netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andrii Nakryiko <andrii.nakryiko@gmail.com>
To: Stanislav Fomichev <sdf@google.com>
Cc: Andrii Nakryiko <andriin@fb.com>, bpf <bpf@vger.kernel.org>,
	Networking <netdev@vger.kernel.org>,
	Alexei Starovoitov <ast@fb.com>,
	Daniel Borkmann <daniel@iogearbox.net>,
	David Ahern <dsahern@gmail.com>, Kernel Team <kernel-team@fb.com>
Subject: Re: [PATCH v4 bpf-next 3/9] bpf, xdp: extract common XDP program attachment logic
Date: Tue, 11 Aug 2020 19:19:23 -0700	[thread overview]
Message-ID: <CAEf4BzbXLg2EogSt1+oqmKY54E1gcVo3FLpY78p9jUrBQST_yA@mail.gmail.com> (raw)
In-Reply-To: <20200811181403.GH184844@google.com>

On Tue, Aug 11, 2020 at 11:14 AM <sdf@google.com> wrote:
>
> On 07/21, Andrii Nakryiko wrote:
> > Further refactor XDP attachment code. dev_change_xdp_fd() is split into
> > two
> > parts: getting bpf_progs from FDs and attachment logic, working with
> > bpf_progs. This makes attachment  logic a bit more straightforward and
> > prepares code for bpf_xdp_link inclusion, which will share the common
> > logic.
> It looks like this patch breaks xdp tests for me:
> * test_xdping.sh
> * test_xdp_vlan.sh
>
> Can you please verify on your side?
>
> Looking at tools/testing/selftests/bpf/xdping.c I see it has:
> static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
>
> And it attaches program two times in the same net namespace,
> so I don't see how it could've worked before the change :-/
> (unless, of coarse, the previous code was buggy).

Ok, so according to the old logic, XDP_FLAGS_UPDATE_IF_NOEXIST flag is
only checked if new program fd is not -1. So if we are installing a
new program and specify XDP_FLAGS_UPDATE_IF_NOEXIST, we'll be allowed
to do this only if there is no BPF program already attached. But we
are uninstalling program, then XDP_FLAGS_UPDATE_IF_NOEXIST is ignored
and we are allowed to uninstall any BPF program.

I can easily fix this by moving the XDP_FLAGS_UPDATE_IF_NOEXIST check
inside `if (new_prog) {}` section. I'm not sure which semantics was
actually originally intended. Maybe XDP folks can chime in here?

  reply	other threads:[~2020-08-12  2:19 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-07-22  6:45 [PATCH v4 bpf-next 0/9] BPF XDP link Andrii Nakryiko
2020-07-22  6:45 ` [PATCH v4 bpf-next 1/9] bpf: make bpf_link API available indepently of CONFIG_BPF_SYSCALL Andrii Nakryiko
2020-07-22  6:45 ` [PATCH v4 bpf-next 2/9] bpf, xdp: maintain info on attached XDP BPF programs in net_device Andrii Nakryiko
2020-07-22 15:26   ` Maciej Fijalkowski
2020-07-22 19:35     ` Andrii Nakryiko
2020-07-22 19:56       ` Maciej Fijalkowski
2020-07-22  6:45 ` [PATCH v4 bpf-next 3/9] bpf, xdp: extract common XDP program attachment logic Andrii Nakryiko
2020-07-22 19:13   ` Maciej Fijalkowski
2020-07-22 19:29     ` Andrii Nakryiko
2020-07-27 12:07   ` Shay Agroskin
2020-07-27 18:51     ` Andrii Nakryiko
2020-08-11 18:14   ` sdf
2020-08-12  2:19     ` Andrii Nakryiko [this message]
2020-07-22  6:45 ` [PATCH v4 bpf-next 4/9] bpf, xdp: add bpf_link-based XDP attachment API Andrii Nakryiko
2020-07-22  6:45 ` [PATCH v4 bpf-next 5/9] bpf, xdp: implement LINK_UPDATE for BPF XDP link Andrii Nakryiko
2020-07-22  6:45 ` [PATCH v4 bpf-next 6/9] bpf: implement BPF XDP link-specific introspection APIs Andrii Nakryiko
2020-07-22  6:46 ` [PATCH v4 bpf-next 7/9] libbpf: add support for BPF XDP link Andrii Nakryiko
2020-07-22  6:46 ` [PATCH v4 bpf-next 8/9] selftests/bpf: add BPF XDP link selftests Andrii Nakryiko
2020-07-22  6:46 ` [PATCH v4 bpf-next 9/9] bpf, xdp: remove XDP_QUERY_PROG and XDP_QUERY_PROG_HW XDP commands Andrii Nakryiko
2020-07-26  3:46 ` [PATCH v4 bpf-next 0/9] BPF XDP link Alexei Starovoitov

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=CAEf4BzbXLg2EogSt1+oqmKY54E1gcVo3FLpY78p9jUrBQST_yA@mail.gmail.com \
    --to=andrii.nakryiko@gmail.com \
    --cc=andriin@fb.com \
    --cc=ast@fb.com \
    --cc=bpf@vger.kernel.org \
    --cc=daniel@iogearbox.net \
    --cc=dsahern@gmail.com \
    --cc=kernel-team@fb.com \
    --cc=netdev@vger.kernel.org \
    --cc=sdf@google.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).