xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Boris Ostrovsky <boris.ostrovsky@oracle.com>
To: xen-devel@lists.xen.org
Cc: jun.nakajima@intel.com, JBeulich@suse.com,
	George.Dunlap@eu.citrix.com, jacob.shin@amd.com,
	eddie.dong@intel.com, dietmar.hahn@ts.fujitsu.com,
	suravee.suthikulpanit@amd.com,
	Boris Ostrovsky <boris.ostrovsky@oracle.com>
Subject: [PATCH v1 00/13] x86/PMU: Xen PMU PV support
Date: Tue, 10 Sep 2013 11:20:57 -0400	[thread overview]
Message-ID: <1378826470-4085-1-git-send-email-boris.ostrovsky@oracle.com> (raw)

The following patch series adds PMU support in Xen for PV guests. There is a
companion patchset for Linux kernel. In addition, another set of changes will
be provided (later) for userland perf code.

This version has following limitations:
* For accurate profiling of dom0/Xen dom0 VCPUs should be pinned.
* Hypervisor code is only profiled on processors that have running dom0 VCPUs
on them.
* No backtrace support.
* Will fail to load under XSM: we ran out of bits in permissions vector and
this needs to be fixed separately


A few notes that may help reviewing:

* A shared data structure (xenpmu_data_t) between each PV VPCU and hypervisor
CPU is used for passing registers' values as well as PMU state at the time of
PMU interrupt.

* PMU interrupts are taken by hypervisor at NMI level for both HVM and PV.

* Guest's interrupt handler does not read/write PMU MSRs directly. Instead, it
accesses xenpmu_data_t and flushes it to HW it before returning.

* PMU mode is controlled at runtime via /sys/hypervisor/pmu/pmu/{pmu_mode,pmu_flags}
in addition to 'vpmu' boot option (which is preserved for back compatibility).
The following modes are provided:
  * disable: VPMU is off
  * enable: VPMU is on. Guests can profile themselves, dom0 profiles itself and Xen
  * priv_enable: dom0 only profiling. dom0 collects samples for everyone. Sampling
    in guests is suspended.

* /proc/xen/xensyms file exports hypervisor's symbols to dom0 (similar to
  /proc/kallsyms)

* VPMU infrastructure is now used for both HVM and PV and therefore has been moved
up from hvm subtree


Boris Ostrovsky (13):
  Export hypervisor symbols
  Set VCPU's is_running flag closer to when the VCPU is dispatched
  x86/PMU: Stop AMD counters when called from vpmu_save_force()
  x86/VPMU: Minor VPMU cleanup
  intel/VPMU: Clean up Intel VPMU code
  x86/PMU: Add public xenpmu.h
  x86/PMU: Make vpmu not HVM-specific
  x86/PMU: Interface for setting PMU mode and flags
  x86/PMU: Initialize PMU for PV guests
  x86/PMU: Add support for PMU registes handling on PV guests
  x86/PMU: Handle PMU interrupts for PV guests
  x86/PMU: Save VPMU state for PV guests during context switch
  x86/PMU: Move vpmu files up from hvm directory

 xen/arch/x86/Makefile                          |   9 +-
 xen/arch/x86/apic.c                            |  13 -
 xen/arch/x86/domain.c                          |  18 +-
 xen/arch/x86/hvm/Makefile                      |   1 -
 xen/arch/x86/hvm/svm/Makefile                  |   1 -
 xen/arch/x86/hvm/svm/entry.S                   |   2 +
 xen/arch/x86/hvm/svm/vpmu.c                    | 494 -------------
 xen/arch/x86/hvm/vmx/Makefile                  |   1 -
 xen/arch/x86/hvm/vmx/entry.S                   |   1 +
 xen/arch/x86/hvm/vmx/vmcs.c                    |  59 ++
 xen/arch/x86/hvm/vmx/vpmu_core2.c              | 894 -----------------------
 xen/arch/x86/hvm/vpmu.c                        | 266 -------
 xen/arch/x86/oprofile/op_model_ppro.c          |   3 +-
 xen/arch/x86/platform_hypercall.c              |  14 +
 xen/arch/x86/traps.c                           |  38 +-
 xen/arch/x86/vpmu.c                            | 545 ++++++++++++++
 xen/arch/x86/vpmu_amd.c                        | 486 +++++++++++++
 xen/arch/x86/vpmu_intel.c                      | 938 +++++++++++++++++++++++++
 xen/arch/x86/x86_64/asm-offsets.c              |   1 +
 xen/arch/x86/x86_64/compat/entry.S             |   4 +
 xen/arch/x86/x86_64/entry.S                    |   4 +
 xen/arch/x86/x86_64/platform_hypercall.c       |   2 +-
 xen/common/event_channel.c                     |   1 +
 xen/common/schedule.c                          |  10 +-
 xen/common/symbols-dummy.c                     |   1 +
 xen/common/symbols.c                           |  78 +-
 xen/include/asm-x86/domain.h                   |   3 +
 xen/include/asm-x86/hvm/vcpu.h                 |   3 -
 xen/include/asm-x86/hvm/vmx/vmcs.h             |   3 +-
 xen/include/asm-x86/hvm/vmx/vpmu_core2.h       |  51 --
 xen/include/asm-x86/hvm/vpmu.h                 | 104 ---
 xen/include/asm-x86/irq.h                      |   1 -
 xen/include/asm-x86/mach-default/irq_vectors.h |   1 -
 xen/include/asm-x86/vpmu.h                     |  97 +++
 xen/include/public/platform.h                  |  21 +
 xen/include/public/xen.h                       |   2 +
 xen/include/public/xenpmu.h                    | 138 ++++
 xen/include/xen/hypercall.h                    |   4 +
 xen/include/xen/softirq.h                      |   1 +
 xen/include/xen/symbols.h                      |   4 +
 xen/tools/symbols.c                            |   4 +
 41 files changed, 2465 insertions(+), 1856 deletions(-)
 delete mode 100644 xen/arch/x86/hvm/svm/vpmu.c
 delete mode 100644 xen/arch/x86/hvm/vmx/vpmu_core2.c
 delete mode 100644 xen/arch/x86/hvm/vpmu.c
 create mode 100644 xen/arch/x86/vpmu.c
 create mode 100644 xen/arch/x86/vpmu_amd.c
 create mode 100644 xen/arch/x86/vpmu_intel.c
 delete mode 100644 xen/include/asm-x86/hvm/vmx/vpmu_core2.h
 delete mode 100644 xen/include/asm-x86/hvm/vpmu.h
 create mode 100644 xen/include/asm-x86/vpmu.h
 create mode 100644 xen/include/public/xenpmu.h

-- 
1.8.1.4

             reply	other threads:[~2013-09-10 15:20 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-09-10 15:20 Boris Ostrovsky [this message]
2013-09-10 15:20 ` [PATCH v1 01/13] Export hypervisor symbols Boris Ostrovsky
2013-09-11  7:51   ` Jan Beulich
2013-09-11 13:55     ` Boris Ostrovsky
2013-09-11 14:12       ` Jan Beulich
2013-09-11 14:57         ` Boris Ostrovsky
2013-09-11 16:01           ` Jan Beulich
2013-09-10 15:20 ` [PATCH v1 02/13] Set VCPU's is_running flag closer to when the VCPU is dispatched Boris Ostrovsky
2013-09-11  7:58   ` Jan Beulich
2013-09-10 15:21 ` [PATCH v1 03/13] x86/PMU: Stop AMD counters when called from vpmu_save_force() Boris Ostrovsky
2013-09-10 15:21 ` [PATCH v1 04/13] x86/VPMU: Minor VPMU cleanup Boris Ostrovsky
2013-09-10 15:21 ` [PATCH v1 05/13] intel/VPMU: Clean up Intel VPMU code Boris Ostrovsky
2013-09-10 15:21 ` [PATCH v1 06/13] x86/PMU: Add public xenpmu.h Boris Ostrovsky
2013-09-11  8:13   ` Jan Beulich
2013-09-11 14:03     ` Boris Ostrovsky
2013-09-11 14:16       ` Jan Beulich
2013-09-11  8:37   ` Ian Campbell
2013-09-10 15:21 ` [PATCH v1 07/13] x86/PMU: Make vpmu not HVM-specific Boris Ostrovsky
2013-09-10 15:21 ` [PATCH v1 08/13] x86/PMU: Interface for setting PMU mode and flags Boris Ostrovsky
2013-09-10 15:21 ` [PATCH v1 09/13] x86/PMU: Initialize PMU for PV guests Boris Ostrovsky
2013-09-10 15:21 ` [PATCH v1 10/13] x86/PMU: Add support for PMU registes handling on " Boris Ostrovsky
2013-09-10 15:21 ` [PATCH v1 11/13] x86/PMU: Handle PMU interrupts for " Boris Ostrovsky
2013-09-10 15:21 ` [PATCH v1 12/13] x86/PMU: Save VPMU state for PV guests during context switch Boris Ostrovsky
2013-09-10 15:21 ` [PATCH v1 13/13] x86/PMU: Move vpmu files up from hvm directory Boris Ostrovsky
2013-09-10 15:34 ` [PATCH v1 00/13] x86/PMU: Xen PMU PV support Jan Beulich
2013-09-10 15:47   ` Boris Ostrovsky
2013-09-11 17:01     ` George Dunlap
2013-09-11 18:22       ` Boris Ostrovsky
2013-09-12  9:39         ` George Dunlap
2013-09-12 14:58           ` Boris Ostrovsky

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=1378826470-4085-1-git-send-email-boris.ostrovsky@oracle.com \
    --to=boris.ostrovsky@oracle.com \
    --cc=George.Dunlap@eu.citrix.com \
    --cc=JBeulich@suse.com \
    --cc=dietmar.hahn@ts.fujitsu.com \
    --cc=eddie.dong@intel.com \
    --cc=jacob.shin@amd.com \
    --cc=jun.nakajima@intel.com \
    --cc=suravee.suthikulpanit@amd.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).