Received-SPF: pass (google.com: domain of linux-kernel+bounces-17462-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:40f1:3f00::1 as permitted sender) client-ip=2604:1380:40f1:3f00::1;
Date: Thu, 04 Jan 2024 20:03:05 -0800
From: "H. Peter Anvin" <hpa@zytor.com>
To: Andrew Cooper <andrew.cooper3@citrix.com>,
        Elizabeth Figura <zfigura@codeweavers.com>,
        Sean Christopherson <seanjc@google.com>
CC: x86@kernel.org, Linux Kernel <linux-kernel@vger.kernel.org>,
        Thomas Gleixner <tglx@linutronix.de>, Ingo Molnar <mingo@redhat.com>,
        Borislav Petkov <bp@alien8.de>,
        Dave Hansen <dave.hansen@linux.intel.com>,
        Ricardo Neri <ricardo.neri-calderon@linux.intel.com>,
        wine-devel@winehq.org
Subject: Re: x86 SGDT emulation for Wine
User-Agent: K-9 Mail for Android
In-Reply-To: <11f2135c-2b64-4a90-95ee-358c3f6117b5@citrix.com>
References: <2285758.taCxCBeP46@uriel> <ZZV65qJuJ67E_n9O@google.com> <1C37C311-CF8A-44EC-89B5-D826EF458708@zytor.com> <2451911.jE0xQCEvom@uriel> <B042BACA-839E-4A42-9BAA-AC7E58624250@zytor.com> <11f2135c-2b64-4a90-95ee-358c3f6117b5@citrix.com>
Message-ID: <AF7CF603-081F-4EA4-B149-3BFEE6AB6319@zytor.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain;
 charset=utf-8
Content-Transfer-Encoding: quoted-printable

On January 4, 2024 6:47:04 PM PST, Andrew Cooper <andrew=2Ecooper3@citrix=
=2Ecom> wrote:
>On 05/01/2024 1:02 am, H=2E Peter Anvin wrote:
>> Note that there is no fundamental reason you cannot run the Unix user s=
pace code inside the VM container, too; you only need to vmexit on an actua=
l system call=2E
>
>I know this is going on a tangent, but getting a VMExit on the SYSCALL
>instruction is surprisingly difficult=2E
>
>The "easy" way is to hide EFER=2ESCE behind the guests back, intercept #U=
D
>and emulate both the SYSCALL and SYSRET instructions=2E=C2=A0 It's slow, =
but it
>works=2E
>
>However, FRED completely prohibits tricks like this, because what you
>cannot reasonably do is clear CR4=2EFRED behind the back of a guest
>kernel=2E=C2=A0 You'd have to intercept and emulate all event sources in =
order
>to catch SYSCALL=2E
>
>I raised this as a concern during early review, but Intel has no
>official feature to take a VMExit on privilege change, and FRED
>(rightly) wasn't an appropriate vehicle to add such a feature, so it was
>deemed not an issue that the FRED design would break the unofficial ways
>that people were using to intercept/monitor/etc system calls=2E
>
>~Andrew
>
>P=2ES=2E Yes, there are more adventurous tricks like injecting a thunk in=
to
>the guest kernel and editing MSR_LSTAR behind the guest's back=2E=C2=A0 I=
n
>principle a similar trick works with FRED, but in order to do this to
>Windows, you also need to hook checkpatch to blind it to the thunk, and
>this is horribly invasive=2E

*In this case* it shouldn't be a problem, since the "guest operating syste=
m" would be virtually nonexistent and entirely puppeted by Wine=2E