xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Andrew Cooper <andrew.cooper3@citrix.com>
To: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: Jan Beulich <JBeulich@suse.com>, Xen-devel <xen-devel@lists.xen.org>
Subject: Re: [PATCH v4 11/26] xen/x86: Improvements to in-hypervisor cpuid sanity checks
Date: Tue, 5 Apr 2016 16:25:39 +0100	[thread overview]
Message-ID: <5703D8F3.4000202@citrix.com> (raw)
In-Reply-To: <20160328152954.GI17944@char.us.oracle.com>

On 28/03/16 16:29, Konrad Rzeszutek Wilk wrote:
> On Wed, Mar 23, 2016 at 04:36:14PM +0000, Andrew Cooper wrote:
>> Currently, {pv,hvm}_cpuid() has a large quantity of essentially-static logic
>> for modifying the features visible to a guest.  A lot of this can be subsumed
>> by {pv,hvm}_featuremask, which identify the features available on this
>> hardware which could be given to a PV or HVM guest.
>>
>> This is a step in the direction of full per-domain cpuid policies, but lots
>> more development is needed for that.  As a result, the static checks are
>> simplified, but the dynamic checks need to remain for now.
>>
>> As a side effect, some of the logic for special features can be improved.
>> OSXSAVE and OSPKE will be automatically cleared because of being absent in the
>> featuremask.  This allows the fast-forward logic to be more simple.
>>
>> In addition, there are some corrections to the existing logic:
>>
>>  * Hiding PSE36 out of PAE mode is architecturally wrong.  It turns out that
>>    it was a bugfix for running HyperV under Xen, which wanted to see PSE36
>>    even after choosing to use PAE paging.  PSE36 is not supported by shadow
>>    paging, so is hidden from non-HAP guests, but is still visible for HAP
>>    guests.
>>  * Changing the visibility of RDTSCP based on host TSC stability or virtual
>>    TSC mode is bogus, so dropped.
> Why is it bogus? It is an PV ABI type CPUID.

The CPUID bit has a well defined meaning, and the vtsc infrastructure
went and diverged the ABI provided by Intel and AMD.

If it were only PV guests, it would be less bad.  However, it breaks HVM
guests as well which is absolutely not ok.

The meaning of the RDTSCP feature bit is well defined.  The presence of
the `rdtscp` instruction, and the TSC_AUX MSR.

The vtsc options control whether rdtsc(p) is intercepted by Xen, and
whether the guest or Xen controls the AUX MSR.  These options are
unrelated, and have no bearing on the availability of the instruction in
the first place.

>
> Independetly of that you would also need to modify tsc_mode.txt file and all uses
> of 'tsc_mode=3'.

A guest kernel cannot use the presence or absence of the rdtscp feature
to identify which vtsc mode is in intended, which necessarily means
there is an extra out-of-band signal controlling its use.

In particular, the current issue fixed by this change is that for the
default case, on migrate, the rdtscp feature disappears from the domain
as the tsc logic decides that the frequency has changed and vtsc mode
should be enabled.

~Andrew

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

  reply	other threads:[~2016-04-05 15:25 UTC|newest]

