-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Friday 14 February 2003 06:37 pm, Mark J Roberts wrote:
> bdonlan:
> > I don't understand. Anyway, why not fork() and restrict the child?
>
> That is hardly easy or efficient.
>
> My point is that I've got this notion of being able to use some
> capability only by calling a predefined function (ie, the one I
> posted that accepts user input from the terminal), but I don't know
> how to translate that into a generic kernel facility.

The function could be overwritten.

> The kernel is concerned with ensuring that the state of the program
> is acceptable before allowing a capability to be used. For instance,
> that the predefined function was called normally, instead of jumping
> into it halfway with a malicious stack.

How can it determine that? The typical state of the stack vaires widely. Have 
the functions been using alloca? Is it in a signal handler? Was this part 
compiler with gcc? This part with icc? All of these can alter the structure 
of the stack.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQE+TX42x533NjVSos4RArD2AJ9PcnfsNOVbg1ma7I3KjKxGxCwABwCdHiY5
Mk4R5ykUGz52i2CxCH2n3YM=
=dOA2
-----END PGP SIGNATURE-----

_______________________________________________
devl mailing list
[EMAIL PROTECTED]
http://hawk.freenetproject.org:8080/cgi-bin/mailman/listinfo/devl

Reply via email to