Re: [Xen-devel] [PATCH] x86/mmuext: don't allow copying/clearing non-RAM pages

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 21.06.17 at 12:51, <andrew.cooper3@xxxxxxxxxx> wrote:
> On 21/06/17 11:10, Jan Beulich wrote:
>> The two operations really aren't meant for anything else.
>>
>> Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
> 
> Reviewed-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>, however...
> 
>>
>> --- a/xen/arch/x86/mm.c
>> +++ b/xen/arch/x86/mm.c
>> @@ -3229,6 +3229,7 @@ long do_mmuext_op(
>>          switch ( op.cmd )
>>          {
>>              struct page_info *page;
>> +            p2m_type_t p2mt;
>>  
>>          case MMUEXT_PIN_L1_TABLE:
>>              type = PGT_l1_page_table;
>> @@ -3528,7 +3529,12 @@ long do_mmuext_op(
>>          }
>>  
>>          case MMUEXT_CLEAR_PAGE:
>> -            page = get_page_from_gfn(pg_owner, op.arg1.mfn, NULL, 
>> P2M_ALLOC);
>> +            page = get_page_from_gfn(pg_owner, op.arg1.mfn, &p2mt, 
>> P2M_ALLOC);
>> +            if ( unlikely(p2mt != p2m_ram_rw) && page )
> 
> ... it would seem more natural to have the null pointer check before the
> p2mt check.

Since the checks are independent, the order doesn't really matter,
and with that it seemed better to put the unlikely() first (to get the
other check off the fast path).

Jan


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel