Hi!
For https://github.com/c1570/Connomore64 , I use several interconnected RP2xxx (four at the moment, will extend to eight soon). In order to make reflashing less of a hassle, and make a small cheap integrated PCB at some point, I'd like to remove all but one flash chip from the setup (everything runs from SRAM anyways), and then, on startup, have one RP2xxx fetch firmwares from its flash, then program the other RP2xxxs via SWD.
1. Is this reasonable, or are there better solutions?
2. Is there code available for this? I found bits and pieces for driving SWD on the net (for other platforms mostly), or I could snoop the SWD data sent by OpenOCD to a PicoProbe when writing to SRAM, but perhaps there's a quicker way than adapting other code and reverse engineering stuff.
3. Currently everything is RP2040s, but I'll likely switch to RP2350 RISC-V soon. I don't quite understand how SWD works there. Any pointers? Is the approach outlined above still workable with RP2350/RISC-V?
Thanks!
For https://github.com/c1570/Connomore64 , I use several interconnected RP2xxx (four at the moment, will extend to eight soon). In order to make reflashing less of a hassle, and make a small cheap integrated PCB at some point, I'd like to remove all but one flash chip from the setup (everything runs from SRAM anyways), and then, on startup, have one RP2xxx fetch firmwares from its flash, then program the other RP2xxxs via SWD.
1. Is this reasonable, or are there better solutions?
2. Is there code available for this? I found bits and pieces for driving SWD on the net (for other platforms mostly), or I could snoop the SWD data sent by OpenOCD to a PicoProbe when writing to SRAM, but perhaps there's a quicker way than adapting other code and reverse engineering stuff.
3. Currently everything is RP2040s, but I'll likely switch to RP2350 RISC-V soon. I don't quite understand how SWD works there. Any pointers? Is the approach outlined above still workable with RP2350/RISC-V?
Thanks!
Statistics: Posted by c1570 — Sat Sep 21, 2024 11:18 pm