From: Igor Mammedov <imammedo@redhat.com>
To: qemu-devel@nongnu.org
Cc: Eduardo Habkost <ehabkost@redhat.com>,
Peter Maydell <peter.maydell@linaro.org>,
Andrew Jones <drjones@redhat.com>,
David Gibson <david@gibson.dropbear.id.au>,
Eric Blake <eblake@redhat.com>,
Paolo Bonzini <pbonzini@redhat.com>,
Shannon Zhao <zhaoshenglong@huawei.com>,
qemu-arm@nongnu.org, qemu-ppc@nongnu.org
Subject: [Qemu-devel] [PATCH v2 23/24] numa: add '-numa cpu, ...' option for property based node mapping
Date: Wed, 3 May 2017 14:57:17 +0200 [thread overview]
Message-ID: <1493816238-33120-24-git-send-email-imammedo@redhat.com> (raw)
In-Reply-To: <1493816238-33120-1-git-send-email-imammedo@redhat.com>
legacy cpu to node mapping is using cpu index values to map
VCPU to node with help of '-numa node,nodeid=node,cpus=x[-y]'
option. However cpu index is internal concept and QEMU users
have to guess /reimplement qemu's logic/ to map it to
a concrete cpu socket/core/thread to make sane CPUs
placement across numa nodes.
This patch allows to map cpu objects to numa nodes using
the same properties as used for cpus with -device/device_add
(socket-id/core-id/thread-id/node-id).
At present valid properties/values to address CPUs could be
fetched using hotpluggable-cpus monitor/qmp command, it will
require user to start qemu twice when creating domain to fetch
possible CPUs for a machine type/-smp layout first and
then the second time with numa explicit mapping for actual
usage. The first step results could be saved and reused to
set/change mapping later as far as machine type/-smp stays
the same.
Proposed impl. supports exact and wildcard matching to
simplify CLI and allow to set mapping for a specific cpu
or group of cpu objects specified by matched properties.
For example:
# exact mapping x86
-numa cpu,node-id=x,socket-id=y,core-id=z,thread-id=n
# exact mapping SPAPR
-numa cpu,node-id=x,core-id=y
# wildcard mapping, all cpu objects that match socket-id=y
# are mapped to node-id=x
-numa cpu,node-id=x,socket-id=y
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
---
v2:
- use new NumaCpuOptions instead of CpuInstanceProperties in
NumaOptions, so that in future we could decouple both
if needed. (Eduardo Habkost <ehabkost@redhat.com>)
- clarify effect of NumaCpuOptions.node-id in qapi-schema.json
---
numa.c | 25 +++++++++++++++++++++++++
qapi-schema.json | 21 +++++++++++++++++++--
qemu-options.hx | 23 ++++++++++++++++++++++-
3 files changed, 66 insertions(+), 3 deletions(-)
diff --git a/numa.c b/numa.c
index 40e9f44..61521f5 100644
--- a/numa.c
+++ b/numa.c
@@ -227,6 +227,7 @@ static int parse_numa(void *opaque, QemuOpts *opts, Error **errp)
NumaOptions *object = NULL;
MachineState *ms = opaque;
Error *err = NULL;
+ CpuInstanceProperties cpu;
{
Visitor *v = opts_visitor_new(opts);
@@ -246,6 +247,30 @@ static int parse_numa(void *opaque, QemuOpts *opts, Error **errp)
}
nb_numa_nodes++;
break;
+ case NUMA_OPTIONS_TYPE_CPU:
+ if (!object->u.cpu.has_node_id) {
+ error_setg(&err, "Missing mandatory node-id property");
+ goto end;
+ }
+ if (!numa_info[object->u.cpu.node_id].present) {
+ error_setg(&err, "Invalid node-id=%" PRId64 ", NUMA node must be "
+ "defined with -numa node,nodeid=ID before it's used with "
