_whitelogger has joined #apicula
<trabucayre> okay it's time for me to order a new trenz...
<trabucayre> pepijndevos: It's possible to change idcode :-/
FabM has joined #apicula
<pepijndevos> trabucayre, what happened?? You bricked your FPGA?
pepijndevos changed the topic of #apicula to: Project Apicula: bitstream documentation and tooling for Gowin FPGAs https://github.com/YosysHQ/apicula -- logs https://freenode.irclog.whitequark.org/apicula
<pepijndevos> Just noticed we have logs now
<trabucayre> pepijndevos: more or less ...
<trabucayre> I've ordered a new one.
<pepijndevos> :( Gowin support was quite helpful when I bricked my Tang Nano fyi
<pepijndevos> But not sure how deeply bricked it is
<trabucayre> VLD sure
<trabucayre> Idcode too
<trabucayre> I will send a mail for VLD in a first time
<trabucayre> When I have something to order from mouser, maybe i'll add a couple of GW1N-9
<pepijndevos> ooh dat file mentions Eflash128kOuts so maybe embedded flash is accessible? Contents is just -1 though...
<trabucayre> try after disabling securebit
<trabucayre> or maybe the readable pattern
<trabucayre> see TN653
<pepijndevos> No I mean... the data in the .dat file is just -1 so I have no idea how it's connected. Just that it's... mentioned.
<trabucayre> a new item to have to the long list of "just mentioned"
<pepijndevos> lolsob
<trabucayre> ...
<pepijndevos> ohhhh, seems like all these weird numbers in the .dat file are reference into some "cru" table.
<pepijndevos> bleg...
<pepijndevos> I was hoping that with my renewed understanding I might find some new bits of info in the dat file but sadly no.
<pepijndevos> For IOB I use it to extract the portmap
<pepijndevos> So I was hoping I could do the same with the center tile clock muxes...
<pepijndevos> instead of some super elaborate and/or incomplete fuzzing
<trabucayre> Gowin ask for screenshot... I'm not sure to have a solution since my FPGA has a wrong idcode ... Snif
<pepijndevos> Sad. What's this idcode used for exactly?
<trabucayre> each gowin (gw1n-1, gw1n-9, ...) has a specific idcode
<trabucayre> 2/ for erase flash timings are not same between 1 and 9
<trabucayre> so 1/ it's used to check if it's the correct bitstream
<pepijndevos> ... how did you manage to change it?
<trabucayre> I don't know
<trabucayre> wrong manip...
<trabucayre> gw1n-6
<trabucayre> maybe it's magic
<pepijndevos> Do you know what the new idcode is? Or it's some magic random number?
<pepijndevos> And uh, if you make a hacked up bitstream with that idcode but tell your programmer to act as if it's 9?
<trabucayre> not working
<trabucayre> I need to try
FabM has quit [Quit: Leaving]
<trabucayre> next time I'll double check before trying stupid thing :)
<trabucayre> with loading a bitstream for GW1N-6 in flash the FPGA seems to accept
<trabucayre> checksum is good
<trabucayre> but since it's the same package ...
<pepijndevos> Wait wat, you can actually load a wrong bitstream now? Or is 6 just a "disabled" 9 and it actually works?
<trabucayre> wrong bitstream with the correct idcode
<trabucayre> I assume
<trabucayre> must doing better test
<trabucayre> I will trying to build a bitstream for gw1n-9 without crc check an just changing idcode to try
<pepijndevos> dead is brain from thinking about fuzzers
<pepijndevos> I added the clock routing data from the vendor file, which is a start but pretty useless.
<pepijndevos> I guess next week I'll take a stab at fuzzing again
<trabucayre> big work