**** BEGIN LOGGING AT Fri Jan 04 02:59:58 2013
Jan 04 04:21:52 <ka6sox-away> emeb R2R DAC
Jan 04 04:22:01 <ka6sox-away> its UGLY but works.
Jan 04 04:29:59 <emeb_mac> ka6sox-away: true.
Jan 04 04:31:18 <ka6sox-away> the impedance of it sucks....but with a proper follower after its fine.
Jan 04 04:32:32 <emeb_mac> big problems are glitches, power noise feedthru, clock feedthru, INL/DNL due to resistor matching.
Jan 04 04:33:15 <ka6sox-away> all of the above are true...but for Q/D it "works"
Jan 04 04:33:28 <emeb_mac> all hail Q/D. :)
Jan 04 04:58:41 <mranostay> Q/D ?
Jan 04 04:59:29 <emeb_mac> quick & dirty
Jan 04 05:17:25 <ka6sox> which is similar to UBE
Jan 04 05:17:37 <ka6sox> (they are related)
Jan 04 05:27:02 <mranostay> ka6sox: so you and emeb_mac have the 3rd place in 'votes' atm
Jan 04 05:27:17 <emeb_mac> buh whut?
Jan 04 05:27:49 <mranostay> emeb_mac: entries page vote == +1 and FB likes
Jan 04 05:28:29 <emeb_mac> huh. don't have a FB for it either.
Jan 04 05:30:30 <emeb_mac> I put it down to Locutus.
Jan 04 05:30:39 <emeb_mac> You can't say no to Locutus.
Jan 04 05:31:22 <emeb_mac> or rather, "Resistance is futile"
Jan 04 05:32:22 <emeb_mac> looks like mranostay is way out ahead with the geiger cape.
Jan 04 05:35:39 <ka6sox> mranostay, what page is this?
Jan 04 05:35:57 <emeb_mac> mranostay: giger? http://1.bp.blogspot.com/-4MvcA9tN8Pw/TZ8l6AmDRcI/AAAAAAAAANo/YJI-um9DpDs/s1600/hr_giger_csg081.jpg
Jan 04 05:36:36 <emeb_mac> ka6sox: http://beagleboard.org/CapeContest/entries
Jan 04 05:36:43 <emeb_mac> scores on left column
Jan 04 05:38:38 <ka6sox> ah
Jan 04 05:45:33 <mranostay> ah great my 12 month intro price for my crappy internet ends
Jan 04 05:45:56 <ka6sox> I see GPS+FM in first, geiger and DMX tied for 2ond
Jan 04 05:47:43 <mranostay> ka6sox: can you not count? :)
Jan 04 05:58:12 <ka6sox> mranostay, apparently either I have OLD information or I can't understand the numbers
Jan 04 06:00:47 <mranostay> ka6sox: http://beagleboard.org/CapeContest/entries/ <-- add the FB + G+ numbers
Jan 04 06:01:42 <ka6sox> now I understand how to read that
Jan 04 06:02:30 <mranostay> btw the solar one is never seeing the day of light :)
Jan 04 06:02:37 <mranostay> hello fire risk
Jan 04 06:02:40 <mph____> just goes to show how many beaglers are there...
Jan 04 06:04:19 <mranostay> hmm 1/14 could be an interesting day
Jan 04 06:04:37 <mranostay> that is when the ELC paper acceptions come in
Jan 04 06:09:18 <mph____> the dmx project is quite clever
Jan 04 06:09:20 <mph____> its very simple
Jan 04 06:09:29 <mph____> but its a great demo of the PRU
Jan 04 06:10:14 <mranostay> better than mine!?!? :)
Jan 04 06:11:05 <mranostay> if my bar VFD tubes ever ship from the Ukraine i can do a cool equalizer demo for ELCE
Jan 04 06:21:32 <mranostay> ka6sox: btw kids these days... hw guy was impressed i could read a schematic :)
Jan 04 06:22:40 <ka6sox> do they teach that on Facebook?
Jan 04 06:23:10 <mranostay> ka6sox: not really in college anymore
Jan 04 06:23:36 <mranostay> i don't have a degree but was surprised a CS major friend of mine never had to take circuits class
Jan 04 06:23:51 <mph____> how many peple go to the ELC?
Jan 04 06:23:58 <ka6sox> not enough
Jan 04 06:24:05 <mranostay> mph____: couple hundred
Jan 04 06:24:17 <mranostay> this year may be a low year
Jan 04 06:24:24 <mph____> its it worth attending?
Jan 04 06:24:30 <mranostay> yeah
Jan 04 06:24:48 <ka6sox> mranostay, CS today is what we called BIS when I went...CE is where you have to read a schematic...
Jan 04 06:24:49 <mph____> i have to head to the US sometim in Feb or March
Jan 04 06:25:04 <ka6sox> make it to SF 20-22 FEB
Jan 04 06:25:25 <mph____> if i turn up, i have already offered to buy mranostray some beer
Jan 04 06:25:32 <mph____> or i could bring some NZ beer with me
Jan 04 06:25:45 <ka6sox> either works I'm sure
Jan 04 06:25:56 <mranostay> mph____: mranostay not mranostray :)
Jan 04 06:26:00 <mph____> after you've had a few, all beer is the same.
Jan 04 06:26:10 <ka6sox> for that I might even leave the Pacific plate!
Jan 04 06:26:29 <ka6sox> mph____, his cat might think he strays
Jan 04 06:26:37 <mranostay> need to check if i can have my friend watch the cat.. i don't want to trek back and front for the 3 days
Jan 04 06:26:39 <mph____> mranostay: i wasnt talking about you, i was talkng about mranostray
Jan 04 06:26:58 <mranostay> ranostay! not ranostray! :)
Jan 04 06:27:14 <ka6sox> mph +2
Jan 04 06:27:16 <mph____> doh doh do..
Jan 04 06:27:22 <mranostay> ka6sox: i swear this cat will live to 30 to spite me :)
Jan 04 06:27:36 <mph____> i'm just wonder of the ELV would be over my head.
Jan 04 06:27:41 <mranostay> nah
Jan 04 06:27:48 <mph____> or if id leave having empllyed another 5 staff
Jan 04 06:27:50 <ka6sox> They let me come!
Jan 04 06:27:51 <mranostay> there are talks on anything
Jan 04 06:28:00 <mph____> i tend to collect people at these thigns
Jan 04 06:28:27 <mranostay> i only do it for networking and working on my 'people skills'
Jan 04 06:29:06 <mph____> beacase there is nothing left for you to learn?
Jan 04 06:29:15 <mranostay> no not that
Jan 04 06:29:30 <mranostay> more that i rarely see people in my field
Jan 04 06:29:41 <mranostay> embedded is a niche world
Jan 04 06:30:11 <ka6sox> I'm so deep they rarely let me out of the tank...
Jan 04 06:30:22 <mph____> so, i'm finding
Jan 04 06:30:38 <mranostay> also have no 'formal' education so i need to be visible to keep getting work
Jan 04 06:30:52 <ka6sox> mranostay, I maybe travelling back and forth...we can maybe carpool
Jan 04 06:31:16 <mph____> my observation is that there are some brillant people out there, but there brillance is in one particaular thing.
Jan 04 06:31:34 <ka6sox> more like laser beams than floodlights?
Jan 04 06:31:35 <mph____> i'm a jack. as in jack of all trades, master of none of them.
Jan 04 06:31:38 <mranostay> ka6sox: maybe.. have to see the schedule.. hope they post that like the day of
Jan 04 06:31:48 <mph____> but i'm good and getting experts to work together.
Jan 04 06:31:50 <mranostay> mph____: yeah that is me
Jan 04 06:31:56 <mranostay> my resume shows it...
Jan 04 06:32:29 <mph____> my business partner loves employing educated people
Jan 04 06:32:35 <mph____> "educated" as in formally trained
Jan 04 06:33:01 <mph____> i love finding drop-outs and never-bothered because it was pointless genius
Jan 04 06:33:04 <ka6sox> like the EE's that don't know which end of a hot soldering iron to pick up?
Jan 04 06:35:32 <mranostay> mph____: yeah that is annoying a few interviews have stopped at "so where is your degree from?"
Jan 04 06:36:08 <ka6sox> College of Life, School of Hard Knocks.
Jan 04 06:36:34 <mranostay> i make it clear on my resume i attend some college but even claim to have graduated
Jan 04 06:36:35 <mph____> yup.. tyhough i have a degree.. its whooppe do.
Jan 04 06:37:34 <Russ> Monsters University? https://www.youtube.com/watch?v=THhRSJC5FX8
Jan 04 06:37:35 <mranostay> mph____: it is all scam really :)
Jan 04 06:37:46 <mranostay> some of the dumbest people i know have degrees
Jan 04 06:37:53 <mranostay> and some of the smartest don't go figure
Jan 04 06:38:05 <mph____> and vice versa
Jan 04 06:38:11 <mranostay> true
Jan 04 06:38:22 <mranostay> but they didn't spend $100k to find out ;)
Jan 04 06:38:34 <mph____> all a undergrad degree tells me, is that someone has the stickablity to do something for 3 years
Jan 04 06:38:44 <mph____> that is a useful marker
Jan 04 06:39:02 <mranostay> mph____: what about holding a tech job for 7 years? :)
Jan 04 06:39:32 <mranostay> one of which for 5 :)
Jan 04 06:41:16 <mph____> woudl depend on how much you morphed hte job from day 1
Jan 04 06:41:25 <mph____> if your job was the same on day1, as it was on day7
Jan 04 06:41:36 <mph____> year 1 / year 7
Jan 04 06:41:39 <mph____> then you fail
Jan 04 06:41:46 <mph____> Buzzz...
Jan 04 06:41:56 <mranostay> not really it varied greatly
Jan 04 06:42:23 <mranostay> jack of all trades master of none .. :)
Jan 04 06:42:39 <mph____> then theres job interviews when they just talk about themselves.
Jan 04 06:42:41 <mph____> mmm
Jan 04 06:42:45 <mph____> sat in on a few of those
Jan 04 06:42:48 <mph____> *idiots*
Jan 04 06:43:03 <mph____> um, were we interviewing the candidate or you?
Jan 04 06:43:36 <mph____> to win at those, the more you make them feel good about them selves the more you win.
Jan 04 06:43:36 <mranostay> we i interview with companies  i know i don't want to work for.. just for practice :)
Jan 04 06:44:05 <mranostay> just as long as you don't ask those crappy interview questions that were posted the other day
Jan 04 06:44:19 <mranostay> i love the greatest weakness one..
Jan 04 06:45:04 <mph____> i once had a guy start by saying " i'm here to work out if i want to work for you .. what are you here for "
Jan 04 06:45:11 <mph____> btw, we hired him
Jan 04 06:45:17 <ka6sox> yup
Jan 04 06:45:24 <ka6sox> I hired an Intern that way
Jan 04 06:46:04 <mranostay> btw i had only 4 interviews ever before this current job.. got 3 jobs out of them :)
Jan 04 06:49:34 <Russ> ya, if you get the interview, you pretty much has the job
Jan 04 06:49:41 <ka6sox> I worked for a "large multinational" for 8yrs...
Jan 04 06:49:50 <ka6sox> er wait...13yrs
Jan 04 06:50:03 <Russ> that part of your soul is just gone
Jan 04 06:50:06 <ka6sox> then started my own business...22yrs ago.
Jan 04 06:50:10 <Russ> :p
Jan 04 06:50:17 <mranostay> ka6sox: need a lab rat? :P
Jan 04 06:50:33 <ka6sox> possiblyh
Jan 04 06:50:49 <ka6sox> er rather...maybe, possibly, we should talk?
Jan 04 06:51:09 <ds2> ka6sox: how bad was the paperwork burden 22yrs agao?
Jan 04 06:51:13 <mranostay> nah not atm at least :)
Jan 04 06:51:27 <mranostay> ka6sox: how expensive is running a business in CA?
Jan 04 06:51:42 <mph____> are you on the market mranostay?
Jan 04 06:51:53 <ka6sox> Bidding War!
Jan 04 06:51:58 <mranostay> mph____: passively looking always :)
Jan 04 06:52:14 <mph____> $1 and a bowl of rice per day?
Jan 04 06:52:26 <mranostay> yeah how about no :)
Jan 04 06:52:36 <mph____> well, its a starting point.
Jan 04 06:52:42 <mph____> lol.
Jan 04 06:52:46 <ka6sox> $2 and a bowl of rice and cup of sake?
