Re: [PATCH v1 3/6] x86: Enable Supervisor Mode Execution Prevention (SMAP) for Xen

[Andrew Cooper <andrew.cooper3@citrix.com>]

On 15/04/14 12:48, Jan Beulich wrote:
>>>> On 15.04.14 at 12:32, <andrew.cooper3@citrix.com> wrote:
>> On 15/04/14 14:01, Feng Wu wrote:
>>> index e9c2c51..09c974d 100644
>>> --- a/xen/arch/x86/setup.c
>>> +++ b/xen/arch/x86/setup.c
>>> @@ -61,6 +61,10 @@ integer_param("maxcpus", max_cpus);
>>>  static bool_t __initdata disable_smep;
>>>  invbool_param("smep", disable_smep);
>>>  
>>> +/* smap: Enable/disable Supervisor Mode Access Prevention (default on). */
>>> +static bool_t __initdata disable_smap;
>>> +invbool_param("smap", disable_smap);
>>> +
>> Please use a positive boolean rather than negative.  Convention is also
>> to prefix the variable with opt_
>>
>> static bool_t __initdata opt_smap = 1;
>> boolean_param("smap", opt_smap);
> Hmm, I'd go for consistency with SMAP here as a first step.
> Converting both may later be an option, but I'm not really sure
> why you think the invbool_param() is bad.
>
> Jan
>

Not invbool_param() per say, but with negative booleans in general.  It
is just unnecessary extra cognitive load when following code.  This is
admittedly a mild example.

~Andrew