#timvideos

tumbleweed: I think Ryan has an Opsis here which reproduces the problem with the fx2 reset

tumbleweed: BTW - Were are we at with packaging the modeswitch tool?

deeprave: xfxf says you've done a bunch more work on the serial daemon thingy?

tumbleweed: I think I figured out your bug and it wasn't actually related to the fx2 at all....

CarlFK: do you have your netboot config/stuff anywhere you use for 16.04 images w/ your newer scripts?

i'm working on it myself, but i just realized i'm probably repeating what you've already done

tumbleweed: ping me when you are around

rohitksingh: Was it this weekend you are at the retreat thing?

mithro: I'm free but not in office. So won't have access to any hardware. I have my laptop with me, and my atlys board plus few other FPGA boards (zybo, saturn and arty) right now with me

rohitksingh: I'm currently working on a couple of small things for PyCon AU

rohitksingh: regarding the vga stuff we should just add it as part of the _video target

mithro: ohkay. I can work on merging vga stuff. And the components to add? 2x HDMI-iI, 2x HDMI-Out, encoder, vga, and pattern, right? (and no eth phy + mac)

rohitksingh: Hrm? Your just adding an extra vga target to the existing targets?

I mean, a vga option to the existing atlys_video target?

mithro: we need to remove ethernet, which comes from Minisoc class in base target...atlys_video inherits that

rohitksingh: so, the XXX_video target can already choose which thing it inherits from

See the very bottom of https://github.com/timvideos/HDMI2USB-misoc-fir...

Title: HDMI2USB-misoc-firmware/atlys_video.py at master · timvideos/HDMI2USB-misoc-firmware · GitHub (at github.com)

rohitksingh: https://github.com/timvideos/HDMI2USB-misoc-fir...

Title: HDMI2USB-misoc-firmware/atlys_hdmi2eth.py at master · timvideos/HDMI2USB-misoc-firmware · GitHub (at github.com)

rohitksingh: actually, it looks like we don't build ethernet by default anyway....

mithro: okay you are saying that I modify the atlys_video target itself. I initially thought, you wanted me to subclass atlys_video...

rohitksingh: Yeah, I think so

mithro: great

morning CARAM__

opps

morning CarlFK

hiya mithro

CarlFK: You where at some event recently?

CarlFK: with tumbleweed?

mithro: more than I can count..

CarlFK: Between Debconf and now?

er, tweed just 1.. me in the last 8 weeks: Write the Docs, Node PDX, OpenCamps, dot Net Fringe, PyGotham, PyOhio

tween at PyOhio

tweed

CarlFK: and you used Atlys boards at all of them?

mix of Atlys and Opsis

er.. and none.

OpenCamps/PyGotham at UN had their own screen capture stuff and gave me 1080i over sdi which is where I got the audio from, so used that for screen too

the one venue system that does what you want it to do

CarlFK: poke. did you see my message to you earlier before you quit re your PXE scripts (particularly late.sh) for 16.04?

also, should i be using/altering your ingest script in your repo, or something else?

xfxf: didn't see anyting about pxe... guessing IRC which is not a good way to get messages to me asynchronously ... get my attention first. otherwise email

ingest in my repo is 'top'

also https://github.com/CarlFK/dvsmon/blob/master/vo...

Title: dvsmon/vocto-prod1.py at master · CarlFK/dvsmon · GitHub (at github.com)

CarlFK: can you point me exactly to what i should be altering my PXE install server with, in terms of stuff you use now?

from looking through the PXE stuff today i assume this is copied off the machine you use to image up your machines?

xfxf: it is kinda a copy from some point in the near past.. I edit the live server and sometimes get the stuff into my git repo

xfxf: give me a sec and I'll update it

github will show you the diffs

xfxf: how soon do you expect to start building machines with it?

CarlFK: I got the front panel IO header working on the Opsis board, so you can get status and reset it using the case

CarlFK: I think the answer is "about 12 hours ago" :P

CarlFK: i am right now and i've already got most of a 16.04 install working, but i figured i'd ask you as well

xfxf: um.. it should just work..

I happened to use it to bring up 4 more machines about 3 weeks ago

xfxf: any point in me getting my repo up to date?

well. any point in me scrambling today

if it's not something you can throw up easily, probably not

i'm mostly through it, just solving some issues with late.sh

and trying to get it to install mode-switch

yeah.. some stuff is crufty, like I install the black magic driver .deb

which means you need to get it, make sure the tar file name is right (changes with driver version...

IIRC the version i have predates you doing that

this is from what you + tumbleweed did back at LCA2016

xfxf: you should use my current stuff

but oh well

so upload it :P

i'll figure out what needs altering

(09:15:18 AM) xfxf: if it's not something you can throw up easily, probably not

also to make this more fun the SSD this was all on seemed to corrupt itself

so i had to install it all from scratch and copy over the files

but that was relatively straight forward

um.. I have an install script

that is described in the readme

oh i have all the stuff from LCA, just the OS itself was corrupt

all of the pxe/http/etc bits were fine

um.. you arn't making much sense now

the filesystem on the avserver got corrupt, it wouldn't boot

i took a backup of it and installed 16.04 onto it on a newer SSD

then reinstalled all of the packages, pushed relevant bits from the backup back in (config for dhcp, nginx, squid, etc etc), put all of the files in /var/www/html/ back in, various symlinks, etc

appears to be working, just installation dies near the end because something isn't quite right, fix that, try install again

xfxf: do you need help?

unless you can upload what you currently use, probably not

the issues i hit are obvious, it'd just be shortcutted by using a working config