Jan 04 06:52:56 <mranostay> multiple that by a few 10 thousands :)
Jan 04 06:52:58 <ka6sox> (doubling down)
Jan 04 06:53:10 <mranostay> well a year not a day :)
Jan 04 06:53:18 <mph____> geepers, ka6sox you'll crash the labour market paying that kind of wage.
Jan 04 06:54:06 <ka6sox> hey, here it might be *still* under the minimum wage!!!
Jan 04 06:54:42 <mranostay> yeah i vaguely remember making min wage
Jan 04 06:55:21 <ka6sox> last time I did that I was 16...and it was $2.35/hr
Jan 04 06:55:29 <mph____> mranostay: did you program all PRU stuff in assembly?
Jan 04 06:55:30 <mranostay> 5.65$...
Jan 04 06:55:50 <mranostay> they upped it to $7.25 a year after i got my first real job
Jan 04 06:56:18 <mranostay> nothing like fast food to convince you need a real job....
Jan 04 06:57:29 <mranostay> joel_: don't be so tame on panto :)
Jan 04 06:59:26 * Russ scrolls up all confused
Jan 04 07:00:12 <mranostay> Russ: in this channel? don't you expect that?
Jan 04 07:00:18 <mph____> mranostay: did you avoid my channel related topic for a more interesting one
Jan 04 07:00:40 <ka6sox> Russ, expect rambling conversations that are waaaaay OT (but interesting to the parties here)
Jan 04 07:01:09 <mranostay> mph____: never on topic
Jan 04 07:01:19 <ka6sox> why do I have 3 beaglebones...all of which the Ethernet is wonky???
Jan 04 07:01:23 <mranostay> the real trolls are holiday still
Jan 04 07:01:31 <mranostay> ka6sox: what rev?
Jan 04 07:01:32 <ka6sox> different revisions even!!!
Jan 04 07:01:38 <mranostay> ah
Jan 04 07:01:46 <ka6sox> a2, a3, and I can't read the last one
Jan 04 07:01:54 <mranostay> i remember was it A2 that needed a resistor popped off
Jan 04 07:01:59 <ka6sox> done
Jan 04 07:02:05 <mph____> mine says 5V
Jan 04 07:02:15 <mranostay> i only have A6's
Jan 04 07:02:24 <ka6sox> okay must be an A6
Jan 04 07:02:30 <mranostay> had a A3 before i smoked that one twice... :)...
Jan 04 07:02:34 <mph____> ha.. thats 5V volts.
Jan 04 07:02:35 <mph____> doh.
Jan 04 07:02:36 <emeb_mac> twice?
Jan 04 07:02:37 <mranostay> A6 works fine
Jan 04 07:02:51 <mph____> this one is a A5A
Jan 04 07:02:52 <emeb_mac> got my A6a today. works great
Jan 04 07:02:54 * ka6sox senses a story
Jan 04 07:03:04 <ka6sox> I got an A6 today too...
Jan 04 07:03:09 <Russ> yes, be very careful with your beaglebone
Jan 04 07:03:11 <ka6sox> so why is it flakey
Jan 04 07:03:13 <Russ> never, ever
Jan 04 07:03:16 <mranostay> well first time was careless nixie testing
Jan 04 07:03:17 <Russ> under any circumstances
Jan 04 07:03:20 <Russ> touch the revision number
Jan 04 07:03:27 <mranostay> seems 40V to a GPIO is a bad idea
Jan 04 07:03:43 <emeb_mac> Russ: you'll go blind? Break your mother's back? what?
Jan 04 07:03:52 <ds2> mranostay: if you put a 1M resistor in line, it might be tolerated
Jan 04 07:03:54 <Russ> tends to wipe off
Jan 04 07:03:55 <ds2> there are ESD diodes
Jan 04 07:04:11 <ka6sox> ds2, *were*
Jan 04 07:04:15 <mranostay> ds2: too late to go back isn't it? :)
Jan 04 07:04:29 * mranostay fires up time machine
Jan 04 07:04:43 <ds2> mranostay:lesson learned
Jan 04 07:04:55 <ds2> there is a microchip app note about applying 120V to the pins of a PIC
Jan 04 07:04:55 <mranostay> yes yes it is
Jan 04 07:05:10 <mranostay> ds2: ah don't do it? :)
Jan 04 07:05:17 <emeb_mac> what rev did that giant 50ohm resistor 2 @ R76 get removed?
Jan 04 07:05:28 <emeb_mac> -2
Jan 04 07:05:32 <ka6sox> A2
Jan 04 07:05:44 <emeb_mac> no - I've got an A3 that has it.
Jan 04 07:05:53 <mph____> im wondering how much work is in debeagling and applicaiton is
Jan 04 07:05:56 <mranostay> yeah A2 was a TI only model iirc
Jan 04 07:06:13 <emeb_mac> debeagling?
Jan 04 07:06:18 <ka6sox> that one is a trooper otherwise
Jan 04 07:06:33 <Russ> also, if you void the warranty, the serial number sticker on the back turns black
Jan 04 07:06:37 <ds2> just use a crane board for ethernet
Jan 04 07:06:44 <ds2> the 3517 isn't that bad
Jan 04 07:06:56 <mranostay> Russ: that stupid sticker rubs off too easily
Jan 04 07:07:09 <mph____> debeagling.. the art of taking a pile of hacked together hardware ( beagles and capes ) and designing somethign that fits on a single board, ready for a production run of "something*
Jan 04 07:07:11 <emeb_mac> put clear tape on it 1st thing.
Jan 04 07:07:26 <mranostay> one i had to RMA i was penning on the last number what i though it was
Jan 04 07:07:33 <ka6sox> oh I painted mine
Jan 04 07:07:37 <mranostay> A6 with bad PMIC inductors
Jan 04 07:09:10 <ds2> I wonder what volume would one need to do a 3517 board with bone capabilities
Jan 04 07:09:24 <mph____> 1
Jan 04 07:09:43 <ds2> including price point
Jan 04 07:14:06 <mph____> i made pad thai for dinner
Jan 04 07:14:10 <mph____> mmmm, gotta love that
Jan 04 07:14:20 <ka6sox> okay I never touched the Ethernet Code in the kernel build :P
Jan 04 07:15:44 <mph____> the beagle is a 4 layer pcb?
Jan 04 07:16:03 <emeb_mac> more like 8
Jan 04 07:16:18 <emeb_mac> the gerbers are out there to look at if you want
Jan 04 07:16:29 <mph____> i could cheat
Jan 04 07:16:39 <mranostay> emeb_mac: the FPGA cape is 8?
Jan 04 07:16:42 <mph____> and ask you
Jan 04 07:16:55 <emeb_mac> mranostay: no the FPGA board is 4
Jan 04 07:17:17 <emeb_mac> could almost fit on 2, but went 4 for better power & ground
Jan 04 07:17:18 <mranostay> heh more than i can handle
Jan 04 07:17:36 <emeb_mac> 4 is easy. just a little more $$ to fab
Jan 04 07:17:43 <mph____> have not manufactued any 4 layer boards here
Jan 04 07:17:44 <mranostay> one reason i don't do the nixie cape for my ELC demo
Jan 04 07:17:49 <mranostay> breadboard works
Jan 04 07:17:57 <mph____> we dont' have a multilayer press
Jan 04 07:18:12 <ds2> Nixie is easy
Jan 04 07:18:25 <emeb_mac> mranostay: mph____ OSHPark does 4 for $10/sqin for 3 boards.
Jan 04 07:18:34 <ka6sox> mranostay, just "bitbangit'
Jan 04 07:18:55 <Russ> highly recommend 4 layer
Jan 04 07:18:55 <ka6sox> my only complaint with OSHPark...no solder tinning
Jan 04 07:18:59 <Russ> you spend a lot less time routing
Jan 04 07:19:03 <mranostay> ka6sox: well that is what the PRU is doing
Jan 04 07:19:04 <mph____> my maylasin factory does lower cost that than
Jan 04 07:19:04 <Russ> ka6sox, that's a damned feature
Jan 04 07:19:10 <emeb_mac> FPGA boards are about $25/ea small qty
Jan 04 07:19:23 <ka6sox> Russ, HEH
Jan 04 07:19:25 <Russ> ka6sox, fine pitch components tend to be more difficult on HASL boards
Jan 04 07:19:30 <Russ> all bumpy
Jan 04 07:19:38 <mranostay> how can i transport a breadboard safely btw?
Jan 04 07:19:41 <Russ> it has a this gold finish
Jan 04 07:19:47 <mph____> yup.. HASL's not such a flash idea
Jan 04 07:19:48 <ka6sox> I think this is .6mm...but I don't remember
Jan 04 07:19:56 <emeb_mac> mranostay: saran wrap. :)
Jan 04 07:20:04 <ka6sox> Plastic wrap
Jan 04 07:20:17 <Russ> mranostay, still plenty of time for an oshpark  board
Jan 04 07:20:21 <ds2> HASL?!?!?
Jan 04 07:20:29 <ds2> who is still using HASL finish?
Jan 04 07:20:31 <Russ> ya, ka6sox thinks HASL is a feature
Jan 04 07:21:16 <ka6sox> if we do more I'm going to make a stencil
Jan 04 07:21:23 <ds2> ENIG
Jan 04 07:21:37 <emeb_mac> HASL isn't bad.
Jan 04 07:21:47 <emeb_mac> ENIG is OK if you use lots of flux.
Jan 04 07:22:05 <Russ> ENIG is so purty http://www.flickr.com/photos/russdill/8263004442
Jan 04 07:22:16 <ds2> HASL isn't perfectly flat
Jan 04 07:22:22 <emeb_mac> oooo - gold.
Jan 04 07:22:33 <Russ> ds2, that's like saying water isn't perfectly dry
Jan 04 07:22:34 <mranostay> Russ: the shift register i'm using is TSSOP-28 :)
Jan 04 07:22:52 <Russ> mranostay, that should make things easy then
Jan 04 07:23:07 <Russ> you know which end of the iron is the hot one, right?
Jan 04 07:23:15 <ds2> blah
Jan 04 07:23:20 <ds2> no need for a stinkin iron
Jan 04 07:23:28 <mranostay> plus the boost converter and EEPROM and mosfet driver
Jan 04 07:23:42 <emeb_mac> ds2: toaster!
Jan 04 07:24:09 <Russ> yes, just place in the slot for one slice, press down
Jan 04 07:24:12 <Russ> press the bagel button
Jan 04 07:24:15 <Russ> and wait
Jan 04 07:24:27 <ds2> yep
Jan 04 07:24:30 <Russ> be sure to turn it up to at least 6 for RoHS
Jan 04 07:24:38 <ds2> want to see some folks do 0.4mm BGAs with an iron ;)
Jan 04 07:25:15 <ka6sox> ds2, you know that cna't be done :p
Jan 04 07:25:32 <mranostay> anyone going to buy a nixie cape? it so i'd probably do it
Jan 04 07:25:34 <ds2> ka6sox: I donno... in theory, via in pad might allow for that to work
Jan 04 07:25:49 <ka6sox> @ .4mm?
Jan 04 07:25:53 <ka6sox> umm....
Jan 04 07:26:03 <mranostay> although for a final product i'd need to fit on a AIN voltage feedback loop
Jan 04 07:26:23 <ds2> 0.4mm pretty much forces via in pad
Jan 04 07:27:00 <ka6sox> BGA BAD.... 0.4mm BGA Muy Mal.
Jan 04 07:28:11 <ds2> not too sure about that
Jan 04 07:28:25 <ds2> the specs for BGA lets you pull off some things you cannot do with a QFP for example
Jan 04 07:28:53 <mph____> i have used a hacked toaster oven for reflow for several years.
Jan 04 07:28:58 <mph____> i now have a real one as well
Jan 04 07:29:07 <ka6sox> mph____, I use a hacked one
Jan 04 07:29:07 <mph____> but the toaster still gets used lots
Jan 04 07:29:16 <mph____> i bought the kit from Sparkfun
Jan 04 07:29:21 <mph____> it works just fine
Jan 04 07:29:37 <ka6sox> if we do more...I will have to get a Stencil for the paste
Jan 04 07:30:00 <ds2> ka6sox: get or make one?
Jan 04 07:30:04 <mph____> my mill 'prints' solder paste.. though its slow.
