Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
To:     Alexander Graf <graf@amazon.com>,
        Siddharth Chandrasekaran <sidcha@amazon.de>,
        Sean Christopherson <seanjc@google.com>,
        Vitaly Kuznetsov <vkuznets@redhat.com>,
        Wanpeng Li <wanpengli@tencent.com>,
        Jim Mattson <jmattson@google.com>,
        Joerg Roedel <joro@8bytes.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
        x86@kernel.org, "H. Peter Anvin" <hpa@zytor.com>
Cc:     Evgeny Iakovlev <eyakovl@amazon.de>, Liran Alon <liran@amazon.com>,
        Ioannis Aslanidis <iaslan@amazon.de>, kvm@vger.kernel.org,
        linux-kernel@vger.kernel.org
References: <20210423090333.21910-1-sidcha@amazon.de>
 <224d266e-aea3-3b4b-ec25-7bb120c4d98a@amazon.com>
 <213887af-78b8-03ad-b3f9-c2194cb27b13@redhat.com>
 <ded8db53-0e58-654a-fff2-de536bcbc961@amazon.com>
From:   Paolo Bonzini <pbonzini@redhat.com>
Subject: Re: [PATCH] KVM: hyper-v: Add new exit reason HYPERV_OVERLAY
Message-ID: <45888d26-89d2-dba6-41cb-de2d58cd5345@redhat.com>
Date:   Fri, 23 Apr 2021 12:15:01 +0200
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.7.0
MIME-Version: 1.0
In-Reply-To: <ded8db53-0e58-654a-fff2-de536bcbc961@amazon.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Precedence: bulk

On 23/04/21 11:58, Alexander Graf wrote:
>> In theory userspace doesn't know how KVM wishes to implement the
>> hypercall page, especially if Xen hypercalls are enabled as well.
> 
> I'm not sure I agree with that sentiment :). User space is the one that 
> sets the xen compat mode. All we need to do is declare the ORing as part 
> of the KVM ABI. Which we effectively are doing already, because it's 
> part of the ABI to the guest, no?

Good point.  But it may change in the future based on KVM_ENABLE_CAP or 
whatever, and duplicating code between userspace and kernel is ugly.  We 
already have too many unwritten conventions around CPUID, MSRs, get/set 
state ioctls, etc.

That said, this definitely tilts the balance against adding an ioctl to 
write the hypercall page contents.  Userspace can either use the 
KVM_SET_MSR or assemble it on its own, and one of the two should be okay.

Paolo

>>
>> But userspace has two plausible ways to get the page contents:
>>
>> 1) add a ioctl to write the hypercall page contents to an arbitrary
>> userspace address
>>
>> 2) after userspace updates the memslots to add the overlay page at the
>> right place, use KVM_SET_MSR from userspace (which won't be filtered
>> because it's host initiated)
>>
>> The second has the advantage of not needing any new code at all, but
>> it's a bit more ugly.
> 
> The more of all of that hyper-v code we can have live in user space, the 
> happier I am :).