public inbox for netdev@vger.kernel.org 
 help / color / mirror / Atom feed
From: Jakub Sitnicki <jakub@cloudflare•com>
To: "Alexei Starovoitov" <alexei.starovoitov@gmail•com>,
	"Toke Høiland-Jørgensen" <toke@redhat•com>,
	"Arthur Fabre" <arthur@arthurfabre•com>
Cc: Network Development <netdev@vger•kernel.org>,
	 bpf <bpf@vger•kernel.org>,
	 Jesper Dangaard Brouer <hawk@kernel•org>,
	 Yan Zhai <yan@cloudflare•com>,
	 jbrandeburg@cloudflare•com, lbiancon@redhat•com,
	 Alexei Starovoitov <ast@kernel•org>,
	 Jakub Kicinski <kuba@kernel•org>,
	 Eric Dumazet <edumazet@google•com>,
	kernel-team@cloudflare•com
Subject: Re: [PATCH RFC bpf-next v2 01/17] trait: limited KV store for packet metadata
Date: Wed, 30 Apr 2025 21:19:51 +0200	[thread overview]
Message-ID: <87ikmle9t4.fsf@cloudflare.com> (raw)
In-Reply-To: <87frhqnh0e.fsf@toke.dk> ("Toke Høiland-Jørgensen"'s message of "Wed, 30 Apr 2025 11:19:29 +0200")

On Wed, Apr 30, 2025 at 11:19 AM +02, Toke Høiland-Jørgensen wrote:
> Alexei Starovoitov <alexei.starovoitov@gmail•com> writes:
>
>> On Fri, Apr 25, 2025 at 12:27 PM Arthur Fabre <arthur@arthurfabre•com> wrote:
>>>
>>> On Thu Apr 24, 2025 at 6:22 PM CEST, Alexei Starovoitov wrote:
>>> > On Tue, Apr 22, 2025 at 6:23 AM Arthur Fabre <arthur@arthurfabre•com> wrote:

[...]

>>> * Hardware metadata: metadata exposed from NICs (like the receive
>>>   timestamp, 4 tuple hash...) is currently only exposed to XDP programs
>>>   (via kfuncs).
>>>   But that doesn't expose them to the rest of the stack.
>>>   Storing them in traits would allow XDP, other BPF programs, and the
>>>   kernel to access and modify them (for example to into account
>>>   decapsulating a packet).
>>
>> Sure. If traits == existing metadata bpf prog in xdp can communicate
>> with bpf prog in skb layer via that "trait" format.
>> xdp can take tuple hash and store it as key==0 in the trait.
>> The kernel doesn't need to know how to parse that format.
>
> Yes it does, to propagate it to the skb later. I.e.,
>
> XDP prog on NIC: get HW hash, store in traits, redirect to CPUMAP
> CPUMAP: build skb, read hash from traits, populate skb hash
>
> Same thing for (at least) timestamps and checksums.
>
> Longer term, with traits available we could move more skb fields into
> traits to make struct sk_buff smaller (by moving optional fields to
> traits that don't take up any space if they're not set).

Perhaps we can have the cake and eat it too.

We could leave the traits encoding/decoding out of the kernel and, at
the same time, *expose it* to the network stack through BPF struct_ops
programs. At a high level, for example ->get_rx_hash(), not the
individual K/V access. The traits_ops vtable could grow as needed to
support new use cases.

If you think about it, it's not so different from BPF-powered congestion
algorithms and scheduler extensions. They also expose some state, kept in
maps, that only the loaded BPF code knows how to operate on.

