From: Martin KaFai Lau <martin.lau@linux.dev>
To: Amery Hung <ameryhung@gmail.com>
Cc: Amery Hung <amery.hung@bytedance.com>,
bpf@vger.kernel.org, netdev@vger.kernel.org,
daniel@iogearbox.net, andrii@kernel.org,
alexei.starovoitov@gmail.com, martin.lau@kernel.org,
sinquersw@gmail.com, toke@redhat.com, jhs@mojatatu.com,
jiri@resnulli.us, stfomichev@gmail.com,
ekarani.silvestre@ccc.ufcg.edu.br, yangpeihao@sjtu.edu.cn,
xiyou.wangcong@gmail.com, yepeilin.cs@gmail.com
Subject: Re: [PATCH bpf-next v1 01/13] bpf: Support getting referenced kptr from struct_ops argument
Date: Thu, 19 Dec 2024 15:06:28 -0800 [thread overview]
Message-ID: <0d09a543-b98c-4618-93e2-eef7bf8aec63@linux.dev> (raw)
In-Reply-To: <CAMB2axOPhr4fMa0J+u_V4TauohzUTwEex6HMwtgdoa8pqT0Mgw@mail.gmail.com>
On 12/18/24 8:57 AM, Amery Hung wrote:
>> At the first glance, the ref_obj_id assignment looks racy because ctx_arg_info
>> is shared by different bpf progs that may be verified in parallel. After another
>> thought, this should be fine because it should always end up having the same
>> ref_obj_id for the same arg-no, right? Not sure if UBSAN can understand this
>> without using the READ/WRITE_ONCE. but adding READ/WRITE_ONCE when using
>> ref_obj_id will be quite puzzling when reading the verifier code. Any better idea?
>>
> It looks like ref_obj_id cannot be reused (id always comes from
> ++env->id_gen), and these will be the earliest references to acquire.
> So, maybe we can assume the ref_obj_id without needing to store it in
> ctx_arg_info? E.g., the first __ref argument's ref_obj_id is always 1.
That seems reasonable to me. Then ctx_arg_info can stay read-only after the very
first initialization during bpf_struct_ops_desc_init().
next prev parent reply other threads:[~2024-12-19 23:06 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-12-13 23:29 [PATCH bpf-next v1 00/13] bpf qdisc Amery Hung
2024-12-13 23:29 ` [PATCH bpf-next v1 01/13] bpf: Support getting referenced kptr from struct_ops argument Amery Hung
2024-12-18 0:58 ` Martin KaFai Lau
2024-12-18 1:24 ` Alexei Starovoitov
2024-12-18 16:09 ` Amery Hung
2024-12-18 17:20 ` Alexei Starovoitov
2024-12-18 1:44 ` Jakub Kicinski
2024-12-18 16:57 ` Amery Hung
2024-12-19 23:06 ` Martin KaFai Lau [this message]
2024-12-13 23:29 ` [PATCH bpf-next v1 02/13] selftests/bpf: Test referenced kptr arguments of struct_ops programs Amery Hung
2024-12-18 1:17 ` Martin KaFai Lau
2024-12-18 16:10 ` Amery Hung
2024-12-19 3:40 ` Yonghong Song
2024-12-19 20:49 ` Amery Hung
2024-12-13 23:29 ` [PATCH bpf-next v1 03/13] bpf: Allow struct_ops prog to return referenced kptr Amery Hung
2024-12-18 22:29 ` Martin KaFai Lau
2024-12-13 23:29 ` [PATCH bpf-next v1 04/13] selftests/bpf: Test returning referenced kptr from struct_ops programs Amery Hung
2024-12-13 23:29 ` [PATCH bpf-next v1 05/13] bpf: net_sched: Support implementation of Qdisc_ops in bpf Amery Hung
2024-12-14 4:51 ` Cong Wang
2024-12-18 23:37 ` Martin KaFai Lau
2024-12-13 23:29 ` [PATCH bpf-next v1 06/13] bpf: net_sched: Add basic bpf qdisc kfuncs Amery Hung
2024-12-18 17:11 ` Amery Hung
2024-12-19 7:37 ` Martin KaFai Lau
2024-12-20 0:32 ` Amery Hung
2024-12-13 23:29 ` [PATCH bpf-next v1 07/13] bpf: net_sched: Add a qdisc watchdog timer Amery Hung
2024-12-19 1:16 ` Martin KaFai Lau
2024-12-20 19:24 ` Amery Hung
2024-12-13 23:29 ` [PATCH bpf-next v1 08/13] bpf: net_sched: Support updating bstats Amery Hung
2024-12-13 23:29 ` [PATCH bpf-next v1 09/13] bpf: net_sched: Support updating qstats Amery Hung
2024-12-13 23:29 ` [PATCH bpf-next v1 10/13] bpf: net_sched: Allow writing to more Qdisc members Amery Hung
2024-12-13 23:29 ` [PATCH bpf-next v1 11/13] libbpf: Support creating and destroying qdisc Amery Hung
2024-12-17 18:32 ` Andrii Nakryiko
2024-12-17 19:08 ` Amery Hung
2024-12-13 23:29 ` [PATCH bpf-next v1 12/13] selftests: Add a basic fifo qdisc test Amery Hung
2024-12-13 23:29 ` [PATCH bpf-next v1 13/13] selftests: Add a bpf fq qdisc to selftest Amery Hung
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=0d09a543-b98c-4618-93e2-eef7bf8aec63@linux.dev \
--to=martin.lau@linux.dev \
--cc=alexei.starovoitov@gmail.com \
--cc=amery.hung@bytedance.com \
--cc=ameryhung@gmail.com \
--cc=andrii@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=ekarani.silvestre@ccc.ufcg.edu.br \
--cc=jhs@mojatatu.com \
--cc=jiri@resnulli.us \
--cc=martin.lau@kernel.org \
--cc=netdev@vger.kernel.org \
--cc=sinquersw@gmail.com \
--cc=stfomichev@gmail.com \
--cc=toke@redhat.com \
--cc=xiyou.wangcong@gmail.com \
--cc=yangpeihao@sjtu.edu.cn \
--cc=yepeilin.cs@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;
as well as URLs for NNTP newsgroup(s).