All of lore.kernel.org
 help / color / mirror / Atom feed
From: Boris Ostrovsky <boris.ostrovsky@oracle.com>
To: Jan Beulich <JBeulich@suse.com>
Cc: kevin.tian@intel.com, keir@xen.org,
	suravee.suthikulpanit@amd.com, andrew.cooper3@citrix.com,
	tim@xen.org, dietmar.hahn@ts.fujitsu.com,
	xen-devel@lists.xen.org, Aravind.Gopalakrishnan@amd.com,
	jun.nakajima@intel.com, dgdegra@tycho.nsa.gov
Subject: Re: [PATCH v12 for-xen-4.5 16/20] x86/VPMU: Handle PMU interrupts for PV guests
Date: Wed, 01 Oct 2014 14:06:50 -0400	[thread overview]
Message-ID: <542C42BA.4050700@oracle.com> (raw)
In-Reply-To: <542C2B4C020000780003B9C2@mail.emea.novell.com>

On 10/01/2014 10:26 AM, Jan Beulich wrote:
>
>>>>> Looking at the separation of hypervisor vs guest context to report
>>>>> again
>>>>>
>>>>>                /* Non-privileged domains are always in XENPMU_MODE_SELF mode */
>>>>>                if ( (vpmu_mode & XENPMU_MODE_SELF) ||
>>>>>                     (!is_hardware_domain(sampled->domain) &&
>>>>>                      !is_idle_vcpu(sampled)) )
>>>>>                    cur_regs = guest_cpu_user_regs();
>>>>>                else
>>>>>                    cur_regs = regs;
>>>>>
>>>>> I now additionally wonder why the condition here isn't just the SELF
>>>>> check: If the interrupt happened while in the hypervisor, why would
>>>>> you override this unconditionally to report a guest sample instead?
>>>>> Shouldn't the profiling domain tell you what it wants in that case
>>>>> (global vs guest local view)?
>>>> The second part of the check (!is_hardware_domain(sampled->domain) &&
>>>> !is_idle_vcpu(sampled)) is to prevent sending hypervisor sample to a
>>>> non-privileged guest. vpmu_mode may be, for example, XENPMU_MODE_HV but
>>>> that only means that dom0 can get hypervisor samples.
>>> Right, but that's not what the code above does: Instead of sending
>>> the hypervisor sample to Dom0 it converts it to a guest mode one.
>> Oh, I see --- when we get interrupted while in a non-privileged guest's
>> context (but in hypervisor) I send guest's registers, not Xen's.
>>
>> I think just SELF check is not sufficient though, we need to make sure
>> that we are not sending hypervisor sample to non-dom0. So
>>       if ( (vpmu_mode & XENPMU_MODE_SELF) ||
>> !is_hardware_domain(sampling->domain) )
> Actually I think instead the determination of sampling needs to
> depend on the register context rather than solely on the current
> domain's ID.

Not sure I follow this --- we do need to take domainID into account to 
avoid sending non-dom0 a hypervisor sample.

Or are you saying that what to send depends on both RIP and domainID?, 
Something like

     if ( (vpmu_mode & XENPMU_MODE_SELF) )
                 cur_regs = guest_cpu_user_regs();
     else if ( (regs->rip >= XEN_VIRT_START) && (regs->rip < 
XEN_VIRT_END) && is_hardware_domain(sampling->domain))
                 cur_regs = regs;
     else
                 cur_regs = guest_cpu_user_regs();

-boris

  reply	other threads:[~2014-10-01 18:06 UTC|newest]

