Artur Skawina wrote:
> > is a downside to doing this. If, for example, timeofday is
> > available in a raw form on one of these data pages, and user-space
> > code can read that page at will (without going to the kernel), you
> > lose your ability to "lie" to user-space about the time.
>
> if you're saying absolutely everything must go through ptrace, even if
> it happens to have userspace-only implementation -- yes, obviously.
>
> preventing a rogue process from accessing the info directly might be a bit
> harder, w/o giving up the main benefit -- speed. There certainly seems to
> be a market for a runtime (global) on/off switch at least.
If you switch it off, the process knows you've switched it off and
may behave differently.
It's probably possible to lie to the process for gettimeofday and signal
operations though -- have the kernel unmap the page for those processes,
and have a ptrace SEGV handler which provides results as if the page
were still mapped.
-- Jamie
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Mar 15 2000 - 21:00:13 EST