Re: [patch 2.1.97] more capabilities support

Linus Torvalds (torvalds@transmeta.com)
21 Apr 1998 17:35:40 GMT


In article <353C7192.44E6F6C1@luz.fe.uni-lj.si>,
Andrej Presern <andrejp@luz.fe.uni-lj.si> wrote:
>
>It is intresting what you say here. I have tried to explain a concept
>much securer than what is being presented here to some people, but have
>failed in doing so because of being unable to provide the complete
>implementation details due to my lack of Linux internals knowledge.
>
>If you are interested, I would very much like to explain it again to
>you.

This concept was using segments to give very low-level access rights on
a per-object basis?

I think I saw it, and never looked closer for a few reasons:
- segments do not exist on anything but a i386+ (even i286- "segments"
are just abominations, and have nothing to do with real segments) in
any popular hardware.
- even where they exist, they are slow.
- protection that finegrained is theoretically nice, but a maintenance
nightmare. In theory it is the best kind of protection, in practice
you won't find anybody who will program a non-trivial program using
them correctly and securely (I claim that even ACL's have this
problem, and ACL's are much simpler)

I personally do not believe in object orientation as a security model
(nor as a general programming paradigm), but feel free to try to
convince me.

[ Or maybe I confuse you with somebody else and some other posting ]

Linus

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu