On Thu, Nov 06, 2014 at 05:09:35PM +0100, Andrew Jones wrote:
> smp_parse has a couple problems. First, it should use max_cpus,
> not smp_cpus when calculating missing topology information.
> Conversely, if maxcpus is not input, then the topology should
> dictate max_cpus, as the topology may suppo
On Fri, Nov 07, 2014 at 01:23:12PM +0100, Andrew Jones wrote:
> On Fri, Nov 07, 2014 at 10:16:06AM -0200, Eduardo Habkost wrote:
> > On Fri, Nov 07, 2014 at 12:21:26PM +0100, Andrew Jones wrote:
> > > On Fri, Nov 07, 2014 at 10:52:31AM +0100, Andrew Jones wrote:
[...]
> > > After talking with Igor,
On Fri, Nov 07, 2014 at 10:16:06AM -0200, Eduardo Habkost wrote:
> On Fri, Nov 07, 2014 at 12:21:26PM +0100, Andrew Jones wrote:
> > On Fri, Nov 07, 2014 at 10:52:31AM +0100, Andrew Jones wrote:
> > > On Fri, Nov 07, 2014 at 10:40:14AM +0100, Paolo Bonzini wrote:
> > > >
> > > >
> > > > On 07/11/
On Fri, Nov 07, 2014 at 12:21:26PM +0100, Andrew Jones wrote:
> On Fri, Nov 07, 2014 at 10:52:31AM +0100, Andrew Jones wrote:
> > On Fri, Nov 07, 2014 at 10:40:14AM +0100, Paolo Bonzini wrote:
> > >
> > >
> > > On 07/11/2014 10:29, Andrew Jones wrote:
> > > >> > I think this would cause too many
On Fri, Nov 07, 2014 at 10:22:39AM +0100, Andrew Jones wrote:
> On Thu, Nov 06, 2014 at 05:17:44PM -0200, Eduardo Habkost wrote:
> > On Thu, Nov 06, 2014 at 05:09:35PM +0100, Andrew Jones wrote:
> > > smp_parse has a couple problems. First, it should use max_cpus,
> > > not smp_cpus when calculatin
On Fri, Nov 07, 2014 at 10:52:31AM +0100, Andrew Jones wrote:
> On Fri, Nov 07, 2014 at 10:40:14AM +0100, Paolo Bonzini wrote:
> >
> >
> > On 07/11/2014 10:29, Andrew Jones wrote:
> > >> > I think this would cause too many failures in the wild. Perhaps error
> > >> > out if it is lower, and warn
On Fri, Nov 07, 2014 at 10:40:14AM +0100, Paolo Bonzini wrote:
>
>
> On 07/11/2014 10:29, Andrew Jones wrote:
> >> > I think this would cause too many failures in the wild. Perhaps error
> >> > out if it is lower, and warn if sockets * cores * threads > max_cpus
> >> > since we actually allow ho
On 07/11/2014 10:29, Andrew Jones wrote:
>> > I think this would cause too many failures in the wild. Perhaps error
>> > out if it is lower, and warn if sockets * cores * threads > max_cpus
>> > since we actually allow hot-plug a thread at a time?
> We'd still have more failures if we choose to
On Thu, Nov 06, 2014 at 11:11:30PM +0100, Paolo Bonzini wrote:
>
>
> On 06/11/2014 17:09, Andrew Jones wrote:
> > +if (sockets * cores * threads != max_cpus) {
> > +fprintf(stderr, "cpu topology: "
> > +"sockets (%u) * cores (%u) * threads (%u) != max_cpus
On Thu, Nov 06, 2014 at 05:17:44PM -0200, Eduardo Habkost wrote:
> On Thu, Nov 06, 2014 at 05:09:35PM +0100, Andrew Jones wrote:
> > smp_parse has a couple problems. First, it should use max_cpus,
> > not smp_cpus when calculating missing topology information.
> > Conversely, if maxcpus is not inpu
On 06/11/2014 17:09, Andrew Jones wrote:
> +if (sockets * cores * threads != max_cpus) {
> +fprintf(stderr, "cpu topology: "
> +"sockets (%u) * cores (%u) * threads (%u) != max_cpus
> (%u)\n",
> +sockets, cores, threads, max_cpus);
> +
On Thu, Nov 06, 2014 at 05:09:35PM +0100, Andrew Jones wrote:
> smp_parse has a couple problems. First, it should use max_cpus,
> not smp_cpus when calculating missing topology information.
> Conversely, if maxcpus is not input, then the topology should
> dictate max_cpus, as the topology may suppo
smp_parse has a couple problems. First, it should use max_cpus,
not smp_cpus when calculating missing topology information.
Conversely, if maxcpus is not input, then the topology should
dictate max_cpus, as the topology may support more than the
input smp_cpus number. Second, smp_parse shouldn't si
13 matches
Mail list logo
