<sb0>
whitequark, I suppose this clears any spurious IRQ that was here before the network driver was initialized
<sb0>
ethmac_sram_reader_ev_pending_write(ETHMAC_EV_SRAM_READER); probably doesn't do anything
<sb0>
since the code isn't even using ethmac_sram_reader_ev_pending_read() or IRQs
<sb0>
_florent_, ^
<cr1901_modern>
whitequark: What's wrong with the m-labs lm32 llvm backend (Pretend that it's in working condition)?
fengling has joined #m-labs
<bb-m-labs>
build #1188 of artiq is complete: Failure [failed python_unittest_1] Build details are at http://buildbot.m-labs.hk/builders/artiq/builds/1188 blamelist: whitequark <whitequark@whitequark.org>, Sebastien Bourdeauducq <sb@m-labs.hk>
<whitequark>
cr1901_modern: well it's not
<whitequark>
that's the problem with it
<cr1901_modern>
Was it never actually working?
<whitequark>
the codebase is a huge mess and from the look of it, misses vital parts
<cr1901_modern>
Okay now it's coming back to me... I seem to recall that as well. Doing that backend would actually be useful to me, since I spend most time w/ the LM32 target. Need to decide whether it's worth learning a broken 2 year old codebase tho.
<whitequark>
probably not hard to write it from scratch anyhow
<whitequark>
start with the or1k one
<cr1901_modern>
Cool, thanks for the hint
<GitHub26>
[misoc] whitequark pushed 1 new commit to master: https://git.io/vMtB9
<cr1901_modern>
Must be the stress of impedance matching :3
<key2>
could be
<cr1901_modern>
You do good work (even if it's not my thing :P)
<key2>
what is your thing
<cr1901_modern>
Mostly vintage stuff, tbh :/... trying to get old systems to do new things. I'm pretty clueless about exploits
<key2>
into video game ?
<cr1901_modern>
Yes! :) Mostly SNES, VB, and Genesis (Mega Drive). I also have a bit of experience with looking at old IC dies (b/c someone else taught me :P)
<cr1901_modern>
And of course, FPGAs are cool. MiSoC and Migen make me like FPGAs again
<key2>
you ever thought about implementing one of those console into a fpga ?
<cr1901_modern>
I have batted it around. Doing that right is an (extremely) nontrivial project, and I don't have the tools (such as a good logic analyzer) to do that properly.
<cr1901_modern>
Well, NES has FPGA impl already. Someone built a Megadrive PCB using off-the-shelf components (minus the video H/W). None open source tho.
<key2>
yeah those stuff are fun to do
<cr1901_modern>
I'm not really sure modern consoles (or anything modern piece of hardware, tbh) will be emulated well via RE. If timing of communication between parts matters (as it often does), anything with multiple clock domains is going to be tough to RE
<cr1901_modern>
But I'd love to be proven wrong
<key2>
well, is it really worth spending 4 month and get something that just buzz 5min and no way to make $ out of it ?
<cr1901_modern>
buzz 5mins?
<cr1901_modern>
key2: No, not really. I don't personally think a FOSS video game console can survive. I *can* imagine scenarios where having an "open and compatible" version of a closed piece of modern hardware would be worth it tho.
<cr1901_modern>
modern being a buzzword for "it's difficult to infer interactions between components from REing alone"
<key2>
well
<key2>
truely speaking
<key2>
the only thing that could be interesting is for example to make a console that could be anything
<key2>
sell the hardware
<key2>
then up to the final user to flash it to be a nes, megadrive, or something else
<key2>
by itself, its just a FPGA
<cr1901_modern>
FPGAArcade is similar to that. There was also the RetroVGS project, but the person who ran that was a pathological liar.
<key2>
anyone has tried the risc128 emulator from Bellard ?
<acathla>
Nope
<acathla>
cr1901_modern, what are doing with migen/misoc?
<cr1901_modern>
acathla: Trying to create a small PCB to drive an old FM synth chip and optionally use a soft-core version of said synth recently released. Prob gonna be two different boards
balrog has quit [Ping timeout: 258 seconds]
<acathla>
misoc is only for the soft core I guess...
<cr1901_modern>
correct :)
<acathla>
I'm trying to use specials to reuse a verilog code but it's not working. Do you have a good example? because the doc is... short on that subject.