+ "-numa cpu,node-id=ID", object->u.cpu.node_id);
+ goto end;
+ }
+
+ memset(&cpu, 0, sizeof(cpu));
+ cpu.has_node_id = object->u.cpu.has_node_id;
+ cpu.node_id = object->u.cpu.node_id;
+ cpu.has_socket_id = object->u.cpu.has_socket_id;
+ cpu.socket_id = object->u.cpu.socket_id;
+ cpu.has_core_id = object->u.cpu.has_core_id;
+ cpu.core_id = object->u.cpu.core_id;
+ cpu.has_thread_id = object->u.cpu.has_thread_id;
+ cpu.thread_id = object->u.cpu.thread_id;
+
+ machine_set_cpu_numa_node(ms, &cpu, &err);
+ break;
default:
abort();
}
diff --git a/qapi-schema.json b/qapi-schema.json
index 76d137d..5baf3a4 100644
--- a/qapi-schema.json
+++ b/qapi-schema.json
@@ -5680,10 +5680,12 @@
##
# @NumaOptionsType:
#
+# @cpu: property based CPU(s) to node mapping (Since: 2.10)
+#
# Since: 2.1
##
{ 'enum': 'NumaOptionsType',
- 'data': [ 'node' ] }
+ 'data': [ 'node', 'cpu' ] }
##
# @NumaOptions:
@@ -5696,7 +5698,8 @@
'base': { 'type': 'NumaOptionsType' },
'discriminator': 'type',
'data': {
- 'node': 'NumaNodeOptions' }}
+ 'node': 'NumaNodeOptions',
+ 'cpu': 'NumaCpuOptions' }}
##
# @NumaNodeOptions:
@@ -5725,6 +5728,20 @@
'*memdev': 'str' }}
##
+# @NumaCpuOptions:
+#
+# Option "-numa cpu" overrides default cpu to node mapping.
+# It accepts the same set of cpu properties as returned by
+# query-hotpluggable-cpus[].props, where node-id could be used to
+# override default node mapping.
+#
+# Since: 2.10
+##
+{ 'struct': 'NumaCpuOptions',
+ 'base': 'CpuInstanceProperties',
+ 'data' : {} }
+
+##
# @HostMemPolicy:
#
# Host memory policy types
diff --git a/qemu-options.hx b/qemu-options.hx
index 787b9c3..e88f534 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -139,13 +139,16 @@ ETEXI
DEF("numa", HAS_ARG, QEMU_OPTION_numa,
"-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node]\n"
- "-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node]\n", QEMU_ARCH_ALL)
+ "-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node]\n"
+ "-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]\n", QEMU_ARCH_ALL)
STEXI
@item -numa node[,mem=@var{size}][,cpus=@var{firstcpu}[-@var{lastcpu}]][,nodeid=@var{node}]
@itemx -numa node[,memdev=@var{id}][,cpus=@var{firstcpu}[-@var{lastcpu}]][,nodeid=@var{node}]
+@itemx -numa cpu,node-id=@var{node}[,socket-id=@var{x}][,core-id=@var{y}][,thread-id=@var{z}]
@findex -numa
Define a NUMA node and assign RAM and VCPUs to it.
+Legacy VCPU assignment uses @samp{cpus} option where
@var{firstcpu} and @var{lastcpu} are CPU indexes. Each
@samp{cpus} option represent a contiguous range of CPU indexes
(or a single VCPU if @var{lastcpu} is omitted). A non-contiguous
@@ -159,6 +162,24 @@ a NUMA node:
-numa node,cpus=0-2,cpus=5
@end example
+@samp{cpu} option is a new alternative to @samp{cpus} option
+which uses @samp{socket-id|core-id|thread-id} properties to assign
+CPU objects to a @var{node} using topology layout properties of CPU.
+The set of properties is machine specific, and depends on used
+machine type/@samp{smp} options. It could be queried with
+@samp{hotpluggable-cpus} monitor command.
+@samp{node-id} property specifies @var{node} to which CPU object
+will be assigned, it's required for @var{node} to be declared
+with @samp{node} option before it's used with @samp{cpu} option.
