Re: [PATCH RFC V4 1/5] xen: Emulate with no writes

["Jan Beulich" <JBeulich@suse.com>]

>>> On 04.08.14 at 16:25, <rcojocaru@bitdefender.com> wrote:
> On 08/04/2014 05:09 PM, Jan Beulich wrote:
>>>>> On 04.08.14 at 13:30, <rcojocaru@bitdefender.com> wrote:
>>> --- a/xen/arch/x86/hvm/emulate.c
>>> +++ b/xen/arch/x86/hvm/emulate.c
>>> @@ -688,6 +688,51 @@ static int hvmemul_write(
>>>      return X86EMUL_OKAY;
>>>  }
>>>  
>>> +static int hvmemul_write_discard(
>>> +    enum x86_segment seg,
>>> +    unsigned long offset,
>>> +    void *p_data,
>>> +    unsigned int bytes,
>>> +    struct x86_emulate_ctxt *ctxt)
>>> +{
>>> +    /* Discarding the write. */
>>> +    return X86EMUL_OKAY;
>>> +}
>> 
>> While this one is okay, ...
>> 
>>> +static int hvmemul_rep_ins_discard(
>>> +    uint16_t src_port,
>>> +    enum x86_segment dst_seg,
>>> +    unsigned long dst_offset,
>>> +    unsigned int bytes_per_rep,
>>> +    unsigned long *reps,
>>> +    struct x86_emulate_ctxt *ctxt)
>>> +{
>>> +    return X86EMUL_OKAY;
>>> +}
>>> +
>>> +static int hvmemul_rep_movs_discard(
>>> +   enum x86_segment src_seg,
>>> +   unsigned long src_offset,
>>> +   enum x86_segment dst_seg,
>>> +   unsigned long dst_offset,
>>> +   unsigned int bytes_per_rep,
>>> +   unsigned long *reps,
>>> +   struct x86_emulate_ctxt *ctxt)
>>> +{
>>> +    return X86EMUL_OKAY;
>>> +}
>> 
>> ... these don't seem to be: I don't think you can just drop the other
>> half of the operation (i.e. the port or MMIO read).
> 
> It's been suggested here:
> 
> http://lists.xen.org/archives/html/xen-devel/2014-07/msg03088.html 
> 
> that we should use modified versions of the rep_ins, rep_movs and
> cmpxchg handlers if we want to make sure absolutely no writes will
> happen. Then again, perhaps the modification were supposed to be more
> subtle than just doing nothing in the handler?

That's what I'm trying to hint at.

Jan