Message-ID: <ad3c51a3-f46e-c559-7ad8-573564f63875@redhat.com>
Date:   Mon, 13 Feb 2023 19:26:41 +0100
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.7.1
Content-Language: en-US
To:     Jeremi Piotrowski <jpiotrowski@linux.microsoft.com>,
        Sean Christopherson <seanjc@google.com>
Cc:     kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
        Tianyu Lan <ltykernel@gmail.com>,
        "Michael Kelley (LINUX)" <mikelley@microsoft.com>
References: <43980946-7bbf-dcef-7e40-af904c456250@linux.microsoft.com>
 <Y+aQyFJt9Tn2PJnC@google.com>
 <9a046de1-8085-3df4-94cd-39bb893c8c9a@linux.microsoft.com>
 <88a89319-a71e-fa90-0dbb-00cf8a549380@redhat.com>
 <21b1ee26-dfd4-923d-72da-d8ded3dd819c@linux.microsoft.com>
From:   Paolo Bonzini <pbonzini@redhat.com>
Subject: Re: "KVM: x86/mmu: Overhaul TDP MMU zapping and flushing" breaks SVM
 on Hyper-V
In-Reply-To: <21b1ee26-dfd4-923d-72da-d8ded3dd819c@linux.microsoft.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 2/13/23 19:05, Jeremi Piotrowski wrote:
> So I looked at the ftrace (all kvm&kvmmu events + hyperv_nested_*
> events) I see the following: With tdp_mmu=0: kvm_exit sequence of
> kvm_mmu_prepare_zap_page hyperv_nested_flush_guest_mapping (always
> follows every sequence of kvm_mmu_prepare_zap_page) kvm_entry
> 
> With tdp_mmu=1 I see: kvm_mmu_prepare_zap_page and
> kvm_tdp_mmu_spte_changed events from a kworker context, but they are
> not followed by hyperv_nested_flush_guest_mapping. The only
> hyperv_nested_flush_guest_mapping events I see happen from the qemu
> process context.
> 
> Also the number of flush hypercalls is significantly lower: a 7second
> sequence through OVMF with tdp_mmu=0 produces ~270 flush hypercalls.
> In the traces with tdp_mmu=1 I now see max 3.
> 
> So this might be easier to diagnose than I thought: the
> HvCallFlushGuestPhysicalAddressSpace calls are missing now.

Can you check if KVM is reusing a nCR3 value?

If so, perhaps you can just add 
hyperv_flush_guest_mapping(__pa(root->spt), NULL) after 
kvm_tdp_mmu_get_vcpu_root_hpa's call to tdp_mmu_alloc_sp()?

Paolo