SB6190 Uncapping in process. Suggestions?

[colin669]

U got a pic of the back of the modem?? I don't see the flash chip anywhere in the front.

here's the pic you're asking for.

Can you tell me the numbers of the 48 pin SMT chip on the back of the board
and can you take a clearer picture of the front and back of the board in high resolution so it can be blown up with clarity and post them to a Dropbox and give me the link i will tell you what you will need to read the chip

---

there is no longer a SPI Flash,all the new modems are nand flash only. and they have Bad blocks to prevent it from being copied so far.

I don't need to copy the firmware, only flash to the NAND. How do I do that? I can compile the firmware from ARRIS, they released it! It's open source, documented link in My first post.

I'm looking into modding & compiling the firmware now. I have an SoC(Jetson TK1) if needed. I should be able to flash to NAND, skipping bad blocks, and the modem will run the My custom firmware just fine. Sounds perfect to Me.

We need some one with coding experience to write a piece of software that can handle the remapping of the bad blocks so we can read the nand correctly just like the guys did for the XBOX so there is hope out there

Experienced coders would be great, but I can write the code Myself, it'll just take more time. That's a guarantee! I will get you a higher res photo set ASAP.

Numbers:
Spansion
S34ML01G200TF100541BB337 A
©12 Spansion

to read this chip correctly you need to use ECC Error correction code
there are many ways to handle this.
Now im wondering if you can see any references to this in the source code in the firmware because the ECC code was written by the manufacturer for that device .
otherwise you will get bite flip every time you read the chip

Each chip has its own bad block or blocks direct from the factory, and bad block occur on the chip after the code is written to it so the code on the chip has to know how to handle the remapping of the chip otherwise you would be trowing these modems into the trash every time it developed a bad block

I'll comb through the firmware but from what I see, BGA uses a limit algorithm and when the limit is satisfied the algorithm restarts causing a remap generating a new memory location and cycles through the NAND that way.