jason0597 / ntrboot_flasher_nds Goto Github PK

I cant get this to work on my sdhc dualcore 2013 it says Flashcart setup failed

I found the official version of ds_ntrboot_flasher.
https://github.com/ntrteam/ds_ntrboot_flasher/releases
What is the difference with it and your version?

Here is the flashcart and error. Attempted to do the flashcart flash using an old DS Lite, Boot9Strapped 3DS XL (Both new and old), and a 2DS. Same error on all of them. Tried with SD formatted as FAT32 and as FAT. Same error.

Ended up using .firm version and doing it on boot9strapped 3DS without issues using the same flash cart.

https://imgur.com/a/m3oXaPO#alYbeCS

For some reason the indentation is messed up in a lot of files. Must be fixed.

Because an ak2i flash dump is 16MB, and the DS only has 4MB of memory, dumping it to flash would crash the program. Modifications must be made to DumpFlash() so that it writes in 1MB chunks (or 0.5MB chunks?).

Apparently #include <fstream> is the reason why file size is massive

Apparently some flashcarts do not like having their SD mounted when trying to do a flash.
We must make it so that it opens fat only when needed, meaning (in the case for ntrboot injection):

1. InjectFIRM() is called
2. Mount FAT temporarily
3. Read the FIRM into memory
4. Unmount FAT
5. Continue with InjectFIRM

We must not have FAT mounted all throughout the program

by splitting the dump procedure up into chunks, each chunk is going to finish at a certain percentage
it happens to be that each successive chunk's final printout is less than the previous chunk
what you're actually seeing is not the overall percentage, it's just the percentage of that chunk
so, you see the percentage of each chunk
each chunk's final printout would be different from one another
and in the end, it makes it seem like that it's decreasing the percentage, while what's happening is that the number of chunks dumped increases, while the last percentage of each chunk is less than the previous chunk

This line from flashcart_core's r4igold.cpp seems to return chipid instead of hwrev, making it so that it would fail this switch statement, and therefore making it say "Flashcart setup failed!" and refuse to proceed. If you add a new case that matches the "hwrev" (really it's the chipid), init will succeed and then read/inject commands will work.

Here's the log it prints out:
[INFO]: R4iGold: Init [NOTICE]: R4iGold: HW Revision = 00000fc2 [NOTICE]: R4iGold: HW Type = a79bca95

I do not know how to go about to fix this issue, or if it's even something that I can fix (flashcart_core issue?)