Re: 8250 serial console fixes -- issue

[Krzysztof Halasa]

Russell King <rmk+lkml@xxxxxxxxxxxxxxxx> writes:

> My point stands - if the user can provide an arbitary string to printk,
> they can fake any kernel message.  That in itself is a security bug.
> If there is an instance of that, then that's the real bug which would
> need fixing.

I think the AT problem is valid - the user doesn't have to be able to
send anything to the console, the system alone can screw it up with
something as simple as ATZ^M (or with almost any string with embedded
[aA][tT].*^M).

_If_ we want to be able to connect a serial console to dial-up modem
we can't send _anything_ to it when DCD is down (except configuration
strings etc). I don't know exact timings but the modem will probably
accept commands hundreds of milliseconds after dropping DCD.


There are/were modems which could be configured using a jumper to
password-protected power-on dial-in mode and which doesn't accept
further Hayes commands - those are safe (they don't even abort
inbound connection negotiation if they receive characters from
the computer).
-- 
Krzysztof Halasa
-
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/