Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
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: [PATCH] ptrace/x86: introduce TS_COMPAT_RESTART to fix get_nr_restart_syscall()
Date:   Fri, 29 Nov 2019 10:19:59 -0800
Message-Id: <CBC57AB1-D92D-44B1-9428-B84B17429C8F@amacapital.net>
References: <20191129173213.GB3992@redhat.com>
Cc:     Linus Torvalds <torvalds@linux-foundation.org>,
        Kees Cook <keescook@chromium.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        Andy Lutomirski <luto@kernel.org>,
        Borislav Petkov <bp@alien8.de>, Ingo Molnar <mingo@kernel.org>,
        Jan Kratochvil <jan.kratochvil@redhat.com>,
        Pedro Alves <palves@redhat.com>, Peter Anvin <hpa@zytor.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        the arch/x86 maintainers <x86@kernel.org>
In-Reply-To: <20191129173213.GB3992@redhat.com>
To:     Oleg Nesterov <oleg@redhat.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


> On Nov 29, 2019, at 9:32 AM, Oleg Nesterov <oleg@redhat.com> wrote:
>=20
> =EF=BB=BFOn 11/28, Andy Lutomirski wrote:
>>=20
>> I think this doesn=E2=80=99t work for x32.
>=20
> Why? get_nr_restart_syscall() can still rely on the "orig_ax & __X32_SYSCA=
LL_BIT"
> check, debugger should restore regs->orig_ax correctly.

Right. Although relying on this is IMO a ridiculous bit of ABI.

>=20
>> Let=E2=80=99s either save the result of syscall_get_arch()
>=20
> We can save the result of syscall_get_arch(), but it doesn't distinguish
> x32/x86_64 tasks, so it doesn't really differ from TS_COMPAT.

Duh. Never mind.

>=20
>> or just actually calculate and save the restart_syscall nr here.
>=20
> sure, we we can do this.

I like this the best unless we can renumber the syscalls.

>=20
>> Before we commit to this, Kees, do you think we can manage to just renumb=
er
>> restart_syscall()?  That=E2=80=99s a much better solution if we can pull i=
t off.
>=20
> Agreed.
>=20
> Oleg.
>=20