Re: [dpdk-dev] [PATCH v2] i40e: fix i40evf_add_mac_addr to permit multicast addresses

["David Harton (dharton)" <dharton@cisco.com>]

Hi Beilei,

> -----Original Message-----
> From: Xing, Beilei [mailto:beilei.xing@intel.com]
> 
> Hi Harton,
> 
> > -----Original Message-----
> > From: David Harton [mailto:dharton@cpp-rtpbld-31.cpprtplab]
> >
> > From: David Harton <dharton@cisco.com>
> >
> > The i40e maintains a single MAC filter table for both unicast and
> > multicast addresses.  The i40e_validate_mac_addr function was
> > preventing multicast addresses from being added to the table via
> > i40evf_add_mac_addr.  Fixed the issue by adjusting the check in
> i40evf_add_mac_addr.
> >
> > Fixes: 4861cde46116 ("i40e: new poll mode driver")
> > Fixes: 97ac72aa71a9 ("i40e: support setting VF MAC address")
> >
> > Signed-off-by: David Harton <dharton@cisco.com>
> > ---
> >
> > v2
> > * Removed multicast check in i40evf_add_mac_addr.
> >
> > v1
> > * Removed multicast check in i40e_validate_mac_addr.
> >
> >  drivers/net/i40e/i40e_ethdev_vf.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/drivers/net/i40e/i40e_ethdev_vf.c
> > b/drivers/net/i40e/i40e_ethdev_vf.c
> > index f6d8293..5916d11 100644
> > --- a/drivers/net/i40e/i40e_ethdev_vf.c
> > +++ b/drivers/net/i40e/i40e_ethdev_vf.c
> > @@ -888,7 +888,7 @@ i40evf_add_mac_addr(struct rte_eth_dev *dev,
> >  	int err;
> >  	struct vf_cmd_info args;
> >
> > -	if (i40e_validate_mac_addr(addr->addr_bytes) != I40E_SUCCESS) {
> > +	if (is_zero_ether_addr(addr) != I40E_SUCCESS) {
> 
> Thanks for the patch, there's some mistake with my last comment.

Actually, I think the logic above works but I do agree it is confusing
but I wanted to use your suggestion to speed up the process. :)

> * @return
>  *   True  (1) if the given ethernet address is filled with zeros;
>  *   false (0) otherwise.
>  */
> According to the comment of is_zero_ether_addr function, return value
> should be 0 or 1.
> So you should use if (!is_zero_ether_addr(addr))  here.

I don't think so.  I think the logic that should be used is
if (is_zero_ether_addr(addr)) {
because we want to report an error if the addr is all 0's.

Please let me know if you agree and I'll post this patch instead.

Thanks,
Dave

> 
> >  		PMD_DRV_LOG(ERR, "Invalid mac:%x:%x:%x:%x:%x:%x",
> >  			    addr->addr_bytes[0], addr->addr_bytes[1],
> >  			    addr->addr_bytes[2], addr->addr_bytes[3],
> > --
> > 2.10.3.dirty