Re: [ckrm-tech] [PATCH 2/7] UBC: core (structures, API)

[Kirill Korotaev <dev@sw.ru>]

Matt Helsley wrote:
> On Fri, 2006-08-18 at 11:35 +0400, Andrey Savochkin wrote:
> 
>>On Thu, Aug 17, 2006 at 10:31:37PM -0700, Andrew Morton wrote:
>>
>>>On Thu, 17 Aug 2006 15:53:40 +0400
>>>Kirill Korotaev <dev@sw.ru> wrote:
>>>
>>>
>>>>>>+struct user_beancounter
>>>>>>+{
>>>>>>+	atomic_t		ub_refcount;
>>>>>>+	spinlock_t		ub_lock;
>>>>>>+	uid_t			ub_uid;
>>>>>
>>>>>
>>>>>Why uid?  Will it be possible to club processes belonging to different
>>>>>users to same bean counter.
>>>>
>>>>oh, its a misname. Should be ub_id. it is ID of user_beancounter
>>>>and has nothing to do with user id.
>>>
>>>But it uses a uid_t.  That's more than a misnaming?
>>
>>It used to be uid-related in ancient times when the notion of container
>>hadn't formed up.
>>"user" part of user_beancounter name has the same origin :)
> 
> 
> Is it similarly irrelevant now? If so perhaps a big rename could be used
> to make the names clearer (s/user_//, s/ub_/bc_/, ...).
hm... let's try :)

Kirill