From: Kui-Feng Lee <sinquersw@gmail.com>
To: Eduard Zingerman <eddyz87@gmail.com>,
bpf@vger.kernel.org, ast@kernel.org
Cc: andrii@kernel.org, daniel@iogearbox.net, martin.lau@linux.dev,
kernel-team@fb.com, yonghong.song@linux.dev, void@manifault.com
Subject: Re: [PATCH bpf-next v1 7/8] libbpf: sync progs autoload with maps autocreate for struct_ops maps
Date: Tue, 27 Feb 2024 14:55:48 -0800 [thread overview]
Message-ID: <ec9d8997-f5a2-44b6-9bc4-2caaf19df8a9@gmail.com> (raw)
In-Reply-To: <20240227204556.17524-8-eddyz87@gmail.com>
On 2/27/24 12:45, Eduard Zingerman wrote:
> Make bpf_map__set_autocreate() for struct_ops maps toggle autoload
> state for referenced programs.
>
> E.g. for the BPF code below:
>
> SEC("struct_ops/test_1") int BPF_PROG(foo) { ... }
> SEC("struct_ops/test_2") int BPF_PROG(bar) { ... }
>
> SEC(".struct_ops.link")
> struct test_ops___v1 A = {
> .foo = (void *)foo
> };
>
> SEC(".struct_ops.link")
> struct test_ops___v2 B = {
> .foo = (void *)foo,
> .bar = (void *)bar,
> };
>
> And the following libbpf API calls:
>
> bpf_map__set_autocreate(skel->maps.A, true);
> bpf_map__set_autocreate(skel->maps.B, false);
>
> The autoload would be enabled for program 'foo' and disabled for
> program 'bar'.
>
> Do not apply such toggling if program autoload state is set by a call
> to bpf_program__set_autoload().
>
> Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
> ---
> tools/lib/bpf/libbpf.c | 35 ++++++++++++++++++++++++++++++++++-
> 1 file changed, 34 insertions(+), 1 deletion(-)
>
> diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
> index b39d3f2898a1..1ea3046724f8 100644
> --- a/tools/lib/bpf/libbpf.c
> +++ b/tools/lib/bpf/libbpf.c
> @@ -446,13 +446,18 @@ struct bpf_program {
> struct bpf_object *obj;
>
> int fd;
> - bool autoload;
> + bool autoload:1;
> + bool autoload_user_set:1;
> bool autoattach;
> bool sym_global;
> bool mark_btf_static;
> enum bpf_prog_type type;
> enum bpf_attach_type expected_attach_type;
> int exception_cb_idx;
> + /* total number of struct_ops maps with autocreate == true
> + * that reference this program
> + */
> + __u32 struct_ops_refs;
>
> int prog_ifindex;
> __u32 attach_btf_obj_fd;
> @@ -4509,6 +4514,28 @@ static int bpf_get_map_info_from_fdinfo(int fd, struct bpf_map_info *info)
> return 0;
> }
>
> +/* Sync autoload and autocreate state between struct_ops map and
> + * referenced programs.
> + */
> +static void bpf_map__struct_ops_toggle_progs_autoload(struct bpf_map *map, bool autocreate)
> +{
> + struct bpf_program *prog;
> + int i;
> +
> + for (i = 0; i < btf_vlen(map->st_ops->type); ++i) {
> + prog = map->st_ops->progs[i];
> +
> + if (!prog || prog->autoload_user_set)
> + continue;
> +
> + if (autocreate)
> + prog->struct_ops_refs++;
> + else
> + prog->struct_ops_refs--;
> + prog->autoload = prog->struct_ops_refs != 0;
> + }
> +}
> +
This part is related to the other patch [1], which allows
a user to change the value of a function pointer field. The behavior of
autocreate and autoload may suprise a user if the user call
bpf_map__set_autocreate() after changing the value of a function pointer
field.
[1]
https://lore.kernel.org/all/20240227010432.714127-1-thinker.li@gmail.com/
> bool bpf_map__autocreate(const struct bpf_map *map)
> {
> return map->autocreate;
> @@ -4519,6 +4546,9 @@ int bpf_map__set_autocreate(struct bpf_map *map, bool autocreate)
> if (map->obj->loaded)
> return libbpf_err(-EBUSY);
>
> + if (map->st_ops && map->autocreate != autocreate)
> + bpf_map__struct_ops_toggle_progs_autoload(map, autocreate);
> +
> map->autocreate = autocreate;
> return 0;
> }
> @@ -8801,6 +8831,7 @@ int bpf_program__set_autoload(struct bpf_program *prog, bool autoload)
> return libbpf_err(-EINVAL);
>
> prog->autoload = autoload;
> + prog->autoload_user_set = 1;
> return 0;
> }
>
> @@ -9428,6 +9459,8 @@ static int bpf_object__collect_st_ops_relos(struct bpf_object *obj,
> return -EINVAL;
> }
>
> + if (map->autocreate)
> + prog->struct_ops_refs++;
> st_ops->progs[member_idx] = prog;
> }
>
next prev parent reply other threads:[~2024-02-27 22:55 UTC|newest]
Thread overview: 61+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-02-27 20:45 [PATCH bpf-next v1 0/8] libbpf: type suffixes and autocreate flag for struct_ops maps Eduard Zingerman
2024-02-27 20:45 ` [PATCH bpf-next v1 1/8] libbpf: allow version suffixes (___smth) for struct_ops types Eduard Zingerman
2024-02-27 21:47 ` Kui-Feng Lee
2024-02-27 21:49 ` Eduard Zingerman
2024-02-28 16:29 ` David Vernet
2024-02-28 17:28 ` Eduard Zingerman
2024-02-28 17:30 ` David Vernet
2024-02-28 23:21 ` Andrii Nakryiko
2024-02-28 23:37 ` Eduard Zingerman
2024-02-27 20:45 ` [PATCH bpf-next v1 2/8] libbpf: tie struct_ops programs to kernel BTF ids, not to local ids Eduard Zingerman
2024-02-28 7:41 ` Martin KaFai Lau
2024-02-28 17:23 ` David Vernet
2024-02-28 17:40 ` Eduard Zingerman
2024-02-28 17:50 ` David Vernet
2024-02-28 23:28 ` Andrii Nakryiko
2024-02-28 23:31 ` Eduard Zingerman
2024-02-28 23:34 ` Andrii Nakryiko
2024-02-27 20:45 ` [PATCH bpf-next v1 3/8] libbpf: honor autocreate flag for struct_ops maps Eduard Zingerman
2024-02-28 17:44 ` David Vernet
2024-02-27 20:45 ` [PATCH bpf-next v1 4/8] selftests/bpf: test struct_ops map definition with type suffix Eduard Zingerman
2024-02-28 18:03 ` David Vernet
2024-02-27 20:45 ` [PATCH bpf-next v1 5/8] selftests/bpf: bad_struct_ops test Eduard Zingerman
2024-02-28 18:15 ` David Vernet
2024-02-28 20:06 ` Eduard Zingerman
2024-02-28 20:11 ` David Vernet
2024-02-28 23:40 ` Andrii Nakryiko
2024-02-28 23:44 ` Eduard Zingerman
2024-02-28 23:56 ` Andrii Nakryiko
2024-02-29 0:06 ` Eduard Zingerman
2024-02-27 20:45 ` [PATCH bpf-next v1 6/8] selftests/bpf: test autocreate behavior for struct_ops maps Eduard Zingerman
2024-02-28 18:29 ` David Vernet
2024-02-28 18:34 ` David Vernet
2024-02-28 19:31 ` Eduard Zingerman
2024-02-28 23:43 ` Andrii Nakryiko
2024-02-28 23:55 ` Eduard Zingerman
2024-02-29 0:02 ` Andrii Nakryiko
2024-02-29 0:56 ` Martin KaFai Lau
2024-03-01 1:28 ` Eduard Zingerman
2024-03-01 18:03 ` Andrii Nakryiko
2024-03-01 18:07 ` Eduard Zingerman
2024-02-27 20:45 ` [PATCH bpf-next v1 7/8] libbpf: sync progs autoload with maps autocreate " Eduard Zingerman
2024-02-27 22:55 ` Kui-Feng Lee [this message]
2024-02-27 23:09 ` Eduard Zingerman
2024-02-27 23:16 ` Kui-Feng Lee
2024-02-27 23:30 ` Eduard Zingerman
2024-02-27 23:40 ` Kui-Feng Lee
2024-02-27 23:43 ` Eduard Zingerman
2024-02-28 0:12 ` Kui-Feng Lee
2024-02-28 0:50 ` Eduard Zingerman
2024-02-28 2:10 ` Martin KaFai Lau
2024-02-28 12:36 ` Eduard Zingerman
2024-02-28 23:55 ` Andrii Nakryiko
2024-02-29 0:04 ` Eduard Zingerman
2024-02-29 0:14 ` Andrii Nakryiko
2024-02-29 0:25 ` Martin KaFai Lau
2024-02-29 0:30 ` Andrii Nakryiko
2024-02-29 0:37 ` Martin KaFai Lau
2024-02-29 0:40 ` Eduard Zingerman
2024-02-27 20:45 ` [PATCH bpf-next v1 8/8] selftests/bpf: tests for struct_ops autoload/autocreate toggling Eduard Zingerman
2024-02-28 18:36 ` David Vernet
2024-02-28 20:10 ` Eduard Zingerman
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=ec9d8997-f5a2-44b6-9bc4-2caaf19df8a9@gmail.com \
--to=sinquersw@gmail.com \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=eddyz87@gmail.com \
--cc=kernel-team@fb.com \
--cc=martin.lau@linux.dev \
--cc=void@manifault.com \
--cc=yonghong.song@linux.dev \
/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