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;
IronPort-SDR: bdsZ1hCimTRB8k03+M+cux2Q1ncpxZdvkUVh7rNZwPw9aEuR9QW14TjIhXHTOy0eccAUfYhash
 pikOSaXJGhBg==
IronPort-SDR: FXwDEa7nooBTcFlSLP73lP+qFeSCeas91rTzJzHFJPLSqYKyddPtPRFqPjmnk58MNK97QHUI30
 EriHrUun9nlQ==
From:   "Bae, Chang Seok" <chang.seok.bae@intel.com>
To:     Jann Horn <jannh@google.com>
CC:     Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@kernel.org>, Borislav Petkov <bp@suse.de>,
        Andy Lutomirski <luto@kernel.org>,
        "the arch/x86 maintainers" <x86@kernel.org>,
        "Brown, Len" <len.brown@intel.com>,
        "Hansen, Dave" <dave.hansen@intel.com>,
        "H.J. Lu" <hjl.tools@gmail.com>, Dave Martin <Dave.Martin@arm.com>,
        Michael Ellerman <mpe@ellerman.id.au>,
        "Luck, Tony" <tony.luck@intel.com>,
        "Shankar, Ravi V" <ravi.v.shankar@intel.com>,
        "libc-alpha@sourceware.org" <libc-alpha@sourceware.org>,
        linux-arch <linux-arch@vger.kernel.org>,
        "Linux API" <linux-api@vger.kernel.org>,
        kernel list <linux-kernel@vger.kernel.org>,
        Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>,
        Roland McGrath <roland@redhat.com>
Subject: Re: [PATCH v2 3/4] x86/signal: Prevent an alternate stack overflow
 before a signal delivery
Thread-Topic: [PATCH v2 3/4] x86/signal: Prevent an alternate stack overflow
 before a signal delivery
Thread-Index: AQHWvqcjVskyJoXuH0ORuPNb6HHBoanRpbwAgAX6YYCAAAWAAIAAIhWA
Date:   Tue, 24 Nov 2020 20:43:45 +0000
Message-ID: <15AB5469-3DBD-4518-9C15-DDCE7C70B1B5@intel.com>
References: <20201119190237.626-1-chang.seok.bae@intel.com>
 <20201119190237.626-4-chang.seok.bae@intel.com>
 <CAG48ez1aKtwYMEHfGX6_FuX9fOruwvCqEGYVL8eLdV8bg-wHCQ@mail.gmail.com>
 <B2D7D498-D118-447E-93C6-DB03D42CBA4E@intel.com>
 <CAG48ez1JK6pMT2UD1v0FwiCQq48FbE5Eb0d3tK=kK4Sg0TG7OQ@mail.gmail.com>
In-Reply-To: <CAG48ez1JK6pMT2UD1v0FwiCQq48FbE5Eb0d3tK=kK4Sg0TG7OQ@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
Content-Type: text/plain; charset="us-ascii"
Content-ID: <798AD0D402EBB74B8A8E3BED902AA4CE@namprd11.prod.outlook.com>
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-Network-Message-Id: b2b6ab1a-1649-423b-09e7-08d890b9a39c
X-MS-Exchange-CrossTenant-originalarrivaltime: 24 Nov 2020 20:43:45.8718
 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 46c98d88-e344-4ed4-8496-4ed7712e255d
X-MS-Exchange-CrossTenant-mailboxtype: HOSTED
X-MS-Exchange-CrossTenant-userprincipalname: 31XZxWASW2Adi7Iy5Cle8XIK4tr8+c1mLGpc5tm0u6+8i9IjJ4L5DmcxeEFL8jzQQ7HkNUpyE0aR/aNh7MFm37Z7CB4RgWPueSlZZL9snGs=
X-MS-Exchange-Transport-CrossTenantHeadersStamped: SJ0PR11MB4909
X-OriginatorOrg: intel.com
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org


