On 09/23/2015 03:35 AM, Juergen Gross wrote:
Depends on the hardware. On some AMD processors one socket covers
multiple NUMA nodes. This is the critical case. set_sched_topology()
will be called on those machines possibly multiple times when bringing
up additional cpus.
I'm asking because
On 09/23/2015 05:36 AM, Juergen Gross wrote:
> On 09/22/2015 06:22 PM, George Dunlap wrote:
>> On 09/22/2015 05:42 AM, Juergen Gross wrote:
>>> One other thing I just discovered: there are other consumers of the
>>> topology sibling masks (e.g. topology_sibling_cpumask()) as well.
>>>
>>> I think
On 09/23/2015 10:30 AM, Dario Faggioli wrote:
On Wed, 2015-09-23 at 06:36 +0200, Juergen Gross wrote:
On 09/22/2015 06:22 PM, George Dunlap wrote:
Juergen / Dario, could one of you summarize your two approaches,
and the
(alleged) advantages and disadvantages of each one?
Okay, I'll have a
On Wed, 2015-09-23 at 06:36 +0200, Juergen Gross wrote:
> On 09/22/2015 06:22 PM, George Dunlap wrote:
> > Juergen / Dario, could one of you summarize your two approaches,
> > and the
> > (alleged) advantages and disadvantages of each one?
>
> Okay, I'll have a try:
>
Thanks for this! ;-)
>
On 09/23/2015 09:24 AM, Dario Faggioli wrote:
On Mon, 2015-09-21 at 07:49 +0200, Juergen Gross wrote:
On 09/15/2015 06:50 PM, Dario Faggioli wrote:
On Thu, 2015-08-20 at 20:16 +0200, Juergen Groß wrote:
On 08/18/2015 05:55 PM, Dario Faggioli wrote:
Hey everyone,
So, as a followup of what we
On Mon, 2015-09-21 at 07:49 +0200, Juergen Gross wrote:
> On 09/15/2015 06:50 PM, Dario Faggioli wrote:
> > On Thu, 2015-08-20 at 20:16 +0200, Juergen Groß wrote:
> > > On 08/18/2015 05:55 PM, Dario Faggioli wrote:
> > > > Hey everyone,
> > > >
> > > > So, as a followup of what we were discussing
On 09/23/2015 05:36 AM, Juergen Gross wrote:
> On 09/22/2015 06:22 PM, George Dunlap wrote:
>> On 09/22/2015 05:42 AM, Juergen Gross wrote:
>>> One other thing I just discovered: there are other consumers of the
>>> topology sibling masks (e.g. topology_sibling_cpumask()) as well.
>>>
>>> I think
On 09/23/2015 03:35 AM, Juergen Gross wrote:
Depends on the hardware. On some AMD processors one socket covers
multiple NUMA nodes. This is the critical case. set_sched_topology()
will be called on those machines possibly multiple times when bringing
up additional cpus.
I'm asking because
On Mon, 2015-09-21 at 07:49 +0200, Juergen Gross wrote:
> On 09/15/2015 06:50 PM, Dario Faggioli wrote:
> > On Thu, 2015-08-20 at 20:16 +0200, Juergen Groß wrote:
> > > On 08/18/2015 05:55 PM, Dario Faggioli wrote:
> > > > Hey everyone,
> > > >
> > > > So, as a followup of what we were discussing
On 09/23/2015 09:24 AM, Dario Faggioli wrote:
On Mon, 2015-09-21 at 07:49 +0200, Juergen Gross wrote:
On 09/15/2015 06:50 PM, Dario Faggioli wrote:
On Thu, 2015-08-20 at 20:16 +0200, Juergen Groß wrote:
On 08/18/2015 05:55 PM, Dario Faggioli wrote:
Hey everyone,
So, as a followup of what we
On Wed, 2015-09-23 at 06:36 +0200, Juergen Gross wrote:
> On 09/22/2015 06:22 PM, George Dunlap wrote:
> > Juergen / Dario, could one of you summarize your two approaches,
> > and the
> > (alleged) advantages and disadvantages of each one?
>
> Okay, I'll have a try:
>
Thanks for this! ;-)
>
On 09/23/2015 10:30 AM, Dario Faggioli wrote:
On Wed, 2015-09-23 at 06:36 +0200, Juergen Gross wrote:
On 09/22/2015 06:22 PM, George Dunlap wrote:
Juergen / Dario, could one of you summarize your two approaches,
and the
(alleged) advantages and disadvantages of each one?
Okay, I'll have a
On 09/22/2015 06:22 PM, George Dunlap wrote:
On 09/22/2015 05:42 AM, Juergen Gross wrote:
One other thing I just discovered: there are other consumers of the
topology sibling masks (e.g. topology_sibling_cpumask()) as well.
I think we would want to avoid any optimizations based on those in
On 09/22/2015 05:42 AM, Juergen Gross wrote:
> One other thing I just discovered: there are other consumers of the
> topology sibling masks (e.g. topology_sibling_cpumask()) as well.
>
> I think we would want to avoid any optimizations based on those in
> drivers as well, not only in the
On 09/22/2015 05:42 AM, Juergen Gross wrote:
> One other thing I just discovered: there are other consumers of the
> topology sibling masks (e.g. topology_sibling_cpumask()) as well.