Thread overview: 92+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-09-25 19:28 [PATCH v12 for-xen-4.5 00/20] x86/PMU: Xen PMU PV(H) support Boris Ostrovsky
2014-09-25 19:28 ` [PATCH v12 for-xen-4.5 01/20] common/symbols: Export hypervisor symbols to privileged guest Boris Ostrovsky
2014-09-26 14:58   ` Konrad Rzeszutek Wilk
2014-09-26 15:10     ` Jan Beulich
2014-09-26 16:49       ` Konrad Rzeszutek Wilk
2014-09-29  6:43         ` Jan Beulich
2014-09-29 13:29           ` Boris Ostrovsky
2014-09-29 13:47             ` Jan Beulich
2014-09-29 14:16               ` Boris Ostrovsky
2014-09-29 14:33                 ` Jan Beulich
2014-09-26 21:43   ` Daniel De Graaf
2014-09-26 22:12     ` Boris Ostrovsky
2014-09-25 19:28 ` [PATCH v12 for-xen-4.5 02/20] x86/VPMU: Manage VPMU_CONTEXT_SAVE flag in vpmu_save_force() Boris Ostrovsky
2014-09-26 14:49   ` Konrad Rzeszutek Wilk
2014-09-25 19:28 ` [PATCH v12 for-xen-4.5 03/20] x86/VPMU: Set MSR bitmaps only for HVM/PVH guests Boris Ostrovsky
2014-09-26 14:59   ` Konrad Rzeszutek Wilk
2014-09-25 19:28 ` [PATCH v12 for-xen-4.5 04/20] x86/VPMU: Make vpmu macros a bit more efficient Boris Ostrovsky
2014-09-25 19:28 ` [PATCH v12 for-xen-4.5 05/20] intel/VPMU: Clean up Intel VPMU code Boris Ostrovsky
2014-09-25 19:28 ` [PATCH v12 for-xen-4.5 06/20] vmx: Merge MSR management routines Boris Ostrovsky
2014-09-26 20:48   ` Tian, Kevin
2014-09-25 19:28 ` [PATCH v12 for-xen-4.5 07/20] x86/VPMU: Handle APIC_LVTPC accesses Boris Ostrovsky
2014-09-25 19:28 ` [PATCH v12 for-xen-4.5 08/20] intel/VPMU: MSR_CORE_PERF_GLOBAL_CTRL should be initialized to zero Boris Ostrovsky
2014-09-25 19:28 ` [PATCH v12 for-xen-4.5 09/20] x86/VPMU: Add public xenpmu.h Boris Ostrovsky
2014-09-26 20:49   ` Tian, Kevin
2014-09-29 14:17   ` Jan Beulich
2014-09-29 14:30     ` Jan Beulich
2014-09-29 15:19       ` Boris Ostrovsky
2014-09-29 15:41         ` Jan Beulich
2014-09-29 15:48           ` Boris Ostrovsky
2014-09-29 14:57     ` Boris Ostrovsky
2014-09-29 15:40       ` Jan Beulich
2014-09-29 15:56         ` Boris Ostrovsky
2014-09-25 19:28 ` [PATCH v12 for-xen-4.5 10/20] x86/VPMU: Make vpmu not HVM-specific Boris Ostrovsky
2014-09-25 19:28 ` [PATCH v12 for-xen-4.5 11/20] x86/VPMU: Interface for setting PMU mode and flags Boris Ostrovsky
2014-09-26 21:04   ` Tian, Kevin
2014-09-26 21:24     ` Boris Ostrovsky
2014-09-26 22:00   ` Daniel De Graaf
2014-09-26 22:26     ` Boris Ostrovsky
2014-09-29 13:25   ` Dietmar Hahn
2014-09-29 13:56     ` Boris Ostrovsky
2014-09-29 14:03       ` Dietmar Hahn
2014-09-29 13:59     ` Jan Beulich
2014-09-29 14:05       ` Dietmar Hahn
2014-09-29 15:14   ` Jan Beulich
2014-09-29 15:34     ` Boris Ostrovsky
2014-10-01  0:48   ` Tian, Kevin
2014-10-01  0:56     ` Boris Ostrovsky
2014-09-25 19:28 ` [PATCH v12 for-xen-4.5 12/20] x86/VPMU: Initialize PMU for PV(H) guests Boris Ostrovsky
2014-09-26 22:16   ` Daniel De Graaf
2014-09-26 22:23     ` Boris Ostrovsky
