[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [f-cpu] little feed-back from the libre softawre meeting

> > > >     read, write, exec bit + at least 3 rings (super user + user +
> > > >     something like for library,...)
> > > 
> > > RWX is ok, because this is how protection is enforced.
> > > i'd say that it is the minimum required TLB feature.
> > 
> > *nod*
> > 
> > What about different permission bits for different protection levels?
> > Like `supervisor may read and write, j. random luser may only read'.
> > That's probably better than a `user/supervisor' bit or an explicit
> > `page protection level'.

>     I have to agree with Michael here. In all the time I've been making
> kernels, I've never EVER found protection levels to be of any use.
> Separate masks for supervisor and user level for the page are the way 
> to go, i.e. user RWX and supervisor RWX.

I totally agree with you. I and nicO didn't find any use of this kind of ring
protection (and in fact that create a stack into the CPU ;-( ). But when
we discuss with Bradley Spencer, he says that somebody was working on a
ring protection for linux and that we must see what we can do with that, but
Bradley say to us too that he didn't see how to use them ! ;-)

I personnally think that we must remove the SR that say that we are superuser
or user, and prefer to us a protection mecanism with the SR. But because
the SR will be really use frequently by the kernel (we didn't have a lot of
entry in our TLB), so we must have a quick protection, I mean for example
that we only make a difference between read-only SR and read/write SR, and
if we trap we put in a SR the number that the user task want to access to and
if it's a write the value it wan't to put in it and that's all.

To unsubscribe, send an e-mail to majordomo@seul.org with
unsubscribe f-cpu       in the body. http://f-cpu.seul.org/