  parent reply	other threads:[~2025-04-30 19:19 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-04-22 13:23 [PATCH RFC bpf-next v2 00/17] traits: Per packet metadata KV store Arthur Fabre
2025-04-22 13:23 ` [PATCH RFC bpf-next v2 01/17] trait: limited KV store for packet metadata Arthur Fabre
2025-04-24 16:22   ` Alexei Starovoitov
2025-04-25 19:26     ` Arthur Fabre
2025-04-29 23:36       ` Alexei Starovoitov
2025-04-30  9:19         ` Toke Høiland-Jørgensen
2025-04-30 16:29           ` Alexei Starovoitov
2025-05-01  7:30             ` Arthur Fabre
2025-04-30 19:19           ` Jakub Sitnicki [this message]
2025-05-01 10:43             ` Toke Høiland-Jørgensen
2025-05-01 14:03               ` Jesper Dangaard Brouer
2025-05-05 10:18               ` Jakub Sitnicki
2025-05-05 12:35                 ` Toke Høiland-Jørgensen
2025-04-22 13:23 ` [PATCH RFC bpf-next v2 02/17] xdp: Track if metadata is supported in xdp_frame <> xdp_buff conversions Arthur Fabre
2025-04-22 13:23 ` [PATCH RFC bpf-next v2 03/17] trait: XDP support Arthur Fabre
2025-04-22 13:23 ` [PATCH RFC bpf-next v2 04/17] trait: XDP selftest Arthur Fabre
2025-04-22 13:23 ` [PATCH RFC bpf-next v2 05/17] trait: XDP benchmark Arthur Fabre
2025-04-22 13:23 ` [PATCH RFC bpf-next v2 06/17] trait: Replace memcpy calls with inline copies Arthur Fabre
2025-04-22 13:23 ` [PATCH RFC bpf-next v2 07/17] trait: Replace memmove calls with inline move Arthur Fabre
2025-04-22 13:23 ` [PATCH RFC bpf-next v2 08/17] skb: Extension header in packet headroom Arthur Fabre
2025-04-22 13:23 ` [PATCH RFC bpf-next v2 09/17] trait: Store traits in sk_buff extension Arthur Fabre
2025-04-22 13:23 ` [PATCH RFC bpf-next v2 10/17] bnxt: Propagate trait presence to skb Arthur Fabre
2025-04-23 16:36   ` Stanislav Fomichev
2025-04-23 20:54     ` Arthur Fabre
2025-04-23 23:45       ` Stanislav Fomichev
2025-04-24  9:49         ` Toke Høiland-Jørgensen
2025-04-24 15:39           ` Stanislav Fomichev
2025-04-24 18:59             ` Jakub Sitnicki
2025-04-25  8:06               ` Toke Høiland-Jørgensen
2025-04-22 13:23 ` [PATCH RFC bpf-next v2 11/17] ice: " Arthur Fabre
2025-04-22 13:23 ` [PATCH RFC bpf-next v2 12/17] veth: " Arthur Fabre
2025-04-22 13:23 ` [PATCH RFC bpf-next v2 13/17] virtio_net: " Arthur Fabre
2025-04-22 13:23 ` [PATCH RFC bpf-next v2 14/17] mlx5: move xdp_buff scope one level up Arthur Fabre
2025-04-22 13:23 ` [PATCH RFC bpf-next v2 15/17] mlx5: Propagate trait presence to skb Arthur Fabre
2025-04-22 13:23 ` [PATCH RFC bpf-next v2 16/17] xdp generic: " Arthur Fabre
2025-04-22 13:23 ` [PATCH RFC bpf-next v2 17/17] trait: Allow socket filters to access traits Arthur Fabre

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=87ikmle9t4.fsf@cloudflare.com \
    --to=jakub@cloudflare$(echo .)com \
    --cc=alexei.starovoitov@gmail$(echo .)com \
    --cc=arthur@arthurfabre$(echo .)com \
    --cc=ast@kernel$(echo .)org \
    --cc=bpf@vger$(echo .)kernel.org \
    --cc=edumazet@google$(echo .)com \
    --cc=hawk@kernel$(echo .)org \
    --cc=jbrandeburg@cloudflare$(echo .)com \
    --cc=kernel-team@cloudflare$(echo .)com \
    --cc=kuba@kernel$(echo .)org \
    --cc=lbiancon@redhat$(echo .)com \
    --cc=netdev@vger$(echo .)kernel.org \
    --cc=toke@redhat$(echo .)com \
    --cc=yan@cloudflare$(echo .)com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox