From: Andrii Nakryiko <andrii.nakryiko@gmail.com>
To: "Toke Høiland-Jørgensen" <toke@redhat.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>,
Alexei Starovoitov <ast@kernel.org>,
Martin KaFai Lau <kafai@fb.com>, Song Liu <songliubraving@fb.com>,
Yonghong Song <yhs@fb.com>,
Jesper Dangaard Brouer <brouer@redhat.com>,
David Miller <davem@davemloft.net>,
Networking <netdev@vger.kernel.org>, bpf <bpf@vger.kernel.org>
Subject: Re: [PATCH bpf-next v2 1/6] libbpf: Unpin auto-pinned maps if loading fails
Date: Fri, 8 Nov 2019 15:35:56 -0800 [thread overview]
Message-ID: <CAEf4BzZSfj98sd=16F4q5m5q0rYbVwsB2L96mmWdTJdAFhix3Q@mail.gmail.com> (raw)
In-Reply-To: <87pni2q6yi.fsf@toke.dk>
On Fri, Nov 8, 2019 at 3:33 PM Toke Høiland-Jørgensen <toke@redhat.com> wrote:
>
> Andrii Nakryiko <andrii.nakryiko@gmail.com> writes:
>
> > On Fri, Nov 8, 2019 at 1:33 PM Toke Høiland-Jørgensen <toke@redhat.com> wrote:
> >>
> >> From: Toke Høiland-Jørgensen <toke@redhat.com>
> >>
> >> Since the automatic map-pinning happens during load, it will leave pinned
> >> maps around if the load fails at a later stage. Fix this by unpinning any
> >> pinned maps on cleanup. To avoid unpinning pinned maps that were reused
> >> rather than newly pinned, add a new boolean property on struct bpf_map to
> >> keep track of whether that map was reused or not; and only unpin those maps
> >> that were not reused.
> >>
> >> Fixes: 57a00f41644f ("libbpf: Add auto-pinning of maps when loading BPF objects")
> >> Acked-by: Song Liu <songliubraving@fb.com>
> >> Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
> >> ---
> >> tools/lib/bpf/libbpf.c | 16 +++++++++++++---
> >> 1 file changed, 13 insertions(+), 3 deletions(-)
> >>
> >> diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
> >> index be4af95d5a2c..cea61b2ec9d3 100644
> >> --- a/tools/lib/bpf/libbpf.c
> >> +++ b/tools/lib/bpf/libbpf.c
> >> @@ -229,6 +229,7 @@ struct bpf_map {
> >> enum libbpf_map_type libbpf_type;
> >> char *pin_path;
> >> bool pinned;
> >> + bool was_reused;
> >
> > nit: just reused, similar to pinned?
> >
> >> };
> >>
> >> struct bpf_secdata {
> >> @@ -1995,6 +1996,7 @@ int bpf_map__reuse_fd(struct bpf_map *map, int fd)
> >> map->def.map_flags = info.map_flags;
> >> map->btf_key_type_id = info.btf_key_type_id;
> >> map->btf_value_type_id = info.btf_value_type_id;
> >> + map->was_reused = true;
> >>
> >> return 0;
> >>
> >> @@ -4007,15 +4009,18 @@ bpf_object__open_buffer(const void *obj_buf, size_t obj_buf_sz,
> >> return bpf_object__open_mem(obj_buf, obj_buf_sz, &opts);
> >> }
> >>
> >> -int bpf_object__unload(struct bpf_object *obj)
> >> +static int __bpf_object__unload(struct bpf_object *obj, bool unpin)
> >> {
> >> size_t i;
> >>
> >> if (!obj)
> >> return -EINVAL;
> >>
> >> - for (i = 0; i < obj->nr_maps; i++)
> >> + for (i = 0; i < obj->nr_maps; i++) {
> >> zclose(obj->maps[i].fd);
> >> + if (unpin && obj->maps[i].pinned && !obj->maps[i].was_reused)
> >> + bpf_map__unpin(&obj->maps[i], NULL);
> >> + }
> >>
> >> for (i = 0; i < obj->nr_programs; i++)
> >> bpf_program__unload(&obj->programs[i]);
> >> @@ -4023,6 +4028,11 @@ int bpf_object__unload(struct bpf_object *obj)
> >> return 0;
> >> }
> >>
> >> +int bpf_object__unload(struct bpf_object *obj)
> >> +{
> >> + return __bpf_object__unload(obj, false);
> >> +}
> >> +
> >> int bpf_object__load_xattr(struct bpf_object_load_attr *attr)
> >> {
> >> struct bpf_object *obj;
> >> @@ -4047,7 +4057,7 @@ int bpf_object__load_xattr(struct bpf_object_load_attr *attr)
> >>
> >> return 0;
> >> out:
> >> - bpf_object__unload(obj);
> >> + __bpf_object__unload(obj, true);
> >
> > giving this is the only (special) case of auto-unpinning auto-pinned
> > maps, why not do a trivial loop here, instead of having this extra
> > unpin flag and extra __bpf_object__unload function?