> On Nov 24, 2020, at 10:41, Jann Horn <jannh@google.com> wrote:
>=20
> On Tue, Nov 24, 2020 at 7:22 PM Bae, Chang Seok
> <chang.seok.bae@intel.com> wrote:
>>> On Nov 20, 2020, at 15:04, Jann Horn <jannh@google.com> wrote:
>>> On Thu, Nov 19, 2020 at 8:40 PM Chang S. Bae <chang.seok.bae@intel.com>=
 wrote:
>>>>=20
>>>> diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c
>>>> index ee6f1ceaa7a2..cee41d684dc2 100644
>>>> --- a/arch/x86/kernel/signal.c
>>>> +++ b/arch/x86/kernel/signal.c
>>>> @@ -251,8 +251,13 @@ get_sigframe(struct k_sigaction *ka, struct pt_re=
gs *regs, size_t frame_size,
>>>>=20
>>>>       /* This is the X/Open sanctioned signal stack switching.  */
>>>>       if (ka->sa.sa_flags & SA_ONSTACK) {
>>>> -               if (sas_ss_flags(sp) =3D=3D 0)
>>>> +               if (sas_ss_flags(sp) =3D=3D 0) {
>>>> +                       /* If the altstack might overflow, die with SI=
GSEGV: */
>>>> +                       if (!altstack_size_ok(current))
>>>> +                               return (void __user *)-1L;
>>>> +
>>>>                       sp =3D current->sas_ss_sp + current->sas_ss_size=
;
>>>> +               }
>>>=20
>>> A couple lines further down, we have this (since commit 14fc9fbc700d):
>>>=20
>>>       /*
>>>        * If we are on the alternate signal stack and would overflow it,=
 don't.
>>>        * Return an always-bogus address instead so we will die with SIG=
SEGV.
>>>        */
>>>       if (onsigstack && !likely(on_sig_stack(sp)))
>>>               return (void __user *)-1L;
>>>=20
>>> Is that not working?
>>=20
>> onsigstack is set at the beginning here. If a signal hits under normal s=
tack,
>> this flag is not set. Then it will miss the overflow.
>>=20
>> The added check allows to detect the sigaltstack overflow (always).
>=20
> Ah, I think I understand what you're trying to do. But wouldn't the
> better approach be to ensure that the existing on_sig_stack() check is
> also used if we just switched to the signal stack? Something like:
>=20
> diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c
> index be0d7d4152ec..2f57842fb4d6 100644
> --- a/arch/x86/kernel/signal.c
> +++ b/arch/x86/kernel/signal.c
> @@ -237,7 +237,7 @@ get_sigframe(struct k_sigaction *ka, struct
> pt_regs *regs, size_t frame_size,
>        unsigned long math_size =3D 0;
>        unsigned long sp =3D regs->sp;
>        unsigned long buf_fx =3D 0;
> -       int onsigstack =3D on_sig_stack(sp);
> +       bool onsigstack =3D on_sig_stack(sp);
>        int ret;
>=20
>        /* redzone */
> @@ -246,8 +246,10 @@ get_sigframe(struct k_sigaction *ka, struct
> pt_regs *regs, size_t frame_size,
>=20
>        /* This is the X/Open sanctioned signal stack switching.  */
>        if (ka->sa.sa_flags & SA_ONSTACK) {
> -               if (sas_ss_flags(sp) =3D=3D 0)
> +               if (sas_ss_flags(sp) =3D=3D 0) {
>                        sp =3D current->sas_ss_sp + current->sas_ss_size;
> +                       onsigstack =3D true;
> +               }
>        } else if (IS_ENABLED(CONFIG_X86_32) &&
>                   !onsigstack &&
>                   regs->ss !=3D __USER_DS &&

Yeah, but wouldn't it better to avoid overwriting user data if we can? The =
old
check raises segfault *after* overwritten.

The old check is still helpful to detect an overflow from the nested signal=
(s)
under sigaltstack.

Thanks,
Chang=