>
> brandon@rhodesmill.org wrote:
>
> >So I think that this architecture:
> >
> > (1) can be made to work today in user space
> > (2) keeps all the complicated rules in user space
> > (3) enables logging with logging mechanisms in user space
> > (4) with a simple mechanism-only kernel patch, this architecture
> > can run about as fast as any possible solution that keeps
> > the complicated rules in user space
>
> It takes two switch context for every trapped system called, isn't it ?
> That would be tooooo slow. Especialy at multiprocessor system.
> For example if you have N processors & N threads then every thread/processor
> have to wait one thread-monitor on 1 processor. Not good.
Can't you use some dummy char device with two ringbuffers, one to put syscalls
into from the kernel (concerned process is then suspended),
a daemon in userland to peek into that first buffer with read() and authorize
them with write() to the second buffer, at which point the concerned process
will be resumed? This does not seem to cause any overhead as opposed to
ptrace in both MP and SP...
> Sergey
>
>
>
> -
> 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/
>
pk.
/*****************************************************************
in a world without walls and borders who needs windows and gates ?
*****************************************************************/
-
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/