**** BEGIN LOGGING AT Thu Jun 06 03:00:22 2013 Jun 06 03:12:27 * mranostay takes jkridner|afk's keyboard Jun 06 03:12:35 Crofton: RF dude Jun 06 03:19:48 mranostay: what are you up to? tesla coils? Jun 06 03:21:13 nothing recently Jun 06 03:29:11 * pru_evtout_2 detects mranostay is a liar. Jun 06 03:33:54 pru_evtout_2: no fun projects Jun 06 03:34:06 outside of work is what i think ds2 is talking about Jun 06 03:51:52 h junkerd Jun 06 03:52:27 *hi even Jun 06 04:02:43 av500: you germans need something to spice up your beers :P Jun 06 04:02:51 to hell with the purity law Jun 06 04:06:56 Now , android angstrom and ubuntu,all of them worked fine on my own board.That's very good. Jun 06 04:38:11 mranostay: here, have another scoop of ice cream in yours Jun 06 04:41:30 * dm8tbr thinks mranostay should be put on an exclusive 'diet' of belgian fruit beer for a month Jun 06 05:08:47 ka6sox: how goes the infection. Jun 06 05:08:52 ? Jun 06 05:10:34 *yawn* Jun 06 05:10:45 whats up there mr lowercase Jun 06 05:15:51 lol pru Jun 06 05:16:29 hello..I have a beagleboard xM. Can anyone help me making it boot Jun 06 05:19:44 pri: what have you tried so far Jun 06 05:21:33 I made the SD card image refering http://elinux.org/BeagleBoardUbuntu . I connected the Serial cable and a HDMI to VGA converter. I am not getting any output at either serial console or vga Jun 06 05:25:50 is that an active hdmi-to-vga converter? Jun 06 05:27:16 I dont know what does that mean. Its written HDMI/VGA 1080 Ultimate Jun 06 05:28:07 pri: do you see something on the serial cable? Jun 06 05:28:29 ok, looks like an active converter. now back to serial Jun 06 05:30:09 pru_evtout_2, feeling rotten but still working Jun 06 05:30:21 stupid time alignment isn't happening :p Jun 06 05:33:40 wow, need popcorn Jun 06 05:34:58 best thing...sitting in room watching GoT for the first time. Jun 06 05:39:21 +/-50mS is sucky Jun 06 05:40:16 ? Jun 06 05:41:13 someone ring? Jun 06 05:41:44 ring, ring, ring, ... bananaphone! Jun 06 05:41:56 dir Jun 06 05:42:03 die even Jun 06 05:45:08 av500, I've been fighting something for a few hours...don't want to drive anywhere to see whats wrong though :P Jun 06 05:46:39 beer? Jun 06 05:49:26 arm-netbook went lively today Jun 06 05:51:45 huh? Jun 06 05:52:39 av500 new toy ? Jun 06 05:52:57 https://lkml.org/lkml/2013/6/5/539 Jun 06 05:53:02 m_billybob: nope Jun 06 05:53:05 well, new toy too Jun 06 05:54:07 x86 is the future Jun 06 05:54:15 mranostay: ack Jun 06 05:54:26 intel is saving us from DT Jun 06 05:54:46 except arm-netbooks == ~$70 usd Jun 06 05:55:05 s/arm-netbooks/crappy arm-netbooks/ Jun 06 05:55:06 they're gettign cheaper by the minute it seems Jun 06 05:55:11 sure Jun 06 05:55:19 I prefer my i7 Jun 06 05:55:40 i prefer my A6 3400M ;) Jun 06 05:55:55 whatever that is Jun 06 05:56:07 AMD Lliano Jun 06 05:56:10 whats the new toy av500 Jun 06 05:56:17 wuad 1.4Ghz laptop Jun 06 05:56:23 quad* Jun 06 05:56:45 ka6sox: 7" tablet Jun 06 05:56:57 a nexus7 killer :) Jun 06 05:57:02 which one ? Jun 06 05:57:19 not on the market yet Jun 06 05:57:31 Nexus7 at the price point was a pretty good tablet Jun 06 05:58:50 sure Jun 06 05:58:52 still is Jun 06 06:01:11 Id like to see something that has a similar setup to the odroid processor in it. I did some reading and apparently one of the odriod setups is an i3 killer. Jun 06 06:01:34 if you squint Jun 06 06:02:32 well performance wise, price point made less sense at ~$450 usd entry fee Jun 06 06:03:05 cant get a decent 15" i3 latop for around that price Jun 06 06:03:08 can get* Jun 06 06:12:13 interesting seeing all the cheap allwinner quad-core micro-pc gizmos on DX. Jun 06 06:12:31 all in the $80-$100 range Jun 06 06:14:07 emeb_mac: I just got that email too Jun 06 06:14:29 emeb_mac: The MEEP kids' tablet is based on the allwinner a10 as well Jun 06 06:14:59 a usb wifi inside, microsd, minihdmi, etc Jun 06 06:15:14 fwiw Jun 06 06:16:49 stuff's getting cheap. Jun 06 06:17:26 <_Sy_> Good morning Jun 06 06:17:54 'lo Jun 06 06:19:13 emeb_mac, yeah i just did a quick search DX is selling on such system for $71 Jun 06 06:20:00 they've got some older lo-spec systems for $40 range Jun 06 06:23:56 false economy . . . Jun 06 06:25:08 my new top tip for BBB Jun 06 06:25:43 ? Jun 06 06:25:52 bend the USB port so it can't be used Jun 06 06:26:01 just reeadin the news groups Jun 06 06:26:10 heh ive had no problems with mine eitherone Jun 06 06:26:19 both USB ports workign fine Jun 06 06:26:21 powering your bone via USB just seems to be a major drama Jun 06 06:26:32 not when you use usb3.0 Jun 06 06:26:46 been powering mine since day one from USB 3.0 Jun 06 06:26:54 I power mine from usb Jun 06 06:26:55 works fine Jun 06 06:27:05 yeah, and not when you run it on port 3 of machine 19, running OS ABC, on a blue moon Jun 06 06:27:10 since the patch to allow the pmic to draw more than 500mA Jun 06 06:27:15 point is, its a support nightmare Jun 06 06:27:33 glad its not my support nightmare Jun 06 06:27:39 true that Jun 06 06:27:49 always going ot have support issues where users are involved Jun 06 06:27:53 but then, how did rpi handle that? Jun 06 06:28:10 lol, the pi Jun 06 06:28:35 funny enough, i was able to power my pi via usb Jun 06 06:28:40 using the usb port that was on the tv Jun 06 06:28:46 intended for a usb-key Jun 06 06:28:56 av500, no idea, maybe because the rpi didnt have users demanding DMA ? Jun 06 06:29:01 * m_billybob shrugs Jun 06 06:29:08 m_billybob: I meant usb power Jun 06 06:29:23 just 1080p Jun 06 06:29:34 av500 but what abotu the fine no upgrade memory point ? Jun 06 06:29:43 ? Jun 06 06:29:56 HAAAALP how do Iz get Rasbian runnong on ma bbb ?! Jun 06 06:30:14 No.. how do i install windows 8 on BBB Jun 06 06:30:35 oh nothing, im just being sily and recounting some of the sillier questions / points made i nhere and on the forums Jun 06 06:31:15 mrpackethead yeah well i want to run WinME on my bbb instruction please ? thanks ! Jun 06 06:32:28 interrestingly though someone was saying that a denser RAm chip could be a replacement away ? or maybe im thinking of the emmc ? Jun 06 06:33:09 might have been the emmc chip, pin compatible with the 8GB chip or something like that Jun 06 06:33:24 sure Jun 06 06:33:32 emmc chips are basically SD cards Jun 06 06:33:38 in a different package Jun 06 06:33:46 the pinout does not change with size Jun 06 06:33:57 you could solder a 64GB Jun 06 06:34:05 av500: really Jun 06 06:34:07 thats interesting Jun 06 06:34:14 what about addressing though ? Jun 06 06:34:25 as long as sd/mmc supports it Jun 06 06:34:34 does the emmc perform better than a normal sd card Jun 06 06:34:35 adressing changed between 2G and 4GB Jun 06 06:34:41 mrpackethead: depends Jun 06 06:34:48 there are good and bad sd cards Jun 06 06:34:51 and good and bad emmcs Jun 06 06:34:52 hats for sure Jun 06 06:34:53 mrpackethead it supposed to but im dubious Jun 06 06:34:59 well its Micron . .. Jun 06 06:35:05 err Jun 06 06:35:08 hmm yeah it is Jun 06 06:35:09 emmc can be faster since it can use 8 bit Jun 06 06:35:11 still takes a few seconds to boot Jun 06 06:35:16 where SD can only use 4 Jun 06 06:36:10 all i know is that iscsi /home is twice as fast as the uSD card im using Jun 06 06:36:25 will test the emmc eventually just for the hell of it Jun 06 06:37:38 would be interresting to see how well the am3359 would handle GbE Jun 06 06:37:47 define "handle" Jun 06 06:38:01 well it maxes 10/100 easily Jun 06 06:38:17 best interface speeds ive seen, actually Jun 06 06:38:20 for 10/100 i mean Jun 06 06:38:33 so GbE to dev/null Jun 06 06:38:48 or from /dev/zero Jun 06 06:39:02 how could i do that ? Jun 06 06:39:08 what? Jun 06 06:39:26 solder and I2C GbE chip..... Jun 06 06:39:28 an* Jun 06 06:39:34 how do i GbE to /dev/zero or reverse ? Jun 06 06:40:05 wouldnt I2C then become the bottleneck ? Jun 06 06:40:12 it was a joke :) Jun 06 06:40:52 ok, well i almost laughed ;) Jun 06 06:42:28 mourning Jun 06 06:43:07 panto: does BBB have fex support? Jun 06 06:43:47 what, no traditional morning greeting? Jun 06 06:44:01 wedont have time for pleasantries Jun 06 06:44:08 the time window is 2 weeks only! Jun 06 06:44:26 fex? Jun 06 06:44:34 my god Jun 06 06:44:40 you greeks live under a rock Jun 06 06:44:50 its in 40% of all tablets Jun 06 06:44:50 my mind is not working without coffee Jun 06 06:44:56 its DT but done right! Jun 06 06:45:04 wtf is fex? Jun 06 06:45:47 https://lkml.org/lkml/2013/6/5/539 Jun 06 06:45:52 grab popcorn and coffee Jun 06 06:46:52 the fscking ini file in allwinner socs? Jun 06 06:47:53 sure I'll get down on writting a fex to dt converter tool Jun 06 06:48:12 no no Jun 06 06:48:14 too late Jun 06 06:48:22 so sad Jun 06 06:48:25 we need fex as an alternative for DT Jun 06 06:48:36 that has the stench of another SoC Jun 06 06:49:07 first of all the name is all wrong Jun 06 06:49:12 let's rename it to sex Jun 06 06:49:26 sex script Jun 06 06:49:32 sex configuration Jun 06 06:49:36 sex compatibility list Jun 06 06:49:39 so much better Jun 06 06:49:44 sexy drivers Jun 06 06:49:46 -@sex Jun 06 06:50:03 pin "polarity".... Jun 06 06:50:22 and for many geeks, that's one opportunity to say they engage in sex regularly Jun 06 06:50:23 morning panto Jun 06 06:50:27 hi m_billybob Jun 06 06:50:41 panto: I edit my sex file every morning Jun 06 06:51:01 I had a bad sex file experience yesterday Jun 06 06:51:02 hands on the keyboard Mister. Jun 06 06:51:09 the port jammed and couldn't accept input Jun 06 06:51:17 heh Jun 06 06:51:18 panto: can you mail me that sex change? Jun 06 06:51:28 sure, sex change coming right up! Jun 06 06:51:45 pusing sex changes now ? Jun 06 06:51:47 it does not apply :( Jun 06 06:51:51 pushing* Jun 06 06:51:59 wow that was almost an unfortunate typo Jun 06 06:52:07 av500, have you diddled with your sex file out of tree? Jun 06 06:52:13 uhm Jun 06 06:52:20 STDs!!!! Jun 06 06:52:24 heh Jun 06 06:53:03 panto: I have to admit, I used one from a vendor kernel Jun 06 06:53:17 so you've dabbled with sex unprotected Jun 06 06:53:46 I wore emacs all the time Jun 06 06:55:38 emacs is an STD by itself Jun 06 06:55:46 not a pleasant idea. loading std's into slots . .. Jun 06 06:57:32 mrpackethead more fun for your USb conudrum -> http://pansenti.wordpress.com/2013/05/27/beaglebone-black-with-two-usb-host-ports-it-can-be-done-but-its-not-easy/ Jun 06 06:58:17 "bbb usb support issues ?! Apply this mod, fry your board, problem solved !" Jun 06 06:59:29 straightforward Jun 06 06:59:57 av500: I love the reply where adding a i2c GPIO expander is "outside the target market" Jun 06 07:00:05 :) Jun 06 07:02:00 ...here are all these allwinner device people could tinker with if they were mainline using fex Jun 06 07:02:02 koen, heh, who said that? Jun 06 07:02:07 ...fex does not support I2C Jun 06 07:02:16 ...thats outside of the use case... Jun 06 07:02:27 panto: Don Quichote Jun 06 07:02:48 what are virtual capes? Jun 06 07:02:50 av500, btw, I spy someone replying in that trainwreck of a thread who looks familiar :) Jun 06 07:03:04 keesj, capes that are not physical Jun 06 07:03:17 panto: :) Jun 06 07:03:20 but affect change in the peripheral configuration Jun 06 07:03:25 as if they were Jun 06 07:03:43 like for the second mmc? Jun 06 07:03:45 i.e. want to enable UART1? There's a virtual cape for that Jun 06 07:03:48 right Jun 06 07:03:53 and the onboard HDMI Jun 06 07:05:23 man, by that thread allwinner mainline doesn't look like it's going to happen Jun 06 07:05:38 and I thought TI was dysfunctional Jun 06 07:06:02 this guy is so deluded Jun 06 07:06:14 which one? I can spot many Jun 06 07:06:15 as if allwinner cared for fex in the long term Jun 06 07:06:28 panto: lkcl said that when someone mentioned needingto patch all driver Jun 06 07:06:29 they can drop that as fast as they made it Jun 06 07:06:43 as soon as Google uses a DT kernel Jun 06 07:06:46 they will too Jun 06 07:06:49 av500, yep Jun 06 07:06:53 it's all over then Jun 06 07:06:58 if the chinese can do one thing good its to copy Jun 06 07:07:09 use DT or no more fruitcake funstuff for you Jun 06 07:07:10 just use uEFI Jun 06 07:07:43 +1 Jun 06 07:08:19 koen, is that sarcasm I detect? Jun 06 07:08:30 pragmatism Jun 06 07:09:00 a pessimist is an optimist with better information afterall Jun 06 07:15:43 hum, btw that hdmi framer can do 7.1 Jun 06 07:15:48 audio I mena Jun 06 07:15:50 *mean Jun 06 07:16:16 nice Jun 06 07:16:20 so, I can watch all my 7.1 full VGA DVDs? Jun 06 07:16:28 too bad no-one works with the new driver :) Jun 06 07:18:37 av500: no, QCIF only Jun 06 07:19:44 * panto actually has a DVD with the quad mix of Dark Side of the Moon Jun 06 07:20:01 also have an HDMI receiver Jun 06 07:20:15 hmm, mplayer can play VOBs right? Jun 06 07:20:35 it should Jun 06 07:20:52 hum, I better fit it then Jun 06 07:23:28 the sucker also supports SACD Jun 06 07:23:55 koen, are there plans of oxygen-free gold plated BBBs? :) Jun 06 07:24:39 of course Jun 06 07:24:47 but you need to demagnitize the DVD first Jun 06 07:24:49 the monster cables edition Jun 06 07:30:49 ok i cant read that any more. seems like kids squabbling over candy or something Jun 06 07:34:04 m_billybob: you should stop reading as soon as you spot lkcl Jun 06 07:34:59 well the post you spoke of is why i went there, pretty hillarious if you ask me Jun 06 07:35:11 didnt find it but yeah got bored Jun 06 07:48:55 koen: about Narcissus, I have some doubts about the arch.conf after the -tune changes in oe (i.e.-iwmmx and -strongarm are gone). If I need up to armv5te can I just start with qemuarm/arch.conf ? Jun 06 07:49:51 yes Jun 06 07:49:59 it's the arch.conf from /etc/opkg Jun 06 07:50:52 I see, thx Jun 06 08:28:10 Hello guys, I'm searching for a way to trigger a shutdown upon pressing the power button on the BBB. In the manual the desired behavior is described, but I can't seem to find a way to configure this. I'm using the Debian Wheezy image. Any hint in the right direction would be much appreciated. Jun 06 08:29:38 I think I read somewhere, that there were recently some kernel patches merged that mapped that button to a proper input event Jun 06 08:30:28 so you might need to check a) if the event is triggered on the kernel you run b) if it's mapped in debian to initiate the shutdown sequence Jun 06 08:33:41 marone, there's an 'power' event triggered Jun 06 08:33:51 how to tie that to user-space is up to you Jun 06 08:34:03 I think it's the display manager's job Jun 06 08:34:13 but you can read it your self I guess Jun 06 08:34:31 Hmm, I figured. So it's safe to say that if I want to use it without a DM, I would need to write a kernel module to catch the event? Jun 06 08:34:50 marone: no, I think you can set that in inittab Jun 06 08:34:59 same as ctrl-alt-del behaviour Jun 06 08:35:29 ahh, interesting Jun 06 08:35:38 thanks for the hint, I'll try it and keep you posted Jun 06 08:36:27 in the old day's some acpi power events where generated Jun 06 08:36:43 (using acpi emulation) Jun 06 08:36:44 marone, no Jun 06 08:37:24 panto: no acpi events you mean? Jun 06 08:37:58 it's not an ACPI event since there's no ACPI on the BBB Jun 06 08:38:05 * panto hopes it will stay that way Jun 06 08:38:16 it's just an input event Jun 06 08:38:37 ok, so no inittab support you think? Jun 06 08:39:47 no idea how you're distro du jour handles it Jun 06 08:40:09 panto: don't forget blacklists for the hdmi cape... Jun 06 08:41:33 * panto kicks mranostay Jun 06 08:41:41 but mranostay is so drunk he can't feel it Jun 06 08:41:54 hey! :P Jun 06 08:42:00 LOL, having a browse of the lkml allwinner thread and R.King talking about how fantastic and faultless device tree is!! Jun 06 08:42:21 was he not point blank refusing to support it 2 weeks ago?! Jun 06 08:45:38 panto: I can't seem to find a way to output the triggered events Jun 06 08:45:51 evtest /dev/input/eventX Jun 06 08:46:17 events from what device? Jun 06 08:47:08 mranostay: I want to catch the power button press Jun 06 08:47:24 panto: I tried that before, evtest lists only my keyboard as a possible event source Jun 06 08:47:59 are you using a recent enough kernel? Jun 06 08:48:09 3.8.12-bone17 Jun 06 08:48:18 root@beaglebone:~# ls -l /dev/input/event* Jun 06 08:48:18 crw-r----- 1 root root 13, 64 Jan 2 17:37 /dev/input/event0 Jun 06 08:48:18 crw-r----- 1 root root 13, 65 Jan 2 17:37 /dev/input/event1 Jun 06 08:48:18 crw-r----- 1 root root 13, 66 Jan 2 17:37 /dev/input/event2 Jun 06 08:48:31 root@beaglebone:~# uname -a Jun 06 08:48:31 Linux beaglebone 3.8.13-00644-g3bdfbe9-dirty #1246 SMP Thu Jun 6 11:42:12 EEST 2013 armv7l GNU/Linux Jun 06 08:49:42 what panto said Jun 06 08:49:55 Aha, interesting. I'm using the debian wheezy image. I'll boot into angstrom and check it there Jun 06 08:50:02 go to bed mranostay you're drunk. Jun 06 08:50:06 lol Jun 06 08:52:05 jackmitchell, rmk is adapting his position as to not having to add anything new to _his_ ARM playground Jun 06 08:54:02 panto: heh, the toys are definitely getting thrown out of the pram on a more regular basis Jun 06 08:54:22 jackmitchell: yes, that confused me too Jun 06 08:55:27 I think he would be happy with an ARM mainline of just PSION organizers Jun 06 08:57:58 wish I still had my old PSION. but glad I've still got my Newton. Must power it up again to see if it's still retaining 1996's calendar. Was there when I last checked ~2011. Jun 06 08:58:18 panto: now he has that cube thing Jun 06 08:58:42 if anyone could advise on BBB+XBMC tips/tricks (if any), would be appreciated :) Jun 06 08:58:50 yes Jun 06 08:58:52 dont do it Jun 06 08:59:32 panto: I have no verified it with angstrom. event0 and event1 is the keyboard, event2 is the mouse. I assume you have the same situation. No board button events. Jun 06 09:00:18 av500: whyso? not that I'm suggesting your advice is faulty. just I have 0 understanding (yet) of BB/BBB and would like to know more :) Jun 06 09:00:24 bleeter, the bone doesn't have a h/w decoder Jun 06 09:00:29 marone, http://pastebin.com/cgX3Sp4Q Jun 06 09:01:03 panto: fair enough :) Jun 06 09:01:18 panto: alright, you served me :) Jun 06 09:02:18 panto: you're using the most recent angstrom distro from the bbb site? Jun 06 09:02:30 I'm using the most recent kernel Jun 06 09:02:36 + some bits of my own Jun 06 09:02:43 distro is irrelevant Jun 06 09:02:45 I see Jun 06 09:03:06 (my own bits have nothing to do with that power button) Jun 06 09:03:14 I'm using the Debian from here http://www.armhf.com/index.php/boards/beaglebone-black/ with Robert Nelson's armhf 3.8 kernel Jun 06 09:06:41 marone, he's been keeping in sync with our kernel Jun 06 09:06:50 eventually he'll update Jun 06 09:06:59 or you can pick up the patch from our tree Jun 06 09:08:15 i'm a bit of a noob kernel-wise. is this the way to go? http://rcn-ee.net/deb/precise-armhf/v3.8.13-bone20/ Jun 06 09:09:40 marone, I have no idea Jun 06 09:09:53 you should ask the guy that built it Jun 06 09:10:10 Ok, thanks Jun 06 09:10:30 can I run seriouslyjs.org on BBB? Jun 06 09:16:45 ok, I'd love to sit here and chat about your sex^Wfex file changes but got a chore to run Jun 06 09:16:45 bbl Jun 06 09:17:26 hi! Jun 06 09:46:34 What is the easiest way to upgrade/change the kernel to the most recent one on the BBB, using the armhf Debian minimal image? Jun 06 11:02:06 Is there anybody that can give me the right push towards CAN-bus? I enabled the dcan1 tx and rx pins using a device tree file but have no clue how to check if I now have a working can setup Jun 06 11:02:51 RagBal: did you load the drivers or did you just use the pinmux settings? Jun 06 11:03:05 I did the pinmux settings Jun 06 11:03:35 But I don't know if I have the correct pull up/down settings Jun 06 11:04:04 0x184 0x2A /* P9_24 = GPIO1_17 -> mode 2,input 0b101010 */ Jun 06 11:04:04 0x180 0x0A /* P9_26 = UART1_RXD -> mode 2,output 0b001010 */ Jun 06 11:04:25 which transceiver are you using? Jun 06 11:04:39 Philips A82C251 Jun 06 11:05:37 How to prepare a SD Card for booting? Jun 06 11:05:50 hi Jun 06 11:07:36 Tey, you need a FAT partition with the MLO and u-boot.img on it Jun 06 11:07:38 RagBal: does it work with 3.3v? Jun 06 11:07:58 after moaning last night about no USB -> Serial... (FTDI) another colleague who has also found this frustrating just told me the next rev will have the FTDI chip and do serial. Woo