Jan 04 07:30:32 <ka6sox> ds2, I haven't figured out how to the the 3d printer to squirt out the right amount of solder paste
Jan 04 07:30:50 <ds2> ka6sox: but can't u laser cut some stenciles?
Jan 04 07:31:03 <emeb_mac> printed paste would be cool
Jan 04 07:31:12 <mph____> all the best stencils are laser cut
Jan 04 07:31:17 <ka6sox> we don't have the laser yet @ the Hacker space
Jan 04 07:31:29 <ds2> oh
Jan 04 07:31:39 <mph____> but a laser to cut stainless is an expensive toy
Jan 04 07:31:50 <emeb_mac> someone was talking the other day about using a craft programmable cutter to make stencils in plastic.
Jan 04 07:32:06 <ka6sox> plastic ones are only good for a little while
Jan 04 07:32:12 <mph____> i have seen them made in mylar sheets for proto's
Jan 04 07:32:18 <mph____> you can also etch them
Jan 04 07:32:19 <emeb_mac> yeah, but they're cheap and you can easily make more.
Jan 04 07:32:32 <ka6sox> mph____, ya, there is a place down south that makes good ones for a reasonable price (if you are willing to wait)
Jan 04 07:32:42 <mph____> the mill can take an hour or more to print even a smallish board.
Jan 04 07:32:55 <mph____> 800-1000 pads per hour
Jan 04 07:32:57 <ka6sox> mph____, thats waht I am worried about
Jan 04 07:32:59 <mph____> dpending on shape and size
Jan 04 07:33:37 <mph____> we have some paste that has 8 hours of life
Jan 04 07:33:40 <mph____> so its not so bad
Jan 04 07:34:01 <ka6sox> the stuff I got from Sparkfun has about 30minutes
Jan 04 07:34:34 <mph____> i am forntate i am on good terms with my assembly contractor
Jan 04 07:34:51 <mph____> i get small quanitys of paste from them, when i need it
Jan 04 07:35:25 <mph____> theres literally 100's of different types
Jan 04 07:35:26 <mph____> its nuts.
Jan 04 07:36:18 <mph____> i am getting way too many ______
Jan 04 07:36:21 <mranostay> dammit panto is here
Jan 04 07:36:22 <mph____> i must register a nick
Jan 04 07:36:25 <ds2> if you are not doing BGA, you can do it w/o paste
Jan 04 07:36:30 <panto> hi there
Jan 04 07:37:14 <Russ> just use some flux
Jan 04 07:37:21 <Russ> mranostay, I was thinking the same thing, I have patches to review
Jan 04 07:37:49 <ka6sox> russ whats a HWA?
Jan 04 07:38:19 <Russ> the board for my thesis, I came up with the name early one morning and forgot the logic hehind it
Jan 04 07:38:20 <ka6sox> HardWareAccelerator?
Jan 04 07:38:31 <Russ> fpga+nand flash
Jan 04 07:38:51 <ka6sox> 2lane PCIe too!
Jan 04 07:38:59 <Russ> 4
Jan 04 07:39:06 <ka6sox> nice
Jan 04 07:39:07 <ka6sox> fast
Jan 04 07:39:16 <mph____> sorry, mind my ignorance, but what is 'allegro' in terms of 'allegro files' for the beagleboard
Jan 04 07:39:27 <ka6sox> board design tool
Jan 04 07:39:29 <Russ> mph____, eda
Jan 04 07:39:34 <ds2> some cheap PoS ;)
Jan 04 07:39:43 <ka6sox> what he said (2 up)
Jan 04 07:39:47 <mph____> eda?
Jan 04 07:39:55 <Russ> its the layout and such
Jan 04 07:39:59 <ka6sox> Electronic Design mumble...
Jan 04 07:40:00 <mph____> pcb cad sytems?
Jan 04 07:40:10 <ka6sox> ya
Jan 04 07:40:15 <mph____> oh
Jan 04 07:41:19 <ka6sox> okay time to attempt sleep....
Jan 04 07:41:38 <mranostay> ka6sox: wimp!
Jan 04 07:41:42 <ka6sox> Interview Prep waaay to early for my liking tomorrow.
Jan 04 07:41:44 <mranostay> you cannot quit us!
Jan 04 07:42:07 <mph____> 6 layers.
Jan 04 07:42:51 <ka6sox> mph____, remember someone plopped a Altoids Tin down and said "make it fit this"
Jan 04 07:43:06 <mranostay> mph____: burrito?
Jan 04 07:43:17 <mph____> ?
Jan 04 07:43:30 <mranostay> i saw 6 layers
Jan 04 07:43:36 <ds2> just make it smaller then a business card
Jan 04 07:43:56 <mph____> i wonder how much stuff i dot' need
Jan 04 07:46:14 <ds2> you don't need any of it
Jan 04 07:46:15 <ds2> :D
Jan 04 07:46:30 <mph____> ds2: how do you know what i need
Jan 04 07:47:26 <ka6sox-away> is it food? is it Air? is it shelter?...Bzzzt...not needed.
Jan 04 07:47:52 <ka6sox-away> sorry...I am overtired...just ignore me...
Jan 04 07:48:08 <Russ> ka6sox-away, who know's maybe he's planning a trip to mars
Jan 04 07:48:32 <ka6sox-away> true
Jan 04 07:48:43 <ka6sox-away> okay I sleep corrected....
Jan 04 07:48:45 <ka6sox-away> nite
Jan 04 07:48:53 <mph____> mars.. nah.. thats easy.. i'm planing a trip to the USA, thats much more complex
Jan 04 07:49:09 <Russ> I have heard mars customs are pretty low key
Jan 04 07:49:19 <Russ> (customs as in document inspection, etc
Jan 04 07:49:24 <mph____> they did'tn even show up on my last trup
Jan 04 07:49:25 <ka6sox-away> OTOH the TSA.....
Jan 04 07:57:50 <mph____> just priced up beagleboards.
Jan 04 07:58:07 <mph____> at 500 off quanitys the pcbs are just over $10 each
Jan 04 07:58:27 <mph____> the bom is about $45
Jan 04 07:58:30 <mph____> thats $55
Jan 04 07:58:42 <mph____> no one is getting rich selling them for $89
Jan 04 07:59:11 <mph____> still got to assemble them
Jan 04 08:00:19 <mph____> you've got to give it to the PI boys that they did it cheap
Jan 04 09:12:24 <jackmitchell> mdp: ah hah, my SPI DMA transfer bug? Which is now fixed with your DMA work :)
Jan 04 09:27:34 * Russ notices the usgs has decided that earthquakes occur sometime in the ear 44980
Jan 04 09:27:37 <Russ> er, year
Jan 04 09:27:54 <Russ> ...and its fixed
Jan 04 09:29:41 * av500 gives mph__ some more _________
Jan 04 09:34:54 <Cykon> fuuu, my beagleboard has mood swings of when it wants to work or not I swear
Jan 04 09:39:28 <aholler_> one shouldn't turn on debug options in an arm kernel. it's like viewing an horror movie
Jan 04 09:41:10 <Cykon> I could actually use some kernel debug messages right about now :(
Jan 04 09:42:15 <aholler_> (at least not on armv5)
Jan 04 09:52:36 <Cykon> Anyone on happen to have a rev c3 beagleboard?
Jan 04 09:52:40 <Cykon> classic
Jan 04 09:55:15 <Cykon> shit it doesn't even have to be c3, any rev will d
Jan 04 09:55:16 <Cykon> o
Jan 04 10:14:01 <aholler_> you could be more verbose about your problem, otherwise no one will ever answer
Jan 04 10:16:04 <Cykon> Wellp, lets see, where to begin
Jan 04 10:16:33 <aholler_> try e.g. the kernel version
Jan 04 10:16:52 <Cykon> Actually, I could probably write a novel about my problem but yeah, i'll start with the kernel version
Jan 04 10:17:18 <Cykon> it's the one in the beagleboard angstrom demo image repo, so 3.0.17
Jan 04 10:17:31 <Cykon> And it's hanging on "booting the kernel"
Jan 04 10:18:05 <Cykon> Around a week ago before I went on vacation I had the same problem, got it working by plugging my usb cable into a wall mount instead of the back of my computer
Jan 04 10:18:15 <Cykon> (and messing with enviornment variables)
Jan 04 10:18:28 <Cykon> and now I come back, excited to mess with a working board, and it does the same thing @_@
Jan 04 10:20:02 <Cykon> I'm not quite sure why the back of my computer vs a wall mounted usb port would have made a difference, then again I'm pretty new to this whole thing so I'm sure there's some reasoning behind it... though perhaps not... it's not working anymore
Jan 04 10:20:09 <av500> stop the boot process and dump the uboot enc
Jan 04 10:20:10 <av500> env
Jan 04 10:20:22 <Cykon> Yep, i'll pastebin it
Jan 04 10:20:25 <av500> also pastebin a full boot log
Jan 04 10:23:19 <Cykon> http://pastebin.com/yha1uEUt
Jan 04 10:23:39 <Cykon> top part are the vars, bottom half is the boot log
Jan 04 10:27:56 <Cykon> Now, something interesting that I did find was under http://processors.wiki.ti.com/index.php/Kernel_-_Common_Problems_Booting_Linux someone mentioned that my problem "can also occur if you power the RTC peripheral on the OMAP-L138 with 3v3 instead of 1v2 (which, incidentally, also makes the chip run a lot hotter). Very strange."
Jan 04 10:28:12 <av500> L138?
Jan 04 10:29:08 <Cykon> Yeah, see, I'm pretty sure that's not the one the beagleboard uses so it was irrelevant, but it still seems weird that it worked after swapping from my computer's usb port to this wall one
Jan 04 10:29:14 <Cykon> then again, it's not working anymore so I have no idea
Jan 04 10:30:11 <aholler_> Image Name:   Angstrom/2.6.32/beagleboard
Jan 04 10:30:16 <aholler_> =
Jan 04 10:30:18 <aholler_> ?
Jan 04 10:30:32 <aholler_> for this you need console on ttyS2
Jan 04 10:30:58 <Cykon> Yeah I did try S2 a few times, but it didn't work at all
Jan 04 10:31:02 <Cykon> I'll give it one more try
Jan 04 10:31:26 <aholler_> anyway get you kernel right, at least the version
Jan 04 10:32:17 <Cykon> :3
Jan 04 10:33:15 <Cykon> Yeah, same thing
Jan 04 10:34:57 <Cykon> Though I guess i'm a bit curious... uImage is the kernel right? So why does the version in the demo repo say 3.0..17
Jan 04 10:35:02 <Cykon> yet mine was 2.6.32
Jan 04 10:35:07 <Cykon> suspicious
Jan 04 10:41:11 <Cykon> derp, I guess it reads from the boot folder in the Angstrom partition rather than the boot partition
Jan 04 10:41:30 <Cykon> wellp, at least I got the kernel to boot... now to figure out why it's freezing up down the line
Jan 04 10:50:50 <av500> Cykon: yes, newer uboots can do ext3 and read from there
Jan 04 10:52:51 <Cykon> Makes sense! Though I wish someone would update the wiki with all this new info ;[
Jan 04 10:52:57 <av500> sure
Jan 04 10:53:01 <av500> that someone is you
Jan 04 10:53:14 <Cykon> Yeah i'll do it once I actually get everything working ;P
Jan 04 10:53:21 <av500> thats what they all say
Jan 04 10:53:25 <Cykon> lol
Jan 04 10:55:40 <Cykon> bleh, now it's stuck on "Waiting for root device /dev/mmcblk0p2..." -- someone mentioned that the SD card could be bad so I tried another one and get the same thing... hmm
Jan 04 10:56:36 <av500> pastebin
Jan 04 10:56:43 <Cykon> okidoki
Jan 04 10:56:45 <av500> pastebin early, pastebin often
Jan 04 10:57:23 <Cykon> xD incoming noob question
Jan 04 10:57:37 <Cykon> is there an easy way to copy all the text from the screen terminal window?
Jan 04 10:57:49 <Cykon> never mind
Jan 04 10:57:53 <Cykon> i'm just silly
Jan 04 10:58:16 <Cykon> no no, that didn't work, you can answer it if you want
Jan 04 11:06:51 <av500> Cykon: ?
