Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

From: Andy Lutomirski <luto@amacapital.net>
To: Andrew Cooper <andrew.cooper3@citrix.com>
Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com>,
	Kyle Huey <me@kylehuey.com>, Thomas Gleixner <tglx@linutronix.de>,
	John Stultz <john.stultz@linaro.org>,
	Ingo Molnar <mingo@redhat.com>, Michal Hocko <mhocko@suse.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	"Michael S. Tsirkin" <mst@redhat.com>,
	Alexander Shishkin <alexander.shishkin@linux.intel.com>,
	Aravind Gopalakrishnan <Aravind.Gopalakrishnan@amd.com>,
	Vlastimil Babka <vbabka@suse.cz>,
	"Luis R. Rodriguez" <mcgrof@kernel.org>,
	Mateusz Guzik <mguzik@redhat.com>,
	Alex Thorlton <athorlton@sgi.com>,
	"Rafael J. Wysocki" <rafael.j.wysocki@intel.com>,
	Dmitry Vyukov <dvyukov@google.com>,
	Vladimir Zapolskiy <vladimir_zapolskiy@mentor.com>,
	Jiri Slaby <jslaby@suse.cz>,
	Andrey Ryabinin <aryabinin@virtuozzo.com>,
	Ben Segall <bsegall@google.com>,
	"maintainer:X86 ARCHITECTURE (32-BIT AND 64-BIT)"
	<x86@kernel.org>, Denys Vlasenko <dvlasenk@redhat.com>,
	Paul Gortmaker <paul.gortmaker@windriver.com>,
	Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>,
	"Robert O'Callahan" <robert@ocallahan.org>,
	"open list:X86 ARCHITECTURE (32-BIT AND 64-BIT)" 
	<linux-kernel@vger.kernel.org>, Juergen Gross <jgross@suse.com>,
	Linux API <linux-api@vger.kernel.org>,
	Fenghua Yu <fenghua.yu@intel.com>,
	Kees Cook <keescook@chromium.org>,
	"Peter Zijlstra (Intel)" <peterz@infradead.org>,
	Borislav Petkov <bp@suse.de>, Len Brown <len.brown@intel.com>,
	Huang Rui <ray.huang@amd.com>, "H. Peter Anvin" <hpa@zytor.com>
Subject: Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.
Date: Wed, 14 Sep 2016 12:36:40 -0700	[thread overview]
Message-ID: <CALCETrUkmiwLDD0V-b5woDncOzdqZaS4Pg5EnneXD4ZEN7ttbg@mail.gmail.com> (raw)
In-Reply-To: <2c7b1052-be5a-f776-60e5-23136cd7443c@citrix.com>

On Wed, Sep 14, 2016 at 12:28 PM, Andrew Cooper
<andrew.cooper3@citrix.com> wrote:
> On 14/09/2016 20:23, Boris Ostrovsky wrote:
>> On 09/14/2016 02:52 PM, Andy Lutomirski wrote:
>>> On Tue, Sep 13, 2016 at 11:13 PM, Kyle Huey <me@kylehuey.com> wrote:
>>>> On Mon, Sep 12, 2016 at 9:56 AM, Andy Lutomirski <luto@amacapital.net> wrote:
>>>>> You should explicitly check that, if the
>>>>> feature is set under Xen PV, then the MSR actually works as
>>>>> advertised.  This may require talking to the Xen folks to make sure
>>>>> you're testing the right configuration.
>>>> This is interesting.  When running under Xen PV the kernel is allowed
>>>> to read the real value of MSR_PLATFORM_INFO and see that CPUID
>>>> faulting is supported.  But as you suggested, writing to
>>>> MSR_MISC_FEATURES_ENABLES doesn't actually enable CPUID faulting, at
>>>> least not in any way that works.
>>>>
>>>> It's not obvious to me how to test this, because when this feature
>>>> works, CPUID only faults in userspace, not in the kernel.  Is there
>>>> existing code somewhere that runs tests like this in userspace?
>>>>
>>> Andrew, Boris: should we expect Xen PV to do anything sensible when we
>>> write to MSR_PLATFORM_INFO to turn on CPUID faulting?  Should the Xen
>>> PV rdmsr hooks or perhaps the hypervisor mask out the feature if it
>>> isn't going to be supported?
>> The hypervisor uses CPUID faulting so we shouldn't advertise this
>> feature to guests.
>
> In the case that the hardware has faulting, or for any HVM guest, the
> extra cost to making the feature available to the guest is a single
> conditional test in the cpuid path.  This is about as close to zero as a
> feature gets.  We really should be offering the feature to guests, and
> have it actually working.  The issue here is that it is leaking when we
> weren't intending to offer it.

As long as Xen can fix this one way or the other in reasonably short
order, I think I'm okay with having Linux incorrectly think it works
on old Xen hypervisors.

--Andy