hoo £30 wasted. Cos i will buy another now! Jun 06 11:08:08 Anything else Special? tryed it like that a few times, it never worked Jun 06 11:09:06 RagBal: it seems, it needs at least 4.5v on txd Jun 06 11:09:09 Rotti, it has 5v vcc Jun 06 11:09:50 RagBal: what voltage is at rxd? it should not be more than 3.3v iirc Jun 06 11:10:04 a wingclipped KotH appears Jun 06 11:10:20 Rotti, at the can-bus side or BBB side? Jun 06 11:10:24 bbb Jun 06 11:10:29 Let me check Jun 06 11:10:35 * KotH roars Jun 06 11:10:43 * KotH JIHADs, too Jun 06 11:10:55 at least the srm says you should never connect 5v signals directly to the gpios Jun 06 11:11:20 http://www.ti.com/lit/ds/symlink/sn65hvd230.pdf this is the one i used Jun 06 11:11:21 Rotti, 4.5v Jun 06 11:11:28 did i mis anything? Jun 06 11:11:47 yes, you mis-spelled Jun 06 11:12:30 Ugh ok, so I'm kinda frying my BBB as we speak? =) Jun 06 11:12:42 maybe Jun 06 11:13:19 i got mine several times rebooted after esd... it still seems to work fine Jun 06 11:13:49 Hmm ok Jun 06 11:13:51 mru: what's an s or two? Jun 06 11:14:28 So I need a 3v3 CAN transceiver? Jun 06 11:14:38 probably Jun 06 11:15:03 well, 'miss spell' would be the winner of a female spelling bee Jun 06 11:15:07 Ok, should be no problem. But I think i also misconfigured something Jun 06 11:15:11 the one from the link above works fine... Jun 06 11:15:15 while 'mis-spell' is quite the opposite Jun 06 11:15:20 did you load the driver? which one? Jun 06 11:15:31 Ah sorry, missed that link Jun 06 11:16:20 Hi! Jun 06 11:17:51 Rotti, do you care to look at my dts for a second, check if it looks ok for the dcan1 muxsettings: http://pastebin.com/1TgcS3dR Jun 06 11:18:59 RagBal: does the kernel module get loaded? Jun 06 11:19:30 any use sd 16gb on BBB black? Jun 06 11:19:36 does can0 appear in /proc/net/dev? Jun 06 11:19:49 Yes can0 appears in the list Jun 06 11:20:21 root@beaglebone:~# echo BB-DCAN1 > /sys/devices/bone_capemgr.9/slots Jun 06 11:20:21 root@beaglebone:~# [ 2976.832882] c_can_platform 481d0000.d_can: invalid resource Jun 06 11:20:54 i think i have seen this once, too Jun 06 11:21:03 haven't used it in kernel 3.8... Jun 06 11:21:09 mru: details! details! Jun 06 11:21:18 Hmm so you're on 3.2? Jun 06 11:21:24 mru: you have to think in the big picture! Jun 06 11:21:24 the devil is in the details Jun 06 11:21:25 ;) Jun 06 11:21:32 nope, i don't use can at the moment Jun 06 11:21:42 av500: get out of my details! ;) Jun 06 11:22:18 RagBal: do you have any can device which transmits data? Jun 06 11:22:18 KotH: meh Jun 06 11:22:21 Ahh ok, so after I load the dts, do I need to configure something else? I did configure the bitrate to 125k and enabled the interface Jun 06 11:22:28 then you should see rx bytes increasing Jun 06 11:22:40 how did you do that? Jun 06 11:22:40 av500: chocolate? Jun 06 11:23:01 +1 Jun 06 11:23:09 ip link set can0 type can bitrate 125000 triple-sampling on Jun 06 11:23:35 RagBal: and it appears in ifconfig, too? Jun 06 11:23:41 Rotti, yes Jun 06 11:23:59 ...but is does not receive any data? Jun 06 11:24:28 Hmm ifconfig reports 3 RX packets Jun 06 11:24:43 But none transmitted Jun 06 11:24:47 KotH: http://youtu.be/1wfamPW3Eaw Jun 06 11:25:00 how did you try to transmit data? Jun 06 11:25:40 With 'cansend' Jun 06 11:25:54 did it report any errors? Jun 06 11:26:23 maybe 3.3v at the input are not enough for the chip... Jun 06 11:26:26 But I didn't actually check it on my scope because I was trying to figure out if I configured it correctly. It seemed to be configured 'to easy' if you know what I mean =) Jun 06 11:26:49 mru: have to watch taht at home Jun 06 11:27:01 it's sfw Jun 06 11:27:09 unless perhaps if you work in a church Jun 06 11:27:10 yeah.. but i dont like watching stuff wat work Jun 06 11:27:14 s/wat/at/ Jun 06 11:27:14 you can use cansniffer or similar to see if any data arrives... if so, guess it is configured correctly Jun 06 11:27:26 Rotti, I'll check the output from the BBB when collegue returns my scope Jun 06 11:27:40 KotH: did you at least check the title? Jun 06 11:27:43 juup Jun 06 11:28:18 he mentions an 'immaculate confection' Jun 06 11:28:45 Rotti, thanks for the help so far Jun 06 11:29:04 np... Jun 06 11:29:23 but i think you should try to get another transceiver... ;) Jun 06 11:30:18 Yea I'm checking your suggestion as we speak =) Jun 06 11:32:06 mru: can't be anything else but chocolate :) Jun 06 11:57:38 back Jun 06 11:59:47 top Jun 06 12:00:00 strange Jun 06 12:01:05 charm Jun 06 12:01:44 cheesecake Jun 06 12:03:29 mmm Jun 06 12:05:47 * georgem_ yawns Jun 06 12:07:10 av500: lol Jun 06 12:23:26 is there a common way to prevent race conditions or implement mutexes/locks when utilitizing both pru cores? Jun 06 12:23:41 yes, use a brain Jun 06 12:23:59 Where can one purchase one of those? Jun 06 12:24:39 don't know, in europe every child is given one at birth Jun 06 12:25:53 mru: I've witnessed counter evidence to that statement. Jun 06 12:26:47 some never learn to use it Jun 06 12:26:50 or even lose it Jun 06 12:27:43 presumptions presumptions Jun 06 12:35:16 hi Jun 06 12:35:38 z Jun 06 12:36:31 Hello, This is my first post here. pardon me if this was already asked. Jun 06 12:36:46 yes Jun 06 12:36:53 I get the Babble interrupt when powering on the USB Modem connected- ti81xx_interrupt 869: CAUTION: musb1: Babble Interrupt Occured Jun 06 12:36:54 many people post first here Jun 06 12:37:01 how to do a long reset Jun 06 12:37:06 long reset? Jun 06 12:37:11 ti81xx_interrupt 870: Please issue long reset to make usb functional !! Jun 06 12:37:37 that doesn't look like a message from a kosher driver Jun 06 12:37:53 https://groups.google.com/forum/?fromgroups#!topic/beagleboard/vVvL_Pe2lQc Jun 06 12:37:54 those are the linux kernel messages. Jun 06 12:38:32 Yes, i did go through the post, but there wasnt any solution. I am facing the same problem. Jun 06 12:38:49 use another modem Jun 06 12:38:56 easiest solution Jun 06 12:39:39 tried that too, but still the same issue. Babble interrupt... Jun 06 12:40:21 with a different modem? Jun 06 12:40:26 using a different chipset? Jun 06 12:40:36 try a working modem Jun 06 12:40:43 I try to avoid modems like the plague. Jun 06 12:41:06 depends on what you mean by modem Jun 06 12:41:27 an adsl connection uses a modem Jun 06 12:41:28 right Jun 06 12:41:45 as does any radio communication Jun 06 12:41:50 Yes, with a different modem. tried with the both CDMA and GSM modems. the Babble interrupt warning message is displayed in the serial console when powering on the modem using the GPIO Jun 06 12:42:13 mru: dialup modems Jun 06 12:42:19 morning Jun 06 12:42:27 you're not powering the modem from the gpio, I hope Jun 06 12:43:30 georgem_: I have one of those in a box somewhere Jun 06 12:51:47 omg Jun 06 12:51:58 this guy is so deluded Jun 06 12:52:10 which guy? Jun 06 12:52:13 lkcl Jun 06 12:52:16 oh Jun 06 12:52:22 tell us something we didn't know Jun 06 12:52:30 lkcl? Jun 06 12:52:52 https://lkml.org/lkml/2013/6/5/539 Jun 06 12:56:57 wtf did they go invent their own device tree alternative Jun 06 12:57:20 nih? Jun 06 12:57:26 ignorance Jun 06 12:57:43 av500: so, the guy came up with something allwinner-specific that is meant to be used instead of devicetree, and expects to get away with it just because allwinner sells a container ship load of unstable chips? Jun 06 12:57:59 pretty much Jun 06 12:58:18 hehe. This should be fun to watch. Jun 06 13:02:18 hi, ich suche eine Dokumentation für das TiWi-5E w Cape, bzw die Stromaufnahme. Jun 06 13:02:33 hi, I am looking for a manual for the TiWi-5E w Cape, or the current record. Jun 06 13:03:34 hi, I am looking for a manual for the TiWi-5E w Cape, or the current record. Jun 06 13:04:07 mru: probably more like they wanted to have a time to market advantage Jun 06 13:04:15 mru: and this clearly paid off Jun 06 13:04:28 don't care, works for us Jun 06 13:04:38 and long-term maintenance is nothing they care about Jun 06 13:04:53 next year they'll have a new chip Jun 06 13:05:06 the software quality and long-term maintenance is overrated :) Jun 06 13:05:33 hi, I am looking for a manual for the TiWi-5E w Cape, or the current Jun 06 13:06:34 hi, I am looking for a manual for the TiWi-5E w Cape, or the current consumption Jun 06 13:07:56 mru: vendor lock-in is also a nice thing (naturally for the vendors) Jun 06 13:08:13 Sebastian_, doesn't work yet in bbb Jun 06 13:08:20 TI wifi driver lacks DT bindings Jun 06 13:08:27 Although... ti didn't do much better if you look at the PSP kernel Jun 06 13:08:43 but, but, all the hardware IP blocks worked! Jun 06 13:08:49 chortle Jun 06 13:13:45 mru: well, I wanted to use a rk3066-based device for my project - alas, the USB support was even worse than on pre-3.8.13 BBB or any RPi Jun 06 13:14:45 mourning Jun 06 13:16:30 ssvb_: there is no vendor lock in Jun 06 13:16:50 kfoltman: rk30 works for us Jun 06 13:18:38 av500: why does this fex thing have such a fervent follower then? :) Jun 06 13:19:09 ssvb_: because lkcl Jun 06 13:19:09 fex? Jun 06 13:19:14 av500: maybe it was a software/kernel issue - the same devices worked with kernel level driver but not userspace driver - but I decided it'd be better to just buy a better board, even if it's slower Jun 06 13:21:09 mourning Jun 06 13:21:30 why does Hotel room coffee suck? Jun 06 13:21:36 never mind...its too early Jun 06 13:21:45 because it's not coffee Jun 06 13:21:48 ka6sox: by the time you discover it, it's too late anyway Jun 06 13:22:03 if you're lucky they give you nespresso Jun 06 13:22:06 maybe time to move up from hotel-6 to something better ka6sox Jun 06 13:22:09 ssvb_: because the guy is deluded Jun 06 13:22:19 ssvb_: his eoma-68 stuff uses Allwinner Jun 06 13:22:33 Rickta59, you would think for $181/night they would do better :P Jun 06 13:22:41 that's not cheap Jun 06 13:22:54 mru: I found out what the "executive" package in novotel is Jun 06 13:22:58 mru, its silicone valley Jun 06 13:23:04 mru: a bath robe and a nespresso machine Jun 06 13:23:11 both you have to leave there Jun 06 13:23:13 av500: I've been on the 'executive floor' at the hilton many times Jun 06 13:23:20 and NO mint on the