From: [IDC]Dragon <idc-dragon_at_gmx.de>
Date: Tue, 2 Dec 2003 09:19:33 +0100 (MET)

Mike Holden wrote:

> It is worth noting that the disk spindown code that is disabled for the
> FMR does actually work, it just has too many side effects to be practical
> on a running box (backlight switches off, power reading unreliable, and
> possibly unit poweroffs). This was disabled during the debugging of FMR
> boxes early this year. It would be worthwhile looking at using this code
> only in this situation on the FMR and see how it works out. We should
> still disable the disk poweroff code during normal operations however!

You mean ide_power_enable() in power.c? It sets PA5 low for Recorders, like
my bootloader.
I've just tried it on my diskless FM wreck using UART boot, hey, the power
goes away! How could I have overlooked that, I've traced the PCB for power
control and found nothing.

So I will work that into the bootloader and need somebody to test it. (Not
dangerous, since the behaviour only changes when the charger is plugged). Then
I hope we can clear away this last drawback of flashing. Rockbox FM would
need some code to power up the disk again when leaving the charging screen.

BTW, by doing a typo I found that PA10 also has an influence. When set to
high, the IDE power goes away for ~60 ms and the current drained by the unit
raises a bit. Probably some kind of collision. What is PA10 used for on FMs? (I
should really start a spreadsheet about port usage.)

Jörg

-- 
HoHoHo! Seid Ihr auch alle schön brav gewesen?
GMX Weihnachts-Special: Die 1. Adresse für Weihnachts-
männer und -frauen! http://www.gmx.net/de/cgi/specialmail
+++ GMX - die erste Adresse für Mail, Message, More! +++

Received on 2003-12-02