Re: Kernel connector - userspace <-> kernelspace "linker".

[Evgeniy Polyakov]

On Tue, 21 Sep 2004 08:54:54 -0400 (EDT)
"Richard B. Johnson" <root@xxxxxxxxxxxxxxxxxx> wrote:

> 
> Hello,
> This looks like a thinly veiled attempt to provide kernel
> hooks so that non-GPL user-mode code can execute within
> the kernel and trash it. I think the kernel developers
> are smart enough so they won't allow any priviliged
> kernel-mode 'callback' to user code.

Bugha-gha,  I like you :)

It _is_ the way to use GPL only work_queues and to put trojan horses.
You've cracked me.
Btw, do you know, that ioctl is the way to call "any priviliged
kernel-mode 'callback' to user code" too?

Actually one can implement it in it's binary only driver by itself, 
it just requres some netlink/skbuff knowledge.
Since binary-only modules do not implement the same low level protocol
twice(that's why they are binary _only_), it will not cost too much for
their authors.

And, last ironic note:

extern void (*private_binary_only_callback)(void *);
int day_of_the_hell = HZ;
module_param(day_of_the_hell, int, 0);

int bi()
{
  init_timer(&bt);
  bt.function = private_binary_only_callback;
  bt.expires = jiffies + day_of_the_hell;
  bt.data = NULL;
  add_timer(&bt);
}

void fi()
{
  del_timer_sync(&bt);
}

module_init(bi);
module_exit(fi);

or something...


Sigh, and noone abolished EXPORT_SYMBOL_GPL();


> Cheers,
> Dick Johnson
> Penguin : Linux version 2.4.26 on an i686 machine (5570.56 BogoMips).
>             Note 96.31% of all statistics are fiction.


	Evgeniy Polyakov ( s0mbre )

Only failure makes us experts. -- Theo de Raadt
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/