Re: [PATCH 0/4] KVM: add KVM_CREATE_VM2 to allow dynamic kvm->vcpus array

From: "Radim Krčmář" <rkrcmar@redhat.com>
To: David Hildenbrand <david@redhat.com>
Cc: linux-kernel@vger.kernel.org, kvm@vger.kernel.org,
	Christoffer Dall <cdall@linaro.org>,
	Marc Zyngier <marc.zyngier@arm.com>,
	Paolo Bonzini <pbonzini@redhat.com>,
	Christian Borntraeger <borntraeger@de.ibm.com>,
	Cornelia Huck <cornelia.huck@de.ibm.com>,
	James Hogan <james.hogan@imgtec.com>,
	Paul Mackerras <paulus@ozlabs.org>,
	Alexander Graf <agraf@suse.com>
Subject: Re: [PATCH 0/4] KVM: add KVM_CREATE_VM2 to allow dynamic kvm->vcpus array
Date: Mon, 24 Apr 2017 19:03:34 +0200	[thread overview]
Message-ID: <20170424170333.GC5713@potion> (raw)
In-Reply-To: <bb0cda85-6222-5cb7-718f-bbc70cd6297c@redhat.com>

2017-04-18 13:11+0200, David Hildenbrand:
> On 13.04.2017 22:19, Radim Krčmář wrote:
>> The basic idea is to let userspace provide the desired maximal number of
>> VCPUs and allocate only necessary memory for them.
>> 
>> The goal is to freeze KVM_MAX_VCPUS at its current level and only increase the
> 
> KVM_MAX_VCPUS might still increase e.g. if hw support for more VCPUs is
> comming.

This patch wanted to make KVM_MAX_VCPUS just a compatibility option for
old userspaces and not looked at in new ones, so we wouldn't have to
touch it from now on.

>> new KVM_MAX_CONFIGURABLE_VCPUS, probably directly to INT_MAX/KVM_VCPU_ID, so we
>> don't have to worry about it for a while.
>> 
>> PPC should be interested in this as they set KVM_MAX_VCPUS to NR_CPUS
>> and probably waste few pages for every guest this way.
> 
> As we just store pointers, this should be a maximum of 4 pages for ppc
> (4k pages). Is this really worth yet another VM creation ioctl? Is there
> not a nicer way to handle this internally?
> 
> An alternative might be to simply realloc the array when it reaches a
> certain size (on VCPU creation, maybe protecting the pointer via rcu).
> But not sure if something like that could work.

Good point.  I'll cover it in the next email.

Thanks.