On Mon, Jul 04, 2005 at 12:59:10AM -0500, Mike Silbersack wrote:
> On Mon, 4 Jul 2005, Andrew Thompson wrote:
> >Both bridges tap their packets off near the end of ether_input(). Since
> >we only need to check alignment when using pfil(9) and each bridge has
> >its own knobs to control filtering, it seems best to leave it where it
> >is at the moment.
> Well, where it was at was the network drivers. Since we know that em is
> the (only?) violator, you're arguing for the backout of your patch when
> you say "leave it where it is."
> The other protocol stacks need aligned headers too, that's why I'm curious
> as to the bridge-specificness of this patch.
Because I have only made a handful of commits so far and they were all
to if_bridge, so im a little narrow minded at the moment ;-)
If there is a better way to do this im all for it.