netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jakub Kicinski <kuba@kernel.org>
To: Jesper Dangaard Brouer <hawk@kernel.org>
Cc: bpf@vger.kernel.org, netdev@vger.kernel.org, lorenzo@kernel.org,
	Alexei Starovoitov <ast@kernel.org>,
	Daniel Borkmann <borkmann@iogearbox.net>,
	Eric Dumazet <eric.dumazet@gmail.com>,
	"David S. Miller" <davem@davemloft.net>,
	Paolo Abeni <pabeni@redhat.com>,
	sdf@fomichev.me, kernel-team@cloudflare.com,
	arthur@arthurfabre.com, jakub@cloudflare.com
Subject: Re: [PATCH bpf-next V1 3/7] net: xdp: Add kfuncs to store hw metadata in xdp_buff
Date: Mon, 16 Jun 2025 14:55:23 -0700	[thread overview]
Message-ID: <20250616145523.63bd2577@kernel.org> (raw)
In-Reply-To: <174897276809.1677018.15753779269046278541.stgit@firesoul>

On Tue, 03 Jun 2025 19:46:08 +0200 Jesper Dangaard Brouer wrote:
> Introduce the following kfuncs to store hw metadata provided by the NIC
> into the xdp_buff struct:
> 
> - rx-hash: bpf_xdp_store_rx_hash
> - rx-vlan: bpf_xdp_store_rx_vlan
> - rx-hw-ts: bpf_xdp_store_rx_ts

My mental model would that these should operate within the "remote XDP".
We have helpers to "get the metadata", now we're adding helpers to "set
the metadata". Should the setting not be simply the inverse of the
getting? Store to driver-specific format?

-> local driver - all the metadata in HW-specific format / descriptors
  -> local XDP - GET and copy data to locally defined md prepend
-> remote drv - CPU map, veth) hooks in its own callbacks and md struct
  -> remote XDP - read the data from md prepend and call SET
-> remote drv - check if XDP set any md in its struct and uses it when
                converting to skb.

Note, this is just the model for the API. We can always introduce
optimizations, like a feature flag that makes the "local" driver output
well known MD struct and have the remote driver pick that up, with
neither XDP calling the helpers. 
But from the model perspective since the GET operations are local to
the driver owning the frame at the time, the SET operations should also
operate on metadata local to the owning driver.

That's my $0.02

  reply	other threads:[~2025-06-16 21:55 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-06-03 17:45 [PATCH bpf-next V1 0/7] xdp: Propagate RX HW hints for XDP_REDIRECTed packets via xdp_frame Jesper Dangaard Brouer
2025-06-03 17:45 ` [PATCH bpf-next V1 1/7] net: xdp: Add xdp_rx_meta structure Jesper Dangaard Brouer
2025-06-03 17:46 ` [PATCH bpf-next V1 2/7] selftests/bpf: Adjust test for maximum packet size in xdp_do_redirect Jesper Dangaard Brouer
2025-06-03 17:46 ` [PATCH bpf-next V1 3/7] net: xdp: Add kfuncs to store hw metadata in xdp_buff Jesper Dangaard Brouer
2025-06-16 21:55   ` Jakub Kicinski [this message]
2025-06-03 17:46 ` [PATCH bpf-next V1 4/7] net: xdp: Set skb hw metadata from xdp_frame Jesper Dangaard Brouer
2025-06-03 17:46 ` [PATCH bpf-next V1 5/7] net: veth: Read xdp metadata from rx_meta struct if available Jesper Dangaard Brouer
2025-06-03 17:46 ` [PATCH bpf-next V1 6/7] bpf: selftests: Add rx_meta store kfuncs selftest Jesper Dangaard Brouer
2025-06-06 21:57   ` Alexei Starovoitov
2025-06-06 22:16     ` Lorenzo Bianconi
2025-06-03 17:46 ` [PATCH bpf-next V1 7/7] net: xdp: update documentation for xdp-rx-metadata.rst Jesper Dangaard Brouer
2025-06-06  2:45   ` Stanislav Fomichev
2025-06-10 13:48     ` Daniel Borkmann
2025-06-10 20:12       ` Toke Høiland-Jørgensen
2025-06-10 22:26         ` Lorenzo Bianconi
2025-06-11  3:40           ` Stanislav Fomichev
2025-06-13 10:59             ` Jesper Dangaard Brouer
2025-06-16 15:34               ` Stanislav Fomichev
2025-06-17 16:15                 ` Jesper Dangaard Brouer
2025-06-17 20:47                   ` Stanislav Fomichev
2025-06-16 12:37             ` Lorenzo Bianconi
2025-06-16 15:45               ` Stanislav Fomichev
2025-06-16 16:40                 ` Lorenzo Bianconi
2025-06-17 11:50                   ` Toke Høiland-Jørgensen
2025-06-17 14:47                     ` Jesper Dangaard Brouer
2025-06-17 15:10                       ` Performance impact of disabling VLAN offload [was: Re: [PATCH bpf-next V1 7/7] net: xdp: update documentation for xdp-rx-metadata.rst] Toke Høiland-Jørgensen
2025-06-19 12:09                         ` Jesper Dangaard Brouer
2025-06-19 12:23                           ` Toke Høiland-Jørgensen
2025-06-13 11:18         ` [PATCH bpf-next V1 7/7] net: xdp: update documentation for xdp-rx-metadata.rst Daniel Borkmann
2025-06-16 11:51           ` Toke Høiland-Jørgensen

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=20250616145523.63bd2577@kernel.org \
    --to=kuba@kernel.org \
    --cc=arthur@arthurfabre.com \
    --cc=ast@kernel.org \
    --cc=borkmann@iogearbox.net \
    --cc=bpf@vger.kernel.org \
    --cc=davem@davemloft.net \
    --cc=eric.dumazet@gmail.com \
    --cc=hawk@kernel.org \
    --cc=jakub@cloudflare.com \
    --cc=kernel-team@cloudflare.com \
    --cc=lorenzo@kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=sdf@fomichev.me \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
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;
as well as URLs for NNTP newsgroup(s).