From: Martin KaFai Lau <martin.lau@linux.dev>
To: Amery Hung <ameryhung@gmail.com>
Cc: Nimrod Oren <noren@nvidia.com>,
bpf@vger.kernel.org, netdev@vger.kernel.org,
alexei.starovoitov@gmail.com, andrii@kernel.org,
daniel@iogearbox.net, kuba@kernel.org, martin.lau@kernel.org,
mohsin.bashr@gmail.com, saeedm@nvidia.com, tariqt@nvidia.com,
mbloch@nvidia.com, maciej.fijalkowski@intel.com,
kernel-team@meta.com, Dragos Tatulea <dtatulea@nvidia.com>
Subject: Re: [RFC bpf-next v1 0/7] Add kfunc bpf_xdp_pull_data
Date: Fri, 5 Sep 2025 10:20:01 -0700 [thread overview]
Message-ID: <9ac027eb-e1fa-43b7-9b36-f5a267e461f8@linux.dev> (raw)
In-Reply-To: <CAMB2axO1Hb=pNMw8p2Ca+4XfVmDTA+TxbbXaD4G6kxUVEsHERg@mail.gmail.com>
On 9/4/25 10:28 AM, Amery Hung wrote:
> On Fri, Aug 29, 2025 at 11:22 AM Martin KaFai Lau <martin.lau@linux.dev> wrote:
>>
>> On 8/28/25 6:39 AM, Nimrod Oren wrote:
>>> I'm currently working on a series that converts the xdp_native program
>>> to use dynptr for accessing header data. If accepted, it should provide
>>> better performance, since dynptr can access without copying the data.
>>
>> The bpf_xdp_adjust_tail is aware of xdp_buff_has_frags. Is there a reason that
>> bpf_xdp_adjust_head cannot handle frags also?
>
> I am not aware of reasons that would stop this.
>
> Are you suggesting another way to pop headers? E.g., use
> bpf_xdp_adjust_head() to shrink the first frag from the front and call
> bpf_xdp_store_bytes() to move the remaining headers
bpf_xdp_pull_data is useful on its own, nothing change there. On top of that,
bpf_xdp_adjust_head() should be useful also for bpf prog that does not care
about the linear/frag layout and stay with the bpf_dynptr helpers (or the
bpf_xdp_{store,load}_bytes you mentioned). Also, if I read it correctly,
bpf_xdp_adjust_head can increase the head of multi buf xdp but not shrinking it
while the bpf_xdp_adjust_tail can do both. It could be a surprise to use. The
adjust_head support can be a followup though. I think some of your work in this
series is pretty close to having adjust_head support also, so I was wondering if
there is reason that adjust_head cannot be supported.
next prev parent reply other threads:[~2025-09-05 17:20 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-08-25 19:39 [RFC bpf-next v1 0/7] Add kfunc bpf_xdp_pull_data Amery Hung
2025-08-25 19:39 ` [RFC bpf-next v1 1/7] net/mlx5e: Fix generating skb from nonlinear xdp_buff Amery Hung
2025-08-27 13:45 ` Dragos Tatulea
2025-08-28 3:44 ` Amery Hung
2025-08-28 16:23 ` Dragos Tatulea
2025-09-04 17:26 ` Amery Hung
2025-08-28 13:41 ` Nimrod Oren
2025-08-25 19:39 ` [RFC bpf-next v1 2/7] bpf: Allow bpf_xdp_shrink_data to shrink a frag from head and tail Amery Hung
2025-08-28 13:43 ` Nimrod Oren
2025-09-04 22:19 ` Amery Hung
2025-09-05 1:52 ` Jakub Kicinski
2025-08-25 19:39 ` [RFC bpf-next v1 3/7] bpf: Support pulling non-linear xdp data Amery Hung
2025-08-25 21:29 ` Stanislav Fomichev
2025-08-25 22:23 ` Amery Hung
2025-08-25 22:29 ` Jakub Kicinski
2025-08-25 22:36 ` Amery Hung
2025-08-25 22:46 ` Stanislav Fomichev
2025-08-25 22:58 ` Jakub Kicinski
2025-08-26 0:12 ` Stanislav Fomichev
2025-08-26 0:30 ` Jakub Kicinski
2025-08-25 22:39 ` Jakub Kicinski
2025-08-26 5:12 ` Amery Hung
2025-08-26 13:20 ` Jakub Kicinski
2025-08-26 13:44 ` Amery Hung
2025-08-25 19:39 ` [RFC bpf-next v1 4/7] bpf: Clear packet pointers after changing packet data in kfuncs Amery Hung
2025-08-25 19:39 ` [RFC bpf-next v1 5/7] bpf: Support specifying linear xdp packet data size in test_run Amery Hung
2025-08-25 19:39 ` [RFC bpf-next v1 6/7] selftests/bpf: Test bpf_xdp_pull_data Amery Hung
2025-08-25 19:39 ` [RFC bpf-next v1 7/7] selftests: drv-net: Pull data before parsing headers Amery Hung
2025-08-25 22:41 ` [RFC bpf-next v1 0/7] Add kfunc bpf_xdp_pull_data Jakub Kicinski
2025-08-26 19:38 ` Gal Pressman
2025-08-28 13:39 ` Nimrod Oren
2025-08-29 7:26 ` Amery Hung
2025-08-30 0:09 ` Jakub Kicinski
2025-08-29 18:21 ` Martin KaFai Lau
2025-09-04 17:28 ` Amery Hung
2025-09-05 17:20 ` Martin KaFai Lau [this message]
2025-09-04 22:16 ` Amery Hung
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=9ac027eb-e1fa-43b7-9b36-f5a267e461f8@linux.dev \
--to=martin.lau@linux.dev \
--cc=alexei.starovoitov@gmail.com \
--cc=ameryhung@gmail.com \
--cc=andrii@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=dtatulea@nvidia.com \
--cc=kernel-team@meta.com \
--cc=kuba@kernel.org \
--cc=maciej.fijalkowski@intel.com \
--cc=martin.lau@kernel.org \
--cc=mbloch@nvidia.com \
--cc=mohsin.bashr@gmail.com \
--cc=netdev@vger.kernel.org \
--cc=noren@nvidia.com \
--cc=saeedm@nvidia.com \
--cc=tariqt@nvidia.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;
as well as URLs for NNTP newsgroup(s).