Eric Jorgensen writes:
> I suppose we could try something drastic like severing the reset pin on
> the cpu. Real hacking starts with an x-acto knife, yaknow ;-)
Cutting CPU reset pins is kind-of bad. Most CPUs rely on the reset pin
to give them the power on reset to get them going from a sensible state.
Without it, you basically start in a purely random state at a random
address with random cache and MMU contents. Meanwhile, all the hardware
is held in a reset state.
No, the correct thing to do is not to try to disable the watchdog (if
you can disable it during run-time, then its not much of a watchdog
IMHO), but to find out where it is, and stroke it during the decompression.
We already have the required hooks for doing this.
Maybe also some tests can be done to try to discover the watchdog timeout
(like disabling all interrupts and timing from that point til when it
reboots, and repeat a few times). It might be milliseconds, or it may
be seconds.
_____
|_____| ------------------------------------------------- ---+---+-
| | Russell King [EMAIL PROTECTED] --- ---
| | | | http://www.arm.linux.org.uk/ / / |
| +-+-+ --- -+-
/ | THE developer of ARM Linux |+| /|\
/ | | | --- |
+-+-+ ------------------------------------------------- /\\\ |
_______________________________________________
http://lists.arm.linux.org.uk/mailman/listinfo/linux-arm