pillow either :P Jun 06 13:23:26 for €20 Jun 06 13:23:42 ka6sox, you should complain, on the INETNERETS! :) Jun 06 13:24:04 bradfa, Jun 06 13:24:06 this doesn't quite fall under 386 though Jun 06 13:24:08 or at least steal all the towels :) Jun 06 13:24:36 bradfa, you mean that sandpaper masquerading as towels? Jun 06 13:25:05 what I'd like to know is how hotel soap manages to feel like sandpaper Jun 06 13:25:06 ka6sox, maybe you have some painting to do at home and need sandpaper? Jun 06 13:25:13 av500: do you know of any dev boards (not android-based consumer electronics wannabes) for rk3066/rk3188, preferably without serious, stability-reducing corner-cutting involved? last time I've looked, the best 'fast' option was to get an odroid board instead (samsung, not rockchip) but that wasn't cheap and I wasn't sure if it would be reliable either Jun 06 13:25:16 red cottage inn, menlo park used to have decent coffee .. no idea now Jun 06 13:25:38 i think it is in that same price range Jun 06 13:26:04 ya, working in South San Jose this time...so a bit far. Jun 06 13:26:45 kfoltman: no idea, I just know our products :) Jun 06 13:29:34 karthik is gone eh? Jun 06 13:30:10 a typical 3G/4G modem takes between 1-2A peak to start up. Jun 06 13:30:31 if they suck at it...and don't put some caps in there...it dies. Jun 06 13:30:52 Hey, do you guys thing I would be able to draw 1280x1024 30fps video using fbdev on BBB? Jun 06 13:32:32 mrpinky: not likely, no Jun 06 13:33:02 define 'video' Jun 06 13:33:18 you can draw frames just fine Jun 06 13:33:26 don't expect decoding h264, though Jun 06 13:35:24 OK, I was hoping to stream video over ethernet, then decode it on the BBB and put it on HDMI using fbdev. I can choose the encoding, ideally something fast/easy to decode. But it's hard to know if the CPU is capable enough. Jun 06 13:36:13 video is only 565 Jun 06 13:36:17 it wont look nice Jun 06 13:36:18 mrpinky, no h/w decoders Jun 06 13:36:46 Hmm.. When using the setup-scripts next branch with MACHINE=beaglebone I don't get a kernel that boots on beaglebone a6. Maybe bbb device tree is bundled in with the kernel? Jun 06 13:37:01 565? as in 5-bit R 6-bit G 5-bit B? where can I read about this... :( Jun 06 13:37:13 in the schematics Jun 06 13:37:26 panto, yeah I am disappointed, i saw that some older beagleboards based on OMAP, TI even provide ready-to-use gstreamer library to hook in to all the decoders Jun 06 13:37:34 yes Jun 06 13:37:39 they still exist Jun 06 13:37:47 they do ~720p Jun 06 13:37:50 not all of it Jun 06 13:38:03 it's an industrial chip, not a multimedia chip Jun 06 13:38:21 it CAN PRU all the things Jun 06 13:39:39 PRU! Jun 06 13:39:40 er.. Jun 06 13:39:42 JIHAD! Jun 06 13:40:30 av500, i believe you about 565, but I want to learn more about it... I looked up the hdmi framer it's NXP TDA19988, but I don't see any restriction like this... i wonder where it comes from :( Jun 06 13:40:31 I wonder how hard it would be to backronym a jihad block into a chip Jun 06 13:40:45 mrpinky: the bbb doesn't have any hw acceleration for video decode, color space conversion, scaling, .. Jun 06 13:40:49 koen, yeah i think it's not right for my purpose :( Jun 06 13:40:55 mru: if you ever have the chance, please do :) Jun 06 13:40:57 mrpinky: see the schematics Jun 06 13:41:08 only 16 lines go to the framer Jun 06 13:41:15 [kernel] philenotfound opened pull request #43: Use DTC_FLAGS also for single dtb target (3.8...3.8) http://git.io/-iS8Qw Jun 06 13:41:23 Jacmet, yeah, I was trying to estimate whether I would have enough CPU for decoding, scaling etc... in software, but i guess not :( Jun 06 13:41:24 mrpinky: the restriction comes from the 16 data lines between the am335x and the tda19988 Jun 06 13:41:26 av500: multiplex them Jun 06 13:42:04 but can the am335x do more? or is it limited to 565? Jun 06 13:42:12 Jacmet, av500, ah of course the 16-data lines clock the color data per-pixel :( yes 565 would look nasty Jun 06 13:42:13 av500: 24bit Jun 06 13:42:19 hi Jun 06 13:42:34 [kernel] koenkooi pushed 2 new commits to 3.8: http://git.io/cQeEnw Jun 06 13:42:34 kernel/3.8 7c064c2 Phil Eichinger: Use DTC_FLAGS also for single dtb target Jun 06 13:42:34 kernel/3.8 bef6c6d Koen Kooi: Merge pull request #43 from philenotfound/3.8... Jun 06 13:43:11 [kernel] koenkooi pushed 1 new commit to 3.8: http://git.io/PxzFOw Jun 06 13:43:11 kernel/3.8 276c671 Koen Kooi: 3.8: more HDMI patches... Jun 06 13:43:17 has anyone tried beaglebone black with gadget DBGP ? Jun 06 13:44:33 wasistday? Jun 06 13:44:59 s/y/s/ Jun 06 13:45:21 Angstrom setup-scripts next branch with MACHINE=beaglebone... Should it produce a 3.8.13 kernel with bundled device tree blob that will boot on beaglebone a6? Someone here has to know this off the top of their head. Jun 06 13:45:25 Jacmet: it has NEON, and NEON is pretty fast (though probably not as good or power-efficient as dedicated hardware) Jun 06 13:45:52 the cpu can decode most 720p h264 Jun 06 13:45:59 but you'll struggle with yuv to rgb conversion Jun 06 13:46:08 is it worth ugrading angstrom Jun 06 13:46:14 mru: what's the problem there? Jun 06 13:46:15 until mru make it run on PRUs, right? Jun 06 13:46:23 kfoltman: neon or not, decoding/colorspace conversion/scaling to 1280x1024 is quite a bit of work for a small a8 Jun 06 13:46:32 aren't PRUs fairly slow? realtime, but slow? Jun 06 13:46:37 "small a8".... Jun 06 13:46:38 KotH: no hardware to do it Jun 06 13:46:54 kfoltman, 200MHz but they are two Jun 06 13:46:58 Jacmet: you're aware that a full blown pc had less horsepower just a couple of years ago? Jun 06 13:47:02 and one clock per insn Jun 06 13:47:05 KotH: yes, a large a8 would do it just fine Jun 06 13:47:23 mru: psssst! dont invalidate my punny argument! Jun 06 13:47:28 KotH: sure, but did you do 30Hz 1280x1024 video playback in pure sw on those PCs? Jun 06 13:47:46 are we talking PRU in here? Jun 06 13:47:49 if you hand tune your algo on fixed h/w you can do wonders Jun 06 13:47:55 Jacmet: been there, done that Jun 06 13:47:56 mranostay: shut up you are drunk! Jun 06 13:48:03 KotH: I doubt that Jun 06 13:48:12 av500: sober now Jun 06 13:48:12 mru: i didnt say i was successfull ;) Jun 06 13:48:16 pc video cards have done colourspace conversion and scaling forever Jun 06 13:48:26 mranostay: you? sober? who is gonna believe that? Jun 06 13:48:29 yes, took forever on mine Jun 06 13:48:34 KotH: let him Jun 06 13:48:34 KotH: it's all relative Jun 06 13:49:10 KotH: there are times :P Jun 06 13:49:28 mru, with dedicated hw though, right? :( Jun 06 13:49:49 what with dedicated hw? Jun 06 13:50:05 mru: pc video card colorspace conversion Jun 06 13:50:17 a pc video card *is* dedicated hw Jun 06 13:50:29 oh yeah Jun 06 13:50:58 I guess OMAP3 is better suited for this purpose, they even provide the gstreamer libraries. but i'm worried OMAP3 is getting old :( Jun 06 13:51:26 one day per day Jun 06 13:51:27 ~ Jun 06 13:51:27 mrpinky: ask av500, he can give you a couple of omap5 Jun 06 13:51:59 they only released the manual last week :) who made the boards already! Jun 06 13:52:07 omap4 works fine Jun 06 13:52:25 mrpinky: fairies Jun 06 13:52:34 mrpinky: sometimes they help trolls Jun 06 13:52:38 mrpinky: last week? Jun 06 13:52:45 anyone have any opinion on odroid boards? Jun 06 13:52:51 yes Jun 06 13:52:53 ooo Jun 06 13:53:01 av500, or very recently, i only saw it on the site last week Jun 06 13:53:10 * KotH nods Jun 06 13:53:13 last week... Jun 06 13:53:20 (well, I'm asking in the wrong channel obviously, but better here than in ##electronics I suppose) Jun 06 13:53:34 kfoltman: what's wrong with #electronics? Jun 06 13:53:40 kfoltman: and why do you always write ## ? Jun 06 13:53:50 brb.. chocolate time Jun 06 13:54:04 KotH: taken over by trolls, I suppose? (Channel is invite-only) Jun 06 13:54:08 KotH, don't some channels actually start with ##? Jun 06 13:54:16 yes, some do Jun 06 13:54:17 trolls trolls trolls Jun 06 13:54:24 bradfa: depends on the network.. yes.. Jun 06 13:54:28 bradfa: but it's unusual Jun 06 13:54:33 KotH, ##arduino does Jun 06 13:54:38 on freenode Jun 06 13:54:40 kfoltman: ah.. and #beagle is free of trolls? Jun 06 13:54:43 bradfa: so ka Jun 06 13:54:46 bradfa: shiranakatta Jun 06 13:54:50 * _troll_ looks around Jun 06 13:55:01 * bradfa somehow got uneaten by _troll_ Jun 06 13:55:02 KotH: the ones here don't try very hard Jun 06 13:55:15 _troll_: your call! Jun 06 13:55:23 quit typing so fast today...I can't keep up. Jun 06 13:55:33 so fast so fast so fast so fast Jun 06 13:55:35 trolling trolling trolling Jun 06 13:57:34 there are some rules around using ## Jun 06 13:57:40 I think its for informal ones Jun 06 13:57:46 beagle should be #### Jun 06 13:58:01 av500, I am about 1/2 way down the cup you posted. Jun 06 13:58:53 i gtg, thanks for all the tips guys, when I can decode HD at full frame rate on BBB then I come back :D Jun 06 13:59:02 have fun Jun 06 13:59:29 the beagle mailinglist is funny Jun 06 13:59:45 "if we define all pins in the EEPROM we don't need to steeeenking DT anymore" Jun 06 13:59:58 heh Jun 06 14:00:08 good luck providing platform data to drivers Jun 06 14:00:17 and if we put some forth code in there too... Jun 06 14:00:32 forth, yes, please Jun 06 14:00:34 or put forth some code Jun 06 14:00:37 koen, idiots... Jun 06 14:00:40 What we need is UEFI, because I'm sure that will make everything better.... Jun 06 14:00:48 georgem_: hey, that's my line! Jun 06 14:00:59 heh Jun 06 14:01:13 koen: didn't know Jun 06 14:01:19 http://www.etsy.com/listing/130326098/coffee-mug-fill-line-mug-coffee-cup-tea Jun 06 14:01:23 if only the problem of driver configuration was only about pinmuxing Jun 06 14:01:37 why dont we use fex? Jun 06 14:01:45 all we need is DOS to edit an ini file Jun 06 14:01:51 and the funny thing is that the EEPROM pinmux data have never even been used for anything Jun 06 14:01:56 even people that sell handbags can do that Jun 06 14:01:58 not in u-boot Jun 06 14:02:02 just got beagleboard black, got debian wheezer running off an SD card. boots up and gets network fine. However if you take down the network, replug the ethernet cable and try and bring