Jan 04 11:13:03 <Cykon> How might I go about copying all of the text from my screen window / serial output
Jan 04 11:13:17 <av500> I use minicom and a log file
Jan 04 11:13:28 <av500> or I mouse select it
Jan 04 11:13:37 <Cykon> Won't let me scroll up :(
Jan 04 11:13:47 <Cykon> Well, unless I enter copy mode
Jan 04 11:14:01 <Cykon> but even in copy mode it wont let me select the whole thing
Jan 04 11:14:13 <av500> for screen no idea
Jan 04 11:14:18 <av500> google will know
Jan 04 11:15:11 <Cykon> Yeah I'm poking around
Jan 04 11:24:28 <woglinde> gm
Jan 04 11:24:31 <woglinde> friday
Jan 04 11:25:08 <woglinde> screen can log too
Jan 04 11:29:50 <Cykon> Yeah, but minicom is so much nicer now that I have it running
Jan 04 11:30:08 <Cykon> At least in the sense that I can actually copy text
Jan 04 11:30:09 <Cykon> http://pastebin.com/5QPeU7MQ
Jan 04 11:31:04 <Cykon> Anyways, that's the output, this is the point to where I got it working last week, came back, and all the sudden it wants to hang on this particular part
Jan 04 11:31:41 <Cykon> The boot hangup from before was because I put a different angstrom image on and didn't know that it read uImage from the boot folder on that partition
Jan 04 11:31:48 <woglinde> hm no screen is nicer
Jan 04 11:32:02 <woglinde> you can detach from the console
Jan 04 11:32:06 <woglinde> retach
Jan 04 11:32:06 <Cykon> It's easier to use, but... but... how do I copy my whole log :(
Jan 04 11:32:19 <woglinde> again use the logging function
Jan 04 11:32:30 <Cykon> screen -L ...?
Jan 04 11:34:16 <woglinde> I do not read the manpage for you
Jan 04 11:34:33 <Cykon> fair enough :P
Jan 04 11:34:46 <aholler_> first, your log is broken (limited to 80 chars), see line 66. 2. your kernel doesn't see an sd-card at all
Jan 04 11:35:48 <aholler_> try a newer kernel.
Jan 04 11:37:21 <Cykon> Where do you see that the kernel isn't seeing an SD card?
Jan 04 11:37:23 <woglinde> yes 3.0.17 is old
Jan 04 11:37:47 <Cykon> boo, it was in the demo repo, but i'm starting to learn that nothing "just works nicely"
Jan 04 11:38:03 <aholler_> I'm unable to see nothing, the question is why you don't see anything
Jan 04 11:38:17 <woglinde> map_hsmmc: probe of omap_hsmmc.1 failed with error -16
Jan 04 11:38:59 <woglinde> ah right koen did not update it in a while
Jan 04 11:39:32 <Cykon> yeah, and the wiki is super out of date too, makes things hard for noobies!
Jan 04 11:39:43 <Cykon> Though I can't say that I'm not slowly figuring everything out I guess
Jan 04 11:40:08 <woglinde> you could try narcissus for an more update to image
Jan 04 11:40:34 <woglinde> http://www.angstrom-distribution.org/builder/
Jan 04 11:40:38 <aholler_> mainline 3.2.x works here
Jan 04 11:40:47 <Cykon> I tried that when I was first trying to figure everything out, though i'm pretty sure I fixed the issue behind it not working correctly
Jan 04 11:40:53 <Cykon> but yeah i'll try it again
Jan 04 11:42:50 <Cykon> Hmm, I'm looking at my tar using Narcissus from a week ago and it has uImage 2.6.32
Jan 04 11:43:02 <Cykon> Perhaps I didn't configure it correctly that time though
Jan 04 11:43:25 <woglinde> okay till later
Jan 04 11:44:10 <Cykon> hmm?
Jan 04 11:46:28 <koen> woglinde: rcn-ee and I are working on 3.7 and 3.8 for beagle right now
Jan 04 11:57:16 * tasslehoff wishes Someone had configured DSI in u-boot so it was Tasslehoff Ready
Jan 04 11:59:09 <av500> DSI?
Jan 04 12:01:19 <tasslehoff> av500: I dumped the dss/dsi/clk regs in linux, and it seems that dispc_fclk_src is is dsi1_pll_fclk. I'm trying to wake that clock in u-boot to get my pixel clock right.
Jan 04 12:01:59 <tasslehoff> now dispc uses DSS1_ALWON_FCLK
Jan 04 12:05:59 <Cykon> Hmmph, I tried compiling uImage via the wiki instructions, but it kept failing -- are there any pre-compiled ones floating around in the git repo?
Jan 04 12:15:20 <av500> tasslehoff: ah, ok, DSI PLL for DPI
Jan 04 12:15:56 <av500> there are uimages on the BB demo page
Jan 04 12:20:49 <tasslehoff> av500: yes. on new ground, so my terminology is bound to be wrong :)
Jan 04 12:24:02 <Cykon> unfortunately they are outdated / not working for me... but I suppose I can just poke at it some more to see if it ends up working
Jan 04 15:17:55 <jsabeaudry> Is koen's tree still the best bet for bbone if I just want to clone a kernel and make uImage?
Jan 04 15:18:39 <mdp> jsabeaudry: https://github.com/beagleboard/kernel/
Jan 04 15:19:25 <mdp> 3.7 or 3.8 branches there have a metric shit-ton of patches collected on top of mainline
Jan 04 15:19:56 <woglinde> hi emeb
Jan 04 15:21:02 <jsabeaudry> mdp, I see that 3.7 and 3.8 do not have different ranches for bboard and bbone, anything special I have to do or will kernel magically work on both?
Jan 04 15:21:13 <mdp> it works on both
Jan 04 15:21:35 <mdp> what happened is that there was beaglebone/board-3.6 when they were separate...but things have since been combined to work on both
Jan 04 15:22:02 <mdp> which is part of jkridner/koen's criteria for having common beagle* releases
Jan 04 15:22:21 <jsabeaudry> mdp, great! I'll give that a spin see how it goes, thanks for the pointer!
Jan 04 15:23:02 <ka6sox> mdp, so 3.7 is now the official party line?
Jan 04 15:23:33 <mdp> ka6sox: I heard yesterday that some beagleboard.org release will be done on 3.7
Jan 04 15:23:38 <jkridner> 3.7 is what is targeted for the next release...
Jan 04 15:23:41 <mdp> I think that was here
Jan 04 15:23:45 <mdp> ok, there you go ;)
Jan 04 15:23:51 <jkridner> but, on-going work is on 3.8.
Jan 04 15:24:09 <jkridner> would love to have everybody give feedback on the 3.7 and file issues.
Jan 04 15:24:13 <ka6sox> mdp, jkridner thanks...
Jan 04 15:24:29 <ka6sox> well, I'll stop playing with this 3.5 then and move up...
Jan 04 15:24:29 <jkridner> but it is in "bug-fix-only" mode.
Jan 04 15:25:33 <mdp> keep in mind that there will be major DT-ish changes going from the beagleboard.org 3.7 release to anything newer..due to the cape support in there only being a first attempt at a solution
Jan 04 15:26:08 <mdp> so don't get married to the 3.7 method of supporting capes..it's already deprecated by DT overlay support
Jan 04 15:26:25 <av500> nothing like a stable API
Jan 04 15:26:37 <ka6sox> av500, +1
Jan 04 15:26:41 <mdp> av500, beagleboard.org is no different than any other product vendor
Jan 04 15:26:49 <av500> mdp: not true
Jan 04 15:26:53 <ka6sox> okay so that means now is NOT a good time to be working on this kernel stuff...gotcha
Jan 04 15:26:59 <av500> if it was a true vendor, it would be on a nice .38 :)
Jan 04 15:27:05 <mdp> lol
Jan 04 15:27:31 <mdp> ka6sox, it means that a version of support was done..not acceptable to upstream, and the reworked version is under review
Jan 04 15:27:45 <mdp> so what do you do? you ship the working version instead of waiting indefinitely
Jan 04 15:28:18 <mdp> same thing we did at every linux vendor I've been at :)
Jan 04 15:28:19 <jkridner> mdp: I agree with av500. while we are working on releasing vendor kernels, we are chasing mainline pretty quickly.
Jan 04 15:28:35 <jkridner> thanks to your efforts
Jan 04 15:28:35 <ka6sox> I don't want to be developing on anymore dead stuff...but if the API is changing (with the DT stuff) then chasing that is futile too.
Jan 04 15:28:59 <jkridner> depends on the API.
Jan 04 15:29:19 <mdp> jkridner, definitely more aggressive...however, the point is, the process is _identical_ regardless of the speed
Jan 04 15:29:49 <ka6sox> mdp, jkridner where is the 3.8 tree you guys are working from?
Jan 04 15:29:53 <mdp> you will always have some out-of-tree feature
Jan 04 15:30:00 <jkridner> http://github.com/beagleboard/kernel
Jan 04 15:30:23 <av500> ka6sox: atm capes are the hotness, wait for some other vendor to have "cloaks" or "daggers"....
Jan 04 15:30:35 <mdp> ka6sox, upstream patches are against kernel.org...and then the posted/review/accepted ones are culled into the tree jkridner noted
Jan 04 15:30:43 <aholler_> ka6sox: there never was a good time ;)
Jan 04 15:30:49 <jkridner> it is arranged as a set of patches in a directory to help organization
Jan 04 15:31:25 <mdp> it's going to be amusing when TI produces a competing tree RSN
Jan 04 15:31:37 <ka6sox> okay so I'll have to forward port the GPMC stuff I've been playing with.
Jan 04 15:32:08 <mdp> ka6sox, gpmc support has been massively overhauled upstream too
Jan 04 15:32:47 <jkridner> only way to stablize stuff and bring in community support is to get the patches accepted upstream
Jan 04 15:33:16 <aholler_> not only community support
Jan 04 15:33:25 <av500> professional support
Jan 04 15:33:33 <mdp> ka6sox, if you wait for things to settle down you'll probably run head-on into the real mayan end-of-the-world date.
Jan 04 15:33:49 <av500> wait for the binary driver API
Jan 04 15:34:33 * ka6sox thinks the Mayan Date will come quicker
Jan 04 15:34:38 <woglinde> so who is unhappy today, hands up
Jan 04 15:34:40 <aholler_> a good time might be 4.0, maybe dt then will be in most trees and changes will become less
Jan 04 15:35:23 <av500> aholler_: they will invent something else by then
Jan 04 15:35:29 <mdp> +1
Jan 04 15:35:32 <aholler_> acpi
Jan 04 15:35:36 <av500> like systemd in kernel
Jan 04 15:35:46 <aholler_> doesn't m$ wants acpi for arm too?
Jan 04 15:35:51 <mdp> there's _always_ been some churn item going on in the kernel
Jan 04 15:35:51 <aholler_> s/wants/requires/
Jan 04 15:36:29 <mdp> pat's device model, system PM, runtime PM, hotplug, blah blah
Jan 04 15:36:30 <av500> aholler_: who cares?
Jan 04 15:37:04 <aholler_> mdp: yes, but having to change sources because of changed pin (numbers) is a pain.
Jan 04 15:37:25 <mdp> all of these things came about to address a problem
Jan 04 15:37:47 <aholler_> missing bios ;)
Jan 04 15:37:54 <ka6sox> the kernel release crap is beginning to look like Firefox releases :P
Jan 04 15:38:18 <mdp> mostly this incarnation of DT came about because all the ARM licensees were poor citizens producing epic crap
Jan 04 15:38:30 <mdp> "pay later" :)
Jan 04 15:38:59 <aholler_> yes, and throwing several hundreds of inexperienced devs at arm-kernels
Jan 04 15:39:09 <aholler_> doesn't help
Jan 04 15:39:27 <mranostay> hi all
Jan 04 15:39:54 <mdp> aholler_: yeah, we continue to unwind really crappy code that made it upstream
Jan 04 15:40:08 <aholler_> some vendor patches look like people just have learned to write "hello world" in C
Jan 04 15:40:28 <mdp> heh
Jan 04 15:40:45 <mdp> stability tomorrow
Jan 04 15:40:58 <av500> aholler_: yes, so?
Jan 04 15:41:03 * mdp goes back to reviewing dt overlay support
Jan 04 15:41:21 <aholler_> of course, C itself is cumbersome. I always wish the kernel would be C++ ;)
Jan 04 15:41:28 <ka6sox> what happened to a development tree and a stable tree...
