linux-kernel - Re: [PATCH v6 05/38] KVM: x86: hyper-v: Handle HVCALL_FLUSH_VIRTUAL_ADDRESS

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] [day] [month] [year] [list]

Message-ID: <87leu0k9ds.fsf@redhat.com>
Date:   Mon, 13 Jun 2022 14:58:39 +0200
From:   Vitaly Kuznetsov <vkuznets@...hat.com>
To:     Maxim Levitsky <mlevitsk@...hat.com>, kvm@...r.kernel.org,
        Paolo Bonzini <pbonzini@...hat.com>
Cc:     Sean Christopherson <seanjc@...gle.com>,
        Wanpeng Li <wanpengli@...cent.com>,
        Jim Mattson <jmattson@...gle.com>,
        Michael Kelley <mikelley@...rosoft.com>,
        Siddharth Chandrasekaran <sidcha@...zon.de>,
        Yuan Yao <yuan.yao@...ux.intel.com>,
        linux-hyperv@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v6 05/38] KVM: x86: hyper-v: Handle
 HVCALL_FLUSH_VIRTUAL_ADDRESS_LIST{,EX} calls gently

Maxim Levitsky <mlevitsk@...hat.com> writes:

...

>>  
>>  void kvm_hv_vcpu_flush_tlb(struct kvm_vcpu *vcpu)
>>  {
>>         struct kvm_vcpu_hv_tlb_flush_fifo *tlb_flush_fifo;
>>         struct kvm_vcpu_hv *hv_vcpu = to_hv_vcpu(vcpu);
>> +       u64 entries[KVM_HV_TLB_FLUSH_FIFO_SIZE];
>> +       int i, j, count;
>> +       gva_t gva;
>>  
>> -       kvm_vcpu_flush_tlb_guest(vcpu);
>> -
>> -       if (!hv_vcpu)
>> +       if (!tdp_enabled || !hv_vcpu) {
> I haven't noticed that in the review I did back then, but
> any reason why !tdp_enabled?

This follows the logic in kvm_vcpu_flush_tlb_guest():

	if (!tdp_enabled) {
		/*
		 * A TLB flush on behalf of the guest is equivalent to
		 * INVPCID(all), toggling CR4.PGE, etc., which requires
		 * a forced sync of the shadow page tables.  Ensure all the
		 * roots are synced and the guest TLB in hardware is clean.
		 */
		kvm_mmu_sync_roots(vcpu);
		kvm_mmu_sync_prev_roots(vcpu);
	}

and as !tdp_enabled should be a rare debug or special case I decided to
take the shortcut and not drag any of this logic into hyperv emulation
code.

-- 
Vitaly