[libre-riscv-dev] daily kan-ban update 12jun2020

[Cole Poirier]

On Jun 12 2020, at 8:53 am, Luke Kenneth Casson Leighton <lkcl at lkcl.net> wrote:
> i believe i did this initially :)
> 
> btw *do* create save files and actually save them regularly.  gtkwave
> has a
> habit of segfaulting, particularly if you use tabs.

Ooof. Thank you for this timely tip, before I've had to experience the
inconvenience of gtkwave segfaulting myself.

> if you do that on code that you have yourself written, and also viewed the
> graphviz diagrams, this cross-association will go faster.
> 
> can i suggest looking at the bperm unit test vcd files, first?

That is an excellent idea. Thank you. I think you suggested it on bug
#325 TRAP pipeline, but I was reading in chronological order and was
working based off of your earlier suggestion in comment #74 "ok. 
suggestion. take alu test_pipe_caller.py and cut out absolutely every
test_* except for one, make sure to pick one with only one assembly
instruction. (or take copy of tpc.py, call it tpc2.py) then run it and
load the generated alu_simulator.gtk in gtkwave." By the way I couldn't
find anything that contained either 'tpc' or 'tpc2' using grep and find.
Will look at bpermd in gtkwave now :)

> my partner marie has a contact at HK University, please do point her
> at it
> when you have it written. (a short wiki page is great).
> 
> we have something on the front page already however it is embedded
> *in* the
> front page and so is not really clear.
> 
> we also cannot call it "jobs" because sigh that implies salaries, per-hour
> payments, employment, that we will cover their insurance costs, holidays
> and so on.  wark-wark.

Appreciated. Created a bug report so that this doesn't get lost in "old
style 'keeping track of things' (an oxymoron) with email".
https://bugs.libre-soc.org/show_bug.cgi?id=379