Jan 04 15:41:56 <av500> it did not work
Jan 04 15:42:26 <av500> you still have "stable" kernels
Jan 04 15:42:30 <av500> that get backports
Jan 04 15:42:34 <av500> the rest is "dev"
Jan 04 15:42:35 * mranostay bear hugs av500 
Jan 04 15:42:48 * av500 blushes
Jan 04 15:42:56 <aholler_> vendors have time constraints and those doesn't fit well into the kernels timeless universe
Jan 04 15:43:28 <rabryn> can anyone tell me the difference between the yocto's meta-ti repository and github's beagleboard repository?  i'm evaluating the AM335x with yocto and an LCD3 cape for a new project.  Yocto's kernel support version A1 of the cape but I have A2 which results in a backlight issue.  Is github's repo the authority for the latest beaglebone kernel?
Jan 04 15:44:08 <aholler_> e.g. I've just got requried oks from 3 different maintainers, so now a patch will end up in 3.9 (summer).
Jan 04 15:45:07 <aholler_> and that problem isn't solvable, except by slowing down hw development
Jan 04 15:46:10 <jsabeaudry> Does ./patch.sh really clone a GB of patches? That's my current estimate at 10% progress
Jan 04 15:47:51 <mranostay> jsabeaudry: well the whole repo
Jan 04 15:48:03 <mranostay> kernel repo then applys the patches
Jan 04 15:48:13 <joel_> mdp: neat that pantelis is forcing phandle generation for all labelled nodes regardless of reference
Jan 04 15:48:24 <joel_> it is quite elegant
Jan 04 15:49:35 <jsabeaudry> mranostay, is there something I can clone with everything already applied?
Jan 04 15:49:54 <koen> prpplague: that link you posted had a hidden gem :)
Jan 04 15:50:17 <prpplague> koen: which one?
Jan 04 15:51:07 <koen> the armbrix one
Jan 04 15:51:20 <koen> prpplague: I highlighted it in my g+ reply
Jan 04 15:52:09 <koen> rabryn: there is *no* yocto support for beaglebone at all, I think you are very confused
Jan 04 15:52:13 <aholler_> av500: everyone makes errors and wrong design decisions, but some things I've seen are really horrible. ;)
Jan 04 15:52:29 <emeb_mac> belated good morning to you woglinde - my little icon wasn't hopping.
Jan 04 15:55:19 <mdp> aholler_: yes, beagleboard.org is a vendor with the same kind of time constraints..my point from earlier..
Jan 04 15:55:40 <aholler_> but of course, getting and keeping an overview over the apis in the kernel isn't easy.
Jan 04 15:55:54 <mdp> joel_: yep..
Jan 04 15:55:55 <rabryn> koen: I agree, very confused.  please straighten me out!  would the right move be to try an integrate the github's beagleboard kernel into my yocto build?
Jan 04 15:56:52 <aholler_> mdp: I don't talk about ti/omap in special, just in case someone might thought that.
Jan 04 15:56:58 <mdp> joel_: the nice thing is to step back and just look at the syntax...pretty simple to create a fragment with __overlay__
Jan 04 15:57:20 <mdp> aholler_: neither do I...my kernel universe is much bigger than TI :)
Jan 04 15:58:21 <koen> rabryn: following the instructions at http://www.angstrom-distribution.org/building-angstrom is all you'll ever need
Jan 04 15:58:54 <mdp> joel_, though I can just see the confusion people will have with the DT dmaengine bindings...at the end of the day, making a board work still requires that you grok the h/w
Jan 04 15:58:57 <aholler_> anyway, the kernel is still an exceptional project. keeping so much patches from devs of all kind without that everything breaks is still some kind of art ;)
Jan 04 16:00:05 <koen> aholler_: will you attend fosdem this year as well?
Jan 04 16:00:31 <joel_> mdp: true
Jan 04 16:00:39 <aholler_> I don't think so. nothing planned and nobody pays me to do so
Jan 04 16:00:45 <mdp> aholler_: agreed...despite some exception cases I like working on it
Jan 04 16:01:13 <joel_> mdp: I feel __local_fixups__ can go away if the phandle was a hash so you don't need to auto-increment local references
Jan 04 16:01:24 <aholler_> and last year with that cold weather on broken train was such a good experience ;)
Jan 04 16:01:28 <rabryn> koen: please allow me to boil my problem down a little.  i'm curious why there's a 'meta-ti' yocto layer over at git.yoctoproject.org and a beagleboard repo at github.  i like the poky distro over angstrom but ultimately my goal is to use a qt embedded with an LCD and the AM335 (2/4, haven't decided yet).  does the angstrom distro support qt embedded without x support?
Jan 04 16:01:29 <aholler_> s/on/and/
Jan 04 16:02:17 <mdp> joel_, that's just one example that's close to home atm, but it's true of just about everything in enabling a device or custom board. DT does nothing to help people understand how their hw interacts with the soc
Jan 04 16:02:52 <mdp> joel_, well, the design constraint was to *not* modify DT format
Jan 04 16:03:01 <mdp> he's pretty clear on that
Jan 04 16:03:08 <joel_> mdp: hehe, add hwmod to the mix and there we have more confusion :P
Jan 04 16:03:28 <joel_> don't think changing way phandles are generated modifies DT format
Jan 04 16:03:48 <mdp> we had the same nightmares on powerpc enablement for "end-users" too...DT does nothing to help the hard part
Jan 04 16:03:50 <joel_> infact it keeps things more like the way they were, by not having to add additional nodes
Jan 04 16:04:10 <prpplague> koen: hehe
Jan 04 16:04:34 <mdp> I can tell you that from the standpoint of having it all clean in the kernel before DT...and then after..it was nice...but still people needed to understand how to hookup their peripherals, dma channels, pci memory map, etc.
Jan 04 16:04:45 <joel_> right
Jan 04 16:05:09 <aholler_> and how to workaround erratas ;)
Jan 04 16:05:10 <mdp> but we had things abstracted such that you change it all in one place before and after
Jan 04 16:05:31 <mdp> aholler_: indeed!
Jan 04 16:06:02 <koen> fwiw: 'errata' is plural, 'erratum' is singular
Jan 04 16:06:08 <koen> </pet peeve>
Jan 04 16:06:15 <mdp> +1
Jan 04 16:06:20 <mdp> datum!
Jan 04 16:06:59 <aholler_> not to speak if they aren't known. finding broken stuff hw is some of the most annoying thing
Jan 04 16:07:19 <mdp> at least you'll never find broken hw from TI
Jan 04 16:07:19 <mranostay> koen: ok that is so OCD
Jan 04 16:08:12 <aholler_> mdp: everyone claims that, and sw-devs are at the end of the line.
Jan 04 16:08:37 <mdp> aholler_: you reminded me of the 'am335x' usb issue when you said that
Jan 04 16:08:54 <aholler_> I don't know about that ;)
Jan 04 16:09:11 <mdp> the cppi 4.1 dma thing
Jan 04 16:09:32 <koen> mranostay: http://lolzombie.com/4886/ocd-cdo/
Jan 04 16:10:18 <aholler_> mdp: I haven't followed that. only thing I've done in the last year is to try a new kernel version by occasion on my beagle classic c4 ;)
Jan 04 16:12:16 <mdp> I have a pile of old beagles here from pre-TI..have to find some use for them
Jan 04 16:12:36 <aholler_> koen: I'm even better with french than latin. And even my french has become very bad ;)
Jan 04 16:14:10 <aholler_> I've learned french instead of dead latin in school ;)
Jan 04 16:14:43 <koen> I learned dead latin and dead greek
Jan 04 16:14:49 <koen> and sucked at french
Jan 04 16:15:02 <aholler_> and the language side of my brain is somehow broken ;)
Jan 04 16:15:16 <ka6sox> okay so with all this talk about 3.7 and 3.8 kernels WHY is the demo using 3.2.28?
Jan 04 16:15:42 <koen> ka6sox: regressions
Jan 04 16:15:58 <mdp> *evil* vendor tree is still superior!
Jan 04 16:16:20 <mdp> the upstream developers are *evil* and regress everything
Jan 04 16:16:20 <koen> mdp: for xM non-vendor 3.2 is still superior
Jan 04 16:16:23 <AndrevS> usb issues..... reminds me of my C3.
Jan 04 16:16:31 <aholler_> ka6sox: maybe because ehci works
Jan 04 16:17:03 <aholler_> ka6sox: I can't boot from ehci on any version above 3.2.x
Jan 04 16:17:46 <aholler_> I've recently tried (again) to find the cause, but still without luck
Jan 04 16:18:12 <aholler_> currently I think it might be something with interrupt or dma.
Jan 04 16:19:43 <aholler_> because it just stalls sometimes on boot up when the disk (ehci) gets really active.
Jan 04 16:19:44 * ka6sox throws up his hands and mutters something about "stability"
Jan 04 16:20:34 <aholler_> and the problem is that bisecting doesn't work here. to many changes between 3.2 and 3.3
Jan 04 16:24:22 <jackmitchell> koen: mdp: I have a fix for SPI which should go, somewhere... it is against the 3.8 branch in github and should probably be applied as a bugfix to 3.7
Jan 04 16:24:45 <jackmitchell> it basically changes the DT SPI pin mux on the CLK pin from OUTPUT to INPUT
Jan 04 16:25:05 <koen> jackmitchell: submit a pullrequest against https://github.com/beagleboard/kernel/
Jan 04 16:27:14 <mdp> jackmitchell: did you see my comment about running into your e2e post the other day? our test team ran into an issue, cited that post, and it led me back to groking that fix.
Jan 04 16:27:29 <mdp> I've got it included in my update to the dmaengine series..not yet posted
Jan 04 16:27:41 <mdp> jackmitchell: thanks!
Jan 04 16:27:51 <jackmitchell> mdp: yes, I noticed it this morning and made me laugh
Jan 04 16:28:10 <jackmitchell> mdp: when did I manage to slip that in, it must have been sometime in summer
Jan 04 16:28:26 <mdp> makes obvious sense now that I realized what you were doing
Jan 04 16:28:50 <mdp> and I had to fix some spidev_test.c issues our test team were having
Jan 04 16:30:19 <jackmitchell> mdp: re: update to dmaengine, that is in regard to the OUTPUT->INPUT spi pin chage above yes?
Jan 04 16:30:46 <mdp> I've got a bunch of updates and rebase to 3.8-rc2..that is included now
Jan 04 16:31:10 <mdp> well, at least in the dts part I include for testing
Jan 04 16:31:34 <mdp> your dts for the beagleboard/kernel will be different
Jan 04 16:31:51 <jackmitchell> ok thats fine, cheers!
Jan 04 16:36:04 <rabryn> meta-beagleboard/common-bsp/linux.inc inherits machine_kernel_pr but bitbake can't find it... it's not in the meta-beagleboard nor openembedded-core/meta... i assume this contains the package revision, but where is it?
Jan 04 16:39:49 <woglinde> angstroem perhaps?
Jan 04 16:49:01 <av500> rabryn: [16:52:09] <koen> rabryn: there is *no* yocto support for beaglebone at all, I think you are very confused
Jan 04 16:49:11 <av500> http://www.angstrom-distribution.org/building-angstrom
Jan 04 16:49:14 <av500> rabryn: ^^^^^
Jan 04 16:52:32 <aholler_> bone != board
Jan 04 16:55:12 <rabryn> av500: understood, but my curiosity overwhelms me so i have to ask... what is 'meta-ti' at 'http://git.yoctoproject.org/' then?
Jan 04 16:56:13 <rabryn> in the meta-ti layer it defines a beaglebone machine.  i'm not trying to be difficult, simply want to understand...
Jan 04 16:57:13 <Crofton|work> it is wording thing
Jan 04 16:57:33 <Crofton|work> The yocot project does not directly support boards
Jan 04 16:57:42 <Crofton|work> (except for a small number of reference cases)
Jan 04 16:57:52 <Crofton|work> it does provide hosting for some bsps though
Jan 04 16:58:04 <denix> rabryn: oh, yes, you are trying to be difficult...
Jan 04 16:58:18 <av500> rabryn: understand that the yocto project is not the beagle project
Jan 04 16:58:29 <av500> yocto might *want* to support the bone
Jan 04 16:58:30 <Crofton|work> people tend to attach the word yocto to things to impress managers
Jan 04 16:58:40 <Crofton|work> :)
Jan 04 16:58:41 <denix> rabryn: fyi, yocto also defines beagleboard machine, as well as meta-ti and meta-beagle...
Jan 04 16:58:52 <ka6sox> Crofton|work, boy is that the truth.
Jan 04 16:59:01 <denix> and av500 is right - yocto soon to define bone as well...
Jan 04 16:59:03 <mdp> denix, meta-ubuntuphone?
