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: umip: AMD Ryzen 3900X, pagefault after emulate SLDT/SIDT instruction
Date:   Wed, 20 May 2020 18:16:37 -0700
Message-Id: <AD88C928-85A9-482C-9EE9-19A82343075A@amacapital.net>
References: <20200521005605.GA11955@ranerica-svr.sc.intel.com>
Cc:     Andy Lutomirski <luto@kernel.org>,
        Brendan Shanks <bshanks@codeweavers.com>,
        Andreas Rammhold <andi@notmuch.email>,
        Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
        X86 ML <x86@kernel.org>, "H. Peter Anvin" <hpa@zytor.com>,
        "Eric W. Biederman" <ebiederm@xmission.com>,
        Sergey Senozhatsky <sergey.senozhatsky@gmail.com>,
        Babu Moger <Babu.Moger@amd.com>,
        Kefeng Wang <wangkefeng.wang@huawei.com>,
        Jason Yan <yanaijie@huawei.com>,
        LKML <linux-kernel@vger.kernel.org>
In-Reply-To: <20200521005605.GA11955@ranerica-svr.sc.intel.com>
To:     Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


> On May 20, 2020, at 5:55 PM, Ricardo Neri <ricardo.neri-calderon@linux.int=
el.com> wrote:
>=20
> =EF=BB=BFOn Tue, May 19, 2020 at 05:54:53PM -0700, Andy Lutomirski wrote:
>>> On Tue, May 19, 2020 at 12:43 PM Ricardo Neri
>>> <ricardo.neri-calderon@linux.intel.com> wrote:
>>>=20
>>> On Tue, May 19, 2020 at 11:56:40AM -0700, Brendan Shanks wrote:
>>>>=20
>>>>> On May 19, 2020, at 7:38 AM, Andreas Rammhold <andi@notmuch.email> wro=
te:
>>>>>=20
>>>>> Hi,
>>>>>=20
>>>>> I've been running into a weird problem with UMIP on a current Ryzen
>>>>> 3900x with kernel 5.6.11 where a process receives a page fault after t=
he
>>>>> kernel handled the SLDT (or SIDT) instruction (emulation).
>>>>>=20
>>>>> The program I am running is run through WINE in 32bit mode and tries t=
o
>>>>> figure out if it is running in a VMWare machine by comparing the resul=
ts
>>>>> of SLDT against well known constants (basically as shown in the
>>>>> [example] linked below).
>>>>>=20
>>>>> In dmesg I see the following log lines:
>>>>>> [99970.004756] umip: Program.exe[3080] ip:4373fb sp:32f3e0: SIDT inst=
ruction cannot be used by applications.
>>>>>> [99970.004757] umip: Program.exe[3080] ip:4373fb sp:32f3e0: For now, e=
xpensive software emulation returns the result.
>>>>>> [99970.004758] umip: Program.exe[3080] ip:437415 sp:32f3e0: SLDT inst=
ruction cannot be used by applications.
>>>>>=20
>>>>> Following that the process terminates with a page fault:
>>>>>> Unhandled exception: page fault on read access to 0xffffffff in 32-bi=
t code (0x0000000000437415).
>>>>>=20
>>>>> Assembly at that address:
>>>>>> 0x0000000000437415: sldt    0xffffffe8(%ebp)
>>>>>=20
>>>>> Running the same executable on the exact same kernel (and userland) bu=
t
>>>>> on a Intel i7-8565U doesn't crash at this point. I am guessing the
>>>>> emulation is supposed to do something different on AMD CPUs?
>>>=20
>>> I am surprised you don't see it on the Intel processor. Maybe it does
>>> not have UMIP. Do you see umip when you do
>>>=20
>>> $ grep umip /proc/cpuinfo
>>>=20
>>> ?
>>>>>=20
>>>>> On the Ryzen the code executes successfully after setting CONFIG_X86_U=
MIP=3Dn.
>>>>=20
>>>> Hi Andreas,
>>>>=20
>>>> The problem is that the kernel does not emulate/spoof the SLDT instruct=
ion, only SGDT, SIDT, and SMSW.
>>>> SLDT and STR weren't thought to be commonly used, so emulation/spoofing=
 wasn=E2=80=99t added.
>>>> In the last few months I have seen reports of one or two (32-bit) Windo=
ws games that use SLDT though.
>>>> Can you share more information about the application you=E2=80=99re run=
ning?
>>>>=20
>>>> Maybe the best path is to add kernel emulation/spoofing for SLDT and ST=
R on 32 and 64-bit, just to cover all the cases. It should be a pretty simpl=
e patch, I=E2=80=99ll start working on it.
>>>=20
>>> I have a patch for this already that I wrote for testing purposes:
>>>=20
>>> https://github.com/ricardon/tip/commit/1692889cb3f8accb523d44b682458e234=
b93be50
>>>=20
>>> Perhaps it can be used as a starting point? Not sure what the spoofing
>>> value should be, though. Perhaps 0?
>>=20
>> Possibly SLDT should return nonzero if there's an LDT.
>=20
> I guess the value should be in the same hole of the x86_64 memory map,
> right? Currently sgdt and sidt return 0xfffffffffffe0000 and
> 0xffffffffffff0000, respectively.

As far as I=E2=80=99m concerned, it makes no difference whether it=E2=80=99s=
 a hole in the memory map.  But sure.

>=20
> Thanks and BR,
> Ricardo