>
> Oh, you mean just do a loop in addition to the call to __unload? Sure, I
> guess we can do that instead...
I think that's cleaner, because it's custom clean up logic in one
place, rather than supported feature of unload.
>
> -Toke
next prev parent reply other threads:[~2019-11-08 23:36 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-11-08 21:33 [PATCH bpf-next v2 0/6] libbpf: Fix pinning and error message bugs and add new getters Toke Høiland-Jørgensen
2019-11-08 21:33 ` [PATCH bpf-next v2 1/6] libbpf: Unpin auto-pinned maps if loading fails Toke Høiland-Jørgensen
2019-11-08 21:41 ` David Miller
2019-11-08 22:40 ` Andrii Nakryiko
2019-11-08 23:33 ` Toke Høiland-Jørgensen
2019-11-08 23:35 ` Andrii Nakryiko [this message]
2019-11-08 21:33 ` [PATCH bpf-next v2 2/6] selftests/bpf: Add tests for automatic map unpinning on load failure Toke Høiland-Jørgensen
2019-11-08 21:41 ` David Miller
2019-11-08 22:00 ` Song Liu
2019-11-08 22:43 ` Andrii Nakryiko
2019-11-08 21:33 ` [PATCH bpf-next v2 3/6] libbpf: Propagate EPERM to caller on program load Toke Høiland-Jørgensen
2019-11-08 21:43 ` David Miller
2019-11-08 22:50 ` Andrii Nakryiko
2019-11-08 23:17 ` Alexei Starovoitov
2019-11-08 23:32 ` Toke Høiland-Jørgensen
2019-11-08 21:33 ` [PATCH bpf-next v2 4/6] libbpf: Use pr_warn() when printing netlink errors Toke Høiland-Jørgensen
2019-11-08 21:43 ` David Miller
2019-11-08 22:52 ` Andrii Nakryiko
2019-11-08 21:33 ` [PATCH bpf-next v2 5/6] libbpf: Add bpf_get_link_xdp_info() function to get more XDP information Toke Høiland-Jørgensen
2019-11-08 21:43 ` David Miller
2019-11-08 23:10 ` Andrii Nakryiko
2019-11-08 21:33 ` [PATCH bpf-next v2 6/6] libbpf: Add getter for program size Toke Høiland-Jørgensen
2019-11-08 21:43 ` David Miller
2019-11-08 23:16 ` Andrii Nakryiko
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='CAEf4BzZSfj98sd=16F4q5m5q0rYbVwsB2L96mmWdTJdAFhix3Q@mail.gmail.com' \
--to=andrii.nakryiko@gmail.com \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=brouer@redhat.com \
--cc=daniel@iogearbox.net \
--cc=davem@davemloft.net \
--cc=kafai@fb.com \
--cc=netdev@vger.kernel.org \
--cc=songliubraving@fb.com \
--cc=toke@redhat.com \
--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;
as well as URLs for NNTP newsgroup(s).