>
> I think we would want to avoid any optimizations based on those in
> drivers as well, not only in the
On 09/22/2015 06:22 PM, George Dunlap wrote:
On 09/22/2015 05:42 AM, Juergen Gross wrote:
One other thing I just discovered: there are other consumers of the
topology sibling masks (e.g. topology_sibling_cpumask()) as well.
I think we would want to avoid any optimizations based on those in
On 09/21/2015 07:49 AM, Juergen Gross wrote:
On 09/15/2015 06:50 PM, Dario Faggioli wrote:
On Thu, 2015-08-20 at 20:16 +0200, Juergen Groß wrote:
On 08/18/2015 05:55 PM, Dario Faggioli wrote:
Hey everyone,
So, as a followup of what we were discussing in this thread:
[Xen-devel] PV-vNUMA
On 09/21/2015 07:49 AM, Juergen Gross wrote:
On 09/15/2015 06:50 PM, Dario Faggioli wrote:
On Thu, 2015-08-20 at 20:16 +0200, Juergen Groß wrote:
On 08/18/2015 05:55 PM, Dario Faggioli wrote:
Hey everyone,
So, as a followup of what we were discussing in this thread:
[Xen-devel] PV-vNUMA
On 09/15/2015 06:50 PM, Dario Faggioli wrote:
On Thu, 2015-08-20 at 20:16 +0200, Juergen Groß wrote:
On 08/18/2015 05:55 PM, Dario Faggioli wrote:
Hey everyone,
So, as a followup of what we were discussing in this thread:
[Xen-devel] PV-vNUMA issue: topology is misinterpreted by the guest
On 09/15/2015 06:50 PM, Dario Faggioli wrote:
On Thu, 2015-08-20 at 20:16 +0200, Juergen Groß wrote:
On 08/18/2015 05:55 PM, Dario Faggioli wrote:
Hey everyone,
So, as a followup of what we were discussing in this thread:
[Xen-devel] PV-vNUMA issue: topology is misinterpreted by the guest
On Wed, 2015-09-02 at 16:30 +0200, Juergen Gross wrote:
> On 09/02/2015 04:08 PM, Boris Ostrovsky wrote:
> > On 09/02/2015 07:58 AM, Juergen Gross wrote:
> >> On 08/31/2015 06:12 PM, Boris Ostrovsky wrote:
> >>> If set_cpu_sibling_map()'s has_mp is false, wouldn't we effectively have
> >>> both
On Thu, 2015-08-20 at 20:16 +0200, Juergen Groß wrote:
> On 08/18/2015 05:55 PM, Dario Faggioli wrote:
> > Hey everyone,
> >
> > So, as a followup of what we were discussing in this thread:
> >
> > [Xen-devel] PV-vNUMA issue: topology is misinterpreted by the guest
> >
On Thu, 2015-08-27 at 11:24 +0100, George Dunlap wrote:
> On 08/18/2015 04:55 PM, Dario Faggioli wrote:
> > *** Intel(R) Xeon(R) X5650 @ 2.67GHz
> > *** pCPUs 48DOM0 vCPUS 16
> > *** RAM393138 MB DOM0 Memory 9955 MB
> > *** NUMA nodes 2
> >
On Wed, 2015-09-02 at 16:30 +0200, Juergen Gross wrote:
> On 09/02/2015 04:08 PM, Boris Ostrovsky wrote:
> > On 09/02/2015 07:58 AM, Juergen Gross wrote:
> >> On 08/31/2015 06:12 PM, Boris Ostrovsky wrote:
> >>> If set_cpu_sibling_map()'s has_mp is false, wouldn't we effectively have
> >>> both
On Thu, 2015-08-20 at 20:16 +0200, Juergen Groß wrote:
> On 08/18/2015 05:55 PM, Dario Faggioli wrote:
> > Hey everyone,
> >
> > So, as a followup of what we were discussing in this thread:
> >
> > [Xen-devel] PV-vNUMA issue: topology is misinterpreted by the guest
> >
On Thu, 2015-08-27 at 11:24 +0100, George Dunlap wrote:
> On 08/18/2015 04:55 PM, Dario Faggioli wrote:
> > *** Intel(R) Xeon(R) X5650 @ 2.67GHz
> > *** pCPUs 48DOM0 vCPUS 16
> > *** RAM393138 MB DOM0 Memory 9955 MB
> > *** NUMA nodes 2
> >
On Thu, Aug 27, 2015 at 11:24 AM, George Dunlap
wrote:
> On 08/18/2015 04:55 PM, Dario Faggioli wrote:
>> Hey everyone,
>>
>> So, as a followup of what we were discussing in this thread:
>>
>> [Xen-devel] PV-vNUMA issue: topology is misinterpreted by the guest
>>
On Thu, Aug 27, 2015 at 11:24 AM, George Dunlap
george.dun...@citrix.com wrote:
On 08/18/2015 04:55 PM, Dario Faggioli wrote:
Hey everyone,
So, as a followup of what we were discussing in this thread:
[Xen-devel] PV-vNUMA issue: topology is misinterpreted by the guest
28 matches
Mail list logo
