[bpf-next,v4,0/9] net: ip: add drop reasons to input route

Message

Menglong Dong Oct. 24, 2024, 9:33 a.m. UTC

In this series, we mainly add some skb drop reasons to the input path of
ip routing, and we make the following functions return drop reasons:

  fib_validate_source()
  ip_route_input_mc()
  ip_mc_validate_source()
  ip_route_input_slow()
  ip_route_input_rcu()
  ip_route_input_noref()
  ip_route_input()
  ip_mkroute_input()
  __mkroute_input()
  ip_route_use_hint()

And following new skb drop reasons are added:

  SKB_DROP_REASON_IP_LOCAL_SOURCE
  SKB_DROP_REASON_IP_INVALID_SOURCE
  SKB_DROP_REASON_IP_LOCALNET
  SKB_DROP_REASON_IP_INVALID_DEST

Changes since v3:
- don't refactor fib_validate_source/__fib_validate_source, and introduce
  a wrapper for fib_validate_source() instead in the 1st patch.
- some small adjustment in the 4-7 patches

Changes since v2:
- refactor fib_validate_source and __fib_validate_source to make
  fib_validate_source return drop reasons
- add the 9th and 10th patches to make this series cover the input route
  code path

Changes since v1:
- make ip_route_input_noref/ip_route_input_rcu/ip_route_input_slow return
  drop reasons, instead of passing a local variable to their function
  arguments.

Menglong Dong (9):
  net: ip: make fib_validate_source() support drop reasons
  net: ip: make ip_route_input_mc() return drop reason
  net: ip: make ip_mc_validate_source() return drop reason
  net: ip: make ip_route_input_slow() return drop reasons
  net: ip: make ip_route_input_rcu() return drop reasons
  net: ip: make ip_route_input_noref() return drop reasons
  net: ip: make ip_route_input() return drop reasons
  net: ip: make ip_mkroute_input/__mkroute_input return drop reasons
  net: ip: make ip_route_use_hint() return drop reasons

 include/net/dropreason-core.h   |  26 ++++
 include/net/ip_fib.h            |  12 ++
 include/net/route.h             |  34 +++---
 net/bridge/br_netfilter_hooks.c |  11 +-
 net/core/lwt_bpf.c              |   6 +-
 net/ipv4/fib_frontend.c         |  17 ++-
 net/ipv4/icmp.c                 |   2 +-
 net/ipv4/ip_fragment.c          |  12 +-
 net/ipv4/ip_input.c             |  20 ++-
 net/ipv4/ip_options.c           |   2 +-
 net/ipv4/route.c                | 210 +++++++++++++++++++-------------
 net/ipv6/seg6_local.c           |  14 +--
 12 files changed, 226 insertions(+), 140 deletions(-)

Comments

Jakub Kicinski Oct. 30, 2024, 12:03 a.m. UTC | #1

On Thu, 24 Oct 2024 17:33:39 +0800 Menglong Dong wrote:
> In this series, we mainly add some skb drop reasons to the input path of
> ip routing, and we make the following functions return drop reasons:
> 
>   fib_validate_source()
>   ip_route_input_mc()
>   ip_mc_validate_source()
>   ip_route_input_slow()
>   ip_route_input_rcu()
>   ip_route_input_noref()
>   ip_route_input()
>   ip_mkroute_input()
>   __mkroute_input()
>   ip_route_use_hint()
> 
> And following new skb drop reasons are added:
> 
>   SKB_DROP_REASON_IP_LOCAL_SOURCE
>   SKB_DROP_REASON_IP_INVALID_SOURCE
>   SKB_DROP_REASON_IP_LOCALNET
>   SKB_DROP_REASON_IP_INVALID_DEST

We're "a bit" behind on patches after my vacation, so no real review
here, but please repost with net-next in the subject. The test
automation trusts the tree designation and bpf-next is no longer
based on net-next. So this doesn't apply.

Menglong Dong Oct. 30, 2024, 1:35 a.m. UTC | #2

On Wed, Oct 30, 2024 at 8:03 AM Jakub Kicinski <[e-mail]> wrote:
>
> On Thu, 24 Oct 2024 17:33:39 +0800 Menglong Dong wrote:
> > In this series, we mainly add some skb drop reasons to the input path of
> > ip routing, and we make the following functions return drop reasons:
> >
> >   fib_validate_source()
> >   ip_route_input_mc()
> >   ip_mc_validate_source()
> >   ip_route_input_slow()
> >   ip_route_input_rcu()
> >   ip_route_input_noref()
> >   ip_route_input()
> >   ip_mkroute_input()
> >   __mkroute_input()
> >   ip_route_use_hint()
> >
> > And following new skb drop reasons are added:
> >
> >   SKB_DROP_REASON_IP_LOCAL_SOURCE
> >   SKB_DROP_REASON_IP_INVALID_SOURCE
> >   SKB_DROP_REASON_IP_LOCALNET
> >   SKB_DROP_REASON_IP_INVALID_DEST
>
> We're "a bit" behind on patches after my vacation, so no real review
> here, but please repost with net-next in the subject. The test
> automation trusts the tree designation and bpf-next is no longer
> based on net-next. So this doesn't apply.

I was wondering how the conflict, which was checked by bpf-ci,
happened, as there was no conflict between this series and
net-next. And now I see, I just tagged a wrong branch for this
series. Sorry about that, and I'll resend it to the right branch.

Thanks!
Menglong Dong


> --
> pw-bot: cr