qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Igor Mammedov <imammedo@redhat.com>
To: Jan Kiszka <jan.kiszka@siemens.com>
Cc: "aliguori@us.ibm.com" <aliguori@us.ibm.com>,
	"ehabkost@redhat.com" <ehabkost@redhat.com>,
	"mst@redhat.com" <mst@redhat.com>,
	"claudio.fontana@huawei.com" <claudio.fontana@huawei.com>,
	"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>,
	"aderumier@odiso.com" <aderumier@odiso.com>,
	"lcapitulino@redhat.com" <lcapitulino@redhat.com>,
	"jfrei@linux.vnet.ibm.com" <jfrei@linux.vnet.ibm.com>,
	"yang.z.zhang@intel.com" <yang.z.zhang@intel.com>,
	"pbonzini@redhat.com" <pbonzini@redhat.com>,
	"afaerber@suse.de" <afaerber@suse.de>,
	"lig.fnst@cn.fujitsu.com" <lig.fnst@cn.fujitsu.com>,
	"rth@twiddle.net" <rth@twiddle.net>
Subject: Re: [Qemu-devel] [PATCH 00/16 v4] target-i386: CPU hot-add with cpu-add QMP command
Date: Tue, 16 Apr 2013 11:43:42 +0200	[thread overview]
Message-ID: <20130416114342.7e7b75f5@thinkpad> (raw)
In-Reply-To: <516D0C3F.7010003@siemens.com>

On Tue, 16 Apr 2013 10:30:55 +0200
Jan Kiszka <jan.kiszka@siemens.com> wrote:

> On 2013-04-16 00:12, Igor Mammedov wrote:
> > Implements alternative way for hot-adding CPU using cpu-add QMP command,
> > wich could be useful until it would be possible to add CPUs via device_add.
> 
> Didn't track the full story: What prevents currently a device_add
> approach? And that so effectively that we have to create a
> to-be-deprecated-again QMP API first?
I guess mostly it's review bandwidth and time concerns, it moves but quite slow
and won't be able to make into 1.5 and possibly miss 1.6 with current speed.

In this series all patches except the last one applicable to device_add as
well so any feedback is appreciated.

To make device_add usable for CPU, I have on my TODO list following
series/topics:

1. convert cpuid features to static properties.
http://lists.gnu.org/archive/html/qemu-devel/2013-02/msg04426.html

2. x86 CPU subclasses
http://lists.gnu.org/archive/html/qemu-devel/2013-02/msg00673.html

Summarized issues/ways to go with CPU subclasses could be found here:
http://wiki.qemu.org/Features/CPUHotplug section "CPU models as CPU subclasses"

[optional]
3. nice to have, unify cpu_model handling and convert it to utilizing global
properties. With it hotplug could look as simple as:
 device_add apic_id=xxx
without specifying all flags that were on -cpu command line.


BTW:
 You were advocating using APIC ID on device_add to identify CPU but Eduardo
 would like avoid its usage on external interfaces.
 It would be nice to have your opinion on subject.
  relevant discussion threads are here:
    1. http://lists.gnu.org/archive/html/qemu-devel/2013-04/msg02205.html
    2. http://lists.gnu.org/archive/html/qemu-devel/2013-04/msg02274.html