Jan 04 16:59:12 <Crofton|work> rofl
Jan 04 16:59:14 <av500> meta-meta-meta?
Jan 04 16:59:14 <denix> mdp, exactly
Jan 04 16:59:26 <mdp> the future is today
Jan 04 16:59:35 <av500> mdp: dont forget ubuntu runs on android phones
Jan 04 16:59:54 <mdp> whoa
Jan 04 17:00:06 <aholler_> as a first class citizen ;)
Jan 04 17:00:10 <mdp> lol
Jan 04 17:00:14 <Crofton|work> we all saw Shuttleworths keynote where he said everything not done by Canonical sucks, right?
Jan 04 17:00:17 <av500> and android phones are linux, but then again they are so not
Jan 04 17:00:25 <av500> so it gets confusing
Jan 04 17:00:34 <av500> whether ubuntu runs on linux or linux on ubuntu
Jan 04 17:00:34 <denix> what are android phones?
Jan 04 17:00:38 <av500> or android on linux on ubuntu
Jan 04 17:00:41 <denix> it's just hardware...
Jan 04 17:00:42 <av500> denix: good ones
Jan 04 17:01:08 <rabryn> hmm... alright, well, thanks for the info.  so, if i may ask for some advice, i'm goal is to build a distro that supports Qt Embedded without X11 support (qte to framebuffer), would you recommend i use angstrom or poky?  size isn't a major concern as i'm booting off sd.
Jan 04 17:01:23 <av500> why not angstrom?
Jan 04 17:01:30 <av500> it has full bone support
Jan 04 17:01:41 <Crofton|work> well, angstrom and poky are distros already
Jan 04 17:01:43 <aholler_> it's all in the cloud, so who cares about the real environment, it just works 99.999 of all time ;)
Jan 04 17:01:44 <woglinde> qte is core
Jan 04 17:01:49 <woglinde> in
Jan 04 17:01:54 <woglinde> so it doesnt matter
Jan 04 17:02:01 <av500> woglinde: stop confusing us with facts
Jan 04 17:02:08 <woglinde> *g*
Jan 04 17:02:12 <woglinde> layer
Jan 04 17:02:12 <mdp> angstrom has a robust app ecosystem too
Jan 04 17:02:13 <woglinde> core
Jan 04 17:02:17 <woglinde> meta-oe hurray
Jan 04 17:02:20 <denix> rabryn: https://lists.yoctoproject.org/pipermail/meta-ti/2012-October/001734.html
Jan 04 17:02:26 <av500> mdp: can I monetize it?
Jan 04 17:02:30 <Crofton|work> where poky is anything that uses meta-yocto (meta-yocto should have been called meta-poky)
Jan 04 17:02:32 <av500> or even monetizen it?
Jan 04 17:02:37 <mdp> either way
Jan 04 17:02:45 <denix> Crofton|work: amen!
Jan 04 17:02:49 <woglinde> titzen
Jan 04 17:02:51 <av500> or. monet-tizen it for mranostay
Jan 04 17:03:02 <denix> woglinde: tit-zen? ok
Jan 04 17:03:27 <woglinde> we will never see this phone outsie jp
Jan 04 17:03:46 <av500> korea
Jan 04 17:03:51 <woglinde> ?
Jan 04 17:03:54 <woglinde> no jp
Jan 04 17:04:08 <woglinde> in korea it might be manufactored
Jan 04 17:04:15 <rabryn> denix: NICE!  that explains it very well!  thanks for that.
Jan 04 17:04:22 <av500> denix: this thread sucks
Jan 04 17:04:25 <av500> there is no flamewar
Jan 04 17:04:27 <av500> at all
Jan 04 17:05:26 <woglinde> av500 soon it will be northkorea
Jan 04 17:05:32 <woglinde> I forgot to mention
Jan 04 17:08:59 <mdp> wow, it even explains publicly why everything is stuck at old kernel releases. impressive.
Jan 04 17:09:31 <woglinde> mdp 3.0.17?
Jan 04 17:09:50 <mdp> psp 3.2
Jan 04 17:11:08 <denix> mdp: https://lists.yoctoproject.org/pipermail/meta-ti/2012-October/001742.html
Jan 04 17:12:12 <mdp> that's what I meant
Jan 04 17:12:21 <mranostay> dmx cape is popular it seems
Jan 04 17:12:58 <av500> dmx > geiger
Jan 04 17:13:00 <denix> mdp: now it's up to you to deliver "full" bone support with 3.8 release... :)
Jan 04 17:13:09 <mdp> I did laugh at that
Jan 04 17:13:12 <av500> mranostay: everybody loves diso
Jan 04 17:13:14 <av500> disco
Jan 04 17:13:26 <av500> nobody loves nucular
Jan 04 17:13:27 <mranostay> everyone?
Jan 04 17:13:33 <mdp> denix, despite all the warning to never throw out kernel versions
Jan 04 17:14:03 <mranostay> av500: girls are scared by radiation i've learned :P
Jan 04 17:14:13 <mdp> denix, I'm probably moving mostly to OMAP soon anyway, not my problem any longer :)
Jan 04 17:14:23 <denix> I know, I know - I was just tired of all the questions
Jan 04 17:14:47 <mdp> yeah, I've mentioned this stuff several times on irc...probably the wrong place. :)
Jan 04 17:15:07 <ka6sox> irc is where it scrolls off...never to be seen again.
Jan 04 17:15:29 <aholler_> sounds like "exact steps"-disease
Jan 04 17:16:16 <denix> mdp: I guess predicting the end of the world is easier than the full bone support... :)
Jan 04 17:16:32 <mdp> ka6sox: yes and no, there is the public logger run here against the recommendation of the freenode staff :)
Jan 04 17:16:44 <av500> against?
Jan 04 17:17:12 <denix> mdp: nobody reads logs
Jan 04 17:17:12 <aholler_> yeah, logging chats isn't nice
Jan 04 17:17:24 <denix> mdp: when was the last time you visited /var/log?
Jan 04 17:17:29 <mdp> "If you're considering publishing channel logs, think it through. The freenode network is an interactive environment. Even on public channels, most users don't weigh their comments with the idea that they'll be enshrined in perpetuity. For that reason, few participants publish logs."
Jan 04 17:18:29 <av500> denix: that was a log time ago :)
Jan 04 17:18:40 <denix> good one
Jan 04 17:18:49 <mdp> log time no see
Jan 04 17:19:14 <ka6sox> http://logs.nslu2-linux.org/livelogs/
Jan 04 17:19:42 <ka6sox> I guess they must HATE me...
Jan 04 17:19:47 <av500> yes
Jan 04 17:19:50 <mdp> infidels
Jan 04 17:19:56 <av500> well, irc is public
Jan 04 17:20:01 <av500> anybody can log  and publish
Jan 04 17:20:31 <av500> randi zuckerberg has to deal with that
Jan 04 17:20:32 <aholler_> big brother ka6sox
Jan 04 17:21:25 <panto> back
Jan 04 17:21:49 <mdp> mranostay, one thing I liked about the dmx cape (before it was ever in the contest) is the very sane use of PRU for that protocol
Jan 04 17:22:33 <aholler_> I would like to buy a bone+ubercape instead of an ubertooth ;)
Jan 04 17:23:23 <mdp> mranostay, the only problem is that I don't think he understands that there's no need to use the SoC GPIO IP to do gpi/gpo from the PRU :(
Jan 04 17:23:39 <mdp> mranostay, the "community" should write a programming guide :)
Jan 04 17:36:03 <rabryn> so i guess i'm confused again... i followed the 'building angstrom' steps, ran 'MACHINE=beaglebone ./oebb.sh config beaglebone' then checked out the sources dir only to find meta-ti vs. meta-beagleboard (which is where i believe the latest kernel for the bone is at).  why wouldn't angstrom use meta-beagleboard for bone support?
Jan 04 17:53:43 <mranostay> heh
Jan 04 17:54:51 <jsabeaudry> How to know if a driver already exists for a i2c chip, AD5627BCPZ for example?
Jan 04 17:55:01 <woglinde> git grep?
Jan 04 17:55:24 <jsabeaudry> woglinde, grep what? the kernel?
Jan 04 17:57:08 <woglinde> sure
Jan 04 17:57:18 <woglinde> you could try git grep google
Jan 04 17:57:19 <rabryn> i assume i was on the wrong branch in the angstrom repo...
Jan 04 17:57:24 <woglinde> but that would last a while
Jan 04 17:57:42 <jsabeaudry> woglinde, I gave the old googster a spin without much luck
Jan 04 17:58:38 <mranostay> google?
Jan 04 17:59:39 <aholler_> git grep -i AD56
Jan 04 18:04:06 <mranostay> aholler_: those are exact seteps!
Jan 04 18:04:09 <mranostay> *steps even
Jan 04 18:04:43 <aholler_> yeah, sometimes I'm lazy
Jan 04 18:05:12 <mranostay> hey prpplague
Jan 04 18:05:31 <prpplague> mranostay: yes?
Jan 04 18:05:32 <mranostay> you guys extended the CFP a week?
Jan 04 18:05:55 <prpplague> mranostay: i dont think we have time to do that, why?
Jan 04 18:06:19 <panto> prpplague, "All submissions must be received by January 11, 2013 at 11:55pm (Pacific Time Zone, Extended Deadline Due to Holiday)."
Jan 04 18:06:23 <Russ> prpplague, I'm going to miss your lunch
Jan 04 18:06:50 <ds2> maybe everyone is going to SCaLe
Jan 04 18:06:51 <ds2> ;)
Jan 04 18:07:01 <mranostay> 3 days to review everything? :)
Jan 04 18:07:06 <Russ> scale could stand to extend their deadline a bit
Jan 04 18:07:29 * Russ looks again, nm, different dave
Jan 04 18:08:07 <prpplague> news to me
Jan 04 18:08:16 <prpplague> i suspect TimB did that
Jan 04 18:08:18 * prpplague looks
Jan 04 18:08:27 <ds2> Review Accelerator blocks
Jan 04 18:08:50 <Russ> apparently the punctuation on the current crop of submissions is that bad
Jan 04 18:09:26 <prpplague> yea had an email from TimB i missed
Jan 04 18:15:42 <mranostay> prpplague: really because of the holiday or is that a cover story? :)
Jan 04 18:16:03 * prpplague has no comment
Jan 04 18:16:27 <mranostay> hehe
Jan 04 18:23:33 * prpplague grumbles
Jan 04 18:23:44 <prpplague> i am starting to get behind on my ces prep
Jan 04 18:24:00 * prpplague needs to go to frys to purchase a couple cheap multimeters
Jan 04 18:24:24 <mranostay> prpplague: you don't have dozens in your flat?
Jan 04 18:24:50 <prpplague> mranostay: they need to be pretty and new for CES
Jan 04 18:25:10 <Russ> prpplague, you can send them to me when they get old and scruffed
Jan 04 18:25:27 <Russ> btw, people at CES only respond to Fluke
Jan 04 18:26:14 <mranostay> Fluke or go home?
Jan 04 18:26:20 <prpplague> hehe
Jan 04 18:26:26 <Russ> no one ever got fired for buying a Fluke
Jan 04 18:26:39 <mranostay> i want that above comment taken out of context!
Jan 04 18:27:19 <mranostay> av500: where the hell is mru?
Jan 04 18:27:39 <mranostay> heh didn't die or something on us, right?
Jan 04 18:28:19 <Russ> I thought he was another one of those trolls that went skiing
Jan 04 18:28:25 <prpplague> http://www.frys.com/product/6894436?site=sr:SEARCH:MAIN_RSLT_PG
Jan 04 18:28:26 <prpplague> hmm
Jan 04 18:28:31 <prpplague> that one looks reasonable
Jan 04 18:28:34 <prpplague> has a big display
Jan 04 18:29:09 <mranostay> seen worse
Jan 04 18:29:35 * prpplague wanders off to go to frys
Jan 04 18:29:37 <prpplague> bbiab
Jan 04 18:38:04 <mdp> jsabeaudry: ad5064.c should cover that one
Jan 04 18:41:55 <jsabeaudry> mdp, thanks, any special trick to find that out or you just happened to know?
Jan 04 18:43:02 <mdp> used the same cmdline as aholler_ suggested on my nightly local kernel.org ref tree.
