On Mon, 07 Mar 2016 17:06:20 +0000 Remy Horton <remy.horton at intel.com> wrote:
> > > On 07/03/2016 16:07, Stephen Hemminger wrote: > >> That version doesn't include a MAC address sanity check - is such a > >> check redundant (or incorrect) in this case? > >> > >> ..Remy > > > > That belongs in common code, not in each driver. > > That differs from previous feedback for a different driver, where such a > check was actually asked to be added.. :/ > > Probably need to audit the upstream functions for such checks. > > ..Remy Probably need checks in both places. Zero is bad for any device, some devices allow multicast addresses as MAC, and some clustering seemed to want it.