2014-09-29 15:25   ` Jan Beulich
2014-09-29 15:41     ` Boris Ostrovsky
2014-09-29 15:42       ` Jan Beulich
2014-09-29 16:04         ` Boris Ostrovsky
2014-09-29 16:10           ` Jan Beulich
2014-10-01  0:16   ` Tian, Kevin
2014-09-25 19:28 ` [PATCH v12 for-xen-4.5 13/20] x86/VPMU: Save VPMU state for PV guests during context switch Boris Ostrovsky
2014-09-29 15:52   ` Jan Beulich
2014-09-25 19:28 ` [PATCH v12 for-xen-4.5 14/20] x86/VPMU: When handling MSR accesses, leave fault injection to callers Boris Ostrovsky
2014-09-25 19:28 ` [PATCH v12 for-xen-4.5 15/20] x86/VPMU: Add support for PMU register handling on PV guests Boris Ostrovsky
2014-09-26 16:34   ` Konrad Rzeszutek Wilk
2014-09-26 16:44     ` Boris Ostrovsky
2014-09-26 16:49       ` Konrad Rzeszutek Wilk
2014-09-29 16:04   ` Jan Beulich
2014-10-01  0:17   ` Tian, Kevin
2014-09-25 19:28 ` [PATCH v12 for-xen-4.5 16/20] x86/VPMU: Handle PMU interrupts for " Boris Ostrovsky
2014-09-26 22:09   ` Daniel De Graaf
2014-09-30  8:11   ` Jan Beulich
2014-09-30 15:07     ` Boris Ostrovsky
2014-09-30 15:44       ` Jan Beulich
2014-09-30 16:37         ` Boris Ostrovsky
2014-10-01  6:49           ` Jan Beulich
2014-10-01 12:53             ` Boris Ostrovsky
2014-10-01 13:18               ` Jan Beulich
2014-10-01 14:08                 ` Boris Ostrovsky
2014-10-01 14:26                   ` Jan Beulich
2014-10-01 18:06                     ` Boris Ostrovsky [this message]
2014-10-02  6:56                       ` Jan Beulich
2014-10-02 13:53                         ` Boris Ostrovsky
2014-09-25 19:28 ` [PATCH v12 for-xen-4.5 17/20] x86/VPMU: Merge vpmu_rdmsr and vpmu_wrmsr Boris Ostrovsky
2014-09-30  8:13   ` Jan Beulich
2014-09-25 19:28 ` [PATCH v12 for-xen-4.5 18/20] x86/VPMU: Add privileged PMU mode Boris Ostrovsky
2014-09-30  8:18   ` Jan Beulich
2014-09-30 15:16     ` Boris Ostrovsky
2014-09-25 19:28 ` [PATCH v12 for-xen-4.5 19/20] x86/VPMU: NMI-based VPMU support Boris Ostrovsky
2014-09-30  8:37   ` Jan Beulich
2014-10-01  0:18     ` Boris Ostrovsky
2014-10-01  7:32       ` Jan Beulich
2014-09-25 19:28 ` [PATCH v12 for-xen-4.5 20/20] x86/VPMU: Move VPMU files up from hvm/ directory Boris Ostrovsky
2014-09-30  8:40   ` Jan Beulich
2014-09-26 17:03 ` [PATCH v12 for-xen-4.5 00/20] x86/PMU: Xen PMU PV(H) support Konrad Rzeszutek Wilk
2014-09-29 13:28 ` Dietmar Hahn

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=542C42BA.4050700@oracle.com \
    --to=boris.ostrovsky@oracle.com \
    --cc=Aravind.Gopalakrishnan@amd.com \
    --cc=JBeulich@suse.com \
    --cc=andrew.cooper3@citrix.com \
    --cc=dgdegra@tycho.nsa.gov \
    --cc=dietmar.hahn@ts.fujitsu.com \
    --cc=jun.nakajima@intel.com \
    --cc=keir@xen.org \
    --cc=kevin.tian@intel.com \
    --cc=suravee.suthikulpanit@amd.com \
    --cc=tim@xen.org \
    --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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.