18:49 <Lofty> ping kgugala, arvindsrinivasan

18:49 <sf-slack4> <arvindsrinivasan> hi

18:49 <Lofty> Would you be willing to test an alternate fix for the ABC9 assert?

18:49 <sf-slack4> <kgugala> yep?

18:50 <sf-slack4> <kgugala> sure

18:50 <_whitenotifier-5> [sphinx-verilog-domain] mithro opened issue #25: Relicense the project to Apache 2.0 - https://git.io/JtpAt

18:50 <Lofty> https://gist.github.com/Ravenslofty/00809d98925cdaf1bc21a8312674bc3b

18:50 <Lofty> You'll have to apply it "manually"

18:51 <Lofty> I think this is a case of "terrible error message" on the part of ABC9

18:53 <Lofty> Hang on, let me find what the file is on GitHub

18:54 <Lofty> https://github.com/SymbiFlow/symbiflow-arch-defs/blob/master/xc/xc7/techmap/cells_sim.v#L88

18:54 <Lofty> This should be lib_whitebox, I believe

18:55 <sf-slack4> <kgugala> seems to be helping

18:56 <sf-slack4> <kgugala> at least in sythesis

18:56 <Lofty> This passes synthesis for me

18:56 <sf-slack4> <kgugala> @Lofty can you open a PR so the CI can run the whole flow with it?

19:00 <Lofty> Will do

19:03 <sf-slack4> <kgugala> thanks

19:13 <_whitenotifier-5> [sphinx-verilog-domain] mithro opened issue #27: Change from Travis to GitHub Actions - https://git.io/JtppH

19:16 <mithro> litghost: BTW You might be interested in https://github.com/marketplace/actions/validate-yaml-schema and https://github.com/marketplace/actions/yaml-schema-validator

19:17 <litghost> There is also https://pypi.org/project/yamllint/

19:17 <tpb> Title: yamllint · PyPI (at pypi.org)

19:17 <litghost> Which is easy to run locally

19:19 <mithro> https://github.com/marketplace/actions/yaml-schema-validator seems kind of interesting from the fact it seems to use the same settings as VSCode

19:42 <sf-slack4> <arvindsrinivasan> @lofty do you still want me to test that change?

19:43 <sf-slack4> <arvindsrinivasan> Whiich ABC assert do you want me to comment back in? the empty cell ones?

19:46 <Lofty> Yes

19:51 <sf-slack4> <arvindsrinivasan> Alright fixing now

20:13 <Lofty> litghost: I think fundamentally the problem we have here is that boxes are implicitly sweepable.

20:15 <litghost> I guess to put this another context, what do you do about things like CFGLUT5? Because of what CFGLUT5 is (re-configurable LUT) it is not legal to sweep it, but if that CFGLUT5 is part of a logic chain, you would want to be able to do path analysis through the element

20:18 <Lofty> I suspect this is might need support adding to ABC [the program] for this kind of thing

20:19 <Lofty> Well, the other (obviously suboptimal) approach is to not invoke the ABC9 sweeper

20:19 <Lofty> Might be time to write an email to Alan, too

20:31 <Lofty> Well, email to Alan sent, let's see how that goes

20:40 <sf-slack4> <arvindsrinivasan> So on my system I pass synthesis but fail alter on

20:42 <sf-slack4> <arvindsrinivasan> ```/home/arvindsrinivasan/opt/symbiflow/xc7/install/bin/vpr_common: line 94: 16262 Bus error (core dumped) vpr ${ARCH_DEF} ${EBLIF} --device ${DEVICE_NAME} ${VPR_OPTIONS} --read_rr_graph ${RR_GRAPH} --read_router_lookahead ${LOOKAHEAD} --read_placement_delay_lookup ${PLACE_DELAY} ${SDC_OPTIONS} $@```

20:42 <sf-slack4> <arvindsrinivasan> @kgugala did you get the same error?

20:44 <sf-slack4> <kgugala> I just checked synthesis (and waiting for CI here https://github.com/SymbiFlow/symbiflow-arch-defs/pull/2061 tot test the rest)

20:45 <sf-slack4> <kgugala> and it looks CI failed there

21:54 <_whitenotifier-5> [nextpnr] litghost opened issue #238: Need to stand up MSAN / ASAN / TSAN test runs - https://git.io/JthYt

22:06 <_whitenotifier-5> [RapidWright] litghost opened issue #1: Largest fabrics consume all memory and die when writing device database - https://git.io/JthOO

22:26 <sf-slack4> <arvindsrinivasan> @lofty any suggestions for other stuff to test to see if we can get a fix?

22:27 <Lofty> Well, fixing this particular case is easy; fixing the problem in general is hard

23:00 <sf-slack4> <arvindsrinivasan> Alright for now, I guess the easy answer is to comment out the asserts

23:19 <mithro> https://byu-cpe.github.io/ComputingBootCamp/ <-- "This bootcamp serves two purposes in terms of topical content. The second is FPGA-specific topics such as Tcl/Vivado and prjxray."

23:19 <tpb> Title: Home • Immerse Computing Bootcamp (at byu-cpe.github.io)