Re: [mmotm:master 141/230] include/linux/kernel.h:744:28: note: in expansion of macro 'min'

[Michal Nazarewicz <mina86@mina86.com>]

On Fri, Jun 20 2014, Andrew Morton <akpm@linux-foundation.org> wrote:
> On Fri, 20 Jun 2014 17:19:55 +0200 Michal Nazarewicz <mina86@mina86.com> wrote:
>
>> On Fri, Jun 20 2014, Fengguang Wu <fengguang.wu@intel.com> wrote:
>> >>> include/linux/kernel.h:744:28: note: in expansion of macro 'min'
>> >     #define clamp(val, lo, hi) min(max(val, lo), hi)
>> >                                ^
>> >>> drivers/net/ethernet/intel/i40e/i40e_debugfs.c:1901:11: note: in expansion of macro 'clamp'
>> >       bytes = clamp(bytes, (u16)1024, (u16)I40E_MAX_AQ_BUF_SIZE);
>> >               ^
>> 
>> The obvious fix:
>> 
>> ----------- >8 --------------------------------------------------------------
>> diff --git a/include/linux/kernel.h b/include/linux/kernel.h
>> index 44649e0..149864b 100644
>> --- a/include/linux/kernel.h
>> +++ b/include/linux/kernel.h
>> @@ -719,8 +719,8 @@ static inline void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) { }
>>         (void) (&_max1 == &_max2);              \
>>         _max1 > _max2 ? _max1 : _max2; })
>>  
>> -#define min3(x, y, z) min(min(x, y), z)
>> -#define max3(x, y, z) max(max(x, y), z)
>> +#define min3(x, y, z) min((typeof(x))min(x, y), z)
>> +#define max3(x, y, z) max((typeof(x))max(x, y), z)
>
> I don't get it.  All the types are u16 so we should be good.
>
> What is the return type of
>
> 	_max1 > _max2 ? _max1 : _max2;

int… Since C promotes it.  (Or unsigned, I never remember, but I think
int if the possible values fit in signed int).

> when both _max1 and _max2 are u16?  Something other than u16 apparently
> - I never knew that.
>
> Maybe we should be fixing min() and max()?

This is also an option.  It would make min() and max() behave more like
a function which takes arguments of type T and returns value of type T.
Currently it behaves as a C operation which undergoes all the promotion
rules other arithmetic operations undergo (i.e. all types smaller than
int get promoted to int).  I don't have opinion either way.

> --- a/include/linux/kernel.h~a
> +++ a/include/linux/kernel.h
> @@ -711,13 +711,13 @@ static inline void ftrace_dump(enum ftra
>  	typeof(x) _min1 = (x);			\
>  	typeof(y) _min2 = (y);			\
>  	(void) (&_min1 == &_min2);		\
> -	_min1 < _min2 ? _min1 : _min2; })
> +	(typeof(x))(_min1 < _min2 ? _min1 : _min2); })
>  
>  #define max(x, y) ({				\
>  	typeof(x) _max1 = (x);			\
>  	typeof(y) _max2 = (y);			\
>  	(void) (&_max1 == &_max2);		\
> -	_max1 > _max2 ? _max1 : _max2; })
> +	(typeof(x))(_max1 > _max2 ? _max1 : _max2); })
>  
>  #define min3(x, y, z) min(min(x, y), z)
>  #define max3(x, y, z) max(max(x, y), z)
>

-- 
Best regards,                                         _     _
.o. | Liege of Serenely Enlightened Majesty of      o' \,=./ `o
..o | Computer Science,  Michał “mina86” Nazarewicz    (o o)
ooo +--<mpn@google.com>--<xmpp:mina86@jabber.org>--ooO--(_)--Ooo--

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>