On 16.08.2017 21:40, Radim Kr?m?? wrote:

The goal is to increase KVM_MAX_VCPUS without worrying about memory
impact of many small guests.

This is a second out of three major "dynamic" options:
 1) size vcpu array at VM creation time
 2) resize vcpu array when new VCPUs are created
 3) use a lockless list/tree for VCPUs

The disadvantage of (1) is its requirement on userspace changes and
limited flexibility because userspace must provide the maximal count on
start.  The main advantage is that kvm->vcpus will work like it does
now.  It has been posted as "[PATCH 0/4] KVM: add KVM_CREATE_VM2 to
allow dynamic kvm->vcpus array",
http://www.mail-archive.com/linux-kernel at vger.kernel.org/msg1377285.html

The main problem of (2), this series, is that we cannot extend the array
in place and therefore require some kind of protection when moving it.
RCU seems best, but it makes the code slower and harder to deal with.
The main advantage is that we do not need userspace changes.

The third option wasn't explored yet.  It would solve the ugly
kvm_for_each_vcpu() of (2), but kvm_get_vcpu() would become linear.
(We could mitigate it by having list of vcpu arrays and A lockless
 sequentially growing "tree" would be logarithmic and not that much more
 complicated to implement.)

That sounds interesting but also too complicated.

Which option do you think is the best?

I actually think the RCU variant doesn't look bad at all. Execution time
should be ok.

As Alex said, doubling the size every time we run out of space could be
done.

I clearly favor a solution that doesn't require user space changes.

-- 

Thanks,

David