[lkml]   [2015]   [May]   [29]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [RFC 3/3] memcg: get rid of mm_struct::owner

On Fri, May 29, 2015 at 03:45:53PM +0200, Michal Hocko wrote:
> Sure but we are talking about processes here. They just happen to share
> mm. And this is exactly the behavior change I am talking about... With

Are we talking about CLONE_VM w/o CLONE_THREAD? ie. two threadgroups
sharing the same VM?

> the owner you could emulate "threads" with this patch you cannot
> anymore. IMO we shouldn't allow for that but just reading the original
> commit message (cf475ad28ac35) which has added mm->owner:
> "
> It also allows several control groups that are virtually grouped by
> mm_struct, to exist independent of the memory controller i.e., without
> adding mem_cgroup's for each controller, to mm_struct.
> "
> suggests it might have been intentional. That being said, I think it was

I think he's talking about implmenting different controllers which may
want to add their own css pointer in mm_struct now wouldn't need to as
the mm is tagged with the owning task from which membership of all
controllers can be derived. I don't think that's something we need to
worry about. We haven't seen even a suggestion for such a controller
and even if that happens we'd be better off adding a separate field
for the new controller.

> a mistake back at the time and we should move on to a saner model. But I
> also believe we should be really vocal when the user visible behavior
> changes. If somebody really asks for the previous behavior I would
> insist on a _strong_ usecase.

I'm a bit lost on what's cleared defined is actually changing. It's
not like userland had firm control over mm->owner. It was already a
crapshoot, no?



 \ /
  Last update: 2015-05-29 16:41    [W:0.047 / U:43.796 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site