On Wed, Dec 12, 2007 at 04:18:15PM +0100, Casper.Dik at Sun.COM wrote:
> >I threw pfzsh together yesterday afternoon using the SFW framework, and
> >would like to request a sponsor to evaluate my work and help me get it
> >I note that the CR is currently assigned to Casper Dik and so he is
> >CC'd, but I don't expect that he will necessarily volunteer for the job.
> >A patch against a sfw-src-b79-20071129.tar.bz2 tree and an x86 package
> >are available at http://cr.opensolaris.org/~ceri/cr/4912090/.
> While you could go ahead and implement this, I am actually working on a
> different way of implementing this as part of the FGAP project.
> The new implementation does away with all the special code in the shells
> needed for pfexec and changes this to essentially become a property of a
> process and the kernel will perform the things needed for pfexec.
> This approach fixes many shortcomings of the current implementations:
> - no need to modify all the shells for what is essentially a
> Solaris specific function (free support for pftcsh, pfbash,
> pfksh93, pfzsh)
> - the ability to intercept exec's which are executed in the guts
> of another program (say /usr/lib/fs/ufs/mount rather than
> /usr/s/bin/mount) to further reduce risk
> - automatically inherit the property for subshells spawned from
> an editor, regardless of $SHELL settings
> In general, if a bug as a responsible engineer listed, that person is
> supposed to be working on the bug and working on it without contacting the
> RE first may very well lead to a duplication of effort.
That's cool; as I said, it was only a few hours yesterday afternoon
which I kind of enjoyed. The new approach sounds cool; I'm
interested how it allows a user to stop exerting a profile but I'll
follow the mailing lists when they appear.
That must be wonderful! I don't understand it at all.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Size: 187 bytes
Desc: not available