On Tue, Feb 01, 2022 at 09:17:47AM +0000, David Laight wrote:
> From: Kees Cook
> > Sent: 01 February 2022 00:10
> ...
> > While the initial code searches[6][7] turned up what appeared to be
> > mostly corner case tests, trying to that just reject argv == NULL
> > (or an immediately terminated pointer list) quickly started tripping[8]
> > existing userspace programs.
> >
> > The next best approach is forcing a single empty string into argv and
> > adjusting argc to match. The number of programs depending on argc == 0
> > seems a smaller set than those calling execve with a NULL argv.
>
> Has anyone considered using the pathname for argv[0]?
> So converting:
> execl(path, NULL);
> into:
> execl(path, path, NULL);
>
> I've not spotted any such suggestion.
It came up on some IRC discussions at some point. I'm personally not a
fan of this because it creates a bit of "new" ABI that has a lot of
variability (depending on "" is one thing, but depending on a "missing"
argv matching the exec path is very different). I think there were also
concerns about dealing with fd-based exec ("what is the 'right' name"),
etc.
I'd prefer we stay as simple as possible for this change.
--
Kees Cook
