On Fri, Jan 06, 2012 at 01:20:58PM -0800, Jesse Gross wrote: > On Mon, Jan 2, 2012 at 9:52 AM, Ben Pfaff <b...@nicira.com> wrote: > > On Fri, Dec 30, 2011 at 12:52:14AM -0500, Jesse Gross wrote: > >> It seems like a bigger problem that this field isn't initialized than > >> just for OVS. ??Should we be proposing to fix AF_PACKET upstream? > > > > I don't know. ??I don't know what the use case is for setting the > > protocol explicitly. ??Maybe it is useful for non-Ethernet media? > > Maybe it is intended only for use with non-"raw" sockets? > > I think it's probably primarily used by non-raw sockets since it's > required in that case. However, it does appear that the intended use > case for it is to initialize skb->protocol. I found this patch from > Patrick McHardy to initialize it for dhcp and he reported that > otherwise packets don't hit filters that look at protocol, as > expected: > http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=474000 > > > I don't know what such a fix would look like. ??AF_PACKET is supposed > > to work with various media, but there's no header_op for parsing a > > packet's protocol, so we'd have to figure out what the right thing to > > do is there. ??Would we need a new header_op? > > My guess is that it was implemented this way to be protocol > independent but it doesn't seem like the right way to go about it, > since you know what device you're sending on and the device should > know how to extract the protocol from it. I think a new header_op > along the lines of dev_parse_header() makes the most sense.
I agree on all points. Let me know if you want me to work on any of this. _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev
