<azonenberg> Wanda[cis]: wrt "empty bitstream capable of damaging the device" that has happened before on other fpgas
<azonenberg> look at the arria 10 transceivers
<azonenberg> So i would not entirely discount the possibility
<vup> wtf, that is cursed. Do you have any idea what would cause that?
FireFly has joined #prjcombine
<Wanda[cis]> ... heh
<Wanda[cis]> funnily enough, virtex 6 GTX transceivers and virtex 4 DCMs have the same problem
<Wanda[cis]> but that's electromigration slowly killing the device if it's powered on without clock toggling
<Wanda[cis]> not missing level shifter, which I'd imagine to be a ... more immediate problem
<Wanda[cis]> you've heard of UltraScale LVS transceivers taking microseconds to tristate. get ready for UltraLite, which one-ups them with plain IO!
<Wanda[cis]> (today in things that broke my SDF parser)
<Wanda[cis]> s/LVS/LVDS/
<whitequark[cis]> rip
<_whitenotifier> [prjunnamed/prjcombine] wanda-phi pushed 1 commit to main [+0/-0/±27] https://github.com/prjunnamed/prjcombine/compare/8933edb7aa40...f2694ecf370b
<_whitenotifier> [prjunnamed/prjcombine] wanda-phi f2694ec - siliconblue: extract SPI/I2C dedicated pins and high-drive IO groups.
<_whitenotifier> [prjcombine] wanda-phi edited issue #2: Finish `prjcombine-siliconblue` - https://github.com/prjunnamed/prjcombine/issues/2
<_whitenotifier> [prjcombine] wanda-phi edited issue #2: Finish `prjcombine-siliconblue` - https://github.com/prjunnamed/prjcombine/issues/2
<_whitenotifier> [prjcombine] wanda-phi edited issue #2: Finish `prjcombine-siliconblue` - https://github.com/prjunnamed/prjcombine/issues/2
<whitequark[cis]> i understand the naming now i think
<Wanda[cis]> fitting
DemiMarieObenour has quit [Quit: Idle timeout reached: 172800s]
<Wanda[cis]> oh. I see why all my attempts to reverse iCE65 IO failed in incomprehensible ways.
<Wanda[cis]> dbIceDeviceInfo::IsIce65 fucking returns false on the iCE65L04 because it compares the wrong thing.
<Wanda[cis]> sigh.
<Wanda[cis]> more binary patches? more binary patches.
<mupuf> ha ha, good luck!
<mupuf> How are you keeping track of your binary patches? hash of the binary, then offsets + new value?
<Wanda[cis]> I'm not
<Wanda[cis]> this wasn't much of a problem when there weren't many of them
<Wanda[cis]> but this is rapidly getting out of hand by now
<mupuf> how many are you at, by now? Is it the third?
<Wanda[cis]> I've lost count
<Wanda[cis]> hm
<Wanda[cis]> I think the SPI/I2C dedicated pins saga required like 5 patches?
<mupuf> oh boy
<mupuf> well, at least I hope you have a copy of the original file 😅
<Wanda[cis]> meh, the original file can be recovered from a fresh install
<Wanda[cis]> it's the patches that I should start actually gathering somewhere
<mupuf> yep. Are you aware of any standard format for that?
<Wanda[cis]> not really, but I guess there should be something?
<mupuf> or will it be a shell script full of dd:s 😃
<mupuf> both look very much overkill