> Thanks,
> Jan
> 
> > 
> > All patches except the last are also applicable to device_add aprroach.
> > 
> > To hot-add CPU use following command from qmp-shell:
> >  cpu-add id=[0..max-cpus - 1)
> > 
> > git tree for testing: https://github.com/imammedo/qemu/tree/cpu_add.v4
> > 
> > based on qom-cpu tree
> > 
> > v4->v3:
> >   * 'id' in cpu-add command will be a thread number instead of APIC ID
> >   * split off resume_vcpu() into separate patch
> >   * move notifier from rtc code into pc.c
> > 
> > v2->v3:
> >   * use local error & propagate_error() instead of operating on
> >     passed in errp in several places
> >   * replace CPUClass.get_firmware_id() with CPUClass.get_arch_id()
> >   * leave IOAPIC creation to board and just set bus to icc-bus
> >   * include kvm-stub.o in cpu libary if no KVM is configured
> >   * create resume_vcpu() stub and include it in libqemustub,
> >     and use it directly instead of CPU method
> >   * acpi_piix4: s/cpu_add_notifier/cpu_added_notifier/
> > 
> > v1->v2:
> >   * generalize cpu sync to KVM, resume and hot-plug notification and
> >     invoke them form CPUClass, to make available to all targets.
> >   * introduce cpu_exists() and CPUClass.get_firmware_id() and use
> >     the last one in acpi_piix to make code target independent.
> >   * move IOAPIC to ICC bus, it was suggested and easy to convert.
> >   * leave kvmvapic as SysBusDevice, it doesn't affect hot-plug and
> >     created only once for all APIC instances. I haven't found yet
> >     good/clean enough way to convert it to ICCDevice. May be follow-up
> >     though.
> >   * split one big ICC patch into several, one per converted device
> >   * add cpu_hot_add hook to machine and implement it for target-i386,
> >     instead of adding stabs. Could be used by other targets to
> >     implement cpu-add.
> >   * pre-allocate links<CPU> for all possible CPUs and make them available
> >     at /machine/icc-bridge/cpu[0..N] QOM path, so users could find out
> >     possible/free CPU IDs to use in cpu-add command.
> > 
> > CC: pbonzini@redhat.com
> > CC: afaerber@suse.de
> > CC: ehabkost@redhat.com
> > 
> > Igor Mammedov (16):
> >   cpu: make kvm-stub.o a part of CPU library
> >   cpu: call cpu_synchronize_post_init() from CPUClass.realize() if
> >     hotplugged
> >   introduce resume_vcpu(), for single CPU
> >   cpu: resume CPU from CPUClass.cpu_common_realizefn() when it is
> >     hot-plugged
> >   introduce CPU hot-plug notifier
> >   target-i386: pc: update rtc_cmos on CPU hot-plug
> >   cpu: introduce get_arch_id() method and override it for target-i386
> >   cpu: add helper cpu_exists(), to check if CPU with specified id exists
> >   acpi_piix4: add infrastructure to send CPU hot-plug GPE to guest
> >   target-i386: introduce apic-id property
> >   introduce ICC bus/device/bridge
> >   target-i386: cpu: attach ICC bus to CPU on its creation
> >   target-i386: replace MSI_SPACE_SIZE with APIC_SPACE_SIZE
> >   target-i386: move APIC to ICC bus
> >   target-i386: move IOAPIC to ICC bus
> >   add cpu-add qmp command and implement CPU hot-add for target-i386
> > 
> >  Makefile.target                    |  14 ++---
> >  cpus.c                             |  11 +++-
> >  default-configs/i386-softmmu.mak   |   1 +
> >  default-configs/x86_64-softmmu.mak |   1 +
> >  hw/acpi/piix4.c                    | 114 ++++++++++++++++++++++++++++++++++++-
> >  hw/cpu/Makefile.objs               |   1 +
> >  hw/cpu/icc_bus.c                   | 104 +++++++++++++++++++++++++++++++++
> >  hw/i386/kvm/apic.c                 |   2 +-
> >  hw/i386/kvm/ioapic.c               |   2 +-
> >  hw/i386/kvmvapic.c                 |   1 +
> >  hw/i386/pc.c                       |  91 +++++++++++++++++++++++++++--
> >  hw/i386/pc_piix.c                  |   7 +++
> >  hw/i386/pc_q35.c                   |   7 +++
> >  hw/intc/apic.c                     |   2 +-
> >  hw/intc/apic_common.c              |  17 ++++--
> >  hw/intc/ioapic_common.c            |  15 +++--
> >  hw/timer/mc146818rtc.c             |   7 +++
> >  hw/xen/xen_apic.c                  |   2 +-
> >  include/exec/memory.h              |  10 ++++
> >  include/hw/boards.h                |   3 +
> >  include/hw/i386/apic_internal.h    |   8 +--
> >  include/hw/i386/icc_bus.h          |  53 +++++++++++++++++
> >  include/hw/i386/ioapic_internal.h  |   6 +-
> >  include/hw/pci/msi.h               |   2 +
> >  include/hw/timer/mc146818rtc.h     |   1 +
> >  include/qom/cpu.h                  |  21 +++++++
> >  include/sysemu/kvm.h               |  22 +++----
> >  include/sysemu/sysemu.h            |   3 +
> >  kvm-all.c                          |   1 +
> >  kvm-stub.c                         |   3 +
> >  memory.c                           |  11 ++++
> >  qapi-schema.json                   |  11 ++++
> >  qmp-commands.hx                    |  23 ++++++++
> >  qmp.c                              |  10 ++++
> >  qom/cpu.c                          |  44 ++++++++++++++
> >  stubs/Makefile.objs                |   1 +
> >  stubs/cpus.c                       |   5 ++
> >  target-i386/cpu.c                  |  75 +++++++++++++++++++-----
> >  target-i386/cpu.h                  |   1 +
> >  vl.c                               |   7 ++-
> >  40 files changed, 656 insertions(+), 64 deletions(-)
> >  create mode 100644 hw/cpu/icc_bus.c
> >  create mode 100644 include/hw/i386/icc_bus.h
> >  create mode 100644 stubs/cpus.c
> > 
> 
> -- 
> Siemens AG, Corporate Technology, CT RTC ITP SDP-DE
> Corporate Competence Center Embedded Linux
> 