the network back up it fails Jun 06 14:02:04 not in any version of the kernel Jun 06 14:02:13 anyone seen this/know the bug ? Jun 06 14:02:15 cornet, be more specific Jun 06 14:02:15 cornet: then dont do that :) Jun 06 14:02:30 fails=? Jun 06 14:02:30 cornet, hastebin? Jun 06 14:03:14 koen: I was noticing a lot of pins aren't setup in DT anyway. I guess they are just assumed to be at the defaults on boot Jun 06 14:03:47 georgem_: yes, that's a bug Jun 06 14:04:01 they are set by uboot Jun 06 14:04:05 like it should be Jun 06 14:04:12 panto: eeprom pinmux data is using in 3.2 for the bebopr cape Jun 06 14:04:17 panto: but that's the only one Jun 06 14:04:22 gee Jun 06 14:04:36 really? Jun 06 14:04:42 yes Jun 06 14:04:48 I'm surprised Jun 06 14:04:55 no one else seems to be using it Jun 06 14:05:21 why not have the DT inside the eprom? Jun 06 14:05:24 bradfa: ok cleanly not a well known thing. I'll do some debugging later and get back to you Jun 06 14:05:25 "do all your capes have a proper EEPROM?" Jun 06 14:05:29 *crickets* Jun 06 14:06:31 cornet, do you have a network management program installed that will detect plug and then bring the interface back up? Jun 06 14:06:46 if not, start by getting one Jun 06 14:07:07 or configure it once and leave it up regardless of link status Jun 06 14:07:09 bradfa: it's just the base wheezer image. so running ifup tries to obtain DHCP address but fails. assigned static IP and still fails Jun 06 14:07:36 cornet, base wheezy image? from who? Jun 06 14:07:39 bradfa: kernel reports link established of 100/Full duplex which is correct Jun 06 14:08:19 cornet, so you're saying that regardless of if you assign a static ip, when you plug ethernet in the network doesn't work? Jun 06 14:08:34 bradfa: yes.. unless the cable is plugged in on boot Jun 06 14:08:48 http://rcn-ee.net/deb/rootfs/wheezy/debian-7.0.0-console-armhf-2013-05-29.tar.xz <-- this image Jun 06 14:08:52 cornet, do the Ethernet link lights come on when you plug in? Jun 06 14:08:55 bradfa: yes Jun 06 14:09:36 cornet, and dmesg shows PHY saying link came up? Jun 06 14:09:41 bradfa: yup Jun 06 14:09:53 cornet, and if you assign a static ip then what exactly doesn't work? Jun 06 14:10:35 bradfa: ifconfig reports it has the IP. No actual packets sent or received Jun 06 14:10:51 cornet, what happens if you send some packets? Jun 06 14:11:26 destination not reachable. Packet counters don't increment on the interface either (yes i've checked th routing too) Jun 06 14:11:54 do you get blinkin lights on the enet jack when you send packets? Jun 06 14:12:23 don't think i did no.. link light is on though Jun 06 14:12:58 what is the other end of the ethernet cable plugged into? Jun 06 14:13:15 this was only a quick 15min test on lunch.. will poke it some more this evening and see if the pre-installed image dpes the same Jun 06 14:13:23 directly into switch Jun 06 14:13:32 anything else plugged into switch? Jun 06 14:13:42 a few things yer Jun 06 14:14:21 even forced the switch to be 100Mbit / full duplex Jun 06 14:14:25 same behaviour Jun 06 14:14:27 if you edit inittab, you can use a ttl serial usb cable to log in Jun 06 14:15:01 ahh, just read up, dont mind me Jun 06 14:18:25 <_troll_> troll__: impostor! Jun 06 14:18:48 rofl Jun 06 14:19:22 the troll sees all Jun 06 14:20:50 How do I write blocks in Haskell? Jun 06 14:21:21 <_troll_> why would you use haskell? Jun 06 14:21:34 <_troll_> I recommend using a real language instead, such as french Jun 06 14:21:46 Crofton: try [] Jun 06 14:21:50 looks blocky to me Jun 06 14:22:02 and {} for pointers Jun 06 14:23:01 http://lists.osmocom.org/pipermail/osmocom-sdr/2013-June/000670.html Jun 06 14:23:27 Crofton: you troll Jun 06 14:24:00 mranostay: we all troll Jun 06 14:24:08 heh Jun 06 14:24:11 that was not me Jun 06 14:24:12 Crofton, my eyes watered Jun 06 14:24:55 * mranostay goes off to x86-the-world Jun 06 14:26:38 mranostay, arming the world is much more fun, the wars are more explosive Jun 06 14:26:59 arm all the bears Jun 06 14:27:00 +1 Jun 06 14:27:14 mru, what about those in new jersey? Jun 06 14:27:27 especially those in new jersey Jun 06 14:27:28 damn friends b-day... so hungover.. Jun 06 14:27:30 :) Jun 06 14:27:30 hi everyone... anyone have any small project/tutorial using the beaglebone? i got one. but I am kind of stuck as I am a totally newbie. =) Jun 06 14:27:50 bradfa: didn't take any fuzzy pics for you i'm sorry Jun 06 14:27:54 two of my friends have their birthdays on saturday Jun 06 14:28:38 but one is on a different continent and the other has a baby, so I'm not expecting any wild parties to happen Jun 06 14:31:18 If anyone gets angstrom next (uImage--3.8.13-r23a) to boot on the regular beaglebone let me know. Looks broken to me. Jun 06 14:32:19 is there a system update, or command line package manager for the default OS on the BBB? Jun 06 14:32:33 opkg Jun 06 14:32:48 opkg update to get the list updated Jun 06 14:32:52 then opkg list Jun 06 14:32:59 thanks av500 Jun 06 14:34:14 this is my first soiree into this distro. Jun 06 14:42:40 Hello. Got my bbb yesterday and I'm amazed about first tries with beaglescript. Before my plans where to install a minimum debian (I feel home there since the 90ies) and a few packages: zabbix, apache. Now I seem to like the javascript magic of preinstalled angstrom. What I don't like is the bloated system with X and other stuff which is not needed. I just ran opkg update && opkg upgrade and it runs for hours... Jun 06 14:43:35 gerritg, heve you rebooted? Jun 06 14:43:37 should I rather live with angstrom and try to install zabbix from source or is there a way to get the same server side javascript magic with debian? Jun 06 14:44:32 rebooted? you mean "# reboot" with root account? Jun 06 14:45:02 no, I just started bbb twice so far Jun 06 14:45:25 opkg upgrade still runs Jun 06 14:46:19 gerritg, what do you mean my "still runs"? Jun 06 14:47:41 it's running currently Jun 06 14:49:27 run like hell Jun 06 14:50:23 even after the reboot? Jun 06 14:50:30 or did you restart it. Jun 06 14:50:32 opkg upgrade? may FSM have mercy on his soul... Jun 06 14:50:46 ka6sox: opkg updgrade is still running so he can't reboot it yet Jun 06 14:50:52 finite state machine? Jun 06 14:51:36 depending on which version of image he has that might result in needing to use the "Flasher" to recover. Jun 06 14:51:49 I see... Jun 06 14:52:37 well, that's always possible as a resort and it would have been the way anyhow to "upgrade" the system Jun 06 14:53:04 using a fresher flasher image... Jun 06 14:53:11 gerritg, the first firmware can fill up the eMMC on a opkg upgrade and render it inoperable, the recovery is to reflash the unit with later firmware. Jun 06 14:53:23 (the first release of the firmware that is) Jun 06 14:53:31 later releases do not have this issue. Jun 06 14:53:45 why do you call it 'firmware'? Jun 06 14:53:54 it's just a linux distro Jun 06 14:53:55 /usr/share/opkg/intercept/depmod: line 4: echo: write error: No space left on device Jun 06 14:53:56 :) Jun 06 14:54:06 gerritg, bingo Jun 06 14:54:07 well, hell :) Jun 06 14:54:12 mru: technically, firmware on many home routers etc. = just a linux distro ;) Jun 06 14:54:43 kfoltman, +1 Jun 06 14:55:01 I think of "images" on SD and "firmware" on internally flashed things. Jun 06 14:55:05 but to the end user, it's still a closed box Jun 06 14:55:14 so a opkg upgrade was not a good idea huh? darn Jun 06 14:55:21 it was in theory Jun 06 14:55:24 but Jun 06 14:55:25 :-) Jun 06 14:55:28 sometimes SW has bugs Jun 06 14:55:30 pbcw, only for the original 4-13 firmware. Jun 06 14:55:32 sometimes? Jun 06 14:55:34 grrrr. Jun 06 14:55:44 ka6sox: I managed to write myself a SPI0 DTO Jun 06 14:55:47 I'll say it again: someone needs to put that in the topic, gerritg is probably 100th person here affected by that problem ;) Jun 06 14:55:52 ssi, excellent Jun 06 14:56:22 kfoltman: I'm not sure people read the topics Jun 06 14:56:23 ka6sox: I can program the fpga successfully, but I haven't had any luck programming the flash Jun 06 14:56:32 I dunno if it's a hardware problem or a software problem :/ Jun 06 14:56:36 i just got my board as well...how can i tell what firmware is on it? Jun 06 14:56:40 don't ask to ask, gets violated daily Jun 06 14:56:45 yes that's in the topic Jun 06 14:56:55 pbcw: just reflash latest Jun 06 14:56:58 http://bugs.elinux.org/issues/13 Jun 06 14:57:05 10-4. Jun 06 14:57:15 I guess what I need to do is somehow put that in Neon Lights. Jun 06 14:57:18 jackmitchell: that doesn't necessarily mean people don't read it, just that their habits get in a way ;) Jun 06 14:57:57 ssi, I don't use it with the flash Jun 06 14:58:04 gotcha Jun 06 14:58:09 neon lights? that is a cortex-a8 feature ka6sox ? Jun 06 14:58:09 since I want to use it as a SDP Jun 06 14:58:22 do you know if there's an existing tool to write the i2c cape config eeprom? Jun 06 14:58:31 Rickta59, only if I use a MSP peripheral controller Jun 06 14:59:16 I've not seen one ssi Jun 06 14:59:17 I might oughta just write one... wouldn't hurt me to understand how it works a little better Jun 06 14:59:22 :) Jun 06 14:59:31 anyway, meeting, brb Jun 06 14:59:44 the link here: http://circuitco.com/support/index.php?title=Updating_The_Software#Procedure for the 2013.5.27 image ends in an XY file format. Is that compressed, or is that the actual image file to be writted to the SD card with the ImageWriter software (Windows) Jun 06 15:00:00 gerritg, sorry, this is one of those once you get past the first "issue" things work better... Jun 06 15:02:31 http://pastebin.com/J0i97N7T i get that message on that page if anyone at circuitco is looking Jun 06 15:16:15 no problem ka6sox, will be fixed soon. And good to know I'm not the only one :) Jun 06 15:17:06 joys of PHP Jun 06 15:17:09 my inital q about debian and bonescript remains: rather stay with bloated angstrom system or bother to equip debian with bonescript Jun 06 15:17:10 mourning Jun 06 15:17:29 moaning mdp Jun 06 15:17:31 gerritg: depends on what you're doing, you might go for lfs/buildroot too Jun 06 15:18:12 I didnt find any information on debian packages for bonescript, one