From: Martin KaFai Lau <martin.lau@linux.dev>
To: thinker.li@gmail.com
Cc: sinquersw@gmail.com, kuifeng@meta.com, bpf@vger.kernel.org,
ast@kernel.org, song@kernel.org, kernel-team@meta.com,
andrii@kernel.org, drosen@google.com
Subject: Re: [PATCH bpf-next v8 06/10] bpf: pass attached BTF to the bpf_struct_ops subsystem
Date: Mon, 30 Oct 2023 18:53:37 -0700 [thread overview]
Message-ID: <ff0e6978-adb5-db47-5968-5af4924aadba@linux.dev> (raw)
In-Reply-To: <20231030192810.382942-7-thinker.li@gmail.com>
On 10/30/23 12:28 PM, thinker.li@gmail.com wrote:
> From: Kui-Feng Lee <thinker.li@gmail.com>
>
> Giving a BTF, the bpf_struct_ops knows the right place to look up type info
> associated with a type ID. This enables a user space program to load a
> struct_ops object linked to a struct_ops type defined by a module, by
> providing the module BTF (fd).
This describes about the struct_ops map creation change (by adding
value_type_btf_obj_fd)? It could be described more clearly in the commit
message, like specify the value_type_btf_obj_fd addition and how it is used in
the struct_ops map creation.
>
> The bpf_prog includes attach_btf in aux which is passed along with the
> bpf_attr when loading the program. The purpose of attach_btf is to
> determine the btf type of attach_btf_id. The attach_btf_id is then used to
> identify the traced function for a trace program. In the case of struct_ops
> programs, it is used to identify the struct_ops type of the struct_ops
> object that a program is attached to.
Does attach_btf_obj_fd also work?
[ ... ]
> diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c
> index 256516aba632..db2bbba50e38 100644
> --- a/kernel/bpf/bpf_struct_ops.c
> +++ b/kernel/bpf/bpf_struct_ops.c
> @@ -694,6 +694,7 @@ static void __bpf_struct_ops_map_free(struct bpf_map *map)
> bpf_jit_uncharge_modmem(PAGE_SIZE);
> }
> bpf_map_area_free(st_map->uvalue);
> + btf_put(st_map->st_ops_desc->btf);
> bpf_map_area_free(st_map);
> }
>
> @@ -735,16 +736,31 @@ static struct bpf_map *bpf_struct_ops_map_alloc(union bpf_attr *attr)
> const struct btf_type *t, *vt;
> struct module *mod = NULL;
> struct bpf_map *map;
> + struct btf *btf;
> int ret;
>
> - st_ops_desc = bpf_struct_ops_find_value(btf_vmlinux, attr->btf_vmlinux_value_type_id);
> - if (!st_ops_desc)
> - return ERR_PTR(-ENOTSUPP);
> + if (attr->value_type_btf_obj_fd) {
> + /* The map holds btf for its whole life time. */
It took me a while to parse this comment and connect it with the
btf_put(st_map->st_ops_desc->btf) in the __bpf_struct_ops_map_free() above.
It is now like "btf" owns "struct_ops_desc" which also stores a pointer pointing
back to itself, like "btf->struct_ops_desc->btf". The struct_ops_desc->btf was
not initialized by st_map but st_map will increment its refcount much later.
Can btf be directly stored in the st_map->btf instead and map_alloc holds the
refcnt of st_map->btf and btf_put(st_map->btf) in map_free?
> + btf = btf_get_by_fd(attr->value_type_btf_obj_fd);
> + if (IS_ERR(btf))
> + return ERR_PTR(PTR_ERR(btf));
> +
> + if (btf != btf_vmlinux) {
> + mod = btf_try_get_module(btf);
> + if (!mod) {
> + ret = -EINVAL;
> + goto errout;
> + }
> + }
> + } else {
> + btf = btf_vmlinux;
> + btf_get(btf);
> + }
>
> - if (st_ops_desc->btf != btf_vmlinux) {
> - mod = btf_try_get_module(st_ops_desc->btf);
> - if (!mod)
> - return ERR_PTR(-EINVAL);
> + st_ops_desc = bpf_struct_ops_find_value(btf, attr->btf_vmlinux_value_type_id);
> + if (!st_ops_desc) {
> + ret = -ENOTSUPP;
> + goto errout;
> }
>
> vt = st_ops_desc->value_type;
next prev parent reply other threads:[~2023-10-31 1:53 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-30 19:28 [PATCH bpf-next v8 00/10] Registrating struct_ops types from modules thinker.li
2023-10-30 19:28 ` [PATCH bpf-next v8 01/10] bpf: refactory struct_ops type initialization to a function thinker.li
2023-10-30 19:28 ` [PATCH bpf-next v8 02/10] bpf, net: introduce bpf_struct_ops_desc thinker.li
2023-10-31 6:40 ` Martin KaFai Lau
2023-10-31 16:00 ` Kui-Feng Lee
2023-10-30 19:28 ` [PATCH bpf-next v8 03/10] bpf: add struct_ops_tab to btf thinker.li
2023-10-31 1:09 ` Martin KaFai Lau
2023-10-31 16:57 ` Kui-Feng Lee
2023-10-30 19:28 ` [PATCH bpf-next v8 04/10] bpf: hold module for bpf_struct_ops_map thinker.li
2023-10-31 1:21 ` Martin KaFai Lau
2023-10-31 17:46 ` Kui-Feng Lee
2023-10-30 19:28 ` [PATCH bpf-next v8 05/10] bpf: validate value_type thinker.li
2023-10-30 19:28 ` [PATCH bpf-next v8 06/10] bpf: pass attached BTF to the bpf_struct_ops subsystem thinker.li
2023-10-31 1:53 ` Martin KaFai Lau [this message]
2023-10-31 20:31 ` Kui-Feng Lee
2023-10-30 19:28 ` [PATCH bpf-next v8 07/10] bpf, net: switch to dynamic registration thinker.li
2023-10-31 6:36 ` Martin KaFai Lau
2023-10-31 23:34 ` Kui-Feng Lee
2023-11-01 0:02 ` Martin KaFai Lau
2023-11-01 0:19 ` Kui-Feng Lee
2023-11-01 0:19 ` Kui-Feng Lee
2023-11-02 0:17 ` Martin KaFai Lau
2023-11-02 0:59 ` Kui-Feng Lee
2023-11-02 1:32 ` Martin KaFai Lau
2023-11-02 4:19 ` Kui-Feng Lee
2023-10-30 19:28 ` [PATCH bpf-next v8 08/10] libbpf: Find correct module BTFs for struct_ops maps and progs thinker.li
2023-10-30 19:28 ` [PATCH bpf-next v8 09/10] bpf: export btf_ctx_access to modules thinker.li
2023-10-30 19:28 ` [PATCH bpf-next v8 10/10] selftests/bpf: test case for register_bpf_struct_ops() thinker.li
2023-10-31 6:59 ` Martin KaFai Lau
2023-11-01 0:30 ` Kui-Feng Lee
2023-11-02 1:43 ` Martin KaFai Lau
2023-11-02 18:26 ` Kui-Feng Lee
2023-10-31 20:45 ` [PATCH bpf-next v8 00/10] Registrating struct_ops types from modules Martin KaFai Lau
2023-11-01 0:48 ` Kui-Feng Lee
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=ff0e6978-adb5-db47-5968-5af4924aadba@linux.dev \
--to=martin.lau@linux.dev \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=drosen@google.com \
--cc=kernel-team@meta.com \
--cc=kuifeng@meta.com \
--cc=sinquersw@gmail.com \
--cc=song@kernel.org \
--cc=thinker.li@gmail.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