4 ms·
Using JTAG to dump parallel NOR flash
- Boardbricker35 3y agoSome questions I recvd (author) : 1) Is there a process by which you can do the same in-circuit, similar to the SOP8 footprints that most people are familiar with? Answer: 1) YES! You totally can. In fact this is a preferred method because of its ease if there is ejtag or a jtag header and the flash is connected.Here are a few ways: A) connecting to a target MCU/FPGA to dump over a preexisting bus. B) using a large clip or soldering leads. Not recommended. May backfeed. 2) Are there any hidden gotchas (similar to NAND flash) that people should know about where reading or writing may not result in a 1:1 copy (which can be used later to revert to a known good working state, if the chip is not write protected)? Answer: 2) Not really reliable with this method. Urjtag is software written in 2006 … so it is kinda dated. Topjtag is also quite dated and both are slightly unreliable. It is up to an engineer to really understand what they are doing and the dangers associated with dumping flash memory. Programmers are fairly inexpensive now so this method is more or less a last resort in most cases (wanted to share after trying this). However can be successful if the engineer does this correctly and takes their time. The first time I tried this i grounded the wrong pin. Power cut through the flash chip and destroyed it and took all the data with it. 3) How do you handle these chips when there are no markings, or the chips themselves have incomplete publicly available specifications? Some parallel memory chips I've run into seem to have fuses built-in to the silicon so if you pass voltage or hook up the wrong pin, the entire chip ceases functioning. Answer: 3) A) Start with a logic analyzer https://www.ebay.com/itm/225946293211?chn=ps&norover=1&mkevt=1&mkrid=711-166974-028196-7&mkcid=2&mkscid=101&itemid=225946293211&targetid=2274564709393&device=m&mktype=pla&googleloc=9008162&poi=&campaignid=21023393889&mkgroupid=159171467659&rlsatarget=pla-2274564709393&abcId=9382268&merchantid=5314075636&gad_source=1&gbraid=0AAAAAD_QDh97h8Ad_8VfL1ZjxQgqtK-qZ&gclid=Cj0KCQjwwYSwBhDcARIsAOyL0fiEPokY5TEZ_iwb1-VfGbWTV00krOhtlg0PbGfo7egIerh-bQrX6dYaAodoEALw_wcB https://www.ebay.com/itm/225946293211?chn=ps&norover=1&mkevt... and look at the package size. More than likely if I look at all 24 pins on one side of the chip with a logic analyzer I can collect enough data to start looking for the pins that are data and map them to a datasheet. For example I see data on pins 1-x and x+2-23 and x+1 is vcc and 24 is vss. Then map that so a package like tsop48 and narrow it down. Incomplete public… that is more difficult. Id use the logic analyzer and see how the MCU is taking to it and see if a protocol exists for it and mimic a read if a protocol exists with pins I find. If not… ill cross that hill when I get to it. B) Oh I wish I had a FIB to edit silicon with. I have not had this issue … ever … so far. Im sure that will change. But ive heard of cracking the lid on a die with a laser and then using a fib to bridge the fuse. Very expensive. Ive also heard of only reading a byte or some small amount of data at a time on powerup, powering down and cycling until full flash is read. I guess it would really depend on the size of the fuse, gap size, and how you can get to it easiest or bypass it. Here is a link to a FIB: https://www.thermofisher.com/us/en/home/electron-microscopy/products/dualbeam-fib-sem-microscopes.html https://www.thermofisher.com/us/en/home/electron-microscopy/... Again, do be careful when doing this kind of stuff. Go to goodwill, get some really solid practice on some old netgear routers or other embedded systems before jumping into the high dollar stuff. It helps keep your cool and also understand the process more. Repetition is king in our field … if you can get it. C) If the target is an FPGA, its a bitstream. And bitstreams are very, very painful to re. The sea of gates is very big. Id recommend looking at something with a tied architecture and older. Thank you all so much for reading my article. I hope it helped and you got something out of it or will try this in your free time. Keep your eyes peeled for more articles.
- catkitcourt 3y agoSeems requiring IO of debugged device is directly connected to scan registers, like a FPGA.
- nneonneo 3y agoFigure 33's hexdump and Figure 35's binwalk output look suspicious to me. The hexdump is too sparse - lots of zero bits but no long runs of 0x00, consistent 0x00, 0x01 or 0x10 in odd-numbered positions, etc. It might be valid, or it might point to some kind of clocking or pinout issue. The binwalk output is even more suspicious - the signatures are likely all spurious. One does not expect a Flash ROM to consist solely of MySQL databases, particularly not random versions; this is likely the result of random bytes happening to match the file signature. As such, there don't seem to be any recoverable files in the dump per binwalk.
- chc4 3y agoYeah, I'd suspect that binwalk output to be junk from the flash tripping heuristics: "MySQL database" is one of those heuristics that a lot of random binary files trip spuriously for who knows what reason. Practically every other read word is `10 00 10 00` which is definitely strange, but also it's hard to judge if that's wrong without looking at the dump more IMO.
- monocasa 3y agoYeah, there's no way it's running MySQl. I'd expect to see a CPIO archive, a UBoot copyright string, or something in that wheelhouse at the top of binwalk given that it's the NOR flash to a mediatek soc.
- Boardbricker34 3y agoThe binwalk is more than likely not aligned for the endianess that is used. This dump was made to just be an example that magic existed and there was data that came from the flash chip using this method. Not necessarily the binary enumeration of the dump. The magic binwalk found is more than likely junk. If one was looking to flip endianess on a dump, use the -swap and it will flip around strings.
- dmitrygr 3y agoThis is a rather convoluted way to do this. Once you have the socket, just connect with wires to a pi pico and dump it over serial.
- monocasa 3y agoYeah, about the only way this would make sense is if they used the boundary scan of the original board's SoC so they didn't have to desolder anything.
- Boardbricker34 3y agoThis also does work. Some devices do not have the ability to connect to a part though (no bsdl, no jtag header)
- monocasa 3y agoGiven that the top bit of every 16bit word looks to be set, I'm not sure that it does work.
- Boardbricker34 3y agoThis would work. And I have done this. You need a jtag port and the flash needs to be directly connected to the SOC/MCU/FPGA. This is similar to how EJTAG works but without ejtag, you need to specify the pins themselves. As long as all 16 data and addr pins are connected, this is possible.
- monocasa 3y agoI know it can be done. NOR boot flashes like this were typically written and verified via boundary scan at manufacturing time. That's probably how this flash was initially written. I question whether you successfully performed that operation.
- Boardbricker34 3y ago