From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-pg1-x535.google.com (mail-pg1-x535.google.com [IPv6:2607:f8b0:4864:20::535]) by mail.toke.dk (Postfix) with ESMTPS id 677669D8995 for ; Tue, 3 Jan 2023 23:23:17 +0100 (CET) Authentication-Results: mail.toke.dk; dkim=pass (2048-bit key; unprotected) header.d=google.com header.i=@google.com header.a=rsa-sha256 header.s=20210112 header.b=XaVLfl+U Received: by mail-pg1-x535.google.com with SMTP id e10so7153120pgc.9 for ; Tue, 03 Jan 2023 14:23:17 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20210112; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:from:to:cc:subject:date:message-id:reply-to; bh=a2SjNr3Z+L3Cp/DNbgmAZXuHJ6agN6FwcAp+3Hg5iyI=; b=XaVLfl+Uz2SAGMIm8MHnc1EOkdySdC0+p02dxb+6tSFKTC27IMm/WSDBzQVKM1r08V Lw32E0gUQLHLEFt8wFVV98f8PMp90YGhMzXksrwEW9oaLF2tQlpj2xsihveAvDJYWrIj A1oQMGAOZ6kW3IRmbgCMDtSl0uGbx3SEAE8EtlVeOiEsTHwnpsWO8KzGHLp8krAFoADo boX+EVA+t7hOmRETa5t+vj1S/PSpmxpaLrEbQRCs/3Og7ZGj6wVsf72AAsnd/8LuzwRD diRccQUZbx+hgK5rKDmSHD2EhqWZzC0UvLXJvUJk1Zy3zCC9hQLrep/877ofbA9QthGG Vo2g== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:x-gm-message-state:from:to:cc:subject:date:message-id :reply-to; bh=a2SjNr3Z+L3Cp/DNbgmAZXuHJ6agN6FwcAp+3Hg5iyI=; b=fA9xNJTv2pwpcqCOK6/jpi7DE8xG+tQ2tKM8RXoRcb6uzAIiFYmlT22m6biCVTLROF D0ITQYiCM2uq8a+DMcuZ3qHE99yC7BuCQEheAnVM05ieBcW4WRdCCWfHXDHZn/7imatZ hYQg3tDBrAWlcD/K60bNB0ElLzLLp44foUbmR/7Ka1sI0gIMfs6lKRB+FP8d9iH0ZxmQ lusaECUD0ttvV+hXQPFGAmLaA1KdpPYW0x+OmmFsiJ4RxOtszxLaPtTHnBwdgJGHtXAp 1btVUOMJslxE3fMFFYDwuugg8ZPhH17LZV4MFmXkfJs6KqfIABBOqmBDIk50pGxD0kg4 ybsA== X-Gm-Message-State: AFqh2kq19qGywO8DNg6xKOordaWtf5DpvAlLR0i9DBF0NW+AFJUgmtRr e5mn49tzjHldzdgloPvCxTLaTqmw4KLIW01HBVOAcg== X-Google-Smtp-Source: AMrXdXtxZlxStVXvaH//rng+QDYaVuAN8FmKPBh5ylM5uibagK+/VgEmrqQWvfUp+YW09kCcaPU6c2FM4Zu9FXLnA+Q= X-Received: by 2002:a65:53c3:0:b0:478:b7ab:2f72 with SMTP id z3-20020a6553c3000000b00478b7ab2f72mr2625285pgr.186.1672784595474; Tue, 03 Jan 2023 14:23:15 -0800 (PST) MIME-Version: 1.0 References: <20221220222043.3348718-1-sdf@google.com> <20221220222043.3348718-2-sdf@google.com> In-Reply-To: From: Stanislav Fomichev Date: Tue, 3 Jan 2023 14:23:03 -0800 Message-ID: To: David Vernet Content-Type: text/plain; charset="UTF-8" Message-ID-Hash: PMTSHKAI6KD3NWM6QCB7TE45QNRGMBLH X-Message-ID-Hash: PMTSHKAI6KD3NWM6QCB7TE45QNRGMBLH X-MailFrom: sdf@google.com X-Mailman-Rule-Misses: dmarc-mitigation; no-senders; approved; emergency; loop; banned-address; member-moderation; nonmember-moderation; administrivia; implicit-dest; max-recipients; max-size; news-moderation; no-subject; digests; suspicious-header CC: bpf@vger.kernel.org, ast@kernel.org, daniel@iogearbox.net, andrii@kernel.org, martin.lau@linux.dev, song@kernel.org, yhs@fb.com, john.fastabend@gmail.com, kpsingh@kernel.org, haoluo@google.com, jolsa@kernel.org, David Ahern , Jakub Kicinski , Willem de Bruijn , Jesper Dangaard Brouer , Anatoly Burakov , Alexander Lobakin , Magnus Karlsson , Maryam Tahhan , xdp-hints@xdp-project.net, netdev@vger.kernel.org X-Mailman-Version: 3.3.7 Precedence: list Subject: [xdp-hints] Re: [PATCH bpf-next v5 01/17] bpf: Document XDP RX metadata List-Id: XDP hardware hints design discussion Archived-At: List-Archive: List-Help: List-Owner: List-Post: List-Subscribe: List-Unsubscribe: On Wed, Dec 28, 2022 at 9:25 AM David Vernet wrote: > > On Tue, Dec 20, 2022 at 02:20:27PM -0800, Stanislav Fomichev wrote: > > Document all current use-cases and assumptions. > > > > Cc: John Fastabend > > Cc: David Ahern > > Cc: Martin KaFai Lau > > Cc: Jakub Kicinski > > Cc: Willem de Bruijn > > Cc: Jesper Dangaard Brouer > > Cc: Anatoly Burakov > > Cc: Alexander Lobakin > > Cc: Magnus Karlsson > > Cc: Maryam Tahhan > > Cc: xdp-hints@xdp-project.net > > Cc: netdev@vger.kernel.org > > Signed-off-by: Stanislav Fomichev > > --- > > Documentation/networking/index.rst | 1 + > > Documentation/networking/xdp-rx-metadata.rst | 107 +++++++++++++++++++ > > 2 files changed, 108 insertions(+) > > create mode 100644 Documentation/networking/xdp-rx-metadata.rst > > > > diff --git