XDP hardware hints discussion mail archive
 help / color / mirror / Atom feed
From: Stanislav Fomichev <sdf@google.com>
To: Jakub Kicinski <kuba@kernel.org>
Cc: "Toke Høiland-Jørgensen" <toke@redhat.com>,
	"Alexei Starovoitov" <alexei.starovoitov@gmail.com>,
	bpf <bpf@vger.kernel.org>, "Alexei Starovoitov" <ast@kernel.org>,
	"Daniel Borkmann" <daniel@iogearbox.net>,
	"Andrii Nakryiko" <andrii@kernel.org>,
	"Martin KaFai Lau" <martin.lau@linux.dev>,
	"Song Liu" <song@kernel.org>, "Yonghong Song" <yhs@fb.com>,
	"John Fastabend" <john.fastabend@gmail.com>,
	"KP Singh" <kpsingh@kernel.org>, "Hao Luo" <haoluo@google.com>,
	"Jiri Olsa" <jolsa@kernel.org>,
	"Saeed Mahameed" <saeedm@nvidia.com>,
	"David Ahern" <dsahern@gmail.com>,
	"Willem de Bruijn" <willemb@google.com>,
	"Jesper Dangaard Brouer" <brouer@redhat.com>,
	"Anatoly Burakov" <anatoly.burakov@intel.com>,
	"Alexander Lobakin" <alexandr.lobakin@intel.com>,
	"Magnus Karlsson" <magnus.karlsson@gmail.com>,
	"Maryam Tahhan" <mtahhan@redhat.com>,
	xdp-hints@xdp-project.net,
	"Network Development" <netdev@vger.kernel.org>
Subject: [xdp-hints] Re: [PATCH bpf-next v3 11/12] mlx5: Support RX XDP metadata
Date: Fri, 9 Dec 2022 09:46:20 -0800	[thread overview]
Message-ID: <CAKH8qBsx4pPuvYenpM18NgdnGCG8QjqnsNY40Uc44EXTUVabMA@mail.gmail.com> (raw)
In-Reply-To: <20221209084524.01c09d9c@kernel.org>

On Fri, Dec 9, 2022 at 8:45 AM Jakub Kicinski <kuba@kernel.org> wrote:
>
> On Fri, 09 Dec 2022 15:42:37 +0100 Toke Høiland-Jørgensen wrote:
> > If we expect the program to do out of band probing, we could just get
> > rid of the _supported() functions entirely?
> >
> > I mean, to me, the whole point of having the separate _supported()
> > function for each item was to have a lower-overhead way of checking if
> > the metadata item was supported. But if the overhead is not actually
> > lower (because both incur a function call), why have them at all? Then
> > we could just change the implementation from this:
> >
> > bool mlx5e_xdp_rx_hash_supported(const struct xdp_md *ctx)
> > {
> >       const struct mlx5_xdp_buff *_ctx = (void *)ctx;
> >
> >       return _ctx->xdp.rxq->dev->features & NETIF_F_RXHASH;
> > }
> >
> > u32 mlx5e_xdp_rx_hash(const struct xdp_md *ctx)
> > {
> >       const struct mlx5_xdp_buff *_ctx = (void *)ctx;
> >
> >       return be32_to_cpu(_ctx->cqe->rss_hash_result);
> > }
> >
> > to this:
> >
> > u32 mlx5e_xdp_rx_hash(const struct xdp_md *ctx)
> > {
> >       const struct mlx5_xdp_buff *_ctx = (void *)ctx;
> >
> >       if (!(_ctx->xdp.rxq->dev->features & NETIF_F_RXHASH))
> >                 return 0;
> >
> >       return be32_to_cpu(_ctx->cqe->rss_hash_result);
> > }
>
> Are there no corner cases? E.g. in case of an L2 frame you'd then
> expect a hash of 0? Rather than no hash?
>
> If I understand we went for the _supported() thing to make inlining
> the check easier than inlining the actual read of the field.
> But we're told inlining is a bit of a wait.. so isn't the motivation
> for the _supported() pretty much gone? And we should we go back to
> returning an error from the actual read?

