On Sat Nov 14, 2020 at 5:44 PM CET, Andrew Lunn wrote: > > > > + * > > > > + * A 3-bit code is used to relay why a particular frame was sent to > > > > + * the CPU. We only use this to determine if the packet was mirrored > > > > + * or trapped, i.e. whether the packet has been forwarded by hardware > > > > + * or not. > > > > > > Maybe add that, not all generations support all codes. > > > > Not sure I have that information. > > I'm not asking you list per code which switches support it. I'm just > think we should add a warning, it cannot be assumed all switches > support all codes. I just looked at the 6161 for example, and it is > missing 5 from its list.
I see, yeah sure I can do that. > > That leaves us with DSA_CODE_IGMP_MLD_TRAP. Here is the problem: > > > > Currenly, tag_dsa.c will set skb->offload_fwd_mark for IGMP/MLD > > packets, whereas tag_edsa.c will exempt them. So we can not unify the > > two without changing the behavior of one. > > I'm not saying that this change is wrong. I'm just afraid as a > behaviour change, it might break something. If something does break, > it will be easier to track down, if it is a change on its own. So > please look if we can add a simple patch to tag_dsa.c which removes > the marking of such frames. And then the next patch can combine the > two into one driver. If it does break, git bisect will then tell us > which patch broke it. Ahh, I think I see what you are saying now. So I would copy the IGMP/MLD exemption from tag_edsa.c to tag_dsa.c first, and then apply the unify patch. Yeah that makes alot of sense, will do!
