linux-kernel - Re: [PATCH v5 2/2] kvm: introduce KVM_MAX_VCPU

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20160504195030.4125b1ce@bahia.huguette.org>
Date:	Wed, 4 May 2016 19:50:30 +0200
From:	Greg Kurz <gkurz@...ux.vnet.ibm.com>
To:	Radim Krčmář <rkrcmar@...hat.com>
Cc:	Paolo Bonzini <pbonzini@...hat.com>, james.hogan@...tec.com,
	mingo@...hat.com, linux-mips@...ux-mips.org, kvm@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	David Hildenbrand <dahi@...ux.vnet.ibm.com>,
	qemu-ppc@...gnu.org, Cornelia Huck <cornelia.huck@...ibm.com>,
	Paul Mackerras <paulus@...ba.org>,
	David Gibson <david@...son.dropbear.id.au>
Subject: Re: [PATCH v5 2/2] kvm: introduce KVM_MAX_VCPU_ID

On Wed, 4 May 2016 18:45:34 +0200
Radim Krčmář <rkrcmar@...hat.com> wrote:

> 2016-05-03 06:52+0200, Greg Kurz:
> > The KVM_MAX_VCPUS define provides the maximum number of vCPUs per guest, and
> > also the upper limit for vCPU ids. This is okay for all archs except PowerPC
> > which can have higher ids, depending on the cpu/core/thread topology. In the
> > worst case (single threaded guest, host with 8 threads per core), it limits
> > the maximum number of vCPUS to KVM_MAX_VCPUS / 8.
> > 
> > This patch separates the vCPU numbering from the total number of vCPUs, with
> > the introduction of KVM_MAX_VCPU_ID, as the maximal valid value for vCPU ids
> > plus one.
> > 
> > The corresponding KVM_CAP_MAX_VCPU_ID allows userspace to validate vCPU ids
> > before passing them to KVM_CREATE_VCPU.
> > 
> > Only PowerPC gets unlimited vCPU ids for the moment. This patch doesn't
> > change anything for other archs.
> > 
> > Suggested-by: Radim Krcmar <rkrcmar@...hat.com>
> > Signed-off-by: Greg Kurz <gkurz@...ux.vnet.ibm.com>
> > ---
> > diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
> > @@ -2272,7 +2272,7 @@ static int kvm_vm_ioctl_create_vcpu(struct kvm *kvm, u32 id)
> >  	int r;
> >  	struct kvm_vcpu *vcpu;
> >  
> > -	if (id >= KVM_MAX_VCPUS)
> > +	if (id >= KVM_MAX_VCPU_ID)
> >  		return -EINVAL;  
> 
> book3s_hv will currently fail with vcpu_id above threads_per_subcore *
> KVM_MAX_VCORES, so userspace cannot use KVM_CAP_MAX_VCPU_ID to limit
> vcpu_id ... 

You're right, I guess powerpc should return threads_per_subcore * KVM_MAX_VCORES
instead of INT_MAX then.

> I thought the check for vcpu_id would move to arch-specific
> code, like the previous version did, to simplify implementation of a
> dynamic limit.
> 
> The dynamic limit was too complicated to be worth it?
> (This version is ok too.)
> 

Actually no, it can be as simple as this in arch/powerpc/include/asm/kvm_host.h:

#include <asm/cputhreads.h>
#define KVM_MAX_VCPU_ID		(threads_per_subcore * KVM_MAX_VCORES)

Thanks !

--
Greg