+
+For example:
+@example
+-M pc \
+-smp 1,sockets=2,maxcpus=2 \
+-numa node,nodeid=0 -numa node,nodeid=1 \
+-numa cpu,node-id=0,socket-id=0 -numa cpu,node-id=1,socket-id=1
+@end example
+
@samp{mem} assigns a given RAM amount to a node. @samp{memdev}
assigns RAM from a given memory backend device to a node. If
@samp{mem} and @samp{memdev} are omitted in all nodes, RAM is
--
2.7.4
next prev parent reply other threads:[~2017-05-03 12:58 UTC|newest]
Thread overview: 94+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-05-03 12:56 [Qemu-devel] [PATCH v2 00/24] numa: add '-numa cpu' option Igor Mammedov
2017-05-03 12:56 ` [Qemu-devel] [PATCH v2 01/24] tests: add CPUs to numa node mapping test Igor Mammedov
2017-05-03 14:32 ` Eduardo Habkost
2017-05-04 13:17 ` Eduardo Habkost
2017-05-03 12:56 ` [Qemu-devel] [PATCH v2 02/24] hw/arm/virt: extract mp-affinity calculation in separate function Igor Mammedov
2017-05-04 9:20 ` Andrew Jones
2017-05-04 13:21 ` Eduardo Habkost
2017-05-03 12:56 ` [Qemu-devel] [PATCH v2 03/24] hw/arm/virt: use machine->possible_cpus for storing possible topology info Igor Mammedov
2017-05-04 9:38 ` Andrew Jones
2017-05-04 12:55 ` Igor Mammedov
2017-05-04 13:16 ` Andrew Jones
2017-05-04 14:33 ` Igor Mammedov
2017-05-03 12:56 ` [Qemu-devel] [PATCH v2 04/24] hw/arm/virt: explicitly allocate cpu_index for cpus Igor Mammedov
2017-05-03 12:56 ` [Qemu-devel] [PATCH v2 05/24] numa: move source of default CPUs to NUMA node mapping into boards Igor Mammedov
2017-05-03 14:42 ` Eduardo Habkost
2017-05-03 15:57 ` Igor Mammedov
2017-05-04 7:32 ` David Gibson
2017-05-05 8:01 ` Igor Mammedov
2017-05-03 14:59 ` Eduardo Habkost
2017-05-03 16:14 ` Igor Mammedov
2017-05-03 15:13 ` Eduardo Habkost
2017-05-04 9:19 ` Igor Mammedov
2017-05-03 12:57 ` [Qemu-devel] [PATCH v2 06/24] spapr: add node-id property to sPAPR core Igor Mammedov
2017-05-03 14:46 ` Eduardo Habkost
2017-05-03 16:12 ` Igor Mammedov
2017-05-04 16:49 ` David Gibson
2017-05-05 8:04 ` Igor Mammedov
2017-05-03 12:57 ` [Qemu-devel] [PATCH v2 07/24] pc: add node-id property to CPU Igor Mammedov
2017-05-05 20:29 ` Eduardo Habkost
2017-05-09 13:14 ` Igor Mammedov
2017-05-03 12:57 ` [Qemu-devel] [PATCH v2 08/24] virt-arm: " Igor Mammedov
2017-05-04 9:57 ` Andrew Jones
2017-05-03 12:57 ` [Qemu-devel] [PATCH v2 09/24] numa: add check that board supports cpu_index to node mapping Igor Mammedov
2017-05-03 15:04 ` Eduardo Habkost
2017-05-03 16:19 ` Igor Mammedov
2017-05-03 17:31 ` Eduardo Habkost
2017-05-04 9:07 ` Igor Mammedov
2017-05-04 10:01 ` Andrew Jones
2017-05-04 16:51 ` David Gibson
2017-05-03 12:57 ` [Qemu-devel] [PATCH v2 10/24] numa: mirror cpu to node mapping in MachineState::possible_cpus Igor Mammedov
2017-05-03 15:20 ` Eduardo Habkost
2017-05-04 8:44 ` Igor Mammedov
2017-05-05 12:16 ` Igor Mammedov
2017-05-05 17:04 ` Eduardo Habkost
2017-05-05 20:00 ` Igor Mammedov
2017-05-04 11:40 ` Andrew Jones
2017-05-04 12:57 ` Igor Mammedov
2017-05-05 11:28 ` Igor Mammedov
2017-05-05 11:47 ` Andrew Jones
