Re: [Xen-devel] [PATCH v2 5/9] x86/pagewalk: Helpers for reserved bit handling

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 23.03.17 at 18:35, <andrew.cooper3@xxxxxxxxxx> wrote:
> On 23/03/17 17:12, Tim Deegan wrote:
>> At 17:02 +0000 on 23 Mar (1490288548), Andrew Cooper wrote:
>>> On 23/03/17 16:55, Tim Deegan wrote:
>>>> At 16:31 +0000 on 16 Mar (1489681899), Andrew Cooper wrote:
>>>>> Some bits are unconditionally reserved in pagetable entries, or reserved
>>>>> because of alignment restrictions.  Other bits are reserved because of 
> control
>>>>> register configuration.
>>>>>
>>>>> Introduce helpers which take an individual vcpu and guest pagetable 
>>>>> entry, 
> and
>>>>> calculates whether any reserved bits are set.
>>>>>
>>>>> While here, add a couple of newlines to aid readability.
>>>>>
>>>>> Signed-off-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
>>>> Reviewed-by: Tim Deegan <tim@xxxxxxx>, although:
>>>>
>>>>> +/* Mask covering the reserved bits from superpage alignment. */
>>>>> +#define SUPERPAGE_RSVD(bit)                                             \
>>>>> +    (((1ULL << (bit)) - 1) & ~(_PAGE_PSE_PAT | (_PAGE_PSE_PAT - 1)))
>>>> I think this will be wrong if we ever get l4 superpages, as the mask
>>>> is only 32 bits wide.
>>> What is 32 bits wide?  1ULL should cause everything else to be suitably
>>> promoted, no?
>>
>> ~(_PAGE_PSE_PAT | (_PAGE_PSE_PAT - 1)) == 0xffffe000.  Promotion comes
>> too late.
> 
> Oh - good point.  Best fix this right now.
> 
> Would you prefer ~((uint64_t)_PAGE_PSE_PAT | (_PAGE_PSE_PAT - 1)) or
> ~(_PAGE_PSE_PAT | (_PAGE_PSE_PAT - 1) | 0ULL)

Why not simply tag the suffix on a constant already there, i.e.

~(_PAGE_PSE_PAT | (_PAGE_PSE_PAT - 1ULL))

(And btw, we commonly use only UL these days.)

Jan


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