Hacking Hardware Picofly - a HWFLY switch modchip

  • Thread starter Thread starter mathew77
  • Start date Start date
  • Views Views 3,675,162
  • Replies Replies 17,052
  • Likes Likes 15
May I ask everyone, when the picofly v2.64 is first powered on, the LED will flash for a long time before displaying "NO SD CARD". What is picofly doing at this time? Has any data been written to NAND? If so, how can it be recovered? And is this step the same for all models?
 
May I ask everyone, when the picofly v2.64 is first powered on, the LED will flash for a long time before displaying "NO SD CARD". What is picofly doing at this time? Has any data been written to NAND? If so, how can it be recovered? And is this step the same for all models?
You're essentially asking for the "secret sauce" that makes the glitch happen. That is what this entire thread has been devoted to, and what Rehius has ultimately released to us as the PicoFly firmware.

While they have been kind enough to share it with us and update it, they have not (and are under no obligation to) released their firmware as opensource to the public.

The generalized concept though, is that the RP2040 is glitching the CPU repeatedly to try and convince it to load an unsigned portion of code. This unsigned portion of code necessarily gets loaded into BOOT0 and is known as the SD Loader. It allows us to put a payload.bin file (and any supporting files) onto an SD card and load it from this SD Loader portion. This is what the "NO SD CARD" screen is, a portion of the BOOT0 which is holding our temporary sideloader.

@rehius or anyone else, please feel free to correct me if I am incorrect, this is just my understanding of RGH style CPU hacks in general.
 
If you follow the forum message, you will understand the pio is already given. So the glitch is actually not a secret.

I already save the pio code, the method of reverse engineer the fw using emulator (its clever one using zig), its all interesting which i might learn when i have spare time.
Yes, I've been following the thread for quite some time as well. I only meant to say that code written out in assembly language written for the RP2040 is hardly easy for a lay-person to recognize what is happening (often times even with the comments being written right into the code, as the PIO for this glitch hack had)

If they just wanted an overview, that's what I was trying to provide.
 
  • Like
Reactions: impeeza
Can I please have alternative 3.3v point for mariko? The cap point from guide seems hard to do
pinout_emmc-jpeg.360378

This graphic is available in the AIO thread. It denotes a 3.3v spot on the EMMC board.
 
  • Love
  • Like
Reactions: Tom8823 and impeeza

Site & Scene News

Popular threads in this forum