-- 
Regards,
  Igor

  reply	other threads:[~2013-04-16  9:44 UTC|newest]

Thread overview: 73+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-04-15 22:12 [Qemu-devel] [PATCH 00/16 v4] target-i386: CPU hot-add with cpu-add QMP command Igor Mammedov
2013-04-15 22:12 ` [Qemu-devel] [PATCH 01/16] cpu: make kvm-stub.o a part of CPU library Igor Mammedov
     [not found]   ` <20130417185844.GA11821@otherpad.lan.raisama.net>
2013-04-17 22:26     ` Paolo Bonzini
2013-04-18  4:23       ` [Qemu-devel] [PATCH 01/16 v2] " Igor Mammedov
2013-04-18 14:24         ` Eduardo Habkost
2013-04-22 13:45           ` Andreas Färber
2013-04-22 14:35             ` Paolo Bonzini
2013-04-22 16:29               ` Gleb Natapov
2013-04-15 22:12 ` [Qemu-devel] [PATCH 02/16] cpu: call cpu_synchronize_post_init() from CPUClass.realize() if hotplugged Igor Mammedov
2013-04-18 17:03   ` Eduardo Habkost
2013-04-22 13:46   ` Andreas Färber
2013-04-22 16:30     ` Gleb Natapov
2013-04-15 22:12 ` [Qemu-devel] [PATCH 03/16] introduce resume_vcpu(), for single CPU Igor Mammedov
2013-04-18 17:04   ` Eduardo Habkost
2013-04-22 10:40   ` Gleb Natapov
2013-04-22 10:54     ` Igor Mammedov
2013-04-22 10:59       ` Gleb Natapov
2013-04-22 11:45         ` Igor Mammedov
2013-04-22 12:34           ` Gleb Natapov
2013-04-22 13:42           ` Andreas Färber
2013-04-22 14:50             ` Igor Mammedov
2013-04-22 14:59               ` Andreas Färber
2013-04-15 22:12 ` [Qemu-devel] [PATCH 04/16] cpu: resume CPU from CPUClass.cpu_common_realizefn() when it is hot-plugged Igor Mammedov
2013-04-18 17:04   ` Eduardo Habkost
2013-04-15 22:12 ` [Qemu-devel] [PATCH 05/16] introduce CPU hot-plug notifier Igor Mammedov
2013-04-22 11:00   ` Gleb Natapov
2013-04-22 11:09     ` Igor Mammedov
2013-04-22 11:24       ` Gleb Natapov
2013-04-22 20:01         ` Igor Mammedov
2013-04-15 22:12 ` [Qemu-devel] [PATCH 06/16] target-i386: pc: update rtc_cmos on CPU hot-plug Igor Mammedov
2013-04-18 17:09   ` Eduardo Habkost
2013-04-22 14:56   ` Andreas Färber
2013-04-22 15:18     ` Igor Mammedov
2013-04-15 22:12 ` [Qemu-devel] [PATCH 07/16] cpu: introduce get_arch_id() method and override it for target-i386 Igor Mammedov
2013-04-18 17:10   ` Eduardo Habkost
2013-04-19  0:05   ` li guang
2013-04-22  9:42   ` Michael S. Tsirkin
2013-04-22 16:33   ` Andreas Färber
2013-04-22 19:10     ` Igor Mammedov
2013-04-15 22:12 ` [Qemu-devel] [PATCH 08/16] cpu: add helper cpu_exists(), to check if CPU with specified id exists Igor Mammedov