has probably to package it Jun 06 15:18:24 I am going to reflash my BeagleBone Black with the latest image from here: http://dominion.thruhere.net/koen/angstrom/beaglebone/?C=M;O=D. (BBB-eMMC-flasher-2013.06.06.img.xz). After the download, I go to unzip file (step 4 here: http://circuitco.com/support/index.php?title=Updating_The_Software#Procedure) and I am getting an error saying that the file is not a zip achive. The command I am using is tar -xvf. My error looks like this: http://pa Jun 06 15:18:24 stebin.com/9xf0cGEp Any thoughts??? Jun 06 15:19:25 unxz was available on my mac Jun 06 15:19:29 mourning mdp Jun 06 15:20:04 pbcw, I just ran unxz *.img.xz Jun 06 15:20:28 I use the linux box not my mac. Jun 06 15:20:30 i am trying unxz right now...honestly, didn't know that existed. Jun 06 15:20:36 so didn't know that Jun 06 15:21:03 unxz seemed to work. baller. Jun 06 15:21:08 thanks gerritg Jun 06 15:25:07 * mranostay is back Jun 06 15:25:37 thanks for the warning :) Jun 06 15:26:00 mranostay: still drunk? Jun 06 15:26:29 no sometime between 2-6a all went well Jun 06 15:27:57 ka6sox: missed your ping..sorry, now in cab to SFO Jun 06 15:28:12 mdp, np Jun 06 15:28:24 enjoy the trip home! Jun 06 15:28:29 busy 3 days :-/ Jun 06 15:28:42 ya, I've been working late here too... Jun 06 15:28:45 ka6sox: no hope..middle seat Jun 06 15:28:53 not going as well as I hoped Jun 06 15:28:57 ugh Middle is sucky Jun 06 15:29:05 ugh..sorry yo hear that Jun 06 15:29:46 I will try to get absorbed in this vendor tree code I have as a reference Jun 06 15:29:54 EVT Jun 06 15:30:55 ka6sox: unless you have two supermodels on each side Jun 06 15:31:07 even if. Jun 06 15:31:27 ka6sox is a old prude :P Jun 06 15:31:52 mranostay, not as much fun as 2 that are attractive and enjoyable to be around... Jun 06 15:34:24 the solution is to do 65k miles a year like my boss and be upgraded most of the time Jun 06 15:34:40 the cure is worse than the illness though Jun 06 15:34:42 mdp: so all or nothing Jun 06 15:36:43 so on topic, I tried to boot up a linaro ubuntu nano rootfs on BBB and failed Jun 06 15:37:03 but... but... Jun 06 15:37:04 what's wrong? :) Jun 06 15:37:06 * ka6sox falls over Jun 06 15:37:23 but... Ubuntu OS is easy and user friendly?! Jun 06 15:37:23 mdp: you are Jun 06 15:37:29 heh Jun 06 15:37:50 jackmitchell: clearly, mdp is not a user Jun 06 15:38:01 are you trying to say Ubuntu is too hard for a kernel developer? Phoronix would love that one... ;) Jun 06 15:38:03 on this page: http://dominion.thruhere.net/koen/angstrom/beaglebone/?C=M;O=D, what is the difference between the eMMC-flasher iamges, and the Angstrom-Cloud9-IDE-GNOME-eglibc-ipk-v2012.12-beaglebone images? Jun 06 15:38:23 oh, fwiw, lava folks are putting BBB into lava publicly...if anybody cares Jun 06 15:38:59 I still have no clue wtf lava is supposed to be Jun 06 15:39:12 is lava the build system/tester thing Jun 06 15:39:16 mru: molten stone? Jun 06 15:39:23 mru: or a type of lamp? Jun 06 15:39:33 jackmitchell: iirc it is automated testing Jun 06 15:39:42 target validation is the buzzword Jun 06 15:40:24 mru, yet another test framework Jun 06 15:40:36 mdp: I know what they _say_ it is Jun 06 15:41:08 mru, in the old days..every student wrote an irc client as their first project..now they write test frameworks Jun 06 15:41:25 pbcw: flasher is what flashes the image files Jun 06 15:41:30 apropos testing... does anyone know a good system for unit testing software that works on bare metal? Jun 06 15:41:42 (system as in approach or software or anything) Jun 06 15:42:10 pbcw: but please use some official images Jun 06 15:42:36 KotH: replace printf with a morse LED Jun 06 15:42:38 done Jun 06 15:42:39 I would/was going to use official ones. imjust kind of confused as to what to use. Jun 06 15:42:48 mdp: i wrote a horribly functioning irc logger iirc :) Jun 06 15:43:20 av500: well.. it should be automated... Jun 06 15:43:40 av500: or are you saying i should put photodiodes on the leds and read those ? :) Jun 06 15:44:38 well Jun 06 15:44:43 if you habe serial on bare metal Jun 06 15:44:45 use printf Jun 06 15:46:59 av500, i am assuming you are meaning these: http://beagleboard.org/latest-images Jun 06 15:48:06 si Jun 06 15:48:46 so, that goes back to my original question. there are two listed there. the cloud9/gnome, and the flasher. do i need both files, or does the cloud9-gnome run off of the SD card, and not write anything to the eMMC? Jun 06 15:49:21 yes, one for BBW and sd card Jun 06 15:49:28 and one for BBB and emmc flashing Jun 06 15:49:46 the sd card one should work on BBB too Jun 06 15:50:00 but then you need to boot from SD Jun 06 15:50:04 with the button Jun 06 15:50:12 10-4. I appreciate the clarification. Jun 06 15:51:05 mranostay: you also wrote a test framework at EA :) Jun 06 15:51:22 oh, that explains a lot Jun 06 15:52:22 mdp: i'll deny that on a stack of bibles :) Jun 06 15:52:32 you're an atheist Jun 06 15:52:52 talk about a meaningless gesture from you Jun 06 15:53:29 short security line today..hurray Jun 06 15:53:36 mru: perjury is still illegal Jun 06 15:54:04 only if you get caught Jun 06 15:54:31 or a president Jun 06 15:54:55 however i think the bigger crime is the level of interns he picked Jun 06 15:55:18 Is there a way to make my BBB boot of the flash card if it's present, but the MMC if not. WITHOUT requiring me to press the button? Jun 06 15:55:44 mranostay: you wanted the job? Jun 06 15:55:55 av500, updating away with the eMMC official image. Thanks again. Jun 06 15:55:59 mru: which? Jun 06 15:56:26 the one the intern did Jun 06 15:57:38 which intern? i'm lost here Jun 06 15:58:00 the one whose level you berated Jun 06 15:58:50 Hello, I want to add to the BBB a clock real time, have verified that it has a device/dev/i2c-1 and /dev/i2c-0, it is activated? Is it necessary to compile the kernel for activate it? Thank you Jun 06 15:58:57 mru: for LAVA? Jun 06 15:58:59 well, at least, maybe the interns aren't drunk all the time Jun 06 15:59:02 although... Jun 06 15:59:05 hehe Jun 06 16:08:25 Focus__, compiling the kernel isn't necessary, you might need to set an Device Tree Overlay via capemanager to enable it Jun 06 16:08:56 ka6sox: all should be enabled Jun 06 16:09:16 okay then you are golden.... ;) Jun 06 16:10:58 ka6sox, Please say me since I it enable Jun 06 16:11:37 -EPA Jun 06 16:11:40 RSE Jun 06 16:14:07 Focus__, what is your native language? maybe we have a speaker here... Jun 06 16:14:27 spanish Jun 06 16:29:32 hiho Jun 06 16:30:51 why isn't this been captioned yet? --> http://www.flickr.com/photos/54974804@N03/8555111709 Jun 06 16:32:11 Does anybody know why pps_gpio is not enabled by default in Angstrom? Jun 06 16:32:36 nobody packaged it? Jun 06 16:33:38 Are kenel modules packaged in Angstrom? (just arrived on BeagleBone Black, so I'm new to platform) Jun 06 16:33:42 ka6sox, can you help me ? Jun 06 16:35:29 topic ^ Jun 06 16:40:08 Focus__, I am @ work now...please email the list. Jun 06 16:42:26 ka6sox: ph0rums Jun 06 17:28:53 I need help with BBB and LCD cape, anyone has worked with both? Jun 06 17:29:40 bertux: Just ask your question. Jun 06 17:29:42 i just flashed the eMMC with the 2013.5.27 image found on http://beagleboard.org/latest-images. I went to go deal with some network stuff. Did the network management change, cause this doesn't make sense to me: http://pastebin.com/jNmJCi1j Jun 06 17:29:49 If someone here knows how to answer, they will. Otherwise, wait Jun 06 17:31:00 The LCD cape doesn't work, the BBB without cape works, but when I put on cape the BBB doesn't starts Jun 06 17:31:23 BBB enter in "kernel loop" Jun 06 17:32:23 I already test with many angstrom images Jun 06 17:33:09 Sounds sort of like maybe the BBB is not getting enough power. Jun 06 17:33:46 The source is enough, supply 2A Jun 06 17:34:32 but when BBB starts the LCD flash and then the system stop Jun 06 17:34:38 OK, then probably not that then. Jun 06 17:35:06 bertux: Are you booting from the sdcard? Jun 06 17:35:11 or the eMMC? Jun 06 17:35:19 eMMC Jun 06 17:35:21 http://www.elinux.org/BeagleBone_Black_Capes Jun 06 17:35:24 says you have to boot from the SD card Jun 06 17:36:08 yeah, but this link is for beaglebone, I have beaglebone black Jun 06 17:36:17 the angtrom is on the eMMC Jun 06 17:36:28 malcom2073 I think network-manager is used or conman Jun 06 17:36:40 bertux: No, that link is for bbb Jun 06 17:36:46 ok Jun 06 17:36:56 Read the page. Jun 06 17:36:58 malcom2073 or you write your own /etc/network/interfaces file Jun 06 17:37:09 woglinde: Not me, pbcw Jun 06 17:37:16 but i don't know starts with SD Jun 06 17:37:18 woglinde, i asked the question, not malcom2073 Jun 06 17:37:19 ups Jun 06 17:37:24 sorrxy Jun 06 17:37:26 :) Jun 06 17:37:28 np. Jun 06 17:37:37 the SD images doesn't boot Jun 06 17:37:48 they are update images Jun 06 17:38:08 I have no experience with angstrom, so I can't help there. Jun 06 17:38:25 only for update, when boot from SD the BBB copy image to eMMC Jun 06 17:38:40 I tested with debian too Jun 06 17:39:25 bertux you need a working kernel first Jun 06 17:39:55 woglinde ? Jun 06 17:39:55 so userland is irrelevant Jun 06 17:40:17 How I do that? Jun 06 17:40:19 why do think it is boot looping Jun 06 17:40:32 because it can not initalize the cape Jun 06 17:40:36 and kernel is crashing Jun 06 17:40:47 woglinde: Apparently SD card works. Jun 06 17:40:51 woglinde: according to the wiki Jun 06 17:40:52 but without cape the system works Jun 06 17:40:56 just not booting from eMMC Jun 06 17:40:58 I do not know which kernel works or what dtb changes you need Jun 06 17:41:24 only on cape the system crash Jun 06 17:41:26 you should examine the boot protocol Jun 06 17:41:36 I have the latest image Jun 06 17:41:50 kernel 3.8.13 Jun 06 17:41:53 Hence me suggesting to get a distro that has a SD image. Jun 06 17:41:55 yes and? Jun 06 17:42:07 examine the boot protocol Jun 06 17:42:23 all other is like looking into crystal balls Jun 06 17:42:59 how examine the boot protocol? Jun 06 17:43:02 there should be more courses explaining what a boot protocol is Jun 06 17:43:07 dmesg output? Jun 06 17:43:33 bertux why did you