Re: "per-process" limits (was: Showstopper list)
Paul Barton-Davis (pbd@Op.Net)
Wed, 14 Oct 1998 21:50:38 -0300
>On 13-Oct-98 Paul Barton-Davis wrote:
>> Indeed. Except that I'm interested in a more general solution to the
>> problem of the missing kernel abstraction. Linux currently has no
>> notion of "a group of tasks"; such an abstraction is both useful for
>> process limits, and lots of other things as well.
>
>I don't know - is a process or group of processes really the central
>abstraction for resource tracking? Sure, processes consume memory and
>processor time, but the consumer of disk space is independent of what process
>did the writing.
Sure. This just underlines my point that resource tracking and user
ID's are occasionally orthogonal, and any mechanisms to implement them
must have the potential to honor this occasional
orthogonality. Conflating resource consumption with a user ID of some
kind, *OR* denying this relationship completely is a mistake either
way. Neither approach is "central" - both must be possible.
--p
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/