Re: [PATCH 1/6] oom: Make find_lock_task_mm() sparse-aware

[KOSAKI Motohiro]

2012/2/6 Greg KH <gregkh@xxxxxxxxxxxxxxxxxxx>:
> On Mon, Feb 06, 2012 at 10:59:09PM +0400, Anton Vorontsov wrote:
>> On Mon, Feb 06, 2012 at 08:35:42AM -0800, Greg KH wrote:
>> [...]
>> > > -extern struct task_struct *find_lock_task_mm(struct task_struct *p);
>> > > +extern struct task_struct *__find_lock_task_mm(struct task_struct *p);
>> > > +
>> > > +#define find_lock_task_mm(p)                                             \
>> > > +({                                                                       \
>> > > + struct task_struct *__ret;                                      \
>> > > +                                                                 \
>> > > + __cond_lock(&(p)->alloc_lock, __ret = __find_lock_task_mm(p));  \
>> > > + __ret;                                                          \
>> > > +})
>> >
>> > Please use the proper "do...while" style thing here for multi-line,
>> > complex #defines like the rest of the kernel does so that you don't end
>> > up debugging horrible problems later.
>>
>> Unfortunately this isn't possible in this case. Unlike '({})' GCC
>> extension, do-while statement does not evaluate to a value, i.e.
>> 'x = do { 123; } while (0);' is illegal.
>
> Ah, you are right, my bad, sorry about that.
>
> greg k-h

Some __cond_lock() caller are inline functions. Is this bad? inline function
is always recommended than macros.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/