Thread overview: 72+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-03-23 16:36 [PATCH v4 00/26] x86: Improvements to cpuid handling for guests Andrew Cooper
2016-03-23 16:36 ` [PATCH v4 01/26] xen/public: Export cpu featureset information in the public API Andrew Cooper
2016-03-24 14:08   ` Jan Beulich
2016-03-24 14:12     ` Andrew Cooper
2016-03-24 14:16       ` Jan Beulich
2016-03-23 16:36 ` [PATCH v4 02/26] xen/x86: Script to automatically process featureset information Andrew Cooper
2016-03-23 16:36 ` [PATCH v4 03/26] xen/x86: Collect more cpuid feature leaves Andrew Cooper
2016-03-23 16:36 ` [PATCH v4 04/26] xen/x86: Mask out unknown features from Xen's capabilities Andrew Cooper
2016-03-23 16:36 ` [PATCH v4 05/26] xen/x86: Annotate special features Andrew Cooper
2016-03-23 16:36 ` [PATCH v4 06/26] xen/x86: Annotate VM applicability in featureset Andrew Cooper
2016-03-23 16:36 ` [PATCH v4 07/26] xen/x86: Calculate maximum host and guest featuresets Andrew Cooper
2016-03-29  8:57   ` Jan Beulich
2016-03-23 16:36 ` [PATCH v4 08/26] xen/x86: Generate deep dependencies of features Andrew Cooper
2016-03-24 16:16   ` Jan Beulich
2016-03-23 16:36 ` [PATCH v4 09/26] xen/x86: Clear dependent features when clearing a cpu cap Andrew Cooper
2016-03-23 16:36 ` [PATCH v4 10/26] xen/x86: Improve disabling of features which have dependencies Andrew Cooper
2016-03-28 15:18   ` Konrad Rzeszutek Wilk
2016-03-23 16:36 ` [PATCH v4 11/26] xen/x86: Improvements to in-hypervisor cpuid sanity checks Andrew Cooper
2016-03-24 15:38   ` Andrew Cooper
2016-03-24 16:47   ` Jan Beulich
2016-03-24 17:01     ` Andrew Cooper
2016-03-24 17:11       ` Jan Beulich
2016-03-24 17:12         ` Andrew Cooper
2016-03-28 15:29   ` Konrad Rzeszutek Wilk
2016-04-05 15:25     ` Andrew Cooper [this message]
2016-03-23 16:36 ` [PATCH v4 12/26] x86/cpu: Move set_cpumask() calls into c_early_init() Andrew Cooper
2016-03-28 15:55   ` Konrad Rzeszutek Wilk
2016-04-05 16:19     ` Andrew Cooper
2016-03-23 16:36 ` [PATCH v4 13/26] x86/cpu: Sysctl and common infrastructure for levelling context switching Andrew Cooper
2016-03-24 16:58   ` Jan Beulich
2016-03-28 16:12   ` Konrad Rzeszutek Wilk
2016-04-05 16:33     ` Andrew Cooper
2016-03-28 17:37   ` Konrad Rzeszutek Wilk
2016-03-23 16:36 ` [PATCH v4 14/26] x86/cpu: Rework AMD masking MSR setup Andrew Cooper
2016-03-28 18:55   ` Konrad Rzeszutek Wilk
2016-04-05 16:44     ` Andrew Cooper
2016-03-23 16:36 ` [PATCH v4 15/26] x86/cpu: Rework Intel masking/faulting setup Andrew Cooper
2016-03-28 19:14   ` Konrad Rzeszutek Wilk
2016-04-05 16:45     ` Andrew Cooper
2016-03-23 16:36 ` [PATCH v4 16/26] x86/cpu: Context switch cpuid masks and faulting state in context_switch() Andrew Cooper
2016-03-28 19:27   ` Konrad Rzeszutek Wilk
2016-04-05 18:34     ` Andrew Cooper
2016-03-23 16:36 ` [PATCH v4 17/26] x86/pv: Provide custom cpumasks for PV domains Andrew Cooper
2016-03-28 19:40   ` Konrad Rzeszutek Wilk
2016-04-05 16:55     ` Andrew Cooper
2016-03-23 16:36 ` [PATCH v4 18/26] x86/domctl: Update PV domain cpumasks when setting cpuid policy Andrew Cooper
2016-03-24 17:04   ` Jan Beulich
2016-03-24 17:05     ` Andrew Cooper
2016-03-28 19:51   ` Konrad Rzeszutek Wilk
2016-03-23 16:36 ` [PATCH v4 19/26] xen+tools: Export maximum host and guest cpu featuresets via SYSCTL Andrew Cooper
2016-03-28 19:59   ` Konrad Rzeszutek Wilk
2016-03-23 16:36 ` [PATCH v4 20/26] tools/libxc: Modify bitmap operations to take void pointers Andrew Cooper
2016-03-28 20:05   ` Konrad Rzeszutek Wilk
2016-03-23 16:36 ` [PATCH v4 21/26] tools/libxc: Use public/featureset.h for cpuid policy generation Andrew Cooper
2016-03-28 20:07   ` Konrad Rzeszutek Wilk
2016-03-23 16:36 ` [PATCH v4 22/26] tools/libxc: Expose the automatically generated cpu featuremask information Andrew Cooper
2016-03-28 20:08   ` Konrad Rzeszutek Wilk
2016-03-23 16:36 ` [PATCH v4 23/26] tools: Utility for dealing with featuresets Andrew Cooper
2016-03-28 20:26   ` Konrad Rzeszutek Wilk
2016-03-23 16:36 ` [PATCH v4 24/26] tools/libxc: Wire a featureset through to cpuid policy logic Andrew Cooper
2016-03-28 20:39   ` Konrad Rzeszutek Wilk
2016-03-23 16:36 ` [PATCH v4 25/26] tools/libxc: Use featuresets rather than guesswork Andrew Cooper
2016-03-23 16:36 ` [PATCH v4 26/26] tools/libxc: Calculate xstate cpuid leaf from guest information Andrew Cooper
2016-03-24 17:20   ` Wei Liu
2016-03-31  7:48   ` Jan Beulich
2016-04-05 17:48     ` Andrew Cooper
2016-04-07  0:16       ` Jan Beulich
2016-04-07  0:40         ` Andrew Cooper
2016-04-07  0:56           ` Jan Beulich
2016-04-07 11:34             ` Andrew Cooper
2016-03-24 10:27 ` [PATCH v4 00/26] x86: Improvements to cpuid handling for guests Jan Beulich
2016-03-24 10:28   ` Andrew Cooper

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=5703D8F3.4000202@citrix.com \
    --to=andrew.cooper3@citrix.com \
    --cc=JBeulich@suse.com \
    --cc=konrad.wilk@oracle.com \
    --cc=xen-devel@lists.xen.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).