*** tpb has joined #yosys | 00:00 | |
*** emeb has quit IRC | 00:01 | |
*** emeb_mac has joined #yosys | 00:04 | |
*** seldridge has quit IRC | 00:32 | |
*** srk has quit IRC | 00:46 | |
*** srk has joined #yosys | 00:51 | |
*** pie_ has quit IRC | 00:51 | |
*** pie_ has joined #yosys | 00:51 | |
*** captain_morgan has quit IRC | 01:05 | |
*** captain_morgan has joined #yosys | 01:05 | |
*** SpaceCoaster has joined #yosys | 01:09 | |
*** X-Scale has quit IRC | 01:21 | |
*** leviathan has joined #yosys | 02:01 | |
*** cemerick_ has joined #yosys | 02:17 | |
*** kmehall has quit IRC | 02:58 | |
*** eduardo has joined #yosys | 03:29 | |
*** kuldeep has quit IRC | 03:35 | |
*** kuldeep has joined #yosys | 03:36 | |
*** m_w has quit IRC | 03:40 | |
*** cemerick_ has quit IRC | 03:46 | |
*** kmehall has joined #yosys | 03:57 | |
*** xrexeon has quit IRC | 04:33 | |
*** leviathan has quit IRC | 04:43 | |
*** proteus-guy has quit IRC | 04:47 | |
*** proteus-guy has joined #yosys | 04:48 | |
*** Gravis has joined #yosys | 04:56 | |
*** Gravis has left #yosys | 04:59 | |
awygle | daveshah: ubuntu still segfaults and arch is a bummer | 05:29 |
---|---|---|
*** proteus-guy has quit IRC | 05:38 | |
*** proteus-guy has joined #yosys | 05:46 | |
daveshah | awygle: can you check the core dump and get a stack trace (might have to rebuild as Debug)? | 05:55 |
awygle | daveshah: it can't find a libboost-python symbol | 05:55 |
daveshah | What happens if you run the tests inside libtrellis/tests? | 05:55 |
awygle | i will do that tho, hang on | 05:55 |
daveshah | awygle: OK, I see | 05:55 |
daveshah | tests won't run then either | 05:55 |
daveshah | Can you post your modified CMake file? | 05:56 |
awygle | https://gist.github.com/awygle/60b40d69bf217e12289061d47ccc7d8e dis | 05:58 |
tpb | Title: gist:60b40d69bf217e12289061d47ccc7d8e · GitHub (at gist.github.com) | 05:58 |
awygle | "ImportError: /usr/lib/x86_64-linux-gnu/libboost_python-py35.so.1.62.0: undefined symbol: PyUnicode_AsUTF8String" | 05:59 |
awygle | (from the tests, i got a different symbol from trying to run a real thing) | 05:59 |
*** GuzTech has joined #yosys | 05:59 | |
awygle | hm maybe i should downgrade | 06:00 |
daveshah | Can you post python --version? | 06:00 |
awygle | 2.7.14 and 3.6.3 | 06:01 |
awygle | so that's probably not idea | 06:01 |
awygle | l | 06:01 |
daveshah | 2.7.14 won't work | 06:01 |
daveshah | 3.6.3 should | 06:01 |
daveshah | But linking with 3.5 seems wrong then | 06:02 |
awygle | yeah idk what i was doing here | 06:04 |
daveshah | Can you put 3.6 after Python Libs? | 06:05 |
daveshah | *PythonLibs? | 06:05 |
daveshah | And possibly PythonInterp | 06:06 |
daveshah | My best guess is that the default Python on Ubuntu is 2.7, and that breaks stuff | 06:07 |
daveshah | Didn't realise that was still the case, living inside an Arch bubble... | 06:07 |
awygle | hm so it's not finding the 3.6 lib or interp for some reason | 06:08 |
awygle | they're there, but libpython3.6 is libpython3.6m for some reason, idk if that matters | 06:08 |
daveshah | Maybe you need libpython3.6-dev? | 06:09 |
daveshah | Hmm, not sure how that ended up bold | 06:09 |
awygle | i have that | 06:10 |
awygle | derp. i had to nuke my cmake stuff apparently. | 06:11 |
daveshah | And python3-dev? | 06:12 |
*** AlexDaniel has quit IRC | 06:12 | |
daveshah | Ah, that makes sense | 06:12 |
awygle | well only kind of lol. it was picking up needing 3.6 but somehow not realizing that meant it had to go look again. | 06:13 |
daveshah | Yep, it caches dependency checks | 06:13 |
daveshah | Now I see why mithro is using conda in symbiflow-arch-defs | 06:14 |
