#timvideos

That's funny... I thought I was in this room

mithro: Are you around right now?

I am kinda

I just checked out symbiflow-arch-defs to my netbook... fresh copy

I get this weird error when running make redir

make/gen.mk:5: *** No configs found for mux. Stop.

Ever see anything like this?

What does "git status" show?

On branch master

Your branch is up-to-date with 'origin/master'.

nothing to commit, working directory clean

Netbook _is_ a Linux, btw

cr1901: I was after the commit version and the status of your submodules

Forgot to check out submodules... again...

and my personal copy isn't up to date w/ yours on master

Okay, still the same problem after git submodule update --init... commit is 1d4f491

commit 84e9ef3d ?

doesn't match any know commit... h/o

hrm? https://github.com/SymbiFlow/symbiflow-arch-def...

Title: Merge pull request #217 from SymbiFlow/dependabot/submodules/third_pa… · SymbiFlow/symbiflow-arch-defs@84e9ef3 · GitHub (at github.com)

oh I was tracking your copy

let's see...

mithro: Same problem.

What does git-status show now?

HEAD detached at 84e9ef3

nothing to commit, working directory clean

>>

submodules checked out

Hrm

netbook is 32 bit

guess I'll just edit over the network and rsync the output files

from home machine

I just did a fresh checkout and got the following

ls: cannot access 'symbiflow-arch-defs-test/third_party/*/.git': No such file or directory

make/inc/common.mk:23: *** Submodules are not initialized! -- Run git submodule update --init. Stop.

Title: Makefile fails in confusing way if Make version 3.82 is used. · Issue #197 · SymbiFlow/symbiflow-arch-defs · GitHub (at github.com)

cr1901: I think that is your issue

Excellent...

cr1901: I'd love you to fix the issue if you know of a good way too

Hopefully we will replace my horrific makefile system with something better soon -> https://github.com/SymbiFlow/symbiflow-arch-def...

Title: CMake build system for architecture definitions and FPGA targets by litghost · Pull Request #215 · SymbiFlow/symbiflow-arch-defs · GitHub (at github.com)

Oh that would explain a lot... my netbook is still running Ubuntu 14.04

mithro: Not offhand... going compile my own copy of make in the interim

cr1901: I think we just need some type of version check

Oh I thought you meant "make it work with 3.82 (or 3.81, my version)"

Funny, right now I'm trying to get Ubuntu 14.04 running on my system. I have some ancient software that doesn't seem happy with anything newer.

I'm heading out now

alright, thanks for the help

xobs: See my comments from earlier

xobs: Well make sure your GNU make is up to date, and if not make sure to fetch it from your neighborhood-friendly ftp mirror :)

mithro: yes, the provider idea is an interesting one.

cr1901_modern: I'm trying to get a VLSI tool from 2015 building, but it wants Boost.Python 2.8, Qt4, and to be compiled in a specific directory. It's not the best build experience.

the hell is python 2.8

Oh, and I just realized this isn't going to work- tfw no 32-bit build of conda

2.7. Phone keyboard, sorry.

xobs: have you looked at qflow ?

Title: Qflow 1.1 (at opencircuitdesign.com)

mithro: Yea, I'm going to have to remotely edit stuff for this, b/c there is _no_ 32-bit Conda :P. I should consider upgrading my netbook- oh wait...

cr1901: if you don't take "make env" it will used stuff from you system

It looks like some stuff is hardcoded to look in conda/env/bin? Or is that only if you run "make env"?

In any case, noted. I have everything but node installed I think. How bad could that be to set up?

Only if you use make env

alright, I'll do git clean -f -d and start over

mithro: I've looked at qflow -- it mostly works, and sometimes I get a synthesized chip out of it, but graywolf is a bit dodgy legally and is rather slow. I'm hoping to use the placer from Coriolis instead.

If it works out, I'll redo the build system to use something more sane. And port it to Python3. And Qt5.

mithro can you check out #tomu please?