Jan 04 18:44:12 <mdp> 5627 failed as it's not explicitly supported..but since 5628 is in the ID list, I suspect it's a minor change to support what I assume is a similar family member
Jan 04 18:44:19 <mdp> that driver supports a ton of parts
Jan 04 18:45:33 <mdp> there's a ton of AD adc/dacs in drivers/iio/ since AD wrote the iio subsystem
Jan 04 18:47:29 <ka6sox> Russ, SCaLE gets pressure to let people early so that they can get the time off of work.
Jan 04 18:47:47 <jsabeaudry> Interesting, I'll give it a try see if 5627 is indirectly supported
Jan 04 18:47:57 <ka6sox> s/let people/let people know/
Jan 04 18:51:06 <jsabeaudry> mdp, Hmmm, 5064 and 5628 look like they are SPI and 5627 is I2C
Jan 04 18:51:26 <mdp> register map might be the same
Jan 04 18:52:02 <jsabeaudry> mdp, Ah, Ok, nothing like trying it out, thanks for the help
Jan 04 18:57:05 <mdp> at least they have a lot of s/w for their adc/dacs in the kernel :)
Jan 04 18:57:08 <mdp> *ahem*
Jan 04 19:00:25 * mranostay pops on some St Anger
Jan 04 19:00:29 * mranostay blares at mdp
Jan 04 19:02:52 <koen> mranostay: post sellout mellow crap
Jan 04 19:03:11 <mdp> hehe
Jan 04 19:04:07 <mdp> I already have my angry music on
Jan 04 19:04:31 <mranostay> koen: true
Jan 04 19:04:33 <mdp> it's like the am335x theme
Jan 04 19:04:48 <mranostay> next is KMFDM
Jan 04 19:05:04 <panto> mranostay, stop it, you're scaring me
Jan 04 19:05:10 <panto> ((snicker))
Jan 04 19:05:13 <mdp> hehe
Jan 04 19:05:17 * mranostay smacks panto 
Jan 04 19:08:02 <mdp> mranostay, speaking of angry music...I got a laugh out of dave mustaine's interwebs rant against....Men's Wearhouse! :)
Jan 04 19:08:02 <mranostay> panto: btw i hate you :P
Jan 04 19:08:03 <mdp> hahaha
Jan 04 19:08:12 <panto> hehe
Jan 04 19:08:40 * panto has flashbacks of the opening scene in office space 
Jan 04 19:08:45 * koen types that into lycos
Jan 04 19:09:37 <koen> mdp: speaking of german music and TGIF: http://www.youtube.com/watch?v=RZPtxGGT0uE
Jan 04 19:10:31 <mdp> mranostay, I generally stick to the first three albums prior to sellout
Jan 04 19:11:18 <mranostay> koen: sfw?
Jan 04 19:11:38 <koen> mranostay: yes
Jan 04 19:12:05 <koen> mdp: black is a sellout?
Jan 04 19:12:21 <mdp> I just tend to stay in the safe zone ;)
Jan 04 19:12:49 <mdp> I don't hate that stuff...just really appreciate the intense level of anger on 1-3
Jan 04 19:13:03 <koen> 2-4?
Jan 04 19:13:17 <mdp> yeah
Jan 04 19:13:36 <mdp> koen, +1 on that video ;)
Jan 04 19:14:24 <koen> wait till the end
Jan 04 19:14:35 <koen> "respect to the man in the icecream van"
Jan 04 19:14:58 <koen> takes me back to being 16 again
Jan 04 19:15:13 <mdp> lol
Jan 04 19:16:01 <mranostay> heh
Jan 04 19:16:21 <mdp> hehe, I love all the android people in denial over the sdk terms change
Jan 04 19:16:34 <panto> oh, what changed?
Jan 04 19:16:35 <mranostay> koen: how high do you have to be to come up those lyrics? :)
Jan 04 19:17:10 <koen> mranostay: it's a bottle blonde german with a megaphone, no drugs needed
Jan 04 19:17:11 <mdp> panto, few months ago...quietly...the anti-fragmentation clause popped into the sdk license terms ;)
Jan 04 19:17:18 <panto> ahh
Jan 04 19:17:29 <woglinde> ieehks scooter
Jan 04 19:17:32 <panto> the end of the handset vendor mods
Jan 04 19:17:43 <woglinde> hyper hyper
Jan 04 19:17:56 <mdp> and, of course, like any good corporate "OSS" project, the commit doesn't say anything about it..just a big whole replace of the file
Jan 04 19:18:12 <koen> woglinde: +1
Jan 04 19:18:25 <mdp> panto, no more sdk derivatives
Jan 04 19:18:48 <panto> yeah, there's goes an independent vendor's android stack
Jan 04 19:19:00 <woglinde> koen sure but they made a lot of money with this shit
Jan 04 19:19:42 <mdp> panto, so you couldn't do a set of industrial automation classes against a post-october sdk and distribute that sdk with your product
Jan 04 19:19:58 <mdp> so you are stuck at the 4.x core api levels when you have to fork
Jan 04 19:21:43 <rabryn> koen: how would i build the meta-beagleboard kernel for the beaglebone while building angstrom?  the beaglebone machine defined in angstrom for beaglebone still uses the kernel out of the meta-ti layer (linux-ti33x-psp-3.2.28)...
Jan 04 19:22:53 <ds2> prpplague: you going to CES?
Jan 04 19:30:18 <mdp> um, seems the enet phy died on one of my A5s...*sigh*
Jan 04 19:41:33 <jsabeaudry> http://narcissus.angstrom-distribution.org/ is infected
Jan 04 19:42:16 <mranostay> koen: heh chrome is blocking it
Jan 04 19:44:41 <Russ> 'Malicious software is hosted on 2 domain(s), including naturalhomecoloncleanse.com/, habercanavari.com/.'
Jan 04 19:44:58 <Russ> '2 domain(s) appear to be functioning as intermediaries for distributing malware to visitors of this site, including aod.org.hk/, habercanavari.com/.'
Jan 04 19:45:23 <Russ> I'm so confused
Jan 04 19:45:36 <Russ> injection?
Jan 04 19:46:39 <mranostay> heh so those sites have a angstrom distro image on it? :)
Jan 04 19:46:42 <bradfa> mdp, did you by chance run an LCD backlight at a PWM other than 0 or 100%?
Jan 04 19:46:49 * Russ looks at a strange script at the bottom of the page source
Jan 04 19:48:05 <mdp> bradfa, I think I only tried a gpio backlight with st7735..I hear there's bugs that come and go in the pwm drivers
Jan 04 19:48:29 <mdp> I think mranostay/koen used a pwm backlight with st7735...maybe on 3.2
Jan 04 19:50:00 <prpplague> ds2: not, not going, just working on the demos
Jan 04 19:50:28 <Russ> yup, narcissus.angstrom-distribution.org is certainly infected
Jan 04 19:50:52 <Russ> '<iframe src="http://aod.org.hk/help/index2.php" width="1" height="3" frameborder="0"></iframe>'
Jan 04 19:51:14 <s5fs> Russ: well thats fantastic
Jan 04 19:51:21 <emeb> mranostay: http://www.goldmine-elec-products.com/prodinfo.asp?number=G19254
Jan 04 19:58:42 <mranostay> emeb: looks fragile
Jan 04 20:00:38 <joel_> panto: fdtdump is not built through Makefiles in the in kernel? I built it from mainline dtc tree though
Jan 04 20:01:10 <panto> joel_, no it is not built in kernel
Jan 04 20:02:17 <bradfa> mdp, ok, I've heard and believe that letting the TPS65217 do backlighting will cause all sorts of fun for the Ethernet PHY
Jan 04 20:02:28 <joel_> panto: ah ok, cool
Jan 04 20:02:44 <emeb> mranostay: you say fragile, I say kewl!
Jan 04 20:02:58 <mdp> bradfa, I heard the same thing regarding that lcd cape using the pmic backlight
Jan 04 20:03:13 <panto> bradfa, I can confirm
Jan 04 20:03:29 <bradfa> I believe the ground loop is the root cause
Jan 04 20:03:32 <panto> using the backlight at any other setting than 100% will cause the ethernet PHY to lose link
Jan 04 20:03:33 <bradfa> no ground in battery header
Jan 04 20:07:41 <mdp> btw, when you said pwm originally, I assume you meant the ehrpwm..which is what has been tested
Jan 04 20:07:50 <mdp> good
Jan 04 20:09:08 <mranostay> bradfa: only ehrpwm
Jan 04 20:09:37 <bradfa> mranostay, ? I'm talking of pmic pwm
Jan 04 20:09:53 <mranostay> yeah i know
Jan 04 20:09:57 <mdp> don't use the pmic pwm
Jan 04 20:10:17 <bradfa> good, now that we're all confused :)
Jan 04 20:10:28 <mranostay> bradfa: if you are designing a cape don't use the PMIC header *ahem* :)
Jan 04 20:10:41 <mdp> doctor, doctor, it hurts when I use the pmic pwm!
Jan 04 20:11:00 <bradfa> mranostay, been using the pmic header for about 6 months, no issues so far, but we don't use anything other than lithium battery
Jan 04 20:11:08 <bradfa> temp sense works, though :)
Jan 04 20:11:11 <bradfa> for battery
Jan 04 20:11:34 <bradfa> hot battery == no chargy
Jan 04 20:11:50 <bradfa> but just simulations, no actual hot batteries yet
Jan 04 20:15:20 <rabryn> is the kernel defined in meta-beagleboard compatible with a beaglebone?
Jan 04 20:42:56 * _av500_ did it
Jan 04 20:43:21 <woglinde> what?
Jan 04 20:43:26 <_av500_> *it*
Jan 04 20:43:30 <panto> sex?
Jan 04 20:43:31 <woglinde> no
Jan 04 20:43:42 <_av500_> better :)
Jan 04 20:43:49 <panto> two girls at the same time?
Jan 04 20:43:56 <_av500_> lol
Jan 04 20:44:05 <woglinde> he bought a new allwinner
Jan 04 20:44:16 <_av500_> woglinde: I never buy stuff
Jan 04 20:44:19 <_av500_> or rarely
Jan 04 20:44:25 <_av500_> I have allwinner for free
Jan 04 20:44:28 <panto> http://www.youtube.com/watch?v=A939QRRSNV4
Jan 04 20:44:31 <_av500_> getting the new 4x A7 soon
Jan 04 20:44:49 <woglinde> hm a7?
Jan 04 20:45:19 <_av500_> woglinde: yes
Jan 04 20:45:26 <_av500_> they are using up all the numbers
Jan 04 20:45:54 <_av500_> A7 is meant as companion to A15 in big.LITTLE setup
Jan 04 20:46:03 <_av500_> but the chinese just take A7 :)
Jan 04 20:48:42 <woglinde> hm ah right
Jan 04 20:48:46 <woglinde> remember now
Jan 04 20:50:01 <mrpackethead> .
Jan 04 20:50:48 <mdp> panto, followed by the timeless advice: http://www.youtube.com/watch?v=sHjlM5A1jME
Jan 04 20:50:49 <woglinde> ..
Jan 04 20:51:17 <mrpackethead> hello folks
Jan 04 20:53:51 <mrpackethead> mph = mrpackethead, now i remembered my password
Jan 04 20:59:50 <mrpackethead> A7 What differences from the currnet A6
Jan 04 21:07:48 <_av500_> A&?
Jan 04 21:07:51 <_av500_> A6?
Jan 04 21:09:29 <bradfa> mrpackethead, no such Cortex-A6 exists in public knowledge
Jan 04 21:09:38 <bradfa> well, on ARM website at least
Jan 04 21:10:34 <mrpackethead> _av500_ claimed he was getting 4 x A7 about 25 minuts ago
Jan 04 21:10:47 <bradfa> mranostay, Cortex-A7 is a thing
Jan 04 21:10:57 <bradfa> ah, gah, sorry mranostay
Jan 04 21:11:04 <bradfa> s/mranostay/mrpackethead/
Jan 04 21:11:16 <bradfa> stupid autocomplete
Jan 04 21:12:30 <bradfa> mrpackethead, does that make sense?
Jan 04 21:13:04 <bradfa> mrpackethead, what A6 are you talking about?
Jan 04 21:13:04 <jsabeaudry> Back in the 3.2 days I had a couple of patches over the board-am335xevm.c file, has this file just been renamed or has the system changed completely?
Jan 04 21:13:23 <bradfa> jsabeaudry, where you're now where and wanting to compare to 3.2?
Jan 04 21:13:49 <bradfa> mainline?
