I just got a Cu V2 which I’d also like to use with iceprog. (On a Mac, so options are limited.) Current versions of iceprog cannot seem to write-enable the EEPROM, but if I write anything first with the Alchitry app, it stays writable till power cycle. I’m perfectly willing to grovel through datasheets, but it seems like a thing someone might have on the top of their head.
As my father told me, “Sometimes a few months work in the laboratory can save you the effort of a few minutes in the library.”
That’s the code in Alchitry Labs/Loader that does the programming of the flash. Specifically line 188 which sends the write enable command (a single byte of value 0x06). I’m not sure what iceprog does.
Oh, good, another language I’ve never seen before. Can’t you guys wait until I’m dead before you invent any more languages? Fortunately, this one is pretty clear. Also, it obviously started as a line-for-line translation of iceprog, so I can probably just look for the parts that are different.
(“Wait until I’m dead” sounded funny when I wrote it. Now I’m not sure. Please accept apologies if it came off as mean.)
I think we all knew the spirit in which you said that. At least those of us who have been doing this for such a very long time now. I am digging into verilog as a kind of emotional rejection for how far off into the weeds C++ seems to be getting these days.
It’s probably wrong to drag a technical question off into casual conversation, but I’m a little dysregulated these days. There’s something really interesting in the way Verilog and VHDL, invented long ago, are still highly functional with only relatively small changes. Meanwhile, programming languages of that era (I learned in Fortran IV, on punch cards) have gone through multiple generations of changes, pretty much all of which have been important and served real needs.
Not enough traffic here to really have to care about topic policing, I think. And since we’re on the (veering) topic, I was getting so dysregulated that I resigned from my job this week! I am so, so burned out. This was an excellent decision though, not a crazy one. I’m going to take at least till the end of the year to recover and take care of myself.
I’ll be spending a lot of that downtime working on my FPGA project!
Hello,
I tested and witnessed that iceprog can’t write to the CuV2.
I checked the source code and it does the write enable command just like Labs does so it’s a mystery…
The good news is that I have two alternatives :
I tried with openfpgaloader instead and it works perfectly fine
It’s been a long time since I had a formal, W-2, show up and do your work type job. But I clearly remember that some of the happiest days of my life were the days on which I quit a job.
Yeah, the SST26VF… has some extra layers of write disable that the Winbond chips iceprog was written around didn’t. Including pins, so it might actually be the initialization of the FT2232 (or whatever) that’s different. Alchitry’s code only special-cases the SST so far as noticing it has a higher top speed; the hardware initialization might be something that doesn’t affect the other chip.
My ambition hasn’t actually risen to the point of doing something with the information I’ve collected yet. If it does, I’ll try to get icestorm to merge it.
