Re: [RFC] on general object IDs again

[Cyrill Gorcunov <gorcunov@gmail.com>]

On Wed, Jan 11, 2012 at 10:22:29PM +0400, Pavel Emelyanov wrote:
> On 01/11/2012 10:19 PM, KOSAKI Motohiro wrote:
> >> Hi Kosaki,
> >>
> >> the idea on user-space is something like
> >>
> >>  - collect all pids to dump
> >>  - collect IDs for every pid
> >>  - sort the IDs obtained
> >>  - find the same IDs (which will be kind of find intersections in a sets of IDs) and
> >>   set up CLONE_ flags on restore procedure as appropriate (for example if
> >>   GEN_OBJ_ID_VM IDs for two or more tasks are the same we need to use CLONE_VM
> >>   at restore time, and so on).
> > 
> > Then, you only need to compare. not any other calculation. i.e. only
> > need id uniqueness.
> > And any resource are referenced from tasks. so, can you reuse pid for
> > this? example,
> > two taska share one mm.
> > 
> > task-a(pid: 100)
> >               |-----------------mm
> > task-b(pid: 200)
> > 
> > 
> > gen_obj_id(task-b, GEN_OBJ_ID_VM) return 100. (youngest pid of referenced tasks)
> 
> We can, but determining the youngest pid for an mm struct is O(N) algo.
> Having N tasks with N mm_structs getting the sharing picture becomes O(N^2).
> 

Yeah, exactly. If not the speed problem we would simply stick
with Andrew's proposal as two-id-are-the-same(pid1, pid2)
syscall. But when we get a number of pids to dump we need the
resource affinity picture over them all.

	Cyrill