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

[jesterday]

non-modified actual firmware (03.01.16F). So far i can see, nothing is changed or even done when hitting the center. 

Anyone played around with the MTABLE, RES_INFO, SYS_CONF sys files or the version.sdk?
The warning in the version.sdk file seems  senseless, as it is restored (like the other files) when changed or removed.

As the files obviously are checked when booting, maybe we have to modifiy them to get the Erase Firmware to work (as example changing the Firmware Text in ERASE or something other)? mentioned one i tested of course.
Just some thoughts

have a nice WE

jesterday

EDIT: Has someone done the work of find out the "meaning" (damm, cant remeber the right word :-) ) of the values in SYS_CONF.SYS? Can this be of help in any way?

[lassowski]

Hi!

Perhaps this feature removes a e200xx.bin firmware image from the disk before the player is upgraded with it?

Just a thought...

Frank

[saratoga]

The erase firmware is probably just a feature that Sandisk never enabled, which is why they don't mention it and why it doesn't seem to do anything.  Firmwares for mp3 players are highly modular, with parts provided by different vendors.  Just because something is included doesn't mean its actually used.

[Domonoky]

Hey,

i just got a sansa m200v2 and want to help out. I have tried the clip bootloader from the git tree, but unfortunatly the buttons are different on the m200v2, so it only starts the OF for me.
 
The SansaV2Firmware wikipage tells the sansa has a 4x4 keymatrix and no directly connected buttons. So i want to try to find those buttons. I have successfully inserted a delay loop, so i know the bootloader is running, and i want to use the this delay to check for buttons.

Unfortunatly i dont know too much asm, so i am still struggeling to set those GPIO pins correctly.
As far as i know, i should set one of A4-A7 to output, then write a 1 to this output, and then check A1-A3 for changes. Could someone help me with this ?  ( i am not afraid of bricking).

EDIT:
with the help of others i found nearly all buttons and put them on the SansaV2HardwareMappings wiki page.
Now i can boot into rb-bootloader if i want, but i get no reaction as lcd and backlight dont seem to work, and there is no led on this player. So next task would be either lcd driver or backlight..

EDIT:
we can control the backlight now. LCD is still unknown. Seems to be different to the one from the Clip.


greets

[funman]

Something shocked me while I was discussing with anythingbutipod people: in the diagnosis mode of the Clip there is a "SDRAM Test" which reads:

bus = [16] bits
SIZE = [2] MB
R/W TEST [PASS]

bertrik also mentioned seeing code related to external memory.

And finally the loading addresses of 2 drm library blocks, plus the "otg_functio" and "usb_functio" blocks are after 0x30000000

In the AS3525 datasheet the memory between 0x30000000 and 0x3000FFFF is "External Memory IF (MPMC Bank 4 - SDRAM)"

All this lets me suspect that the assumption that there is no SDRAM on the Clip (V1) is wrong.

By the way why do you suspect there is no SDRAM on the Clip ?


I tried writing to 0x30000000 and reading back but it failed, I suppose we need to initialize the SDRAM before using it.

EDIT: it seems the Clip ships with the 224 BGA AS3525 (with memory controller ARM PL172)

[Domonoky]

Another update for the m200v2:
Funmans merge of the ssd1815 driver with the dbop commands works, and i can see the rockbox logo on the m200v2 !! *yeah*
So the LCD for the m200v2 is really the  ssd1815 :-)

Edit:
you can see rockbox on m200v2 here: http://www.retrospektiwe.de/sansa-rockbox.JPG :-)

[funman]

Hi !

I can know read and write to the 2MBytes of SDRAM embedded in the Clip

I've put my code in the flamed git repository: see this commit and this one

Now I'll speak with the rockbox developers to commit that in the rockbox svn tree.

[saratoga]

Hi !

I can know read and write to the 2MBytes of SDRAM embedded in the Clip

I've put my code in the flamed git repository: see this commit and this one

Now I'll speak with the rockbox developers to commit that in the rockbox svn tree.

Great news.  This will save us a lot of trouble rewriting how the codecs use memory for the clip port.

[Domonoky]

Hi,

i can confirm that this sdram code also works on the m200v2. (which has 2MB external SDRAM).

[Hillshum]

Hi,

i can confirm that this sdram code also works on the m200v2. (which has 2MB external SDRAM).

and the e200v2

[funman]

Hello !

I was accepted into the committers, and I have committed the SDRAM code in rockbox SVN.

Next step: commit the SD driver embryo so people who hadn't noticed our git tree yet can look at it.

Once it's done, everything will be in SVN and there will be no need to commit code in the git repository.
If you want to work on the SVN tree just ask the developers or send me your patches!

By the way you can work on the rockbox SVN from a git-svn repository, there are instructions in the wiki on how to do so.

[ChiefAlex]

hi together.
I own a Sandisk Fuze and e260 (V1, with rockbox running on it) .
I would love to see Rockbox on the Fuze as well. If i can help you with my player (but without risk to "kill" it) let me know it. 

(edit: i'm german, so any commands in simply english)

[kugel.]

Hey, I'd gladly accept help on creating a feedback mechanism for testing on the Fuze.

As of now, neither LCD nor button light nor backlight works (i.e. no visual response at all). Buttons are unknown too. The only feedback mechanism is the 5s delay loop which tends to be annoying and time intensive to work with.
Anyone with greater experience in OF disassembling than me have a look at the Fuze OF and help me getting at least the button light to work; of course LCD would be nice too

I've really not enough time to spend much time on the port (which means I cannot really catch up to your knowledge at this time), so I'd really need help to get the Fuze further.
 

saratoga: How's your JTAG going? Did you already receive the correct hardware (the first one you got was too big IIRC).

[linuxstb]

There are still things that people not involved with the low-level hacking can do to help the ports along:

1) Add the Fuze (and c200v2) to the build system.  This would just be a matter of copy/paste of the existing code for targets already in SVN (clip, e200v2, m200v2).

2) Start work on all the UI simulators for these targets. 

The UI sims will all need a photo/scan of the externals of the device (resized so that the area of the LCD is the real LCD's resolution) for use as the background image (or possibly just copied from the V1 targets, where applicable).

The e200v2 sim should be the easiest to get working, as this uses the same keypad definition as the e200v1.  Other targets may need keymaps defining for the Rockbox apps/ code, plus all the plugins.

Making the UI simulators work means doing the majority of the work needed in getting the main Rockbox code working (after the low-level drivers).

Patches welcome (on Flyspray)...

[ChiefAlex]

Fuze could use the same keypad because buttons and wheel are the same just "menu" button is top right (botton left on e200).