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: <CAABZP2xJRGhPmfB-PrfesQKzP7fsuZsj+3TewAiLLW8u=YK4dg@mail.gmail.com>
 <CAEXW_YSSGYgqTpxqbYikCFS9t=2f+L-0phbU+gAAngB5z-FbyA@mail.gmail.com>
 <ZEXOMC2casTlobE1@boqun-archlinux> <87fs8pzalj.fsf@mail.concordia>
 <20230424151351.GP19790@gate.crashing.org> <ZEagN1jJwg+rUzX4@boqun-archlinux>
 <CAEXW_YRfetnhgCw5OgnwhgZF_U+UkHN=uy=L8ovGLqn1UCtfTg@mail.gmail.com>
 <20230425101324.GD1331236@hirez.programming.kicks-ass.net>
 <CAABZP2ypJ98T3XAqPnLrxxzrYckSQ6sn3woEmpigQ+cRRaw=Zw@mail.gmail.com> <CAEXW_YQEarLt7YGQZdwmcSyZcGRCGKf89ovxjQdXBO-TgXAk-w@mail.gmail.com>
In-Reply-To: <CAEXW_YQEarLt7YGQZdwmcSyZcGRCGKf89ovxjQdXBO-TgXAk-w@mail.gmail.com>
From:   Zhouyi Zhou <zhouzhouyi@gmail.com>
Date:   Tue, 25 Apr 2023 11:12:25 +0800
Message-ID: <CAABZP2x5UMxuNYEr6ATsbZPywdnHYYMvi7Pyy80aSqpcHk9yTw@mail.gmail.com>
Subject: Re: BUG : PowerPC RCU: torture test failed with __stack_chk_fail
To:     Joel Fernandes <joel@joelfernandes.org>
Cc:     Christophe Leroy <christophe.leroy@c-s.fr>,
        Peter Zijlstra <peterz@infradead.org>,
        Boqun Feng <boqun.feng@gmail.com>,
        Segher Boessenkool <segher@kernel.crashing.org>,
        Michael Ellerman <mpe@ellerman.id.au>,
        linuxppc-dev <linuxppc-dev@lists.ozlabs.org>,
        rcu <rcu@vger.kernel.org>,
        linux-kernel <linux-kernel@vger.kernel.org>, lance@osuosl.org,
        "Paul E. McKenney" <paulmck@kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Tue, Apr 25, 2023 at 7:06=E2=80=AFPM Joel Fernandes <joel@joelfernandes.=
org> wrote:
>
> On Tue, Apr 25, 2023 at 6:58=E2=80=AFAM Zhouyi Zhou <zhouzhouyi@gmail.com=
> wrote:
> >
> > hi
> >
> > On Tue, Apr 25, 2023 at 6:13=E2=80=AFPM Peter Zijlstra <peterz@infradea=
d.org> wrote:
> > >
> > > On Mon, Apr 24, 2023 at 02:55:11PM -0400, Joel Fernandes wrote:
> > > > This is amazing debugging Boqun, like a boss! One comment below:
> > > >
> > > > > > > Or something simple I haven't thought of? :)
> > > > > >
> > > > > > At what points can r13 change?  Only when some particular funct=
ions are
> > > > > > called?
> > > > > >
> > > > >
> > > > > r13 is the local paca:
> > > > >
> > > > >         register struct paca_struct *local_paca asm("r13");
> > > > >
> > > > > , which is a pointer to percpu data.
> > > > >
> > > > > So if a task schedule from one CPU to anotehr CPU, the value gets
> > > > > changed.
> > > >
> > > > It appears the whole issue, per your analysis, is that the stack
> > > > checking code in gcc should not cache or alias r13, and must read i=
ts
> > > > most up-to-date value during stack checking, as its value may have
> > > > changed during a migration to a new CPU.
> > > >
> > > > Did I get that right?
> > > >
> > > > IMO, even without a reproducer, gcc on PPC should just not do that,
> > > > that feels terribly broken for the kernel. I wonder what clang does=
,
> > > > I'll go poke around with compilerexplorer after lunch.
> > > >
> > > > Adding +Peter Zijlstra as well to join the party as I have a feelin=
g
> > > > he'll be interested. ;-)
> > >
> > > I'm a little confused; the way I understand the whole stack protector
> > > thing to work is that we push a canary on the stack at call and on
> > > return check it is still valid. Since in general tasks randomly migra=
te,
> > > the per-cpu validation canary should be the same on all CPUs.
> > >
> > > Additionally, the 'new' __srcu_read_{,un}lock_nmisafe() functions use
> > > raw_cpu_ptr() to get 'a' percpu sdp, preferably that of the local cpu=
,
> > > but no guarantees.
> > >
> > > Both cases use r13 (paca) in a racy manner, and in both cases it shou=
ld
> > > be safe.
> > New test results today: both gcc build from git (git clone
> > git://gcc.gnu.org/git/gcc.git) and Ubuntu 22.04 gcc-12.1.0
> > are immune from the above issue. We can see the assembly code on
> > http://140.211.169.189/0425/srcu_gp_start_if_needed-gcc-12.txt
> >
> > while
> > Both native gcc on PPC vm (gcc version 9.4.0), and gcc cross compiler
> > on my x86 laptop (gcc version 10.4.0) will reproduce the bug.
>
> Do you know what fixes the issue? I would not declare victory yet. My
> feeling is something changes in timing, or compiler codegen which
> hides the issue. So the issue is still there but it is just a matter
> of time before someone else reports it.
I am going to try bisect on GCC, hope we can find some clue.
>
> Out of curiosity for PPC folks, why cannot 64-bit PPC use per-task
> canary? Michael, is this an optimization? Adding Christophe as well
> since it came in a few years ago via the following commit:
>
> commit 06ec27aea9fc84d9c6d879eb64b5bcf28a8a1eb7
> Author: Christophe Leroy <christophe.leroy@c-s.fr>
> Date:   Thu Sep 27 07:05:55 2018 +0000
>
>     powerpc/64: add stack protector support
>
>     On PPC64, as register r13 points to the paca_struct at all time,
>     this patch adds a copy of the canary there, which is copied at
>     task_switch.
>     That new canary is then used by using the following GCC options:
>     -mstack-protector-guard=3Dtls
>     -mstack-protector-guard-reg=3Dr13
>     -mstack-protector-guard-offset=3Doffsetof(struct paca_struct, canary)=
)
>
>     Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
>     Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
>
>  - Joel