awygle | btw are you building from a build/ directory with cmake ../ or in libtrellis? | 06:15 |
daveshah | TBH I'm just building in libtrellis | 06:15 |
daveshah | But could be worth adding a command to copy the library at the end so out of tree builds can be done too | 06:16 |
awygle | oh this looks like fun. "dynamic module does not define init function". | 06:18 |
daveshah | What PyInit is it looking for? | 06:20 |
mithro | daveshah: lol :-P | 06:21 |
awygle | "initpytrellis" | 06:21 |
daveshah | Ah, stupid me | 06:22 |
daveshah | Change python in run_all.sh to python3 | 06:23 |
awygle | ahh yes | 06:23 |
awygle | okay passed! sweeeeet | 06:23 |
daveshah | Sweeeet | 06:23 |
daveshah | I think 90% of the issues were Python 2/3 related, and should be fairly fixable | 06:24 |
daveshah | But the CMake will now need a bit of work to not require one exact version | 06:24 |
awygle | yup | 06:24 |
awygle | also, devices.json should be in database/ according to the fuzzer script | 06:26 |
daveshah | Database should be a clone of prjtrellis-db, as in prjxray | 06:27 |
awygle | ah | 06:27 |
daveshah | There will be scripts to do that, and also create it from scratch | 06:27 |
daveshah | You will also need to create skeleton database files using create_empty_bitdbs | 06:28 |
*** GuzTech has quit IRC | 06:31 | |
*** anindita has joined #yosys | 06:35 | |
anindita | Hi, we are facing an issue with Yosys synthesis. We have a 4-bits state register. After synthesis, somehow we get 12-bits state register. On simulating gate-level synthesized file, we also see that this register does get reset to binary 12'b0 but to 12'b1. Any clues to what may be happening? Thanks. | 06:39 |
daveshah | anindita: if you have a fsm pass in your script, then yosys may recode the FSM to one-hot | 06:43 |
daveshah | Which is what is happening here | 06:43 |
daveshah | Either pass -norecode to fsm, or set fsm_encoding to binary | 06:44 |
*** proteusguy has quit IRC | 06:46 | |
*** dys has quit IRC | 06:56 | |
*** GuzTech has joined #yosys | 07:13 | |
*** emeb_mac has quit IRC | 07:18 | |
*** proteusguy has joined #yosys | 07:19 | |
*** kraiskil has joined #yosys | 07:33 | |
*** dmin7 has joined #yosys | 07:38 | |
*** jwhitmore has joined #yosys | 07:39 | |
*** proteusguy has quit IRC | 07:45 | |
*** proteusguy has joined #yosys | 08:00 | |
*** proteusguy has quit IRC | 08:01 | |
*** proteusguy has joined #yosys | 08:03 | |
*** proteusguy has quit IRC | 08:07 | |
*** proteusguy has joined #yosys | 08:08 | |
*** m_w has joined #yosys | 08:40 | |
*** jwhitmore has quit IRC | 08:41 | |
*** m_w has quit IRC | 08:45 | |
*** promach_ has quit IRC | 08:53 | |
*** fsasm has joined #yosys | 09:26 | |
*** sklv has quit IRC | 09:35 | |
*** sklv has joined #yosys | 09:36 | |
*** X-Scale has joined #yosys | 09:49 | |
*** leviathan has joined #yosys | 10:05 | |
*** leviathan has quit IRC | 10:08 | |
*** leviathanch has joined #yosys | 10:08 | |
*** dys has joined #yosys | 10:08 | |
*** SpaceCoaster has quit IRC | 11:05 | |
dmin7 | hi, quick question: is that a pure SPI slave implementation? https://github.com/cliffordwolf/icotools/blob/master/icosoc/mod_spi/mod_spi.v | 11:09 |
tpb | Title: icotools/mod_spi.v at master · cliffordwolf/icotools · GitHub (at github.com) | 11:10 |
*** proteusguy has quit IRC | 11:10 | |
*** SpaceCoaster has joined #yosys | 11:11 | |
dmin7 | (it does seem to write spi_sclk (and also CS when asked to), so i guess it shouldn't do that as a slave x) | 11:11 |
dmin7 | s/slave/master in the question* | 11:12 |
*** jwhitmore has joined #yosys | 11:31 | |
daveshah | dmin7: it's a pure master | 11:34 |
