Re: [Qemu-devel] Regression (?) due to c4177479 ('spapr: make sure RMA is in first mode of first memory node')

[Alexey Kardashevskiy]

On 04/19/2014 05:58 AM, Benjamin Herrenschmidt wrote:
> On Fri, 2014-04-18 at 23:29 +1000, Alexey Kardashevskiy wrote:
>> Which would be what in the current QEMU?
>>
>> If I read the current QEMU code correctly, NUMA nodes and device tree
>> nodes
>> are the same thing for SPAPR now, see spapr_populate_memory()
>> function.
> 
> So the trick here would be to find a way to make the node "number"
> exposed to the guest be different from the node "index" in qemu...
> 
> More work...

I already have in plan to fix non-power-of-two sized memory nodes so I will
this too.

What exactly is the point in having NUMA memoryless nodes now? We do not
support memory hotplug yet and balloon is not memory hotplug at all.

And I still fail to see how the patch is wrong. May be the idea of having
DT memory nodes the same thing as NUMA memory nodes is not the best one,
but the patch is not changing that.


-- 
Alexey