Seems fair, we can always bring those _supported() calls back in the
future when the inlining is available and having those separate calls
shows clear benefit.
Then let's go back to a more conventional form below?

int mlx5e_xdp_rx_hash(const struct xdp_md *ctx, u32 *timestamp)
{
      const struct mlx5_xdp_buff *_ctx = (void *)ctx;

       if (!(_ctx->xdp.rxq->dev->features & NETIF_F_RXHASH))
                 return -EOPNOTSUPP;

       *timestamp = be32_to_cpu(_ctx->cqe->rss_hash_result);
       return 0;
 }

> Is partial inlining hard? (inline just the check and generate a full
> call for the read, ending up with the same code as with _supported())

I'm assuming you're suggesting to do this partial inlining manually
(as in, writing the code to output this bytecode)?
This probably also falls into the "manual bpf asm generation tech debt" bucket?
LMK if I missed your point.

  reply	other threads:[~2022-12-09 17:46 UTC|newest]

Thread overview: 61+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-12-06  2:45 [xdp-hints] [PATCH bpf-next v3 00/12] xdp: hints via kfuncs Stanislav Fomichev
2022-12-06  2:45 ` [xdp-hints] [PATCH bpf-next v3 01/12] bpf: Document XDP RX metadata Stanislav Fomichev
2022-12-08  4:25   ` [xdp-hints] " Jakub Kicinski
2022-12-08 19:06     ` Stanislav Fomichev
2022-12-06  2:45 ` [xdp-hints] [PATCH bpf-next v3 02/12] bpf: Rename bpf_{prog,map}_is_dev_bound to is_offloaded Stanislav Fomichev
2022-12-08  4:26   ` [xdp-hints] " Jakub Kicinski
2022-12-06  2:45 ` [xdp-hints] [PATCH bpf-next v3 03/12] bpf: XDP metadata RX kfuncs Stanislav Fomichev
2022-12-07  4:29   ` [xdp-hints] " Alexei Starovoitov
2022-12-07  4:52     ` Stanislav Fomichev
2022-12-07  7:23       ` Martin KaFai Lau
2022-12-07 18:05         ` Stanislav Fomichev
2022-12-08  2:47   ` Martin KaFai Lau
2022-12-08 19:07     ` Stanislav Fomichev
2022-12-08 22:53       ` Martin KaFai Lau
2022-12-08 23:45         ` Stanislav Fomichev
2022-12-08  5:00   ` Jakub Kicinski
2022-12-08 19:07     ` Stanislav Fomichev
2022-12-09  1:30       ` Jakub Kicinski
2022-12-09  2:57         ` Stanislav Fomichev
2022-12-08 22:39   ` Toke Høiland-Jørgensen
2022-12-08 23:46     ` Stanislav Fomichev
2022-12-09  0:07       ` Toke Høiland-Jørgensen
2022-12-09  2:57         ` Stanislav Fomichev
2022-12-10  0:42           ` Martin KaFai Lau
2022-12-10  1:12             ` Martin KaFai Lau
2022-12-09 11:10   ` Jesper Dangaard Brouer
2022-12-09 17:47     ` Stanislav Fomichev
2022-12-11 11:09       ` Jesper Dangaard Brouer
2022-12-06  2:45 ` [xdp-hints] [PATCH bpf-next v3 04/12] veth: Introduce veth_xdp_buff wrapper for xdp_buff Stanislav Fomichev
2022-12-06  2:45 ` [xdp-hints] [PATCH bpf-next v3 05/12] veth: Support RX XDP metadata Stanislav Fomichev
2022-12-06  2:45 ` [xdp-hints] [PATCH bpf-next v3 06/12] selftests/bpf: Verify xdp_metadata xdp->af_xdp path Stanislav Fomichev
2022-12-06  2:45 ` [xdp-hints] [PATCH bpf-next v3 07/12] mlx4: Introduce mlx4_xdp_buff wrapper for xdp_buff Stanislav Fomichev
2022-12-08  6:11   ` [xdp-hints] " Tariq Toukan
2022-12-08 19:07     ` Stanislav Fomichev
2022-12-06  2:45 ` [xdp-hints] [PATCH bpf-next v3 08/12] mxl4: Support RX XDP metadata Stanislav Fomichev
2022-12-08  6:09   ` [xdp-hints] " Tariq Toukan
2022-12-08 19:07     ` Stanislav Fomichev
2022-12-08 20:23       ` Tariq Toukan
2022-12-06  2:45 ` [xdp-hints] [PATCH bpf-next v3 09/12] xsk: Add cb area to struct xdp_buff_xsk Stanislav Fomichev
2022-12-06  2:45 ` [xdp-hints] [PATCH bpf-next v3 10/12] mlx5: Introduce mlx5_xdp_buff wrapper for xdp_buff Stanislav Fomichev
2022-12-06  2:45 ` [xdp-hints] [PATCH bpf-next v3 11/12] mlx5: Support RX XDP metadata Stanislav Fomichev
2022-12-08 22:59   ` [xdp-hints] " Toke Høiland-Jørgensen
2022-12-08 23:45     ` Stanislav Fomichev
2022-12-09  0:02       ` Toke Høiland-Jørgensen
2022-12-09  0:07         ` Alexei Starovoitov
2022-12-09  0:29           ` Toke Høiland-Jørgensen
2022-12-09  0:32             ` Alexei Starovoitov
2022-12-09  0:53               ` Toke Høiland-Jørgensen
2022-12-09  2:57                 ` Stanislav Fomichev
2022-12-09  5:24                   ` Saeed Mahameed
2022-12-09 12:59                     ` Jesper Dangaard Brouer
2022-12-09 14:37                       ` Toke Høiland-Jørgensen
2022-12-09 15:19                       ` Dave Taht
2022-12-09 14:42                   ` Toke Høiland-Jørgensen
2022-12-09 16:45                     ` Jakub Kicinski
2022-12-09 17:46                       ` Stanislav Fomichev [this message]
2022-12-09 22:13                         ` Jakub Kicinski
2022-12-06  2:45 ` [xdp-hints] [PATCH bpf-next v3 12/12] selftests/bpf: Simple program to dump XDP RX metadata Stanislav Fomichev
2022-12-08 22:28 ` [xdp-hints] Re: [PATCH bpf-next v3 00/12] xdp: hints via kfuncs Toke Høiland-Jørgensen
2022-12-08 23:47   ` Stanislav Fomichev
2022-12-09  0:14     ` 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

