SanDisk Sansa c200v2, m200v4, clipv1, clipv2, clip+, and fuzev2

[kugel.]

Thanks for the help kugel. Cannot really understand the explanation re the ports and how they are addressed. I take it each pin is not a single bit within a word, but there are seperate words for each pin?

Each pin is a byte. But only 1 bit of those bytes is significant, the other 7 are not. But yea, you always read/write 1 byte for each pin.

I really recomment reading the data sheet, it helps.

[RockRabbit]

I can confirm that GPIOD pin 7 is the button light for the Sansa c250.

Thanks Kugel. The code worked fine.

Could you add it to the hardware mappings page for me please.

 

[gabe565]

Hmmmm...Whenever I try to run Rockbox, the bootloader starts and I see the Rockbox logo, but then, it says

*PANIC*
SD : DATA TIMEOUT,

I'm using the latest in SVN, and I'm not sure why this is happening! I'm trying to figure out what the problem is, does it mean internal SD? Even if I take out my SD card, I still get the same error. I just don't get it.

I'll add it to the hardware mappings page for you, RockRabbit.

[FlynDice]

I was getting the SD : DATA TIMEOUT until I took all the music off my e280, then It booted all the way into the menu just fine.  Buttons didn't work though so I tried Domonoky's button patch and buttons sort of work.  The down button won't move to a lower menu.

I did get an ogg file to play quite  nicely though by using a fixed.cfg file.  I had rockbox start in the db menu and right arrowed my way to a music file.  After the one file played though the screen goes dark and I need to reboot.  Any other patches to get the down arrow working?

[kugel.]

I can confirm that GPIOD pin 7 is the button light for the Sansa c250.

Thanks Kugel. The code worked fine.

Could you add it to the hardware mappings page for me please.

 

Great! I'll add it.  I'll maybe have a look at the c200v2 OF for the LCD if I have time.

Another thing: I think it's essential for safety that you find your USB "button". Turn the player on with USB and see if a delay happens or not compared to starting with the power button.

It would be very helpful if start creating a c200v2 target in the Rockbox tree then. So we can  build a basic bootloader using dual boot and test code *easier and safer*.

Edit: Uhm, now that I think about it. Finding USB  (and maybe some other buttons) should be essential for dual booting anyway, since I can imagine that dual boot with the power button will be somewhat difficult.

[RockRabbit]

Where does the arm object dump program come from? I cannot find it in Ubuntu. I would like to dump the OF and have a look at it myself.

[kugel.]

It should be there once you have compiled the rockbox toolchain with rockboxdev.sh.

Look for arm-elf-objdump

[Hillshum]

The dual boot on a release will use USB to boot OF anyway

[gabe565]

Kugel: I already added it to the hardware mappings page for him.

FlynDice: That's odd... I think I'll try what you said and see if it works. Thanks!

[RockRabbit]

Point taken about the power button - I will try to find it. However, am currently waiting till I get a datasheet before doing any more hacking. I have requested one so hopefully wont be long now till I can get things moving. In the meantime I will investigate the main ROckbox code and see what I can pick up about how it works.

Most of my life ive worked in IT writing software, usually for large companies. I have to say that doing Rockbox is a thousand times more interesting and satisfying. The pay is'nt as good, but there is lots of job satisfaction and how often do you get to listen to your code? Would'nt it be great if the Rockbox hardware could get made too?

[kugel.]

Point taken about the power button - I will try to find it. However, am currently waiting till I get a datasheet before doing any more hacking. I have requested one so hopefully wont be long now till I can get things moving. In the meantime I will investigate the main ROckbox code and see what I can pick up about how it works.

Most of my life ive worked in IT writing software, usually for large companies. I have to say that doing Rockbox is a thousand times more interesting and satisfying. The pay is'nt as good, but there is lots of job satisfaction and how often do you get to listen to your code? Would'nt it be great if the Rockbox hardware could get made too?

I have a datasheet. If you want to know something I can give you some information. Also, our Rockbox source code is a fine source of information already.
The base addresses of the GPIO ports:
GPIOA_BASE           0xC80B0000
GPIOB_BASE           0xC80C0000
GPIOC_BASE           0xC80D0000
GPIOD_BASE           0xC80E0000

You can read each pin using this formula:

GPIOx_BASE+(1<<(X+2))

where X is your pin and x A, B, C or D (the signficant pins are PAD_ADDR[9:2], that's why the +2).

To set pins use the mentioned formula and do
GPIOx_BASE+(1<<(X+2)) = (1<<X)

To clear just
GPIOx_BASE+(1<<(X+2)) = 0

You can set/clear GPIOx_DIR for pins using this
GPIOx_BASE+0x400 = (1<<7|..)

I hope this gets you somewhere until you have the datasheet.

[RockRabbit]

Thanks hugel.

But could you explain what you mean by BASE and DIR?

[kugel.]

GPIOx_BASE just stands for the base adress, i.e. the starting address of a port.

DIR (stands for direction), as already said, decides whether a pin is configured as output or input. output means we can safely write to it (with our software), input means the hardware can safely write to it.

[RockRabbit]

If i address a GPIO pin using the code "ldr   r1, [r0, #0x8] " where r0 has the base address of the IO port A, am I right in thinking that this is pin A1? Similarly would "ldr   r1, [r0, #0x10] " be A2?

Also, I notice that some of the pins seem to detect on bit unset (0), others on bit set (non zero). Is this normal?

I notice that in the hardware mappings twiki, no mention is made of which logic level is used for each port/button combination. Is that becuase it is not necessary or because I have got confused somewhere?

Thanks, 

[kugel.]

If i address a GPIO pin using the code "ldr   r1, [r0, #0x8] " where r0 has the base address of the IO port A, am I right in thinking that this is pin A1? Similarly would "ldr   r1, [r0, #0x10] " be A2?

This is true. And Pin 7 is (1<<7+2) -> (1<<9) -> 512 -> 0x200. I think you got it now

Also, I notice that some of the pins seem to detect on bit unset (0), others on bit set (non zero). Is this normal?

Yes, the states at boot up don't have to be the same for all pins.

Is that becuase it is not necessary or because I have got confused somewhere?

I don't know. Probably the creator haven't thought of it (maybe because he didn't have enough inforomation at that time) when he created it.