bought a bbb when you do not know about boot protocol and how to get it Jun 06 17:44:09 Becouse I try to learn Jun 06 17:44:20 so next task for you Jun 06 17:44:21 I've never head to deal with a "boot protocol" before, and I've been tinkering with my BB for a year now Jun 06 17:44:42 try to setup serial console to get a uboot and kernel messages Jun 06 17:45:39 ok, I never used boot serial Jun 06 17:45:51 hi shoragan Jun 06 17:46:03 I'll read about this Jun 06 17:46:13 malcom2073 boot protocol == out from serial console from MLO, u-boot, kernel Jun 06 17:46:26 Oh, never heard it called that before, ok thanks Jun 06 17:46:33 yes sorry Jun 06 17:46:35 *g* Jun 06 17:46:48 hi shoragan Jun 06 17:46:48 thanks Jun 06 17:46:50 oh neat, you can write the cape eeprom directly using sysfs Jun 06 17:46:53 fwiw, he's best off installign to SD card, THEN doign that. Since the eMMC won't work anyway with the shield. Jun 06 17:47:04 err cape Jun 06 17:48:42 hi woglinde :) Jun 06 17:49:19 koen, i finally figured out how to use the am335x's bandgap temp sensor Jun 06 17:50:05 write 0x1c to 0x44e10448, then read 0x44e10448[15:8] Jun 06 17:50:21 please tell me you aren't mmap'ing it Jun 06 17:50:41 no, it's already mapped Jun 06 17:50:51 (in the bootloader that it ;) Jun 06 17:50:58 s/it/is/ Jun 06 17:51:04 * mranostay puts away the cluebat Jun 06 17:53:00 mranostay, because it lives in a dangerous neighborhood? Jun 06 17:54:39 hmm the bbb seems to initialize the trim register differently Jun 06 17:55:49 mranostay, but how would you read it in the kernel without ioremaping it? Jun 06 18:04:00 shoragan_: well as long as you don't do a open("/dev/mem", O_RDWR) in userspace :) Jun 06 18:04:25 heh, only for testing Jun 06 18:04:51 but it seems those regs are not accessible from usermode... Jun 06 18:05:17 would real mode help here? Jun 06 18:05:22 * mranostay trolls away Jun 06 18:05:24 ahrg Jun 06 18:05:40 load HIMEM.SYS Jun 06 18:07:19 ah, no it's accessible all right, it just doesn't like byte-reads Jun 06 18:07:38 Spirilis: +1 Jun 06 18:07:58 shoragan_: mmio registers usually must be accessed as their native size Jun 06 18:08:37 mru, yes, i know, just some tools count the amount to read in words, others in bytes Jun 06 18:09:17 ok so jumping into the harm's way of the cluebat, once you get pinmux set up for a GPIO pin and you want to go twiddle it on/off from a C program, let's say..... how do you do it? mmap /dev/mem and play with that? Jun 06 18:10:07 use a proper driver goddammit :) Jun 06 18:10:14 +1 Jun 06 18:10:18 ^^^ Jun 06 18:10:35 so there's still some sense left in mranostay... Jun 06 18:10:55 ah k, so jump into kernel land with your fun and expose a proper interface to userland. Jun 06 18:10:57 shoragan_: you wouldn't have an hwmon driver for that ready, would you? Jun 06 18:11:54 koen, i just figured out where the dont-just-report-zero bit is 30 minutes ago :) Jun 06 18:12:49 if you find the dwim bit, let us know Jun 06 18:13:11 and i'd need to do some calibration first against an external thermometer, as i have no idea on how to map the register value to a °C scale Jun 06 18:13:29 that should be in the trm Jun 06 18:13:31 or the datasheet Jun 06 18:13:49 mru, should, but isn't Jun 06 18:16:50 more testing tomorrow :) Jun 06 18:16:51 cya Jun 06 18:17:59 hmm yeah ok so basically if you're ghetto or a bonescript user, you use sysfs to play with GPIO's, if you're smart you write a kernel driver and if you're crazy godlike you make the PRU do it. Jun 06 18:18:22 pru for the win Jun 06 18:20:06 mru: would the following statement be accurate - for a given arm core, say cortex-a8 of a given revision, the performance for a purely cpu bound code is the same at a given clock rate regardless of manufacture of teh SoC the core is in? Jun 06 18:20:24 ds2: no Jun 06 18:20:32 there are implementation options Jun 06 18:20:40 like cache size and speed Jun 06 18:20:47 mru: speed? Jun 06 18:20:59 how's that different from clock rate? Jun 06 18:21:00 L2 cache isn't necessarily clocked at full speed Jun 06 18:21:08 Ohhhh of the cache Jun 06 18:21:10 gotcha Jun 06 18:21:35 haha did there were some beagles which run with slow level cache rate? Jun 06 18:24:56 been too long to remmeber that ;) Jun 06 18:25:16 there was some very early code rev for the xm that didn't enable L2 at all Jun 06 18:26:53 hmmm is L2 always disableable? Jun 06 18:27:29 it's usually off at reset Jun 06 18:28:10 disabling it can be useful Jun 06 18:30:04 Does anyone know which cable I should buy from Monoprice.com to get the BBB to work on this monitor? S22C300H Jun 06 18:30:17 Monitor has HDMI and D-Sub Jun 06 18:30:29 Here are the specs if it helps: http://www.samsung.com/hk_en/support/model/LS22C300HS/XK-techspecs Jun 06 18:33:14 reads of a volatile qualified variable should not be re-orderd by gcc, right? Jun 06 18:33:51 that's probably a overly broad statement Jun 06 18:34:01 read the C spec Jun 06 18:34:54 trying to make sure that I grab a copy of the variable at the time i want Jun 06 18:35:13 i'll try it and see. C spec comes later ;) Jun 06 18:35:19 ugh Jun 06 18:35:32 what are your actual requirements? Jun 06 18:35:40 I tend to use '#define barrier() __asm__ __volatile__("": : :"memory")' Jun 06 18:35:42 ISR is updating a counter Jun 06 18:36:02 i want to make sure I have 1 and only 1 snapshot Jun 06 18:36:13 the counter is volatile qualified Jun 06 18:36:33 that's probably both unnecessary and insufficient Jun 06 18:36:35 MikeB: VGA on the d-sub, or DVI? Jun 06 18:36:45 'Operations on volatile variables are not atomic, nor do they establish a proper happens-before relationship for threading.' Jun 06 18:36:50 MikeB: In general, I think you want microHDMI to HDMI for the BBB/. Jun 06 18:36:52 Russ: that's a hint to the processor not to re-order not to the compiler, right? Jun 06 18:36:53 agmlego: VGA Jun 06 18:36:54 what Russ said Jun 06 18:37:06 ds2: read the goddamn spec Jun 06 18:37:14 agmlego: Can I use a microHDMI to HDMI? Jun 06 18:37:15 Russ: how can reading an int (on arm) not be atomic? Jun 06 18:37:24 ds2: if it's not aligned, for instance Jun 06 18:37:31 I think all volatile does is prevent things from getting optimized away Jun 06 18:37:34 but that's not allowed by C Jun 06 18:37:44 like if you have while (!(status_reg & READY_BIT)); Jun 06 18:37:51 read the spec dammit, both of you Jun 06 18:37:53 volatile means something somewhere else may change it Jun 06 18:37:57 nothing more Jun 06 18:38:07 anything else, use barrier(); Jun 06 18:38:24 mru, this is more fun, cargo cult programming just sounds like it should be cruise control for cool Jun 06 18:38:25 and you probably need a real barrier, not just the fake one Russ quoted Jun 06 18:38:29 this comese down to trying to strip away spinlocks Jun 06 18:38:38 MikeB: ...I just said that I believe so, yes. Jun 06 18:38:54 what you're probably really after is __atomic_foo() stuff Jun 06 18:39:10 I started with that Jun 06 18:39:23 then decided...this is an int, and all I am doing is reading it... Jun 06 18:39:33 agmlego: sorry, I misread it! Thank you for that. Jun 06 18:39:35 if only reality were that simple Jun 06 18:39:37 let see if this blows up Jun 06 18:40:04 MikeB: Check the SRM to make sure that port can spaek HDMI--it may be DVI-over-HDMI. Jun 06 18:40:22 CPU is not keeping up with a 8K/sec interrupt rate hence optimizing Jun 06 18:40:23 if all you care about is getting some value that was probably current not too long ago, you'll likely get away with just reading it normally Jun 06 18:40:53 agmlego: Very new.. SRM? Jun 06 18:40:58 yes, that is all I care about. I don't need the latest. I just need 1 copy of it Jun 06 18:41:01 MikeB: System Reference manual. Jun 06 18:41:11 MikeB: The datasheet for the whole board. Jun 06 18:46:07 agmlego: Thanks... looking into it now. Jun 06 18:46:55 No problem. Jun 06 18:47:31 Should I see beagle as a "product" which is "done" or something with which I should expect many issues? Jun 06 18:47:49 Yes. Jun 06 18:47:51 jdoles: see bbb as a development board Jun 06 18:48:05 agmlego: which of the two? Jun 06 18:48:10 jdoles: it's _NOT_ an enduser product Jun 06 18:48:20 jdoles: but there aren't many issues either Jun 06 18:48:38 jdoles: it's basically just a lumb of clay... it becomes whatever you make out of it Jun 06 18:49:18 jdoles: Both, actually. Jun 06 18:49:28 i.e. doorstop, paperweight, ballast Jun 06 18:49:53 It is a finished product if what you are looking for is the beagle. It is a device with many issues if you are expecting it to be something else. Jun 06 18:49:59 Mostly it is a reference design. Jun 06 18:50:02 KotH: for everything you can buy and modify a line is drawn somewhere as to what you can modify about it. Jun 06 18:50:08 eval board Jun 06 18:50:14 woglinde: That too. Jun 06 18:50:18 hey woglinde Jun 06 18:50:19 KotH: if everything above the line is working, that's OK. Jun 06 18:50:27 woglinde: long time no see! Jun 06 18:50:31 yiha Jun 06 18:50:31 d Jun 06 18:50:41 KotH: but the line needs to be drawn somewhere and I would like to see some documentation then saying what is guaranteed to work. Jun 06 18:50:42 yes I was watching lets play videos Jun 06 18:50:42 * KotH hands woglinde a proper JIHAD Jun 06 18:50:54 jdoles: you draw the line where you draw the line Jun 06 18:51:07 jdoles: we cannot darw it for you... Jun 06 18:51:11 everything is modifiable Jun 06 18:51:11 * mru draws a circle around jdoles Jun 06 18:51:29 mru: is he inside or outside? Jun 06 18:51:36 got a decaping setup? you can modify the CPU die Jun 06 18:51:49 KotH: topological details do not concern me Jun 06 18:52:20 ds2: might void the warranty though Jun 06 18:52:40 if you can recap it, who will know?;) Jun 06 18:52:58 ds2: they could easily encode a number in the amount of glue for example. Jun 06 18:53:06 ds2: my crystal ball will tell it! Jun 06 18:53:11 ds2: or whatever the way it works. Jun 06 18:53:40 There is a warranty on the beagleboard? I thought those were sold as-is. Jun 06 18:53:44 like they'll check Jun 06 18:53:46 ;) Jun 06 18:54:03 * ds2 notes that Gerald does not IRC }:-) Jun 06 18:54:16 hardware people can't work the irc Jun 06 18:54:25 