From: Kui-Feng Lee <sinquersw@gmail.com>
To: Martin KaFai Lau <martin.lau@linux.dev>, thinker.li@gmail.com
Cc: 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 v5 3/9] bpf: hold module for bpf_struct_ops_map.
Date: Thu, 19 Oct 2023 22:07:42 -0700 [thread overview]
Message-ID: <02e2a704-4939-4f8c-b465-473c3a2eae1c@gmail.com> (raw)
In-Reply-To: <7ea8ebf7-3349-4461-b204-be106e3b547a@gmail.com>
On 10/19/23 09:29, Kui-Feng Lee wrote:
>
>
> On 10/18/23 17:36, Martin KaFai Lau wrote:
>> On 10/17/23 9:23 AM, thinker.li@gmail.com wrote:
>
>>
>>> }
>>> void bpf_struct_ops_init(struct btf *btf, struct bpf_verifier_log
>>> *log)
>>> @@ -215,7 +218,7 @@ void bpf_struct_ops_init(struct btf *btf, struct
>>> bpf_verifier_log *log)
>>> for (i = 0; i < ARRAY_SIZE(bpf_struct_ops); i++) {
>>> st_ops = bpf_struct_ops[i];
>>> - bpf_struct_ops_init_one(st_ops, btf, log);
>>> + bpf_struct_ops_init_one(st_ops, btf, NULL, log);
>>> }
>>> }
>>> @@ -630,6 +633,7 @@ static void __bpf_struct_ops_map_free(struct
>>> bpf_map *map)
>>> bpf_jit_uncharge_modmem(PAGE_SIZE);
>>> }
>>> bpf_map_area_free(st_map->uvalue);
>>> + module_put(st_map->st_ops->owner);
>>> bpf_map_area_free(st_map);
>>> }
>>> @@ -676,9 +680,18 @@ static struct bpf_map
>>> *bpf_struct_ops_map_alloc(union bpf_attr *attr)
>>> if (!st_ops)
>>> return ERR_PTR(-ENOTSUPP);
>>> + /* If st_ops->owner is NULL, it means the struct_ops is
>>> + * statically defined in the kernel. We don't need to
>>> + * take a refcount on it.
>>> + */
>>> + if (st_ops->owner && !btf_try_get_module(st_ops->btf))
>>
>> This just came to my mind. Is the module refcnt needed during map
>> alloc/free or it could be done during the reg/unreg instead?
>
>
> Sure, I can move it to reg/unreg.
Just found that we relies type information in st_ops to update element
and clean up maps.
We can not move get/put modules to reg/unreg except keeping a redundant
copy in
st_map or somewhere. It make the code much more complicated by
introducing get/put module here and there.
I prefer to keep as it is now. WDYT?
>
next prev parent reply other threads:[~2023-10-20 5:07 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-17 16:22 [PATCH bpf-next v5 0/9] Registrating struct_ops types from modules thinker.li
2023-10-17 16:22 ` [PATCH bpf-next v5 1/9] bpf: refactory struct_ops type initialization to a function thinker.li
2023-10-17 16:22 ` [PATCH bpf-next v5 2/9] bpf: add struct_ops_tab to btf thinker.li
2023-10-19 0:00 ` Martin KaFai Lau
2023-10-19 0:33 ` Kui-Feng Lee
2023-10-19 2:28 ` Martin KaFai Lau
2023-10-19 16:15 ` Kui-Feng Lee
2023-10-17 16:23 ` [PATCH bpf-next v5 3/9] bpf: hold module for bpf_struct_ops_map thinker.li
2023-10-19 0:36 ` Martin KaFai Lau
2023-10-19 16:29 ` Kui-Feng Lee
2023-10-20 5:07 ` Kui-Feng Lee [this message]
2023-10-20 21:37 ` Martin KaFai Lau
2023-10-20 22:28 ` Kui-Feng Lee
2023-10-17 16:23 ` [PATCH bpf-next v5 4/9] bpf: validate value_type thinker.li
2023-10-17 16:23 ` [PATCH bpf-next v5 5/9] bpf: pass attached BTF to the bpf_struct_ops subsystem thinker.li
2023-10-17 16:23 ` [PATCH bpf-next v5 6/9] bpf, net: switch to dynamic registration thinker.li
2023-10-19 1:49 ` Martin KaFai Lau
2023-10-20 15:12 ` Kui-Feng Lee
2023-10-20 17:53 ` Kui-Feng Lee
2023-10-17 16:23 ` [PATCH bpf-next v5 7/9] libbpf: Find correct module BTFs for struct_ops maps and progs thinker.li
2023-10-17 21:49 ` Andrii Nakryiko
2023-10-18 2:25 ` Kui-Feng Lee
2023-10-19 2:43 ` Martin KaFai Lau
2023-10-19 16:31 ` Kui-Feng Lee
2023-10-17 16:23 ` [PATCH bpf-next v5 8/9] bpf: export btf_ctx_access to modules thinker.li
2023-10-17 16:23 ` [PATCH bpf-next v5 9/9] selftests/bpf: test case for register_bpf_struct_ops() thinker.li
2023-10-17 18:03 ` kernel test robot
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=02e2a704-4939-4f8c-b465-473c3a2eae1c@gmail.com \
--to=sinquersw@gmail.com \
--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=martin.lau@linux.dev \
--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