Re: [PATCH V4 5/5] xen: Handle resumed instruction based on previous mem_event reply

[Razvan Cojocaru <rcojocaru@bitdefender.com>]

On 09/08/14 18:50, Jan Beulich wrote:
>>>> On 05.09.14 at 12:01, <rcojocaru@bitdefender.com> wrote:
>> --- a/xen/arch/x86/hvm/vmx/vmx.c
>> +++ b/xen/arch/x86/hvm/vmx/vmx.c
>> @@ -1695,6 +1695,21 @@ static void vmx_enable_msr_exit_interception(struct domain *d)
>>                                           MSR_TYPE_W);
>>  }
>>  
>> +static bool_t vmx_exited_by_nested_pagefault(void)
>> +{
>> +    unsigned long exit_qualification, exit_reason;
>> +
>> +    __vmread(VM_EXIT_REASON, &exit_reason);
>> +    ASSERT(exit_reason == EXIT_REASON_EPT_VIOLATION);
>> +
>> +    __vmread(EXIT_QUALIFICATION, &exit_qualification);
>> +
>> +    if ( (exit_qualification & EPT_GLA_FAULT) == 0 )
>> +        return 0;
>> +
>> +    return 1;
>> +}
> 
> So I said a number of times that I think this new function can go
> away with the recent changes introducing struct npfec, making
> the interesting bit of information already available to
> p2m_mem_access_check() if I'm not mistaken. And the previous
> excuse of that code not being committed yet is no longer valid.

I'll change that and test the code first thing tomorrow.


Thanks,
Razvan Cojocaru