2013-04-22 10:28   ` Michael S. Tsirkin
2013-04-22 10:45     ` Igor Mammedov
2013-04-22 11:15     ` Andreas Färber
2013-04-15 22:12 ` [Qemu-devel] [PATCH 09/16] acpi_piix4: add infrastructure to send CPU hot-plug GPE to guest Igor Mammedov
2013-04-22  9:55   ` Michael S. Tsirkin
2013-04-15 22:12 ` [Qemu-devel] [PATCH 10/16] target-i386: introduce apic-id property Igor Mammedov
2013-04-22  9:49   ` Michael S. Tsirkin
2013-04-22 14:05   ` Andreas Färber
2013-04-22 16:30     ` Igor Mammedov
2013-04-26 16:35       ` Andreas Färber
2013-04-15 22:12 ` [Qemu-devel] [PATCH 11/16] introduce ICC bus/device/bridge Igor Mammedov
2013-04-22 11:39   ` Peter Maydell
2013-04-22 12:27     ` Paolo Bonzini
2013-04-22 13:22   ` Andreas Färber
2013-04-22 15:08     ` Igor Mammedov
2013-04-15 22:12 ` [Qemu-devel] [PATCH 12/16] target-i386: cpu: attach ICC bus to CPU on its creation Igor Mammedov
2013-04-22 15:02   ` Andreas Färber
2013-04-22 15:20     ` Igor Mammedov
2013-04-22 15:27       ` Andreas Färber
2013-04-22 19:14         ` Igor Mammedov
2013-04-15 22:12 ` [Qemu-devel] [PATCH 13/16] target-i386: replace MSI_SPACE_SIZE with APIC_SPACE_SIZE Igor Mammedov
2013-04-22 15:03   ` Andreas Färber
2013-04-15 22:12 ` [Qemu-devel] [PATCH 14/16] target-i386: move APIC to ICC bus Igor Mammedov
2013-04-22 15:18   ` Andreas Färber
2013-04-22 15:59     ` Igor Mammedov
2013-04-15 22:12 ` [Qemu-devel] [PATCH 15/16] target-i386: move IOAPIC " Igor Mammedov
2013-04-15 22:12 ` [Qemu-devel] [PATCH 16/16] add cpu-add qmp command and implement CPU hot-add for target-i386 Igor Mammedov
2013-04-15 22:20   ` Eric Blake
2013-04-16 20:04     ` Igor Mammedov
2013-04-23 16:17       ` Eric Blake
2013-04-16  8:30 ` [Qemu-devel] [PATCH 00/16 v4] target-i386: CPU hot-add with cpu-add QMP command Jan Kiszka
2013-04-16  9:43   ` Igor Mammedov [this message]
2013-04-16 13:49     ` Eduardo Habkost

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=20130416114342.7e7b75f5@thinkpad \
    --to=imammedo@redhat.com \
    --cc=aderumier@odiso.com \
    --cc=afaerber@suse.de \
    --cc=aliguori@us.ibm.com \
    --cc=claudio.fontana@huawei.com \
    --cc=ehabkost@redhat.com \
    --cc=jan.kiszka@siemens.com \
    --cc=jfrei@linux.vnet.ibm.com \
    --cc=lcapitulino@redhat.com \
    --cc=lig.fnst@cn.fujitsu.com \
    --cc=mst@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=rth@twiddle.net \
    --cc=yang.z.zhang@intel.com \
    /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).