From: Boris Gjenero <boris.gjenero_at_gmail.com>
Date: Sun, 05 Feb 2012 09:09:11 -0500

On 12-02-04 11:28 PM, David Hall wrote:
> Ipod Nano 1st Generation, 4GB, original everything
> HW rev: 0x000C0006
> LCD type: 1
> PP version: PP5022C
> Est. clock (kHz) 81802 (30677 next look)
>
>
> Running unaltered.ipod:
> Pass: 0
> Exiting, 37 mismatches
> Press PLAY to quit
> (FWIW on clean shutdown the screen was left with alternating horizontal
> lines (1 pixel high it appears) of black and white. Haven't NOTICED
> that before).

This is as if lcd_shutdown() didn't properly shut down the LCD. It
certainly is called from shutdown_hw(). I guess it could just be due to
memory corruption.

> Running (default) modified build:
> Ran for ~6 minutes, 38 passes. Will run again and longer if you want.

That is sufficient.

> Running dma.ipod: (I believe you misspoke as to the name in your readme)
> Ran for ~6 minutes, 41 passes. Will run again and longer if you want.

There is need to test dma.* or nodma.* via pp502x_cache_test.rock
separately. They have exactly the same workaround as /.rockbox/rockbox.*.

Yeah, I misspoke regarding the 1G Nano in the exceptions section. I
meant pp502x_cache_test/dma.ipod not p502x_cache_test/dma.mi4.

> I will use this build as my normal build, yadda yadda yadda, but won't
> init database unless you say it would be helpful. This device just
> shuffles shit while I'm running, I never look at it.

Note that if you want to test DMA on the 1G Nano during normal
operation, you should replace /.rockbox/rockbox.ipod with
/pp502x_cache_test/dma.ipod. On the GoGear HDD6330, you need to replace
/rockbox/rockbox.mi4 with /pp502x_cache_test/dma.mi4. All others have
DMA enabled in /.rockbox/rockbox.*, but for these two I thought they
have device-specific problems with DMA which could make a build with DMA
totally unusable. I did it this way to ensure the normal boot works,
mismatches can be tested, and DMA can first be tested by loading
dma.ipod via rolo.

Thanks for testing!

Boris
Received on 2012-02-05