Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
From:   Andy Lutomirski <luto@amacapital.net>
Mime-Version: 1.0 (1.0)
Subject: Re: io_uring vs in_compat_syscall()
Date:   Mon, 20 Jul 2020 10:28:55 -0700
Message-Id: <8987E376-6B13-4798-BDBA-616A457447CF@amacapital.net>
References: <b754dad5-ee85-8a2f-f41a-8bdc56de42e8@kernel.dk>
Cc:     Christoph Hellwig <hch@lst.de>, linux-arch@vger.kernel.org,
        linux-api@vger.kernel.org, linux-kernel@vger.kernel.org,
        io-uring@vger.kernel.org
In-Reply-To: <b754dad5-ee85-8a2f-f41a-8bdc56de42e8@kernel.dk>
To:     Jens Axboe <axboe@kernel.dk>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


> On Jul 20, 2020, at 10:02 AM, Jens Axboe <axboe@kernel.dk> wrote:
>=20
> =EF=BB=BFOn 7/20/20 10:58 AM, Andy Lutomirski wrote:
>>=20
>>>> On Jul 20, 2020, at 9:37 AM, Jens Axboe <axboe@kernel.dk> wrote:
>>>=20
>>> =EF=BB=BFOn 7/20/20 12:10 AM, Christoph Hellwig wrote:
>>>> Hi Jens,
>>>>=20
>>>> I just found a (so far theoretical) issue with the io_uring submission
>>>> offloading to workqueues or threads.  We have lots of places using
>>>> in_compat_syscall() to check if a syscall needs compat treatmenet.
>>>> While the biggest users is iocttl(), we also have a fair amount of
>>>> places using in_compat_task() in read and write methods, and these
>>>> will not do the wrong thing when used with io_uring under certain
>>>> conditions.  I'm not sure how to best fix this, except for making sure
>>>> in_compat_syscall() returns true one way or another for these cases.
>>>=20
>>> We can probably propagate this information in the io_kiocb via a flag,
>>> and have the io-wq worker set TS_COMPAT if that's the case.
>>>=20
>>=20
>> Is TS_COMPAT actually a cross-arch concept for which this is safe?
>> Having a real arch helper for =E2=80=9Cset the current syscall arch for t=
he
>> current kernel thread=E2=80=9D seems more sensible to me.=20
>=20
> Sure, I'd consider that implementation detail for the actual patch(es)
> for this issue.

There=E2=80=99s a corner case, though: doesn=E2=80=99t io_uring submission f=
requently do the work synchronously in the context of the calling thread?  I=
f so, can a thread do a 64-bit submit with 32-bit work or vice versa?

Sometimes I think that in_compat_syscall() should have a mode in which calli=
ng it warns (e.g. not actually in a syscall when doing things in io_uring). =
 And the relevant operations should be properly wired up to avoid global sta=
te like this.

>=20
> --=20
> Jens Axboe
>=20