stable-tracking test image

Werner Almesberger werner at
Mon Nov 17 18:31:23 CET 2008

Andy Green wrote:
> I have read it somewhere I am sure... but the partial reset thing is
> definitely a fact in the 6410 datasheet, they explicitly describe three
> levels of hard reset.  That's how come GPIO state are preserved on resume.

Good point. GPIOs are indeed treated separately. (Phew, when I read
your question, I briefly wondered "and what if they aren't preserved
all the way through".)

However, regarding Steppingstone. The 6410 manual says:

| The [SLEEP] exiting sequence is as follows:
| 4. NFCON copies boot code from an external NAND device to the stepping
|    stone if the boot device is NAND.
|    (XOM[4:3] == 2'b00)

So I wouldn't bet on SD boot to be different. Anyway, I just wanted
be sure we don't accidently depend on a feature that may not be

- Werner

More information about the openmoko-kernel mailing list