54 100644
>> --- a/lib/librte_ether/rte_ethdev.c
>> +++ b/lib/librte_ether/rte_ethdev.c
>> @@ -2643,11 +2643,11 @@ rte_eth_dev_default_mac_addr_set(uint16_t
>> port_id, struct ether_addr *addr)
>> dev = &rte_eth_devices[port_id];
>> RTE_FUNC_PTR_OR_ERR_RET(*dev->dev_ops->mac_addr_set, -ENOTSUP);
>> + (*dev->dev_ops->mac_addr_set)(dev, addr);
>> +
>> /* Update default address in NIC data structure */
>> ether_addr_copy(addr, &dev->data->mac_addrs[0]);
>> - (*dev->dev_ops->mac_addr_set)(dev, addr);
>> -
>> return 0;
>> }
>>
>
> NACK, unfortunately it will break net/sfc in one of branches when a new MAC
> is set using restart. It relies on the fact that a new MAC is already
> available in
> device data.
>
More information about the dev
mailing list