Rockbox mail archiveSubject: Re: 5.5G iPod shutdown confusing the boot code?
Re: 5.5G iPod shutdown confusing the boot code?
From: Boris Gjenero <boris.gjenero_at_gmail.com>
Date: Sat, 21 Mar 2009 17:16:32 -0400
On Fri, Feb 27, 2009 at 5:17 AM, Alex Bennee <kernel-hacker_at_bennee.com> wrote:
> I've noticed that sometimes when Rockbox shuts down it leave the boot
> code confused as to the charge state of the battery. The iPod will
> refuse to boot with the usual Menu key and when plugged in to the USB
> reports very low battery. Once I give it the two finger salute and
> reboot into Rockbox everything is fine.
> Has anyone else seen this behaviour before?
Very rarely after I connect the FireWire charger my 5G 30GB iPod
starts booting, then resets with an emergency head retract noise and
restarts the boot process. Last time this happened, it reminded me of
how the PCF50606 automatically transitions to standby state 8 seconds
after the battery voltage monitor detects a low battery condition.
This makes sense to me because I can't see any way for the software to
transition to standby without the ability to wake in the usual way.
Lowering THRSHLD and enabling debouncing in the BVMC register when
shutting down Rockbox did not seem to help, maybe because firmware in
flash sets those on startup.
Other research and experimentation led to:
With the patch, power_off() on 5G/5.5G iPods writes a string to IRAM
and then resets the iPod. Then, OF in flash acts as if the startup
process failed because the battery is low. If there is no charger
connected, this leads to a shutdown. You will see a brief flash of
the low battery icon at shutdown.
So far, I've never had to reset my iPod after turning it off this way,
but I'm not sure I've done enough testing. Right now I feel a weird
hack like this isn't worthy of being in the tracker. My opinion may
change if it seems like the patch helps, though even then the right
solution would be finding what the OF does differently.
-- BorisReceived on 2009-03-21