Re: [PATCH net-next v3 3/3] net: icmp: add reasons of the skb drops to icmp protocol

[Menglong Dong <menglong8.dong@gmail.com>]

On Fri, Mar 18, 2022 at 12:10 PM David Ahern <dsahern@kernel.org> wrote:
>
> On 3/17/22 7:37 PM, Menglong Dong wrote:
> > On Thu, Mar 17, 2022 at 10:48 PM David Ahern <dsahern@kernel.org> wrote:
> >>
> >> On 3/16/22 10:05 PM, Jakub Kicinski wrote:
> >>> On Wed, 16 Mar 2022 21:35:47 -0600 David Ahern wrote:
> >>>> On 3/16/22 9:18 PM, Jakub Kicinski wrote:
> >>>>>
> >>>>> I guess this set raises the follow up question to Dave if adding
> >>>>> drop reasons to places with MIB exception stats means improving
> >>>>> the granularity or one MIB stat == one reason?
> >>>>
> >>>> There are a few examples where multiple MIB stats are bumped on a drop,
> >>>> but the reason code should always be set based on first failure. Did you
> >>>> mean something else with your question?
> >>>
> >>> I meant whether we want to differentiate between TYPE, and BROADCAST or
> >>> whatever other possible invalid protocol cases we can get here or just
> >>> dump them all into a single protocol error code.
> >>
> >> I think a single one is a good starting point.
> >
> > Ok, I'll try my best to make a V4 base this way...Is there any inspiration?
> >
> > Such as we make SKB_DROP_REASON_PTYPE_ABSENT to
> > SKB_DROP_REASON_L2_PROTO, which means the L2 protocol is not
> > supported or invalied.
>
> not following. PTYPE is a Linux name. That means nothing to a user.
>
> I am not sure where you want to use L2_PROTO.

Yeah, PTYPE seems not suitable. I mean that replace SKB_DROP_REASON_PTYPE_ABSENT
that is used in __netif_receive_skb_core() with L3_PROTO, which means no L3
protocol handler (or other device handler) is not found for the
packet. This seems more
friendly and not code based.

>
> >
> > And use SKB_DROP_REASON_L4_PROTO for the L4 protocol problem,
> > such as GRE version not supported, ICMP type not supported, etc.

Is this L4_PROTO followed by anyone?

Thanks!
Menglong Dong

> >
> > Sounds nice, isn't it?
>