  List information: https://lists.xdp-project.net/postorius/lists/xdp-hints.xdp-project.net/

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

  git send-email \
    --in-reply-to=CAKH8qBsx4pPuvYenpM18NgdnGCG8QjqnsNY40Uc44EXTUVabMA@mail.gmail.com \
    --to=sdf@google.com \
    --cc=alexandr.lobakin@intel.com \
    --cc=alexei.starovoitov@gmail.com \
    --cc=anatoly.burakov@intel.com \
    --cc=andrii@kernel.org \
    --cc=ast@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=brouer@redhat.com \
    --cc=daniel@iogearbox.net \
    --cc=dsahern@gmail.com \
    --cc=haoluo@google.com \
    --cc=john.fastabend@gmail.com \
    --cc=jolsa@kernel.org \
    --cc=kpsingh@kernel.org \
    --cc=kuba@kernel.org \
    --cc=magnus.karlsson@gmail.com \
    --cc=martin.lau@linux.dev \
    --cc=mtahhan@redhat.com \
    --cc=netdev@vger.kernel.org \
    --cc=saeedm@nvidia.com \
    --cc=song@kernel.org \
    --cc=toke@redhat.com \
    --cc=willemb@google.com \
    --cc=xdp-hints@xdp-project.net \
    --cc=yhs@fb.com \
    /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