On Tue, May 09, 2017 at 05:58:29PM +0200, Igor Mammedov wrote:
> On Mon, 8 May 2017 15:40:04 +1000
> David Gibson wrote:
>
> > On Wed, May 03, 2017 at 02:57:17PM +0200, Igor Mammedov wrote:
> > > legacy cpu to node mapping is using cpu index values to map
> > > VCPU to node with help of '-numa no
On Mon, 8 May 2017 15:40:04 +1000
David Gibson wrote:
> On Wed, May 03, 2017 at 02:57:17PM +0200, Igor Mammedov wrote:
> > 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 concep
On Mon, May 08, 2017 at 03:40:04PM +1000, David Gibson wrote:
> On Wed, May 03, 2017 at 02:57:17PM +0200, Igor Mammedov wrote:
> > 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
On Wed, May 03, 2017 at 02:57:17PM +0200, Igor Mammedov wrote:
> 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 m
On Wed, 3 May 2017 14:58:02 -0300
Eduardo Habkost wrote:
> On Wed, May 03, 2017 at 02:38:41PM -0300, Eduardo Habkost wrote:
> > On Wed, May 03, 2017 at 11:39:10AM -0500, Eric Blake wrote:
> > > On 05/03/2017 11:35 AM, Eduardo Habkost wrote:
> > >
> > > >> +
> > > >> +memset(&cpu, 0,
On Wed, May 03, 2017 at 02:38:41PM -0300, Eduardo Habkost wrote:
> On Wed, May 03, 2017 at 11:39:10AM -0500, Eric Blake wrote:
> > On 05/03/2017 11:35 AM, Eduardo Habkost wrote:
> >
> > >> +
> > >> +memset(&cpu, 0, sizeof(cpu));
> > >> +cpu.has_node_id = object->u.cpu.has_node_id;
On Wed, May 03, 2017 at 11:39:10AM -0500, Eric Blake wrote:
> On 05/03/2017 11:35 AM, Eduardo Habkost wrote:
>
> >> +
> >> +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_sock
On 05/03/2017 11:35 AM, Eduardo Habkost wrote:
>> +
>> +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
On Wed, May 03, 2017 at 02:57:17PM +0200, Igor Mammedov wrote:
[...]
> 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;
> Mach
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
10 matches
Mail list logo
