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: <20221124094845.1907443-1-debug@rivosinc.com> <20230720001852.GA572618@google.com>
 <CAKC1njSaR2d-T_UnVJDZYbROT5OLEjBJ+Aps-UHPFTefDc8=6g@mail.gmail.com>
In-Reply-To: <CAKC1njSaR2d-T_UnVJDZYbROT5OLEjBJ+Aps-UHPFTefDc8=6g@mail.gmail.com>
From:   Deepak Gupta <debug@rivosinc.com>
Date:   Thu, 20 Jul 2023 11:42:38 -0700
Message-ID: <CAKC1njTb4=mnvNzvwU-afQvXbr+i-BzJBtp-tO91cEFh2QN6Vw@mail.gmail.com>
Subject: Re: [PATCH v2] riscv: VMAP_STACK overflow detection thread-safe
To:     Sami Tolvanen <samitolvanen@google.com>
Cc:     palmer@dabbelt.com, linux-kernel@vger.kernel.org,
        linux-riscv@lists.infradead.org, Guo Ren <guoren@kernel.org>,
        Jisheng Zhang <jszhang@kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Thu, Jul 20, 2023 at 11:38=E2=80=AFAM Deepak Gupta <debug@rivosinc.com> =
wrote:
>
> On Wed, Jul 19, 2023 at 5:18=E2=80=AFPM Sami Tolvanen <samitolvanen@googl=
e.com> wrote:
>>
>> Hi Deepak,
>>
>> On Thu, Nov 24, 2022 at 01:48:45AM -0800, Deepak Gupta wrote:
>> > commit 31da94c25aea ("riscv: add VMAP_STACK overflow detection") added
>> > support for CONFIG_VMAP_STACK. If overflow is detected, CPU switches t=
o
>> > `shadow_stack` temporarily before switching finally to per-cpu
>> > `overflow_stack`.
>> >
>> > If two CPUs/harts are racing and end up in over flowing kernel stack, =
one
>> > or both will end up corrupting each other state because `shadow_stack`=
 is
>> > not per-cpu. This patch optimizes per-cpu overflow stack switch by
>> > directly picking per-cpu `overflow_stack` and gets rid of `shadow_stac=
k`.
>>
>> Are you planning on resending this patch? I see it didn't gain much
>> traction last time, but this looks like a much cleaner solution for
>> selecting the overflow stack than having a `shadow_stack` and calling
>> to C to compute the per-CPU offset. The asm_per_cpu macro also would
>> come in handy when implementing CONFIG_SHADOW_CALL_STACK, which we'd
>> like to have on RISC-V too.
>
>
> I personally like the approach of using percpu because I found it cleaner=
 and easy to maintain.
>
> There wasn't any willingness to accept that patch because list members / =
maintainers thought the best way to fix an error prone mechanism is
> to introduce another clunky mechanism which is difficult to maintain. So =
no plan on resending that patch.
>
> If there is motivation to use the asm_per_cpu macro for other stuff, plea=
se feel free to use it from my patch.
> I believe it will be handy.
>
> If there is motivation to accept this patch on the list or provide sugges=
tions on this patch, I can resend it.

Sorry for the last html formatted email.
Somehow settings in my gmail got reset and plaintext mode was set to html a=
gain.

>
> -Deepak
>
>
>
>>
>>
>> Sami