Re: [Xen-devel] [PATCH v8] x86/emulate: Send vm_event from emulate

[Alexandru Stefan ISAILA <aisaila@xxxxxxxxxxxxxxx>]

On 06.09.2019 18:46, Jan Beulich wrote:
> On 03.09.2019 16:01, Alexandru Stefan ISAILA wrote:
>> A/D bit writes (on page walks) can be considered benign by an introspection
>> agent, so receiving vm_events for them is a pessimization. We try here to
>> optimize by filtering these events out.
>> Currently, we are fully emulating the instruction at RIP when the hardware 
>> sees
>> an EPT fault with npfec.kind != npfec_kind_with_gla. This is, however,
>> incorrect, because the instruction at RIP might legitimately cause an
>> EPT fault of its own while accessing a _different_ page from the original 
>> one,
>> where A/D were set.
>> The solution is to perform the whole emulation, while ignoring EPT 
>> restrictions
>> for the walk part, and taking them into account for the "actual" emulating of
>> the instruction at RIP. When we send out a vm_event, we don't want the 
>> emulation
>> to complete, since in that case we won't be able to veto whatever it is 
>> doing.
>> That would mean that we can't actually prevent any malicious activity, 
>> instead
>> we'd only be able to report on it.
>> When we see a "send-vm_event" case while emulating, we need to first send the
>> event out and then suspend the emulation (return X86EMUL_RETRY).
>> After the emulation stops we'll call hvm_vm_event_do_resume() again after the
>> introspection agent treats the event and resumes the guest. There, the
>> instruction at RIP will be fully emulated (with the EPT ignored) if the
>> introspection application allows it, and the guest will continue to run past
>> the instruction.
>>
>> A common example is if the hardware exits because of an EPT fault caused by a
>> page walk, p2m_mem_access_check() decides if it is going to send a vm_event.
>> If the vm_event was sent and it would be treated so it runs the instruction
>> at RIP, that instruction might also hit a protected page and provoke a 
>> vm_event.
>>
>> Now if npfec.kind != npfec_kind_with_gla and 
>> d->arch.monitor.inguest_pagefault_disabled
>> is true then we are in the page walk case and we can do this emulation 
>> optimization
>> and emulate the page walk while ignoring the EPT, but don't ignore the EPT 
>> for the
>> emulation of the actual instruction.
> 
> Instead of comparing against npfec_kind_with_gla, wouldn't you
> better compare against npfec_kind_in_gpt to positively identify
> page table accesses by the guest? Both VMX and SVM may leave the
> field at npfec_kind_unknown after all.

I think it is safer to check against npfec_kind_in_gpt. I have to run 
some tests with this but it looks good from the code.

> 
>> In the first case we would have 2 EPT events, in the second case we would 
>> have
>> 1 EPT event if the instruction at the RIP triggers an EPT event.
>>
>> We use hvmemul_map_linear_addr() to intercept r/w access and
>> __hvm_copy() to intercept exec access.
>>
>> hvm_emulate_send_vm_event() can return false if there was no violation,
>> if there was an error from monitor_traps() or p2m_get_mem_access().
>> Returning false if p2m_get_mem_access() fails if the entry was not found
>> in the EPT in which case it is unrestricted.
> 
> I'm afraid I can't interpret what the second sentence is supposed to
> clarify.

It wants to clarify why returning false, if p2m_get_mem_access() fails, 
is needed in hvm_emulate_send_vm_event(). It is sort of unclear and it 
has to be reformulated like

"Returning false if p2m_get_mem_access() fails is needed because the EPT 
entry will have rwx memory access rights."

> 
>> @@ -531,6 +533,72 @@ static int hvmemul_do_mmio_addr(paddr_t mmio_gpa,
>>       return hvmemul_do_io_addr(1, mmio_gpa, reps, size, dir, df, ram_gpa);
>>   }
>>   
>> +/*
>> + * Send memory access vm_events based on pfec. Returns true if the event was
>> + * sent and false for p2m_get_mem_access() error, no violation and event 
>> send
>> + * error. Assumes the caller will check arch.vm_event->send_event.
>> + *
>> + * NOTE: p2m_get_mem_access() can fail if the entry was not found in the EPT
>> + * (in which case access to it is unrestricted, so no violations can occur).
>> + * In this cases it is fine to continue the emulation.
>> + */
>> +bool hvm_emulate_send_vm_event(unsigned long gla, gfn_t gfn,
>> +                               uint32_t pfec)
>> +{
>> +    xenmem_access_t access;
>> +    vm_event_request_t req = {};
>> +    paddr_t gpa = (gfn_to_gaddr(gfn) | (gla & ~PAGE_MASK));
>> +
>> +    ASSERT(current->arch.vm_event->send_event);
>> +
>> +    current->arch.vm_event->send_event = false;
>> +
>> +    if ( p2m_get_mem_access(current->domain, gfn, &access,
>> +                            altp2m_vcpu_idx(current)) != 0 )
>> +        return false;
>> +
>> +    switch ( access )
>> +    {
>> +    case XENMEM_access_x:
>> +    case XENMEM_access_rx:
>> +        if ( pfec & PFEC_write_access )
>> +            req.u.mem_access.flags = MEM_ACCESS_R | MEM_ACCESS_W;
>> +        break;
>> +
>> +    case XENMEM_access_w:
>> +    case XENMEM_access_rw:
>> +        if ( pfec & PFEC_insn_fetch )
>> +            req.u.mem_access.flags = MEM_ACCESS_X;
>> +        break;
>> +
>> +    case XENMEM_access_r:
>> +    case XENMEM_access_n:
>> +        if ( pfec & PFEC_write_access )
>> +            req.u.mem_access.flags |= MEM_ACCESS_R | MEM_ACCESS_W;
>> +        if ( pfec & PFEC_insn_fetch )
>> +            req.u.mem_access.flags |= MEM_ACCESS_X;
>> +        break;
>> +
>> +    case XENMEM_access_wx:
>> +    case XENMEM_access_rwx:
>> +    case XENMEM_access_rx2rw:
>> +    case XENMEM_access_n2rwx:
>> +    case XENMEM_access_default:
>> +        break;
>> +    }
>> +
>> +    if ( !req.u.mem_access.flags )
>> +        return false; /* no violation */
>> +
>> +    req.reason = VM_EVENT_REASON_MEM_ACCESS;
>> +    req.u.mem_access.gfn = gfn_x(gfn);
>> +    req.u.mem_access.flags |= MEM_ACCESS_FAULT_WITH_GLA | 
>> MEM_ACCESS_GLA_VALID;
>> +    req.u.mem_access.gla = gla;
>> +    req.u.mem_access.offset = gpa & ~PAGE_MASK;
>> +
>> +    return monitor_traps(current, true, &req) >= 0;
>> +}
> 
> Since it's non-static anyway, wouldn't this better live in
> hvm/vm_event.c, putting it under VM EVENT maintainership? In any
> event I don't feel qualified eventually giving an ack for this
> function.

