Re: [Xen-devel] [PATCH 1/1] xen:include:asm-x86:e820: just covert nr_map with unsigned int

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 08.08.14 at 09:54, <tiejun.chen@xxxxxxxxx> wrote:
> On 2014/8/8 15:38, Jan Beulich wrote:
>>>>> On 08.08.14 at 09:26, <tiejun.chen@xxxxxxxxx> wrote:
>>> On 2014/8/8 15:15, Jan Beulich wrote:
>>>>>>> On 08.08.14 at 09:00, <tiejun.chen@xxxxxxxxx> wrote:
>>>>> This shouldn't ever be negative.
>>>>
>>>> This means nothing for the actual change, and makes me imply you did
>>>> the change without checking it actually _is_ never negative in any of
>>>> the uses.
>>>>
>>>> Also please cut down your patch title prefixes (not just here) to
>>>> something reasonable.
>>>>
>>>
>>> What about this?
>>>
>>> xen:e820map: just convert nr_map with unsigned int
>>
>> x86: make struct e820map's nr_map unsigned
> 
> Okay.
> 
>>
>>> nr_map is used to represent the amount of e820entry in e820 so it
>>> shouldn't be negative.
>>
>> You still say "shouldn't" instead of "isn't", making me still imply you
>> didn't verify that fact. I.e. I'd be fine with a suitable variant of the
> 
> Maybe I'm poor to describe that main point here.
> 
> x86: make struct e820map's nr_map unsigned
> 
> nr_map is used to represent the amount of e820entry in e820 so its
> not always negative.

"not always negative"?

Based on your initial one, "This is never negative" would be fine,
_provided you verified_ this to be the case.

Jan


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