From: Bharata B Rao <bharata@linux.vnet.ibm.com>
To: qemu-devel@nongnu.org
Cc: ehabkost@redhat.com, aik@ozlabs.ru,
Bharata B Rao <bharata@linux.vnet.ibm.com>,
agraf@suse.de, armbru@redhat.com, pbonzini@redhat.com,
imammedo@redhat.com, afaerber@suse.de,
david@gibson.dropbear.id.au
Subject: [Qemu-devel] [RFC PATCH v0 0/8] cpu-package hotplug
Date: Mon, 22 Feb 2016 10:31:17 +0530 [thread overview]
Message-ID: <1456117285-22273-1-git-send-email-bharata@linux.vnet.ibm.com> (raw)
Hi,
This is an attempt to implement David Gibson's RFC that was posted at
https://lists.gnu.org/archive/html/qemu-ppc/2016-02/msg00000.html
I am not sure if I have followed all the aspects of the RFC fully, but we
can make changes going forward.
An example cpu-package implementation is done for sPAPR in this patchset.
Hot removal is not yet done in this patchset.
For the command line,
-smp 8,sockets=1,cores=1,threads=8,maxcpus=16 -numa node,nodeid=0,cpus=0-7 -numa node,nodeid=1,cpus=8-15
the HMP query looks like this:
(qemu) info cpu-packages
CPU Package: ""
type: "spapr-cpu-package"
qom_path: "/machine/cpu-package[0]"
realized: true
nr_cpus: 8
CPU: 0
Type: "host-powerpc64-cpu"
Arch ID: 0
Thread: 0
Core: 0
Socket: 0
Node: 0
CPU: 1
Type: "host-powerpc64-cpu"
Arch ID: 1
Thread: 1
Core: 0
Socket: 0
Node: 0
CPU: 2
Type: "host-powerpc64-cpu"
Arch ID: 2
Thread: 2
Core: 0
Socket: 0
Node: 0
CPU: 3
Type: "host-powerpc64-cpu"
Arch ID: 3
Thread: 3
Core: 0
Socket: 0
Node: 0
CPU: 4
Type: "host-powerpc64-cpu"
Arch ID: 4
Thread: 4
Core: 0
Socket: 0
Node: 0
CPU: 5
Type: "host-powerpc64-cpu"
Arch ID: 5
Thread: 5
Core: 0
Socket: 0
Node: 0
CPU: 6
Type: "host-powerpc64-cpu"
Arch ID: 6
Thread: 6
Core: 0
Socket: 0
Node: 0
CPU: 7
Type: "host-powerpc64-cpu"
Arch ID: 7
Thread: 7
Core: 0
Socket: 0
Node: 0
CPU Package: ""
type: "spapr-cpu-package"
qom_path: "/machine/cpu-package[1]"
realized: false
nr_cpus: 8
As can be seen from above, all the cores upto max_cpus are created upfront
here and hot plug is done in the following manner:
(qemu) qom-set /machine/cpu-package[1] realized true
This will result in the 2nd cpu-package consisting of a core with 8 threads
to become available.
I am not fully sure if the QMP emumeration here works for all archs, but
just wanted to share what I currently.
Bharata B Rao (8):
cpu: Store CPU typename in MachineState
cpu: Don't realize CPU from cpu_generic_init()
cpu: CPU package abstract device
spapr: Introduce CPU core device
spapr: Convert boot CPUs into CPU core device initialization
spapr: CPU hotplug support
qmp: Implement query cpu-packages
hmp: Implement 'info cpu-slots'
hmp-commands-info.hx | 14 +++
hmp.c | 50 ++++++++
hmp.h | 1 +
hw/cpu/Makefile.objs | 1 +
hw/cpu/package.c | 85 +++++++++++++
hw/ppc/Makefile.objs | 1 +
hw/ppc/spapr.c | 246 +++++++++++++++++++++++++++++++++++--
hw/ppc/spapr_cpu_package.c | 50 ++++++++
hw/ppc/spapr_events.c | 3 +
hw/ppc/spapr_rtas.c | 24 ++++
include/hw/boards.h | 2 +
include/hw/cpu/package.h | 27 ++++
include/hw/ppc/spapr.h | 1 +
include/hw/ppc/spapr_cpu_package.h | 27 ++++
qapi-schema.json | 48 ++++++++
qom/cpu.c | 6 -
target-arm/helper.c | 16 ++-
target-cris/cpu.c | 16 ++-
target-lm32/helper.c | 16 ++-
target-moxie/cpu.c | 16 ++-
target-openrisc/cpu.c | 16 ++-
target-ppc/translate_init.c | 16 ++-
target-sh4/cpu.c | 16 ++-
target-tricore/helper.c | 16 ++-
target-unicore32/helper.c | 16 ++-
25 files changed, 707 insertions(+), 23 deletions(-)
create mode 100644 hw/cpu/package.c
create mode 100644 hw/ppc/spapr_cpu_package.c
create mode 100644 include/hw/cpu/package.h
create mode 100644 include/hw/ppc/spapr_cpu_package.h
--
2.1.0
next reply other threads:[~2016-02-22 5:02 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-02-22 5:01 Bharata B Rao [this message]
2016-02-22 5:01 ` [Qemu-devel] [RFC PATCH v0 1/8] cpu: Store CPU typename in MachineState Bharata B Rao
2016-02-22 8:04 ` David Gibson
2016-02-22 5:01 ` [Qemu-devel] [RFC PATCH v0 2/8] cpu: Don't realize CPU from cpu_generic_init() Bharata B Rao
2016-02-22 5:01 ` [Qemu-devel] [RFC PATCH v0 3/8] cpu: CPU package abstract device Bharata B Rao
2016-02-22 5:01 ` [Qemu-devel] [RFC PATCH v0 4/8] spapr: Introduce CPU core device Bharata B Rao
2016-02-22 6:44 ` Andreas Färber
2016-02-22 7:47 ` David Gibson
2016-02-22 15:58 ` Andreas Färber
2016-02-22 23:24 ` David Gibson
2016-02-22 8:05 ` Bharata B Rao
2016-02-22 15:48 ` Andreas Färber
2016-02-22 5:01 ` [Qemu-devel] [RFC PATCH v0 5/8] spapr: Convert boot CPUs into CPU core device initialization Bharata B Rao
2016-02-22 5:01 ` [Qemu-devel] [RFC PATCH v0 6/8] spapr: CPU hotplug support Bharata B Rao
2016-02-22 5:01 ` [Qemu-devel] [RFC PATCH v0 7/8] qmp: Implement query cpu-packages Bharata B Rao
2016-02-22 16:49 ` Eric Blake
2016-02-23 8:37 ` Igor Mammedov
2016-02-22 5:01 ` [Qemu-devel] [RFC PATCH v0 8/8] hmp: Implement 'info cpu-slots' Bharata B Rao
2016-02-22 5:10 ` [Qemu-devel] [RFC PATCH v0 0/8] cpu-package hotplug Bharata B Rao
2016-02-22 15:32 ` Andreas Färber
2016-02-22 23:28 ` David Gibson
2016-02-23 6:11 ` Bharata B Rao
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=1456117285-22273-1-git-send-email-bharata@linux.vnet.ibm.com \
--to=bharata@linux.vnet.ibm.com \
--cc=afaerber@suse.de \
--cc=agraf@suse.de \
--cc=aik@ozlabs.ru \
--cc=armbru@redhat.com \
--cc=david@gibson.dropbear.id.au \
--cc=ehabkost@redhat.com \
--cc=imammedo@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.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).