linux-kernel - Re: [PATCH 01/16] KVM: x86/mmu: Exit to userspace with -EFAULT if private fault hits emulation

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: Tue, 12 Mar 2024 10:59:25 +0800
From: Binbin Wu <binbin.wu@...ux.intel.com>
To: "Gupta, Pankaj" <pankaj.gupta@....com>,
 Sean Christopherson <seanjc@...gle.com>, Paolo Bonzini <pbonzini@...hat.com>
Cc: kvm@...r.kernel.org, linux-kernel@...r.kernel.org,
 Yan Zhao <yan.y.zhao@...el.com>, Isaku Yamahata <isaku.yamahata@...el.com>,
 Michael Roth <michael.roth@....com>, Yu Zhang <yu.c.zhang@...ux.intel.com>,
 Chao Peng <chao.p.peng@...ux.intel.com>, Fuad Tabba <tabba@...gle.com>,
 David Matlack <dmatlack@...gle.com>
Subject: Re: [PATCH 01/16] KVM: x86/mmu: Exit to userspace with -EFAULT if
 private fault hits emulation



On 3/7/2024 8:52 PM, Gupta, Pankaj wrote:
>> Exit to userspace with -EFAULT / KVM_EXIT_MEMORY_FAULT if a private 
>> fault
>> triggers emulation of any kind, as KVM doesn't currently support 
>> emulating
>> access to guest private memory.  Practically speaking, private faults 
>> and
>> emulation are already mutually exclusive, but there are edge cases upon
>> edge cases where KVM can return RET_PF_EMULATE, and adding one last 
>> check
>
> edge cases upon edge cases?
>
> Just curious about the details of the edge cases scenarios?

Same question out of curiosity.

>
>> to harden against weird, unexpected combinations is inexpensive.
>>
>> Suggested-by: Yan Zhao <yan.y.zhao@...el.com>
>> Signed-off-by: Sean Christopherson <seanjc@...gle.com>
>> ---
>>   arch/x86/kvm/mmu/mmu.c          |  8 --------
>>   arch/x86/kvm/mmu/mmu_internal.h | 13 +++++++++++++
>>   2 files changed, 13 insertions(+), 8 deletions(-)
>>
>> diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
>> index e4cc7f764980..e2fd74e06ff8 100644
>> --- a/arch/x86/kvm/mmu/mmu.c
>> +++ b/arch/x86/kvm/mmu/mmu.c
>> @@ -4309,14 +4309,6 @@ static inline u8 kvm_max_level_for_order(int 
>> order)
>>       return PG_LEVEL_4K;
>>   }
>>   -static void kvm_mmu_prepare_memory_fault_exit(struct kvm_vcpu *vcpu,
>> -                          struct kvm_page_fault *fault)
>> -{
>> -    kvm_prepare_memory_fault_exit(vcpu, fault->gfn << PAGE_SHIFT,
>> -                      PAGE_SIZE, fault->write, fault->exec,
>> -                      fault->is_private);
>> -}
>> -
>>   static int kvm_faultin_pfn_private(struct kvm_vcpu *vcpu,
>>                      struct kvm_page_fault *fault)
>>   {
>> diff --git a/arch/x86/kvm/mmu/mmu_internal.h 
>> b/arch/x86/kvm/mmu/mmu_internal.h
>> index 0669a8a668ca..0eea6c5a824d 100644
>> --- a/arch/x86/kvm/mmu/mmu_internal.h
>> +++ b/arch/x86/kvm/mmu/mmu_internal.h
>> @@ -279,6 +279,14 @@ enum {
>>       RET_PF_SPURIOUS,
>>   };
>>   +static inline void kvm_mmu_prepare_memory_fault_exit(struct 
>> kvm_vcpu *vcpu,
>> +                             struct kvm_page_fault *fault)
>> +{
>> +    kvm_prepare_memory_fault_exit(vcpu, fault->gfn << PAGE_SHIFT,
>> +                      PAGE_SIZE, fault->write, fault->exec,
>> +                      fault->is_private);
>> +}
>> +
>>   static inline int kvm_mmu_do_page_fault(struct kvm_vcpu *vcpu, 
>> gpa_t cr2_or_gpa,
>>                       u32 err, bool prefetch, int *emulation_type)
>>   {
>> @@ -320,6 +328,11 @@ static inline int kvm_mmu_do_page_fault(struct 
>> kvm_vcpu *vcpu, gpa_t cr2_or_gpa,
>>       else
>>           r = vcpu->arch.mmu->page_fault(vcpu, &fault);
>>   +    if (r == RET_PF_EMULATE && fault.is_private) {
>> +        kvm_mmu_prepare_memory_fault_exit(vcpu, &fault);
>> +        return -EFAULT;
>> +    }
>> +
>>       if (fault.write_fault_to_shadow_pgtable && emulation_type)
>>           *emulation_type |= EMULTYPE_WRITE_PF_TO_SP;
>
>