From mboxrd@z Thu Jan  1 00:00:00 1970
From: Radim =?utf-8?B?S3LEjW3DocWZ?= <rkrcmar@redhat.com>
Subject: Re: [PATCH v4 2/2] KVM: x86: KVM_CAP_SYNC_REGS
Date: Fri, 2 Feb 2018 22:04:35 +0100
Message-ID: <20180202210434.GC27896@flask>
References: <20180201000336.154170-1-hofsass@google.com>
 <20180201000336.154170-3-hofsass@google.com>
 <ac1118b1-6f44-be85-f4c4-7c9e2db2b750@redhat.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: David Hildenbrand <david@redhat.com>, kvm@vger.kernel.org,
        pbonzini@redhat.com, jmattson@google.com
To: Ken Hofsass <hofsass@google.com>
Return-path: <kvm-owner@vger.kernel.org>
Received: from mx1.redhat.com ([209.132.183.28]:7520 "EHLO mx1.redhat.com"
        rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
        id S1751878AbeBBVFB (ORCPT <rfc822;kvm@vger.kernel.org>);
        Fri, 2 Feb 2018 16:05:01 -0500
Content-Disposition: inline
In-Reply-To: <ac1118b1-6f44-be85-f4c4-7c9e2db2b750@redhat.com>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

2018-02-01 14:05+0100, David Hildenbrand:
> > @@ -7351,6 +7363,17 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
> >  		goto out;
> >  	}
> >  
> > +	if (vcpu->run->kvm_valid_regs) {
> 
> You can check directly, without this if. (and not sure if the extra
> function is needed here).

Right, it's unlikely that we'll be doing more checks on kvm_valid_regs,
so I've moved it here and replaced the block with

  if (vcpu->run->kvm_valid_regs & ~KVM_SYNC_X86_VALID_FIELDS) {
  	r = -EINVAL;
  	goto out;
  }

You can see the result in kvm/queue.

Btw. the userspace doesn't know what the next userspace exit is going to
be, so it the purpose of kvm_valid_regs to avoid needless writes of
registers on old userspaces?

Thanks.

> > +		r = check_valid_regs(vcpu);
> > +		if (r != 0)
> > +			goto out;
> > +	}