qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Igor Mammedov <imammedo@redhat.com>
To: Eduardo Habkost <ehabkost@redhat.com>
Cc: aliguori@us.ibm.com, claudio.fontana@huawei.com,
	qemu-devel@nongnu.org, aderumier@odiso.com,
	lcapitulino@redhat.com, jfrei@linux.vnet.ibm.com,
	yang.z.zhang@intel.com, pbonzini@redhat.com, afaerber@suse.de,
	lig.fnst@cn.fujitsu.com, rth@twiddle.net
Subject: Re: [Qemu-devel] [PATCH 22/22] add cpu-add qmp command and implement CPU hot-add for target-i386
Date: Thu, 11 Apr 2013 17:17:29 +0200	[thread overview]
Message-ID: <20130411171729.2fb431cc@thinkpad> (raw)
In-Reply-To: <20130405172435.GD6862@otherpad.lan.raisama.net>

On Fri, 5 Apr 2013 14:24:35 -0300
Eduardo Habkost <ehabkost@redhat.com> wrote:

> On Fri, Apr 05, 2013 at 02:10:54PM -0300, Eduardo Habkost wrote:
> > On Fri, Apr 05, 2013 at 04:37:16PM +0200, Igor Mammedov wrote:
> > [...]
> > > diff --git a/qapi-schema.json b/qapi-schema.json
> > > index db542f6..a760ed5 100644
> > > --- a/qapi-schema.json
> > > +++ b/qapi-schema.json
> > > @@ -1387,6 +1387,17 @@
> > >  { 'command': 'cpu', 'data': {'index': 'int'} }
> > >  
> > >  ##
> > > +# @cpu-add
> > > +#
> > > +# Adds CPU with specified id
> > > +#
> > > +# @id: cpu id of CPU to be created
> > 
> > Can we have the semantics/constraints of "id" documented here? Is it an
> > arbitrary ID chosen by the caller? Does it have to be the APIC ID? Does
> > it have to be the index of the CPU in the CPU list? How the IDs of
> > existing CPUs set using "-smp" are allocated?
> > 
> > I am looking at the code right now to understand how this implementation
> > works, but the documentation could contain or point to documentation on
> > how the "id" parameter is used and interpreted.
> 
> So, the answer to my own question seems to be on patch 21/22. Let me
> check if I understand this correctly:
> 
>  @id: cpu id of the CPU to be created. It must be one of the the
>       available IDs listed at the "/machine/icc-bridge/cpu[0..N]" links.
> 
> Is the above correct?
yes.

> 
> Now, my question is: suppose a caller starts QEMU as:
> 
>  $ qemu -smp 18,cores=3,threads=3,maxcpus=36 \
>         -numa node,cpus=0-8   -numa node,cpus=9-17 \
>         -numa node,cpus=18-26 -numa node,cpus=27-35
> 
> and the caller wants to hot-add all VCPUs in the last CPU socket (that
> means: all the VCPUs in the last NUMA node, that means: CPU indexes
> 27-35). What should the caller do to find out which of the
> /machine/icc-bridge/cpu[0..N] links/IDs really correspond to those
> VCPUs?
If one wants plug in a specific CPU, It won't work with -numa yet.

to make it work we need to specify sockets on -numa cmd. line and then
construct synthetic QOM containers tree that could looklike:

/machine/numa_nodes/[0..N]/sockets/[0..M]/cpus/[0..X]

then command line could look like:
$ qemu -smp 18,cores=3,threads=3,maxcpus=36 \
       -numa node,sockets=0-1 -numa node,sockets=2-3
> 
> 
> > 
> > > +#
> > > +# Returns: Nothing on success
> > > +##
> > > +{ 'command': 'cpu-add', 'data': {'id': 'int'} }
> > > +
> > > +##
> > -- 
> > Eduardo
> > 
> 
> -- 
> Eduardo
> 


-- 
Regards,
  Igor

  reply	other threads:[~2013-04-11 15:19 UTC|newest]

Thread overview: 77+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-04-05 14:36 [Qemu-devel] [PATCH 00/22 v2] target-i386: CPU hot-add with cpu-add QMP command Igor Mammedov
2013-04-05 14:36 ` [Qemu-devel] [PATCH 01/22] target-i386: consolidate error propagation in x86_cpu_realizefn() Igor Mammedov
2013-04-09 17:42   ` Andreas Färber
2013-04-05 14:36 ` [Qemu-devel] [PATCH 02/22] target-i386: split APIC creation from initialization " Igor Mammedov
2013-04-08  2:26   ` li guang
2013-04-08 18:16   ` Eduardo Habkost
2013-04-09 18:52   ` Andreas Färber
2013-04-05 14:36 ` [Qemu-devel] [PATCH 03/22] target-i386: split out CPU creation and features parsing into cpu_x86_create() Igor Mammedov
2013-04-08 18:30   ` Eduardo Habkost
2013-04-09 10:30     ` Paolo Bonzini
2013-04-09 10:33       ` Igor Mammedov
2013-04-09 14:02         ` Andreas Färber
2013-04-05 14:36 ` [Qemu-devel] [PATCH 04/22] cpu: Pass CPUState to *cpu_synchronize_post*() Igor Mammedov
2013-04-08 19:38   ` Eduardo Habkost
2013-04-05 14:36 ` [Qemu-devel] [PATCH 05/22] cpu: call cpu_synchronize_post_init() from CPUClass.realize() if hotplugged Igor Mammedov
2013-04-08 19:45   ` Eduardo Habkost
2013-04-09 10:13     ` Igor Mammedov
2013-04-09 11:17       ` Paolo Bonzini
2013-04-09 11:15   ` Paolo Bonzini
2013-04-05 14:36 ` [Qemu-devel] [PATCH 06/22] cpu: introduce CPUClass.resume() method Igor Mammedov
2013-04-08  2:27   ` li guang
2013-04-08 20:13   ` Eduardo Habkost
2013-04-09 10:26     ` Igor Mammedov
2013-04-09 13:21       ` Andreas Färber
2013-04-09 11:20     ` Paolo Bonzini
2013-04-10 12:57       ` Igor Mammedov
2013-04-05 14:36 ` [Qemu-devel] [PATCH 07/22] target-i386: kvmvapic: replace FROM_SYSBUS() with QOM type cast Igor Mammedov
2013-04-10 17:54   ` Andreas Färber
2013-04-05 14:37 ` [Qemu-devel] [PATCH 08/22] target-i386: ioapic: " Igor Mammedov
2013-04-08  2:13   ` li guang
2013-04-08 11:32     ` Igor Mammedov
2013-04-09 11:36       ` Paolo Bonzini
2013-04-10  0:21         ` li guang
2013-04-10  8:06           ` Paolo Bonzini
2013-04-10 16:12           ` Igor Mammedov
2013-04-10 18:12             ` Andreas Färber
2013-04-10 17:58   ` Andreas Färber
2013-04-05 14:37 ` [Qemu-devel] [PATCH 09/22] introduce CPU hot-plug notifier Igor Mammedov
2013-04-09 11:23   ` Paolo Bonzini
2013-04-05 14:37 ` [Qemu-devel] [PATCH 10/22] rtc: update rtc_cmos on CPU hot-plug Igor Mammedov
2013-04-05 14:37 ` [Qemu-devel] [PATCH 11/22] cpu: introduce get_firmware_id() method and override it for target-i386 Igor Mammedov
2013-04-08  2:02   ` li guang
2013-04-08 11:41     ` Igor Mammedov
2013-04-05 14:37 ` [Qemu-devel] [PATCH 12/22] cpu: add helper cpu_exists(), to check if CPU with specified id exists Igor Mammedov
2013-04-09 11:25   ` Paolo Bonzini
2013-04-05 14:37 ` [Qemu-devel] [PATCH 13/22] acpi_piix4: add infrastructure to send CPU hot-plug GPE to guest Igor Mammedov
2013-04-08  2:24   ` li guang
2013-04-08 11:47     ` Igor Mammedov
2013-04-05 14:37 ` [Qemu-devel] [PATCH 14/22] target-i386: introduce apic-id property Igor Mammedov
2013-04-09 11:26   ` Paolo Bonzini
2013-04-05 14:37 ` [Qemu-devel] [PATCH 15/22] introduce ICC bus/device/bridge Igor Mammedov
2013-04-05 14:37 ` [Qemu-devel] [PATCH 16/22] target-i386: cpu: attach ICC bus to CPU on its creation Igor Mammedov
2013-04-05 14:37 ` [Qemu-devel] [PATCH 17/22] target-i386: replace MSI_SPACE_SIZE with APIC_SPACE_SIZE Igor Mammedov
2013-04-05 14:37 ` [Qemu-devel] [PATCH 18/22] target-i386: move APIC to ICC bus Igor Mammedov
2013-04-05 16:15   ` Eduardo Habkost
2013-04-05 22:23     ` Igor Mammedov
2013-04-05 22:31   ` Igor Mammedov
2013-04-09 11:29   ` Paolo Bonzini
2013-04-05 14:37 ` [Qemu-devel] [PATCH 18/22] target-i386: move IOAPIC " Igor Mammedov
2013-04-09 11:33   ` Paolo Bonzini
2013-04-09 12:51     ` Igor Mammedov
2013-04-05 14:37 ` [Qemu-devel] [PATCH 19/22] target-i386: move APIC " Igor Mammedov
2013-04-09 12:47   ` Igor Mammedov
2013-04-05 14:37 ` [Qemu-devel] [PATCH 19/22] target-i386: move IOAPIC " Igor Mammedov
2013-04-05 14:37 ` [Qemu-devel] [PATCH 20/22] qdev: set device's parent before calling realize() down inheritance chain Igor Mammedov
2013-04-09 11:34   ` Paolo Bonzini
2013-04-05 14:37 ` [Qemu-devel] [PATCH 21/22] target-i386: expose all possible CPUs as /machine/icc-bridge/cpu[0..N] links Igor Mammedov
2013-04-05 14:37 ` [Qemu-devel] [PATCH 22/22] add cpu-add qmp command and implement CPU hot-add for target-i386 Igor Mammedov
2013-04-05 17:10   ` Eduardo Habkost
2013-04-05 17:24     ` Eduardo Habkost
2013-04-11 15:17       ` Igor Mammedov [this message]
2013-04-11 15:49         ` Eduardo Habkost
2013-04-09 20:19     ` Igor Mammedov
2013-04-09 20:46       ` Eduardo Habkost
2013-04-09 21:05         ` Igor Mammedov
2013-04-11 15:12           ` Eduardo Habkost
2013-04-11 15:37             ` Igor Mammedov

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=20130411171729.2fb431cc@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=jfrei@linux.vnet.ibm.com \
    --cc=lcapitulino@redhat.com \
    --cc=lig.fnst@cn.fujitsu.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).