[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-ppc] [PATCH v2 05/24] numa: move source of default CPUs to NUM
From: |
David Gibson |
Subject: |
Re: [Qemu-ppc] [PATCH v2 05/24] numa: move source of default CPUs to NUMA node mapping into boards |
Date: |
Thu, 4 May 2017 17:32:13 +1000 |
User-agent: |
Mutt/1.8.0 (2017-02-23) |
On Wed, May 03, 2017 at 11:42:40AM -0300, Eduardo Habkost wrote:
> On Wed, May 03, 2017 at 02:56:59PM +0200, Igor Mammedov wrote:
> > Originally CPU threads were by default assigned in
> > round-robin fashion. However it was causing issues in
> > guest since CPU threads from the same socket/core could
> > be placed on different NUMA nodes.
> > Commit fb43b73b (pc: fix default VCPU to NUMA node mapping)
> > fixed it by grouping threads within a socket on the same node
> > introducing cpu_index_to_socket_id() callback and commit
> > 20bb648d (spapr: Fix default NUMA node allocation for threads)
> > reused callback to fix similar issues for SPAPR machine
> > even though socket doesn't make much sense there.
> >
> > As result QEMU ended up having 3 default distribution rules
> > used by 3 targets /virt-arm, spapr, pc/.
> >
> > In effort of moving NUMA mapping for CPUs into possible_cpus,
> > generalize default mapping in numa.c by making boards decide
> > on default mapping and let them explicitly tell generic
> > numa code to which node a CPU thread belongs to by replacing
> > cpu_index_to_socket_id() with @cpu_index_to_instance_props()
> > which provides default node_id assigned by board to specified
> > cpu_index.
> >
> > Signed-off-by: Igor Mammedov <address@hidden>
>
> Reviewed-by: Eduardo Habkost <address@hidden>
>
> Just two extra comments below:
>
> [...]
> > +static CpuInstanceProperties
> > +virt_cpu_index_to_props(MachineState *ms, unsigned cpu_index)
> > +{
> > + MachineClass *mc = MACHINE_GET_CLASS(ms);
> > + const CPUArchIdList *possible_cpus = mc->possible_cpu_arch_ids(ms);
> > +
> > + assert(cpu_index < possible_cpus->len);
> > + return possible_cpus->cpus[cpu_index].props;;
> > +}
> > +
> [...]
> > +static CpuInstanceProperties
> > +pc_cpu_index_to_props(MachineState *ms, unsigned cpu_index)
> > {
> > + MachineClass *mc = MACHINE_GET_CLASS(ms);
> > + const CPUArchIdList *possible_cpus = mc->possible_cpu_arch_ids(ms);
> > +
> > + assert(cpu_index < possible_cpus->len);
> > + return possible_cpus->cpus[cpu_index].props;;
> > }
>
> The fact that these two implementations look exactly the same
> made me wonder:
>
> 1) Why this isn't the default implementation;
> 2) Why exactly spapr needs a different implementation.
>
> Then I noticed that there's nothing in the common machine code
> that specifies that possible_cpus->cpus[] is indexed by
> cpu_index. This means it is indeed safer to require each machine
> to provide its own cpu_index_to_props implementation than having
> a default implementation that can unexpectedly break (e.g. if
> granularity at possible_cpus is not at VCPU/thread level).
>
> I would still like to have an abstraction that wouldn't require
> writing machine-specific code (e.g. cpu_index ranges to
> possible_cpus like David suggested), but that's for a follow-up
> series.
Yeah, that similarity bothered me to, but like you I realised the
problem is that spapr simply doesn't have the same granularity of
information as x86 and ARM - there's only one entry per core for PAPR
instead of one per thread.
So, we do need a machine specific mapping of cpu_index to location
properties, which is what the callback is for.
It does occur to me that another way of accomplishing that would be
for possible_cpu_arch_ids() to create a cpu_index->props mapping as a
simple array ofpointers, in addition to the list of possiblee props
structures.
Not sure if that would end up looking better or not.
> [...]
> > +static CpuInstanceProperties
> > +spapr_cpu_index_to_props(MachineState *machine, unsigned cpu_index)
> > {
> > + CPUArchId *core_slot;
> > + MachineClass *mc = MACHINE_GET_CLASS(machine);
> > +
> > + /* make sure possible_cpu are intialized */
> > + mc->possible_cpu_arch_ids(machine);
> > + core_slot = spapr_find_cpu_slot(machine, cpu_index, NULL);
> > + assert(core_slot);
> > + return core_slot->props;
> > }
>
> If you need to submit v3, maybe a comment here explaining why
> spapr needs a different cpu_index_to_props implementation would
> be helpful. I took a while to figure it out.
>
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
signature.asc
Description: PGP signature
- Re: [Qemu-ppc] [PATCH v2 02/24] hw/arm/virt: extract mp-affinity calculation in separate function, (continued)
[Qemu-ppc] [PATCH v2 04/24] hw/arm/virt: explicitly allocate cpu_index for cpus, Igor Mammedov, 2017/05/03
[Qemu-ppc] [PATCH v2 05/24] numa: move source of default CPUs to NUMA node mapping into boards, Igor Mammedov, 2017/05/03
Re: [Qemu-ppc] [PATCH v2 05/24] numa: move source of default CPUs to NUMA node mapping into boards, Eduardo Habkost, 2017/05/03
Re: [Qemu-ppc] [PATCH v2 05/24] numa: move source of default CPUs to NUMA node mapping into boards, Eduardo Habkost, 2017/05/03
[Qemu-ppc] [PATCH v2 07/24] pc: add node-id property to CPU, Igor Mammedov, 2017/05/03
[Qemu-ppc] [PATCH v2 08/24] virt-arm: add node-id property to CPU, Igor Mammedov, 2017/05/03