a/Documentation/networking/index.rst b/Documentation/networking/index.rst > > index 4f2d1f682a18..4ddcae33c336 100644 > > --- a/Documentation/networking/index.rst > > +++ b/Documentation/networking/index.rst > > @@ -120,6 +120,7 @@ Refer to :ref:`netdev-FAQ` for a guide on netdev development process specifics. > > xfrm_proc > > xfrm_sync > > xfrm_sysctl > > + xdp-rx-metadata > > > > .. only:: subproject and html > > > > diff --git a/Documentation/networking/xdp-rx-metadata.rst b/Documentation/networking/xdp-rx-metadata.rst > > new file mode 100644 > > index 000000000000..37e8192d9b60 > > --- /dev/null > > +++ b/Documentation/networking/xdp-rx-metadata.rst > > Hey Stanislav, > > This is looking excellent. Left a few more minor comments and > suggestions. > > > @@ -0,0 +1,107 @@ > > +=============== > > +XDP RX Metadata > > +=============== > > + > > +This document describes how an XDP program can access hardware metadata > > In similar fashion to LWN articles, can we spell out what XDP means the > first time it's used, e.g.: > > ...describes how an eXpress Data Path (XDP) program... > > In general this applies to other acronyms unless they're super obvious, > like "CPU" (thanks for already having done it for XSK). Sure. Hopefully no need to explain RX below? Don't see anything else.. LMK if I missed something > > +related to a packet using a set of helper functions, and how it can pass > > +that metadata on to other consumers. > > + > > +General Design > > +============== > > + > > +XDP has access to a set of kfuncs to manipulate the metadata in an XDP frame. > > +Every device driver that wishes to expose additional packet metadata can > > +implement these kfuncs. The set of kfuncs is declared in ``include/net/xdp.h`` > > +via ``XDP_METADATA_KFUNC_xxx``. > > + > > +Currently, the following kfuncs are supported. In the future, as more > > +metadata is supported, this set will grow: > > + > > +- ``bpf_xdp_metadata_rx_timestamp`` returns a packet's RX timestamp > > +- ``bpf_xdp_metadata_rx_hash`` returns a packet's RX hash > > So, I leave this up to you as to whether or not you want to do this, but > there is a built-in mechanism in sphinx that converts doxygen comments > to rendered documentation for a function, struct, etc, and also > automatically links other places in documentation where the function is > referenced. See [0] for an example of this in code, and [1] for an > example of how it's rendered. > > [0]: https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git/tree/Documentation/bpf/kfuncs.rst#n239 > [1]: https://docs.kernel.org/bpf/kfuncs.html#c.bpf_task_acquire > > So you would do something like add function headers to the kfuncs, and > then do: > > .. kernel-doc:: net/core/xdp.c > :identifiers: bpf_xdp_metadata_rx_timestamp bpf_xdp_metadata_rx_hash > > At some point we will need a consistent story for how we document > kfuncs. That's not set in stone yet, which is why I'm saying it's up to > you whether or not you want to do this or just leave it as teletype with > a written description next to it. Later on when we settle on a > documentation story for kfuncs, we can update all of them to be > documented in the same way. Let me try and see how it looks in the html doc. I like the idea of referencing the code directly, hopefully less chance it goes stale. > > +The XDP program can use these kfuncs to read the metadata into stack > > s/The/An Ack, ty! > > +variables for its own consumption. Or, to pass the metadata on to other > > +consumers, an XDP program can store it into the metadata area carried > > +ahead of the packet. > > + > > +Not all kfuncs have to be implemented by the device driver; when not > > +implemented, the default ones that return ``-EOPNOTSUPP`` will be used. > > + > > +Within the XDP frame, the metadata layout is as follows:: > > s/the/an Thx! > > + > > + +----------+-----------------+------+ > > + | headroom | custom metadata | data | > > + +----------+-----------------+------+ > > + ^ ^ > > + | | > > + xdp_buff->data_meta xdp_buff->data > > + > > +The XDP program can store individual