-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
> The trick is very simple: with JTAG, write a program to RAM and run it.
> That program connects somehow to a port (serial, USB or a custom
> protocol over JTAG!) and reflashes the NAND.
JTAG cannot write to SDRAM. Remember that JTAG boundary-scan is VERY SLOW ...
way too slow to refresh the RAM properly! You can try loading to instruction
cache, but you still need a stub in flash to jump to the appropriate area.
> Is that possible? Yes it is, because how would the upgrade application
> upgrade your NAND flash with another version of PPC? So the CPU can
> reflash
Hm? The upgrade application just writes directly to NAND flash. The CPU never
executes natively from NAND flash, so that's not the problem.
> So, I'm not worried at all to kill the flash contents of my spare board.
> That's where it is for and I'm really convinced that I can rewrite it.
I'm sure you can rewrite it, but there is no software to do it.
> The only point is that I don't have a proper JTAG device. So I'm toying
> around with a Wiggler device :-(
Ah, Wiggler is fine. It's pretty standard. I'm using an Insight IJC-2 here,
which is pretty much the same thing, albeit with a different pinout.
> Hopefully, I get that to work soon.
Best of luck
> Bas.
Joshua
- --
Joshua Wise | www.joshuawise.com
GPG Key | 0xEA80E0B3
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (GNU/Linux)
iD8DBQFAfuUsPn9tWOqA4LMRAv7dAJ43dR05x15WHgT+O81lOU4V/OMEPgCff/ia
rN3ZkLFGLJTWcRUFZ3svSbg=
=TZt1
-----END PGP SIGNATURE-----
Received on Thu Apr 15 2004 - 19:40:29 EDT
This archive was generated by hypermail 2.2.0 : Mon Jul 25 2005 - 17:19:27 EDT