Search the Community
Showing results for tags 'jtag'.
In this well descripted analysis, Markus Gaasedelen has managed to successfully dump the console's secret boot ROM by intercepting the Intel CPU's JTAG functionality with a custom made adapter. Its amazing to see discoveries appear here and there to this very day in 2023, since this can open the full capabilities of the hardware to many hobbyists and hardcore developers. Detailed explanation can be found here: https://blog.ret2.io/2023/08/09/jtag-hacking-the-original-xbox-2023/
I'm trying to write a bitstream to a spare Aladdin XT2 Plus modchip with a knock-off USB Blaster. The USB Blaster in question is one that came with an AliExpress FPGA dev board. It looks like an official product, literally says "USB Blaster" and "Altera" on the sticker. One of those knock-offs. I only mention this because it could pertain to the issue. The github page linked says to externally power the modchip, which I am doing. I've found that if I connect 3.3v from an Adafruit trinket and leave the 3.3v line from the USB Blaster connected, I am able to run `urjtag` on Linux, connect the USB Blaster, and run the `detect` command. This reads out the device ID and hangs. I give the process a `SIGINT` and it segfaults. Seems promising, but I don't trust this to write the configuration to the chip. Possibly an issue with `urjtag`, but I can always debug that myself. Should the USB blaster be providing 3.3v while there is an external source providing 3.3v? This doesn't make sense to me, but it's the only way I've been able to get a partially successful ID from the Lattice chip. I also have this power supply. It takes my 12v wall wart and gives me some nice 3.3v headers. I've double checked with a multimeter and it seems to do what it says. When I power with this, it does bad things to the USB Blaster. I tried first without the 3.3v line from the USB Blaster and again with the 3.3v line connected. Both times, the USB Blaster is not even detected by my computer. It takes a minute or two of being unplugged before the USB Blaster is recognized by my computer again. This particular power supply has an on/off switch. I can switch it to off, plug in power leads to the modchip, and this still messes up the USB Blaster. It won't be detected even though the external power source is turned off. What could be causing this? Completely out of scope, I've heard that the Xecuter 3 uses a Lattice chip in the same LC4XXX family. From what I understand, they scrubbed the markings on the CPLD and made it look like an Actel chip to make cloning more difficult. There's a schematic publicly available, but from what I can tell, it requires transplanting chips from a working Xecuter 3. I seem to recall hearing that they locked the chips during manufacturing so the configuration / bitstream couldn't be copied. Does anyone have any recollection of this?
So whatever code is running on the xenium clones I have does not like anything being on the flash other than their own modified xenium os 1.0. They're still happily recognized by the flasher tool, even after flashing 2.3.1 to them. but I accidentally overwrote the flash.bin when I went to write it back. Learn my ass for not fully reading shit and madly clicking. TL:DR: I'm an idiot and need a JTAG programmer to flash the openxenium code to my xenium clones. Would this work?(page auto-translated from Russian) Got all the parts to piece it together, so its a helluva lot cheaper and quicker than buying a programmer online.
Board Life Status
Board startup date: April 23, 2017 12:45:48