daveshah | it has no wiring to read the clock even | 11:34 |
dmin7 | yes, thx for confirming (: .. looking at other SPI slaves, i think i might be able to write myself one of those :) | 11:36 |
* dmin7 or hope so at least | 11:36 | |
daveshah | dmin7: SPI slave is not too bad if you just need a simple custom application | 11:37 |
daveshah | a generic SPI slave for all applications would be harderr | 11:37 |
dmin7 | (actually, i'ld rather use it as master, but it seems there is no "hardware" SPI slave on the pi :/) | 11:37 |
*** proteusguy has joined #yosys | 11:41 | |
*** jwhitmore has quit IRC | 11:53 | |
*** cemerick_ has joined #yosys | 12:33 | |
*** eduardo_ has joined #yosys | 12:43 | |
*** pie_ has quit IRC | 12:43 | |
*** eduardo has quit IRC | 12:46 | |
*** promach_ has joined #yosys | 12:55 | |
*** cemerick_ has quit IRC | 13:00 | |
*** AlexDaniel has joined #yosys | 13:02 | |
*** pocha has joined #yosys | 13:42 | |
*** pocha has quit IRC | 14:17 | |
*** pie_ has joined #yosys | 14:28 | |
*** kraiskil has quit IRC | 14:28 | |
*** jwhitmore has joined #yosys | 14:29 | |
*** emeb has joined #yosys | 14:42 | |
*** seldridge has joined #yosys | 14:55 | |
*** Lucas_ has joined #yosys | 15:03 | |
*** anindita has quit IRC | 15:17 | |
*** GuzTech has quit IRC | 15:27 | |
*** xrexeon has joined #yosys | 15:38 | |
*** xrexeon has joined #yosys | 15:38 | |
*** fsasm has quit IRC | 15:43 | |
*** m_t has joined #yosys | 15:45 | |
*** GuzTech has joined #yosys | 16:17 | |
*** pocha has joined #yosys | 16:31 | |
*** indy has quit IRC | 17:16 | |
*** indy has joined #yosys | 17:19 | |
*** leviathanch has quit IRC | 17:23 | |
*** pocha has quit IRC | 17:56 | |
*** jwhitmore has quit IRC | 18:48 | |
*** fsasm has joined #yosys | 18:56 | |
*** dys has quit IRC | 19:01 | |
keesj | https://hackaday.com/2018/05/18/arduino-just-introduced-an-fpga-board-announces-debugging-and-better-software/ | 19:09 |
tpb | Title: Arduino Just Introduced an FPGA Board, Announces Debugging and Better Software | Hackaday (at hackaday.com) | 19:09 |
*** ravenexp has quit IRC | 19:19 | |
*** ravenexp has joined #yosys | 19:24 | |
shapr | looks like an Intel Cyclone 10 | 20:22 |
qu1j0t3 | yeah i think Max 10 .. the product page is better than this article | 20:22 |
qu1j0t3 | no prices yet | 20:22 |
qu1j0t3 | but looks pretty nice | 20:22 |
shapr | 10CLO16 if the LUT count is right | 20:23 |
shapr | qu1j0t3: what's Max 10? | 20:23 |
qu1j0t3 | shapr: https://store.arduino.cc/arduino-vidor-4000 hm, weird, i thought it was named here. | 20:25 |
tpb | Title: Arduino Vidor 4000 (at store.arduino.cc) | 20:25 |
qu1j0t3 | but i don't see it now | 20:25 |
qu1j0t3 | eh well ignore me, i don't know where i read that it was a Max 10. seems it is actually a Cyclone. | 20:28 |
shapr | qu1j0t3: I just magnified the chip on the board until I could read the text | 21:29 |
shapr | couldn't find any other info | 21:29 |
*** m_t has quit IRC | 21:29 | |
qu1j0t3 | shapr: ah ok! | 21:30 |
qu1j0t3 | some releases use the word Cyclone so i'm sure you're right | 21:30 |
awygle | probably the second least free fpga | 21:44 |
*** seldridge has quit IRC | 21:45 | |
*** GuzTech has quit IRC | 21:51 | |
*** mjoldfield has joined #yosys | 21:54 | |
*** fsasm has quit IRC | 21:55 | |
*** quigonjinn has quit IRC | 21:57 | |
*** quigonjinn has joined #yosys | 22:21 | |
*** Ristovski has joined #yosys | 22:53 | |
*** dmin7 has quit IRC | 23:00 | |
*** jkiv has joined #yosys | 23:15 | |
*** m_w has joined #yosys | 23:18 | |
*** cr1901_modern has quit IRC | 23:44 | |
*** cr1901_modern has joined #yosys | 23:45 |
Generated by irclog2html.py 2.13.1 by Marius Gedminas - find it at mg.pov.lt!