Re: [PATCH 1/2] rcu: introduce list_last_or_null_rcu

[Dan Streetman]

On Thu, May 28, 2015 at 5:30 PM, Dan Streetman <ddstreet@xxxxxxxx> wrote:
> On Thu, May 28, 2015 at 5:28 PM, Steven Rostedt <rostedt@xxxxxxxxxxx> wrote:
>> On Thu, 28 May 2015 17:19:40 -0400
>> Dan Streetman <ddstreet@xxxxxxxx> wrote:
>>
>>> >> oh, ok.  how do we do type-generic inline funcs?  return void*?
>>> >
>>> > I was hoping that you would tell me.  I use macros in that case.
>>>
>>> ha, if I ever get to the point where i think i know more than you,
>>> i'll let you know ;-)
>>
>> static inline struct list_head *
>> list_last_or_null_rcu(struct list_head *ptr)
>> {
>>         struct list_head *entry;
>>         struct list_head *last;
>>
>>         for (entry = list_next_rcu(ptr);
>>              entry != ptr; entry = list_next_rcu(entry))
>>                 last = __entry;
>>         if (last != ptr)
>>                 return last;
>>         return NULL;
>> }
>>
>> #define list_last_or_null_entry_rcu(ptr, type, member) \
>> ({ \
>>         struct list_head *__ptr = list_last_or_null_rcu(ptr); \
>>         __ptr ? list_entry_rcu(__ptr, type, member) : NULL; \
>> })
>
> well sure :-)
>
> however, the rcu list function currently doesn't include "entry" in
> the function name, but returns the list entry (not the list_head).
> that could be changed, but all callers would need to change, too.

to clarify, list_first_or_null_rcu() returns the list entry, not the
list_head; list_last_or_null_rcu() should likewise return the same
type.

git grep only shows 12 users of list_first_or_null_rcu(), so maybe
changing the name to list_first_entry_or_null_rcu() makes sense, which
would then use a new inline list_first_or_null_rcu() that returns
list_head instead of the entry type...

>
>>
>> -- Steve
--
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/