linux-kernel - Re: [PATCH v4 3/3] KVM: nVMX: Emulate EPTP switching for the L1 hypervisor

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

Date:   Thu, 13 Jul 2017 13:08:49 -0400
From:   Bandan Das <bsd@...hat.com>
To:     David Hildenbrand <david@...hat.com>
Cc:     kvm@...r.kernel.org, pbonzini@...hat.com,
        linux-kernel@...r.kernel.org,
        Radim Krčmář <rkrcmar@...hat.com>
Subject: Re: [PATCH v4 3/3] KVM: nVMX: Emulate EPTP switching for the L1 hypervisor

David Hildenbrand <david@...hat.com> writes:

>>>> +	/*
>>>> +	 * If the (L2) guest does a vmfunc to the currently
>>>> +	 * active ept pointer, we don't have to do anything else
>>>> +	 */
>>>> +	if (vmcs12->ept_pointer != address) {
>>>> +		if (address >> cpuid_maxphyaddr(vcpu) ||
>>>> +		    !IS_ALIGNED(address, 4096))
>>>
>>> Couldn't the pfn still be invalid and make kvm_mmu_reload() fail?
>>> (triggering a KVM_REQ_TRIPLE_FAULT)
>> 
>> If there's a triple fault, I think it's a good idea to inject it
>> back. Basically, there's no need to take care of damage control
>> that L1 is intentionally doing.
>
> I quickly rushed over the massive amount of comments. Sounds like you'll
> be preparing a v5. Would be great if you could add some comments that
> were the result of this discussion (for parts that are not that obvious
> - triple faults) - thanks!

Will do. Basically, we agreed that we don't need to do anything with mmu_reload() faillures
because the invalid eptp that mmu_unload will write to root_hpa will result in an ept
violation.

Bandan