lkml.org 
[lkml]   [2015]   [Apr]   [17]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
SubjectRe: [PATCH v9 3/4] cgroups: allow a cgroup subsystem to reject a fork
From
>> >> Do you also want me to completely drop the COUNT macro? IMO it makes
>> >> the CGROUP_<TAG>_COUNT consolidation much nicer.
>> >
>> > What's wrong with simply having start and end tags?
>>
>> Because you'd have to write (CGROUP_TAG_END - CGROUP_TAG_START) every
>> time? It's a small addition and it makes referencing the range of a
>> tagged section much easier.
>
> Wouldn't loops look more like
>
> for (subsys = CGROUP_TAG_START; subsys < CGROUP_TAG_END; subsys++)

Sorry, I meant for defining arrays. `state[CGROUP_TAG_END -
CGROUP_TAG_START]` is just more annoying to type and read than
`state[CGROUP_TAG_COUNT]`.

> And even if not, just define a separate macro for the length. It's
> not like we're gonna have a lot of tags.

Do you mean like this?

#define SUBSYS_TAG_COUNT(_tag) (CGROUP_ ## _tag ## _END - CGROUP_ ##
_tag ## _START)

That's fine I guess, I just wanted to match CGROUP_SUBSYS_COUNT in
semantics, but I'll do that if you prefer it that way.

--
Aleksa Sarai (cyphar)
www.cyphar.com


\
 
 \ /
  Last update: 2015-04-17 23:21    [W:0.039 / U:0.048 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site