metadata items into this data_meta > > Can we teletype ``data_meta``? Also, s/The XDP program/An XDP program Sure. > > +area in whichever format it chooses. Later consumers of the metadata > > +will have to agree on the format by some out of band contract (like for > > +the AF_XDP use case, see below). > > + > > +AF_XDP > > +====== > > + > > +``AF_XDP`` use-case implies that there is a contract between the BPF program > > +that redirects XDP frames into the ``AF_XDP`` socket (``XSK``) and the final > > +consumer. Thus the BPF program manually allocates a fixed number of > > Can we have ``AF_XDP`` link to the ``AF_XDP`` docs page for any reader > that's not familiar with it? I think you can just do the following and > it should just work: > > s/``AF_XDP``/:doc:`af_xdp` Will try. > > +bytes out of metadata via ``bpf_xdp_adjust_meta`` and calls a subset > > +of kfuncs to populate it. The userspace ``XSK`` consumer computes > > +``xsk_umem__get_data() - METADATA_SIZE`` to locate its metadata. > > s/to locate its metadata/to locate that metadata SG. > to make it clear that the consumer is consuming the metadata that was > populated by bpf_xdp_adjust_meta()? It's pretty clear from context but I > think "its metadata" may confuse some people as the metadata is not [..] > really owned by the consumer. Also, should we mention that > xsk_umem__get_data() is defined in libxdp? Add something like: Note, ``xsk_umem__get_data`` is defined in ``libxdp`` and ``METADATA_SIZE`` is an application-specific constant. ? > One other probably dumb question: is METADATA_SIZE static? If so, should > we provide a libxdp wrapper to access it? (added a note above) > > + > > +Here is the ``AF_XDP`` consumer layout (note missing ``data_meta`` pointer):: > > + > > + +----------+-----------------+------+ > > + | headroom | custom metadata | data | > > + +----------+-----------------+------+ > > + ^ > > + | > > + rx_desc->address > > + > > +XDP_PASS > > +======== > > + > > +This is the path where the packets processed by the XDP program are passed > > +into the kernel. The kernel creates the ``skb`` out of the ``xdp_buff`` > > +contents. Currently, every driver has custom kernel code to parse > > +the descriptors and populate ``skb`` metadata when doing this ``xdp_buff->skb`` > > +conversion, and the XDP metadata is not used by the kernel when building > > +skbs. However, TC-BPF programs can access the XDP metadata area using > > +the data_meta pointer. > > ``data_meta`` Ack. > > + > > +In the future, we'd like to support a case where an XDP program > > +can override some of the metadata used for building skbs. > > + > > +bpf_redirect_map > > +================ > > + > > +``bpf_redirect_map`` can redirect the frame to a different device. > > +Some devices (like virtual ethernet links) support running a second XDP > > +program after the redirect. However, the final consumer doesn't have > > +access to the original hardware descriptor and can't access any of > > +the original metadata. The same applies to XDP programs installed > > +into devmaps and cpumaps. > > + > > +This means that for redirected packets only custom metadata is > > +currently supported, which has to be prepared by the initial XDP program > > +before redirect. If the frame is eventually passed to the kernel, the > > +skb created from such a frame won't have any hardware metadata populated > > +in its skb. And if such a packet is later redirected into an ``XSK``, > > s/And if/If > > Also, can you add teletype ``skb`` in this paragraph? Sure, thanks. > > +that will also only have access to the custom metadata. > > + > > + > > +bpf_tail_call > > +============= > > + > > +Adding programs that access metadata kfuncs to the ``BPF_MAP_TYPE_PROG_ARRAY`` > > +is currently not supported. > > + > > +Example > > +======= > > + > > +See ``tools/testing/selftests/bpf/progs/xdp_metadata.c`` and > > +``tools/testing/selftests/bpf/prog_tests/xdp_metadata.c`` for an example of > > +BPF program that handles XDP metadata. > > -- > > 2.39.0.314.g84b9a713c41-goog > > Looks great otherwise, thanks.