2017-05-03 12:57 ` [Qemu-devel] [PATCH v2 11/24] numa: do default mapping based on possible_cpus instead of node_cpu bitmaps Igor Mammedov
2017-05-04 11:45 ` Andrew Jones
2017-05-03 12:57 ` [Qemu-devel] [PATCH v2 12/24] numa: add numa_[has_]node_id() wrappers Igor Mammedov
2017-05-04 12:30 ` Andrew Jones
2017-05-05 1:45 ` David Gibson
2017-05-05 8:09 ` Igor Mammedov
2017-05-05 9:06 ` Andrew Jones
2017-05-05 17:12 ` Eduardo Habkost
2017-05-05 20:04 ` Igor Mammedov
2017-05-03 12:57 ` [Qemu-devel] [PATCH v2 13/24] pc: get numa node mapping from possible_cpus instead of numa_get_node_for_cpu() Igor Mammedov
2017-05-04 12:30 ` Andrew Jones
2017-05-03 12:57 ` [Qemu-devel] [PATCH v2 14/24] spapr: " Igor Mammedov
2017-05-05 3:15 ` David Gibson
2017-05-03 12:57 ` [Qemu-devel] [PATCH v2 15/24] virt-arm: " Igor Mammedov
2017-05-04 12:33 ` Andrew Jones
2017-05-03 12:57 ` [Qemu-devel] [PATCH v2 16/24] QMP: include CpuInstanceProperties into query_cpus output output Igor Mammedov
2017-05-03 12:57 ` [Qemu-devel] [PATCH v2 17/24] tests: numa: add case for QMP command query-cpus Igor Mammedov
2017-05-05 3:22 ` David Gibson
2017-05-03 12:57 ` [Qemu-devel] [PATCH v2 18/24] numa: remove no longer used numa_get_node_for_cpu() Igor Mammedov
2017-05-04 12:34 ` Andrew Jones
2017-05-03 12:57 ` [Qemu-devel] [PATCH v2 19/24] numa: remove no longer need numa_post_machine_init() Igor Mammedov
2017-05-04 12:35 ` Andrew Jones
2017-05-03 12:57 ` [Qemu-devel] [PATCH v2 20/24] machine: call machine init from wrapper Igor Mammedov
2017-05-04 12:43 ` Andrew Jones
2017-05-05 3:23 ` David Gibson
2017-05-08 14:31 ` Eduardo Habkost
2017-05-03 12:57 ` [Qemu-devel] [PATCH v2 21/24] numa: use possible_cpus for not mapped CPUs check Igor Mammedov
2017-05-04 12:43 ` Andrew Jones
2017-05-04 13:06 ` Igor Mammedov
2017-05-03 12:57 ` [Qemu-devel] [PATCH v2 22/24] numa: remove node_cpu bitmaps as they are no longer used Igor Mammedov
2017-05-04 12:45 ` Andrew Jones
2017-05-03 12:57 ` Igor Mammedov [this message]
2017-05-03 16:35 ` [Qemu-devel] [PATCH v2 23/24] numa: add '-numa cpu, ...' option for property based node mapping Eduardo Habkost
2017-05-03 16:39 ` Eric Blake
2017-05-03 17:38 ` Eduardo Habkost
2017-05-03 17:58 ` Eduardo Habkost
2017-05-04 9:52 ` Igor Mammedov
2017-05-08 5:40 ` David Gibson
2017-05-08 14:47 ` Eduardo Habkost
2017-05-09 15:58 ` Igor Mammedov
2017-05-10 1:15 ` David Gibson
2017-05-03 12:57 ` [Qemu-devel] [PATCH v2 24/24] tests: check -numa node, cpu=props_list usecase Igor Mammedov
2017-05-08 6:35 ` David Gibson
2017-05-04 13:41 ` [Qemu-devel] [PATCH v2 00/24] numa: add '-numa cpu' option Eduardo Habkost
2017-05-04 14:34 ` 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=1493816238-33120-24-git-send-email-imammedo@redhat.com \
--to=imammedo@redhat.com \
--cc=david@gibson.dropbear.id.au \
--cc=drjones@redhat.com \
--cc=eblake@redhat.com \
--cc=ehabkost@redhat.com \
--cc=pbonzini@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-arm@nongnu.org \
--cc=qemu-devel@nongnu.org \
--cc=qemu-ppc@nongnu.org \
--cc=zhaoshenglong@huawei.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).