DMARC-Filter: OpenDMARC Filter v1.3.2 mx1.redhat.com D6E024A71F
Date: Fri, 4 Aug 2017 23:56:13 +0200
From: Radim =?utf-8?B?S3LEjW3DocWZ?= <rkrcmar@redhat.com>
To: Bandan Das <bsd@redhat.com>
Cc: kvm@vger.kernel.org, pbonzini@redhat.com, david@redhat.com,
        jmattson@google.com, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v7 0/3] Expose VMFUNC to the nested hypervisor
Message-ID: <20170804215612.GC2119@potion>
References: <20170803195443.24911-1-bsd@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20170803195443.24911-1-bsd@redhat.com>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 2304
Lines: 68

Queued, thanks.

2017-08-03 15:54-0400, Bandan Das:
> v7:
>  3/3:
>   Fix check for AD
>   Use kvm_vcpu_read_guest_page()
> 
> v6:
>  https://lkml.org/lkml/2017/8/1/1015
>  3/3:
>    Fix check for memory type in address
>    Change check function name as requested in the review
>    Move setting of mmu->ept_ad to after calling mmu_unload
>    and also reset base_role.ad_disabled appropriately
>    Replace IS_ALIGN with page_address_valid()
> 
> v5:
>  https://lkml.org/lkml/2017/7/28/621
>  1/3 and 2/3 are unchanged but some changes in 3/3. I left
>  the mmu_load failure path untouched because I am not sure what's
>  the right thing to do here.
>  3/3:
>     Move the eptp switching logic to a different function
>     Add check for EPTP_ADDRESS in check_vmentry_prereq
>     Add check for validity of ept pointer
>     Check if AD bit is set and set ept_ad
>     Add TODO item about mmu_unload failure
> 
> v4:
>  https://lkml.org/lkml/2017/7/10/705
>  2/3:  Use WARN_ONCE to avoid logging dos
> 
> v3:
>  https://lkml.org/lkml/2017/7/10/684
>  3/3: Add missing nested_release_page_clean() and check the
>  eptp as mentioned in SDM 24.6.14
> 
> v2:
>  https://lkml.org/lkml/2017/7/6/813
>  1/3: Patch to enable vmfunc on the host but cause a #UD if
>       L1 tries to use it directly. (new)
>  2/3: Expose vmfunc to the nested hypervisor, but no vm functions
>       are exposed and L0 emulates a vmfunc vmexit to L1. 
>  3/3: Force a vmfunc vmexit when L2 tries to use vmfunc and emulate
>       eptp switching. Unconditionally expose EPTP switching to the
>       L1 hypervisor since L0 fakes eptp switching via a mmu reload.
> 
> These patches expose eptp switching/vmfunc to the nested hypervisor.
> vmfunc is enabled in the secondary controls for the host and is
> exposed to the nested hypervisor. However, if the nested hypervisor
> decides to use eptp switching, L0 emulates it.
> 
> v1:
>  https://lkml.org/lkml/2017/6/29/958
> 
> Bandan Das (3):
>   KVM: vmx: Enable VMFUNCs
>   KVM: nVMX: Enable VMFUNC for the L1 hypervisor
>   KVM: nVMX: Emulate EPTP switching for the L1 hypervisor
> 
>  arch/x86/include/asm/vmx.h |   9 +++
>  arch/x86/kvm/vmx.c         | 185 ++++++++++++++++++++++++++++++++++++++++++++-
>  2 files changed, 192 insertions(+), 2 deletions(-)
> 
> -- 
> 2.9.4
>