Rockbox mail archiveSubject: Re: PP502x cache bug and ATA DMA test build
Re: PP502x cache bug and ATA DMA test build
From: Boris Gjenero <boris.gjenero_at_gmail.com>
Date: Sun, 05 Feb 2012 09:33:51 -0500
On 12-02-05 07:57 AM, David Hall wrote:
> Sansa E200 2GB, original everything. NO SD card inserted if that is
> PP Version: PP5022C
> Est. clock (kHz): 30675
> Running unaltered.mi4:
> Pass: 0
> Exiting, 30 mismatches
> Running (default) modified build:
> Hmm, after a few minutes, always when I'm not looking, it shuts down.
> Idle power off is set for 30 minutes and it's happening much before that.
> Passes come much faster than on my iPods.
I wonder why it's faster. I guess that means it's boosted. The CPU
shouldn't be more efficient at the same clock speed compared to iPod
CPUs. You should be able to see the clock speed in the "CPU frequency"
debug screen. On my 5G iPod if I quickly enter "CPU frequency" after
scrolling, I see 80000000 for a fraction of a second from GUI boost, and
then it goes back to 30000000. This is the proper behaviour that I expect.
> Device is unresponsive to holding the "play" button to quit.
> If I keep tapping the "play" button to keep the backlight on I get up to
> 105 passes before I tire of it.
In sansa-e200/button-target.h, there's no BUTTON_PLAY, so BUTTON_SELECT
should be used instead. The message at startup inaccurately says "PLAY".
The second message (when exiting after finding mismatches) properly says
> Hmm, maybe it isn't shutting down, just VERY slow to turn the backlight
> back on. Nope - somewhere within 5 minutes of the backlight going off
> it shuts down.
From your other message:
> Shuts down after 24 passes and no touching even when backlight set to
> always on. Idle power off is set for 30 minutes.
Could this just be due to all the CPU activity using a lot of power and
causing a low battery shutdown?
Thanks for testing!
Received on 2012-02-05