Re: [RFC][PATCH 1/5] uts namespaces: Implement utsname namespaces

On Saturday 08 April 2006 22:28, Serge E. Hallyn wrote:

This is something we've been discussing - whether to use a single
"container" structure pointing to all the namespaces, or put everything
into the task_struct. Using container structs means more cache misses
and refcounting issues, but keeps task_struct smaller as you point out.

The more cache misses argument seems bogus to me. If you consider
the case of a lot of processes with lots of shared name spaces
the overall foot print should be in fact considerable less.


The consensus so far has been to start putting things into task_struct
and move if needed. At least the performance numbers show that so far
there is no impact.

Performance is not the only consider consideration here. Overall
memory consumption is important too.

Sure for a single namespace like utsname it won't make much difference,
but it likely will if you have 10-20 of these things.


iirc container patches have been sent before. Should those be resent,
then, and perhaps this patchset rebased on those?

I think so.

-Andi
-
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/

Relevant Pages

  • Re: [PATCH 0/7] containers (V7): Generic Process Containers
    ... then "container" seems like a good name for my ... The aggregation happens at the task struct when you apply different ... to have some namespaces, possibly not aggregated within the nsproxy ... other new resource controllers and process control modules. ...
    (Linux-Kernel)
  • Re: Which of the virtualization approaches is more suitable for kernel?
    ... If to drop VPID virtualization which caused many disputes, ... The proposed solution is similar to fs namespaces. ... container is involved unlike a second approach. ... functions (e.g. Eric's patch for networking is 1.5 bigger than openvz). ...
    (Linux-Kernel)
  • Which of the virtualization approaches is more suitable for kernel?
    ... If to drop VPID virtualization which caused many disputes, ... The proposed solution is similar to fs namespaces. ... container is involved unlike a second approach. ... fine grained namespaces are actually an obfuscation, since kernel ...
    (Linux-Kernel)
  • Lazy container
    ... the container pulls items on the fly from an unspecified source through a function and returns an instance of a given class over the pulled item. ... The problem with this implementation is that the cache never decreases in length as Foo instances are no longer referenced by the lc accessor since they're all referenced by the internal cache. ...
    (comp.lang.python)
  • Re: The issues for agreeing on a virtualization/namespaces implementation.
    ... Where do we put the references to the different namespaces? ... - Do we put the references in a struct container that we reference ... but first approach should always not to cache and go through the container. ...
    (Linux-Kernel)