Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20230516111823.2103536-1-starmiku1207184332@gmail.com>
 <e37c0a65-a3f6-e2e6-c2ad-367db20253a0@meta.com> <CALyQVax8X63qekZVhvRTmZFFs+ucPKRkBB7UnRZk6Hu3ggi7Og@mail.gmail.com>
 <57dc6a0e-6ba9-e77c-80ac-6bb0a6e2650a@meta.com> <CALyQVazb=D1ejapiFdTnan6JbjFJA2q9ifhSsmF4OC9MDz3oAw@mail.gmail.com>
 <d027cb6b-e32c-36ad-3aba-9a7b1177f89f@meta.com> <CALyQVayW7e4FPbaMNNuOmYGYt5pcd47zsx2xVkrekEDaVm7H2g@mail.gmail.com>
 <113dc8c1-0840-9ee3-2840-28246731604c@meta.com>
In-Reply-To: <113dc8c1-0840-9ee3-2840-28246731604c@meta.com>
From:   Alexei Starovoitov <alexei.starovoitov@gmail.com>
Date:   Wed, 24 May 2023 12:44:16 -0700
Message-ID: <CAADnVQ+5GXu8Q1_awiHExhBB9_LmGrcPTvjQEjQU58pzX3WbQQ@mail.gmail.com>
Subject: Re: [bug] kernel: bpf: syscall: a possible sleep-in-atomic bug in __bpf_prog_put()
To:     Yonghong Song <yhs@meta.com>
Cc:     Teng Qi <starmiku1207184332@gmail.com>,
        Alexei Starovoitov <ast@kernel.org>,
        Daniel Borkmann <daniel@iogearbox.net>,
        John Fastabend <john.fastabend@gmail.com>,
        Andrii Nakryiko <andrii@kernel.org>,
        Martin KaFai Lau <martin.lau@linux.dev>,
        Song Liu <song@kernel.org>, Yonghong Song <yhs@fb.com>,
        KP Singh <kpsingh@kernel.org>,
        Stanislav Fomichev <sdf@google.com>,
        Hao Luo <haoluo@google.com>, Jiri Olsa <jolsa@kernel.org>,
        "David S. Miller" <davem@davemloft.net>,
        Jakub Kicinski <kuba@kernel.org>,
        Jesper Dangaard Brouer <hawk@kernel.org>,
        bpf <bpf@vger.kernel.org>, LKML <linux-kernel@vger.kernel.org>,
        Network Development <netdev@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Wed, May 24, 2023 at 12:34=E2=80=AFPM Yonghong Song <yhs@meta.com> wrote=
:
>
>
>
> On 5/24/23 5:42 AM, Teng Qi wrote:
> > Thank you.
> >
> >> We cannot use rcu_read_lock_held() in the 'if' statement. The return
> >> value rcu_read_lock_held() could be 1 for some configurations regardle=
ss
> >> whether rcu_read_lock() is really held or not. In most cases,
> >> rcu_read_lock_held() is used in issuing potential warnings.
> >> Maybe there are other ways to record whether rcu_read_lock() is held o=
r not?
> >
> > Sorry. I was not aware of the dependency of configurations of
> > rcu_read_lock_held().
> >
> >> If we cannot resolve rcu_read_lock() presence issue, maybe the conditi=
on
> >> can be !in_interrupt(), so any process-context will go to a workqueue.
> >
> > I agree that using !in_interrupt() as a condition is an acceptable solu=
tion.
>
> This should work although it could be conservative.
>
> >
> >> Alternatively, we could have another solution. We could add another
> >> function e.g., bpf_prog_put_rcu(), which indicates that bpf_prog_put()
> >> will be done in rcu context.
> >
> > Implementing a new function like bpf_prog_put_rcu() is a solution that =
involves
> > more significant changes.
>
> Maybe we can change signature of bpf_prog_put instead? Like
>     void bpf_prog_put(struct bpf_prog *prog, bool in_rcu)
> and inside bpf_prog_put we can add
>     WARN_ON_ONCE(in_rcu && !bpf_rcu_lock_held());

bpf_rcu_lock_held() is used for different cases.

Here s/in_irq/in_interrupt/ inside bpf_prog_put() is enough
to address this theoretical issue.