Jan 04 21:14:00 <jsabeaudry> bradfa, I'm trying to update to 3.8, I'm not sure I understand your question fully
Jan 04 21:14:27 <bradfa> jsabeaudry, it wasn't clear to me where you are looking now and not finding a board file
Jan 04 21:15:03 <bradfa> On 3.8 you should use device tree
Jan 04 21:15:06 <bradfa> no more board file
Jan 04 21:15:23 <bradfa> jsabeaudry, but device tree solves everything, so you should be in good hands :)
Jan 04 21:15:24 <jsabeaudry> Ah ok, new system, that's what I feared
Jan 04 21:15:43 <bradfa> am335x was one of the first ARM devices to be required to use device tree
Jan 04 21:15:55 <bradfa> slowly all will be moving also
Jan 04 21:16:06 <bradfa> (thus the "solves everything" meme)
Jan 04 21:16:15 <Russ> can't wait to port tuxscreen to do
Jan 04 21:16:17 <Russ> er, dt
Jan 04 21:16:35 <Russ> anyone have a cat system to test on?
Jan 04 21:16:52 <bradfa> Russ, er? tabby or some other kind?
Jan 04 21:17:00 <Russ> sorry, cats
Jan 04 21:17:06 <jsabeaudry> bradfa, Device tree as in www.devicetree.org ?
Jan 04 21:17:42 <Russ> http://www.simtec.co.uk/products/EB110ATX/resources.html
Jan 04 21:17:50 <bradfa> jsabeaudry, yeah
Jan 04 21:18:06 <bradfa> see examples, one of which should be for am335x
Jan 04 21:18:15 <bradfa> or take a look at koen's github
Jan 04 21:18:53 <bradfa> Russ, ah, ok, now I see
Jan 04 21:19:13 <bradfa> jsabeaudry, examples in kernel sources, there's a few
Jan 04 21:19:44 <jsabeaudry> bradfa, great, I'll take a look, shouldn't be hard just chaning a few pin muxes and adding a couple of i2c devices
Jan 04 21:19:58 <Russ> I'm just naming old now arcane systems
Jan 04 21:20:24 <bradfa> jsabeaudry, basically it's the board file but in a new format and some new ideas to grok
Jan 04 21:21:41 <bradfa> and there's some things you could do in a board file that you can't do with dt
Jan 04 21:22:21 <bradfa> but many of those things you can't do in dt were kind of hack-ish in board files
Jan 04 21:22:49 <bradfa> but there's a few gripes that are valid about how dt doesn't actually solve the problems that board files did even though it's being forced to be a replacement for them
Jan 04 21:23:00 <bradfa> it's a fun topic :)
Jan 04 21:24:18 <jsabeaudry> bradfa, looks controversial, at least from all the jokes thrown at it here
Jan 04 21:25:03 <panto> bed time trolls
Jan 04 21:25:07 <panto> cya
Jan 04 21:25:32 <_av500_> yo
Jan 04 21:25:36 <_av500_> sleepytime
Jan 04 21:26:03 <mrpackethead> sleep? its just morning.
Jan 04 21:27:37 <mdp> bradfa, DT solves some problems..just not all problems and introduces new problems
Jan 04 21:32:17 <ka6sox> mrpackethead, ~ugt
Jan 04 21:32:28 <mrpackethead> ~ugt?
Jan 04 21:38:08 <bradfa> mdp, you're delusional, dt solves everything, including world hunger
Jan 04 21:38:29 <mdp> bah, I'm just worried about my hunger atm
Jan 04 21:40:44 * Crofton|work is concerned about his sobriety
Jan 04 21:41:40 <mranostay> Crofton|work: worried that you are sober?
Jan 04 21:41:58 <Crofton|work> exactly
Jan 04 21:49:20 <mdp> joel_ is asking philosophical questions about DT elsewhere so I'm afraid I need to start drinking now ;)
Jan 04 21:49:35 * mdp thanks Crofton|work's power of suggestion
Jan 04 21:49:50 <djlewis> mMMmm, beer . . .
Jan 04 21:50:53 <djlewis> last night I looked in the fridge for a cold beer == none,
Jan 04 21:51:03 <djlewis> turned out beer in closet was plenty cold :)
Jan 04 21:53:15 <emeb> +1 for winter time.
Jan 04 21:53:40 <mranostay> mdp: you mean peer pressure
Jan 04 21:53:58 <emeb> beer pressure
Jan 04 21:54:03 <mdp> mmmm
Jan 04 21:59:27 <mranostay> yeah shake the can for that
Jan 04 21:59:57 <mdp> *can*
Jan 04 22:00:38 <mranostay> bottle
Jan 04 22:01:00 <mrpackethead> i got an interesting email from TI today
Jan 04 22:01:09 <mrpackethead> about the ICE and the ethercat drivers
Jan 04 22:01:33 <mrpackethead> turns out that they did all their validation testing for the ICE and the Ethercat in linux
Jan 04 22:01:51 <mrpackethead> then created their SYS/BIOS RT thing
Jan 04 22:02:13 <mrpackethead> so all the work to get my ethercat interfaces running has alrady been done.
Jan 04 22:02:30 <mrpackethead> but they dont' know if they can release it.
Jan 04 22:02:36 <mrpackethead> palm_head_splat
Jan 04 22:03:01 <mranostay> mrpackethead: welcome to the real world :)
Jan 04 22:03:21 * mdp sips
Jan 04 22:03:24 <mrpackethead> i'll threaten them with 10000 units
Jan 04 22:04:42 <mranostay> add a another 0
Jan 04 22:04:57 <mrpackethead> yeah..
Jan 04 22:05:09 <mrpackethead> 10,000 units they may belive
Jan 04 22:05:15 <mrpackethead> but thats probalby not enough to make them interested
Jan 04 22:05:27 <mdp> that surprises me...i thought there was a supported release of ethercat for linux, oh well
Jan 04 22:05:55 <mrpackethead> mdp: do you? where do you think you saw it?
Jan 04 22:05:56 <mdp> the ICE folks are essentially on another planet though
Jan 04 22:06:27 <mdp> I just assumed since I knew we did all the ICE validation with linux already
Jan 04 22:06:33 <mdp> that's how it's been done on new parts
Jan 04 22:07:02 <mrpackethead> mdp: your a TI employee as well
Jan 04 22:07:06 <mrpackethead> ?
Jan 04 22:07:40 <mrpackethead> the guys at TI-india told me last night that the ICE validation was done with linux
Jan 04 22:07:47 <prpplague> mranostay: being TI is more of a state of mind than an actual fact
Jan 04 22:08:02 <prpplague> opps that was for mrpackethead
Jan 04 22:08:49 <mdp> mrpackethead: yes, that's what I mean...that's how bringup is done..with linux. sounds like they just choose not to support a release
Jan 04 22:09:05 <mdp> my guess is there are no large customers for it on linux
Jan 04 22:09:20 <mrpackethead> they did admit that my use-case was somethign that they had not anticipated
Jan 04 22:09:33 <mdp> sounds typical
Jan 04 22:09:46 <mrpackethead> and that using linux rather than SYS/BIOS was a compelling argumetn
Jan 04 22:10:55 <mdp> heh
Jan 04 22:11:51 <mrpackethead> there is a huge variance in processor cost across distributoers.
Jan 04 22:11:56 <mrpackethead> even at 1 off
Jan 04 22:12:41 <ds2> anyone know off hand what are the big differences between a 3507 and the 33x in the bone?
Jan 04 22:12:43 <mdp> the advertised prices aren't real anyway
Jan 04 22:13:03 <mdp> buyers get down to the real price when they put a little effort into it
Jan 04 22:13:27 <mdp> and if you are a .cn company...you can get stuff for nearly free ;)
Jan 04 22:14:17 <mdp> ds2, an exhaustive list?
Jan 04 22:14:33 <ds2> mdp: no... just main high lights
Jan 04 22:14:36 <mrpackethead> yes..
Jan 04 22:14:41 <ds2> from a high level, there isn't that much of a difference
Jan 04 22:14:52 <ds2> A9, ethernet MAC
Jan 04 22:15:33 <ds2> mdp: trying a mental experiment to see if a bone equiv can be done with the 3507 or 3517
Jan 04 22:15:33 <mdp> am3505/3517?
Jan 04 22:15:45 <ds2> it is am P/N
Jan 04 22:15:53 <mdp> am35x vs am33x...here we go
Jan 04 22:15:57 <ds2> it is a 3505? arrrggggg
Jan 04 22:16:00 <mdp> dss vs lcdc
Jan 04 22:16:10 <mrpackethead> why do you want to clone the bone
Jan 04 22:16:12 <mdp> ... vs pruss
Jan 04 22:16:23 <mdp> omappwm vs. ehrpwm
Jan 04 22:16:28 <mdp> ... vs ecap
Jan 04 22:16:41 <ds2> oh sounds like different revs/flavors of the peripherals?
Jan 04 22:16:50 <mdp> sdma vs edma
Jan 04 22:16:56 <prpplague> mdp: hey your 6502 pru stuff has got nothing on these RPi folks......
Jan 04 22:16:59 <prpplague> mdp: http://hackaday.com/2013/01/04/turning-a-raspberry-pi-into-an-snes/
Jan 04 22:17:41 <ds2> it sounds like a bone equiv that can share the classic kernels can be done with the 3505
Jan 04 22:17:46 <ds2> or even the 3517
Jan 04 22:17:55 <mdp> davinci emac on am35x ... cpsw gige/switch on am33xx
Jan 04 22:18:21 <mdp> ds2, yes, except it's been orphaned afa kernel support ...
Jan 04 22:18:32 <mdp> prpplague: lol, yes, saw that :)
Jan 04 22:18:55 <mdp> depending on what you need am35x is still useful
Jan 04 22:19:24 <mdp> it has really bad PM versus am37x
Jan 04 22:19:33 <ds2> mdp: really? didn't seem like it... just did a project with it
Jan 04 22:19:46 <ds2> mdp: the PM on the 33x isn't exactly working either
Jan 04 22:19:54 <mdp> am37x is lower power, highest performance
Jan 04 22:19:59 <mdp> I said, "versus am37x" :)
Jan 04 22:20:12 <ds2> am37x is also $$$$
Jan 04 22:20:15 <mdp> yep
Jan 04 22:20:43 <ds2> love the am/dm37x... but then you hear the BOM cost complaints
Jan 04 22:21:46 <mdp> even if PM was working on am33x...by design and theorectical numers it's worse than am37x's system suspend number in practice
Jan 04 22:22:01 * mdp typos like crazy
Jan 04 22:22:28 <ds2> yes but in reality, unless your email is kh......@...., it doesn't exactly mean much
Jan 04 22:22:30 <mdp> but my point is that they have a lot of differences...if you value the differences
Jan 04 22:22:37 <mdp> hrm
Jan 04 22:22:49 <ds2> yes, but the flip side is the 33x just doesn't look that mature yet
Jan 04 22:22:59 <mdp> in my previous role here, the customer I supported cared very much...one reason they couldn't move to am33x
Jan 04 22:23:39 <mdp> so many different use cases in this world ;)
Jan 04 22:24:00 <rabryn> PLEASE if someone knows about building angstrom for beaglebone help... is it normal to have to add the meta-beagleboard/common-bsp layer to setup-scripts/conf/bblayers.conf to get the latest kernel?  also, i had to edit setup-scripts/sources/meta-ti/conf/machine/beaglebone.conf to change the preferred kernel, again, is this right?
Jan 04 22:32:05 <mrpackethead> i better go and do a runner
Jan 04 22:35:16 <mdp> ds2, I wouldn't argue about the maturity at all
Jan 04 22:35:56 <mdp> most of the drivers for the IP it inherited were not mature on davinci or bitrotted in terms of linux functionality
Jan 04 22:36:21 <mdp> am35x has it's own issues..upstream has gotten a little bit better
Jan 04 22:38:05 <ds2> mdp: yes. been through some of the am35x issues
Jan 04 22:38:14 <ds2> along with device errata an conflicting datasheet info ;)
Jan 05 01:29:33 <mranostay> hum de hum
Jan 05 01:35:19 <emeb> mranostay: what you doing here - should be out bar hopping or something.
Jan 05 01:36:53 <mranostay> only 5:30
Jan 05 02:33:14 <mrpackethead> ...
Jan 05 02:55:58 <mranostay> mrpackethead: dialup?
**** ENDING LOGGING AT Sat Jan 05 02:59:59 2013