I agree, I will move it in vm_event.c.

> 
>> @@ -544,10 +612,11 @@ static void *hvmemul_map_linear_addr(
>>       struct hvm_emulate_ctxt *hvmemul_ctxt)
>>   {
>>       struct vcpu *curr = current;
>> -    void *err, *mapping;
>> +    void *err = NULL, *mapping;
>>       unsigned int nr_frames = ((linear + bytes - !!bytes) >> PAGE_SHIFT) -
>>           (linear >> PAGE_SHIFT) + 1;
>>       unsigned int i;
>> +    gfn_t gfn;
>>   
>>       /*
>>        * mfn points to the next free slot.  All used slots have a page 
>> reference
>> @@ -582,7 +651,7 @@ static void *hvmemul_map_linear_addr(
>>           ASSERT(mfn_x(*mfn) == 0);
>>   
>>           res = hvm_translate_get_page(curr, addr, true, pfec,
>> -                                     &pfinfo, &page, NULL, &p2mt);
>> +                                     &pfinfo, &page, &gfn, &p2mt);
>>   
>>           switch ( res )
>>           {
>> @@ -596,7 +665,6 @@ static void *hvmemul_map_linear_addr(
>>               goto out;
>>   
>>           case HVMTRANS_bad_gfn_to_mfn:
>> -            err = NULL;
>>               goto out;
>>   
>>           case HVMTRANS_gfn_paged_out:
>> @@ -619,14 +687,19 @@ static void *hvmemul_map_linear_addr(
>>               }
>>   
>>               if ( p2mt == p2m_ioreq_server )
>> -            {
>> -                err = NULL;
>>                   goto out;
>> -            }
>>   
>>               ASSERT(p2mt == p2m_ram_logdirty || !p2m_is_readonly(p2mt));
>>           }
>> +
>> +        if ( curr->arch.vm_event &&
> 
> Just like you have in __hvm_copy(), may I ask for unlikely() here?

I will add this.

> 
>> +             curr->arch.vm_event->send_event &&
>> +             hvm_emulate_send_vm_event(addr, gfn, pfec) )
>> +            err = ERR_PTR(~X86EMUL_RETRY);
> 
> In the description you say this takes care of r/w, and __hvm_copy()
> takes care of exec, but the function we're in doesn't get called
> on the read path (but it may be down the road). Additionally the
> function now also gets called from hvmemul_cache_op(), which I'm
> not sure you actually want to send events for.

I was aiming for hvmemul_write/hvmemul_rmw but if a access error occurs 
when the option (curr->arch.vm_event->send_event) is active, IMO it is 
ok to send a vm_event. Doesn't matter where it originated.

> 
>>       }
>> +    /* Check if eny vm_event was sent */
> 
> "any" and please add blank line ahead of your addition.

I will correct this.

> 
>> +    if ( err )
>> +        goto out;
> 
> And wait - why does this sit after the loop? Is that a re-basing
> mistake from when you had put on top of my uncommitted patch?

This is done to skip the mapping part down the line. If there is an 
error then we have to return _it_ and not the mapping.

> 
>> --- a/xen/arch/x86/hvm/hvm.c
>> +++ b/xen/arch/x86/hvm/hvm.c
>> @@ -3224,6 +3224,14 @@ static enum hvm_translation_result __hvm_copy(
>>               return HVMTRANS_bad_gfn_to_mfn;
>>           }
>>   
>> +        if ( unlikely(v->arch.vm_event) &&
>> +             v->arch.vm_event->send_event &&
>> +             hvm_emulate_send_vm_event(addr, gfn, pfec) )
>> +        {
>> +            put_page(page);
>> +            return HVMTRANS_gfn_paged_out;
>> +        }
> 
> I'm pretty sure I had previously asked for there to either be
> a change to the return value, or for there to be a comment
> justifying the (apparent) abuse of the one used.

Sorry for not clarifying this in the last version. I looked now and I 
missed to respond. I return  HVMTRANS_gfn_paged_out so that later it 
will be treated and return X86EMUL_RETRY.
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel