From: "Björn Töpel" <bjorn.topel@intel.com>
To: "Toke Høiland-Jørgensen" <toke@redhat.com>,
"Maciej Fijalkowski" <maciej.fijalkowski@intel.com>,
daniel@iogearbox.net, ast@kernel.org, bpf@vger.kernel.org,
netdev@vger.kernel.org
Cc: andrii@kernel.org, magnus.karlsson@intel.com, ciara.loftus@intel.com
Subject: Re: [PATCH bpf-next 1/3] libbpf: xsk: use bpf_link
Date: Mon, 15 Feb 2021 18:38:46 +0100 [thread overview]
Message-ID: <fe0c957e-d212-4265-a271-ba301c3c5eca@intel.com> (raw)
In-Reply-To: <87eehhcl9x.fsf@toke.dk>
On 2021-02-15 18:07, Toke Høiland-Jørgensen wrote:
> Maciej Fijalkowski <maciej.fijalkowski@intel.com> writes:
>
>> Currently, if there are multiple xdpsock instances running on a single
>> interface and in case one of the instances is terminated, the rest of
>> them are left in an inoperable state due to the fact of unloaded XDP
>> prog from interface.
>>
>> To address that, step away from setting bpf prog in favour of bpf_link.
>> This means that refcounting of BPF resources will be done automatically
>> by bpf_link itself.
>>
>> When setting up BPF resources during xsk socket creation, check whether
>> bpf_link for a given ifindex already exists via set of calls to
>> bpf_link_get_next_id -> bpf_link_get_fd_by_id -> bpf_obj_get_info_by_fd
>> and comparing the ifindexes from bpf_link and xsk socket.
>
> One consideration here is that bpf_link_get_fd_by_id() is a privileged
> operation (privileged as in CAP_SYS_ADMIN), so this has the side effect
> of making AF_XDP privileged as well. Is that the intention?
>
We're already using, e.g., bpf_map_get_fd_by_id() which has that
as well. So we're assuming that for XDP setup already!
> Another is that the AF_XDP code is in the process of moving to libxdp
> (see in-progress PR [0]), and this approach won't carry over as-is to
> that model, because libxdp has to pin the bpf_link fds.
>
I was assuming there were two modes of operations for AF_XDP in libxdp.
One which is with the multi-program support (which AFAIK is why the
pinning is required), and one "like the current libbpf" one. For the
latter Maciej's series would be a good fit, no?
> However, in libxdp we can solve the original problem in a different way,
> and in fact I already suggested to Magnus that we should do this (see
> [1]); so one way forward could be to address it during the merge in
> libxdp? It should be possible to address the original issue (two
> instances of xdpsock breaking each other when they exit), but
> applications will still need to do an explicit unload operation before
> exiting (i.e., the automatic detach on bpf_link fd closure will take
> more work, and likely require extending the bpf_link kernel support)...
>
I'd say it's depending on the libbpf 1.0/libxdp merge timeframe. If
we're months ahead, then I'd really like to see this in libbpf until the
merge. However, I'll leave that for Magnus/you to decide!
Bottom line; I'd *really* like bpf_link behavior (process scoped) for
AF_XDP sooner than later! ;-)
Thanks for the input!
Björn
> -Toke
>
> [0] https://github.com/xdp-project/xdp-tools/pull/92
> [1] https://github.com/xdp-project/xdp-tools/pull/92#discussion_r576204719
>
next prev parent reply other threads:[~2021-02-15 17:41 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-15 15:46 [PATCH bpf-next 0/3] Introduce bpf_link in libbpf's xsk Maciej Fijalkowski
2021-02-15 15:46 ` [PATCH bpf-next 1/3] libbpf: xsk: use bpf_link Maciej Fijalkowski
2021-02-15 17:07 ` Toke Høiland-Jørgensen
2021-02-15 17:38 ` Björn Töpel [this message]
2021-02-15 19:35 ` Toke Høiland-Jørgensen
2021-02-16 2:01 ` Maciej Fijalkowski
2021-02-16 9:15 ` Björn Töpel
2021-02-16 10:27 ` Toke Høiland-Jørgensen
2021-02-16 20:15 ` Maciej Fijalkowski
2021-02-15 20:22 ` John Fastabend
2021-02-15 21:38 ` Toke Høiland-Jørgensen
2021-02-16 0:18 ` John Fastabend
2021-02-16 2:23 ` Maciej Fijalkowski
2021-02-16 9:23 ` Björn Töpel
2021-02-16 10:36 ` Toke Høiland-Jørgensen
2021-02-23 1:15 ` Andrii Nakryiko
2021-02-17 2:23 ` Dan Siemon
2021-02-17 7:16 ` Magnus Karlsson
2021-02-17 7:36 ` Magnus Karlsson
2021-02-16 2:10 ` Maciej Fijalkowski
2021-02-15 20:49 ` John Fastabend
2021-02-16 2:38 ` Maciej Fijalkowski
2021-02-16 18:19 ` John Fastabend
2021-02-16 20:10 ` Maciej Fijalkowski
2021-02-16 9:20 ` Björn Töpel
2021-02-16 10:39 ` Toke Høiland-Jørgensen
2021-02-16 19:15 ` John Fastabend
2021-02-16 20:50 ` Maciej Fijalkowski
2021-02-16 21:17 ` John Fastabend
2021-02-15 15:46 ` [PATCH bpf-next 2/3] libbpf: clear map_info before each bpf_obj_get_info_by_fd Maciej Fijalkowski
2021-02-15 20:33 ` John Fastabend
2021-02-16 2:42 ` Maciej Fijalkowski
2021-02-15 15:46 ` [PATCH bpf-next 3/3] samples: bpf: do not unload prog within xdpsock Maciej Fijalkowski
2021-02-15 20:24 ` John Fastabend
2021-02-16 9:22 ` Björn Töpel
2021-02-16 14:15 ` Maciej Fijalkowski
2021-02-15 16:07 ` [PATCH bpf-next 0/3] Introduce bpf_link in libbpf's xsk Björn Töpel
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=fe0c957e-d212-4265-a271-ba301c3c5eca@intel.com \
--to=bjorn.topel@intel.com \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=ciara.loftus@intel.com \
--cc=daniel@iogearbox.net \
--cc=maciej.fijalkowski@intel.com \
--cc=magnus.karlsson@intel.com \
--cc=netdev@vger.kernel.org \
--cc=toke@redhat.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).