bradfa is the rare exception Jun 06 18:54:34 I don't see the point of buying a device without any warranty. Jun 06 18:54:38 I am a hardware person Jun 06 18:54:43 * KotH glares at mranostay Jun 06 18:54:51 but his photo taking skills suffer in turn Jun 06 18:54:52 jdoles: then dont Jun 06 18:55:05 jdoles: the bbb is not for everyone Jun 06 18:55:23 and some of like me and koen are 'perverts' Jun 06 18:55:40 KotH: I am sure it's something for me, but if I cannot complain and have someone listen to my complaints when whoever created it made a mistake, I am not going to fund failure. Jun 06 18:55:41 mranostay: and i thought, you were mostly drunk Jun 06 18:56:08 jdoles: then read up on the beagle and learn what other people encountered Jun 06 18:56:09 in that case, you might consider leaving the planet Jun 06 18:56:24 $45 really going to hurt your wallet that much? Jun 06 18:56:26 jdoles: the internet exists for a reason! Jun 06 18:56:26 ds2: +1 Jun 06 18:56:33 jdoles: do an informed decision Jun 06 18:57:01 jdoles: bitching here about what might be and what might not be wont help anyone Jun 06 18:57:06 jdoles: What are you loking for? Jun 06 18:57:11 jdoles: it only serves as an after work amusment for me Jun 06 18:57:13 before the beagle you got the boards for 10k Jun 06 18:57:16 one Jun 06 18:57:26 KotH: drunk on life maybe Jun 06 18:57:27 woglinde: not that expensive.. but yes Jun 06 18:57:29 jdoles: This IRC channel is pretty much your support, and it works well. Why do you want to pay for something that is free? Jun 06 18:57:34 mranostay: oh.. met a nice girl? Jun 06 18:57:40 mranostay: or boy? Jun 06 18:57:46 or troll Jun 06 18:58:11 agmlego: I was interested in the pi first, but all I would like to do is control an electric pump for example. Jun 06 18:58:27 agmlego it works for people you ask smart questions Jun 06 18:58:31 have you consider c? it is faster then pi Jun 06 18:58:33 ups who Jun 06 18:58:34 agmlego: at first a small one, and later a big industrial one. Jun 06 18:58:35 at least larger Jun 06 18:58:35 on the rpi channel they ban you if you complain Jun 06 18:58:50 mru really? Jun 06 18:59:07 KotH: hey! Jun 06 18:59:12 agmlego: I expect that the beagle can do something simple like that. Jun 06 18:59:18 jdoles for that a beagle is a bit over powered Jun 06 18:59:34 better use an andrino Jun 06 18:59:36 it could overload the pump! Jun 06 18:59:37 mranostay: you once told me, that you dont swing "that" way, but you never told me which way you swing ;) Jun 06 18:59:53 jdoles: Since the beaglebone is in every way more capable than the rpi...yes, I would think so. Jun 06 18:59:56 jdoles: what you want is probably an arduino Jun 06 18:59:57 woglinde: like someone said 45$ isn't going to hurt me. Jun 06 19:00:21 jdoles: or a 2CHF switch from the hardware shop Jun 06 19:00:27 jdoles you did not say what you want to do with it Jun 06 19:00:45 woglinde: well, I likely also want to hook it up to the Internet. Jun 06 19:01:02 oh.. cool! industrial size pumps on the intarwebz! Jun 06 19:01:08 jdoles and you see we are not marketing which tells you can not do the task without at am355s Jun 06 19:01:11 ups x Jun 06 19:01:28 jdoles stuxnet Jun 06 19:01:29 now we can finaly make whole factories explode by 5min hacking on MovieOS! Jun 06 19:02:09 KotH: you used to be able to flood an entire factory in shit by knowing just 4 numbers, unencrypted at that... Jun 06 19:02:26 yay for things going boom Jun 06 19:02:45 jdoles: you'd be surprised how many systems still work that way Jun 06 19:03:09 what kind of factory has that much shit? Jun 06 19:03:24 ds2: the one that makes the windows tablets Jun 06 19:03:25 ds2: a shit processing plant. Jun 06 19:03:27 manurefactory ;) Jun 06 19:03:39 dunno, I was at the zoo, and I saw they had a bunch of paper produces made from zoo animal shit Jun 06 19:03:46 pretty sure they'd be knee deep Jun 06 19:03:55 er, paper products Jun 06 19:04:12 and people were buying that? 0_o Jun 06 19:04:25 yup Jun 06 19:04:37 which country do you live in again? Jun 06 19:04:38 koth its paper Jun 06 19:04:38 its from animals that eat grasses and the like Jun 06 19:04:44 not shit Jun 06 19:04:47 If I know how to compile OpenWrt and write my own packages, would that be a good background to start with the beagle? Jun 06 19:05:07 Russ: didn't need to know that Jun 06 19:05:12 woglinde: we do a lot of medical trials on animals... i know what shit contains....or rather dont want to know Jun 06 19:05:26 http://www.elephantdungpaper.com/process.html Jun 06 19:06:19 Does OpenWRT even run on the beaglebone? Jun 06 19:06:20 Is the difference between the beagle and more expensive devices that it isn't rated for industrial environments? Jun 06 19:06:32 jdoles: yes Jun 06 19:06:37 jdoles: read the specs Jun 06 19:06:54 jdoles: Generally, folk run things like Angstrom and debian-armhf on the beagle. I would think *those* would be good places to start, especially if you have little Linux expereicne. Jun 06 19:06:58 jdoles: they tell you _everything_ you need to know...ok nearly almost everything Jun 06 19:07:07 hmm, i have yet to be able to successfully install and update a kernel for the beaglebone black... using robertCnelson's ubuntu 13.04 script using --uboot bone_dtb and his linux-dev tree just results in an sd card image that doesn't boot... Jun 06 19:07:17 jdoles: If you want a device that is industrially-rated...buy a PLC. Jun 06 19:07:35 cyronin_: pated 2.3? Jun 06 19:07:44 jdoles: and just for the record, "industrial environment" means different things to different people. depending on what you are doing, the beagle might not be reliable enough Jun 06 19:07:51 agmlego: are there any PLCs that are running full Linux systems these days? Jun 06 19:08:13 cyronin_: The script uses parted to make the card. If you look carefully at the output from the script, if you have parted 2.3, the script actually crashes because parted 2.3 has a major bug in it/. Jun 06 19:08:30 jdoles: Not that I am aware of. Why do you need Linux? Jun 06 19:09:16 im not getting a crash, and the script builds a proper sd card image that boots on the BBW Jun 06 19:09:40 KotH: what does industrial environment mean to you Jun 06 19:10:02 agmlego: I suppose PLCs typically just run one simple process with such instructions to set voltage levels periodically based on some first-order environment. Jun 06 19:10:07 im getting "/boot/zImage" not found errors and it switches back to booting the emmc Jun 06 19:10:25 even though zImage is definitely there on the boot partition Jun 06 19:10:40 mrpackethead: you know that our last "industrial" project ended up below sea level :) Jun 06 19:11:00 back Jun 06 19:11:00 jdoles: I guess you have not really spent any time looking at PLCs. ;-P Jun 06 19:11:16 KotH: better then lost on mars ;) Jun 06 19:11:17 agmlego: I programmed a few PLCs. Jun 06 19:11:23 cyronin_: Ah, different problem from what I have seen. Jun 06 19:11:26 agmlego: but never did the product selection. Jun 06 19:11:43 jdoles: Sounds like you have a lot of research ahead of you then.  Jun 06 19:11:46 ds2: uhm... there is this one project we are currently doing... which is supposed to pass mercur on its way :) Jun 06 19:11:50 agmlego: in those cases there was just a manual, a way to upload code to the device and a way to run it. Jun 06 19:11:58 and I know far too little about uboot to figure out if its some configuration issue Jun 06 19:12:14 agmlego: can you give a definition of what you think of as a PLC? Jun 06 19:12:42 but that wasn't a lost beagle Jun 06 19:12:45 ./tools/install_kernel.sh runs successfully from linux-dev tree on the sd card built with his ubuntu script as well, and I have tried booting before i mess with installing the kernel Jun 06 19:12:58 jdoles: Sure. https://en.wikipedia.org/wiki/Programmable_logic_controller Jun 06 19:13:30 ds2: the beagle is not radiation hard enough, otherwise i would make sure to include one :) Jun 06 19:13:36 I guess all im really looking for is a distro that has all of the /lib/firmware/capes/*dtbo files already there Jun 06 19:13:50 KotH: https://en.wikipedia.org/wiki/Beagle_2 Jun 06 19:13:56 :) Jun 06 19:13:57 stuff like BB-UART1.dtbo Jun 06 19:14:33 cyronin_: might be a good exercise to build the kernel yourself with everything you need https://github.com/beagleboard/kernel/tree/3.8 Jun 06 19:15:19 I've been able to build the kernel pretty well, using that tree and rcn's linux-dev tree, i just havent been able to get the thing to boot on a ubuntu sd Jun 06 19:15:35 ah ok, can I ask why you want ubuntu so much? Jun 06 19:15:52 seems like an easy image to build with rcn's script Jun 06 19:16:02 i guess i could go debian as well Jun 06 19:16:05 ds2: remember that one probe that crashed into mars because of double metric to imperial conversion? Jun 06 19:16:05 <_av500_> because of convergence Jun 06 19:16:29 <_av500_> trev-hub_: ^^^ Jun 06 19:16:52 ^^^^^ that was an extrenely dissappointing ready Jun 06 19:16:57 read Jun 06 19:17:21 that the GUI or something? Jun 06 19:17:30 (bit out of the nix distro loop) Jun 06 19:17:49 nvm found it, phone desktop? Jun 06 19:18:32 <_av500_> sorry? Jun 06 19:19:43 whats the major bug in parted 2.3 anyways? Jun 06 19:21:04 ah, just running ubuntu across multiple devices? (phone, desktop, beaglebone etc). Is that you mean by convergence av500? Jun 06 19:24:14 <_av500_> dont forget cloud supercomputers Jun 06 19:24:34 _av500_: elce talk? Jun 06 19:24:35 _av500_: the ones they make weaather forecasts with? Jun 06 19:25:09 agmlego: it seems that the major difference is that there is a certain amount of hard-real time constraints assumed in some of those languages associated with PLCs. Jun 06 19:25:51 agmlego: at least, that's the computational difference, hardware wise they are just more robust (which is mostly why people buy them). Jun 06 19:27:39 (and I can imagine that in the implementation AND blocks are mapped to real hardware making it all run in parallel) Jun 06 19:28:07 I looked at a manual quickly and it seemed very familiar and straight forward. Jun 06 19:28:19 A beagle would basically then just be an experimentation platform. Jun 06 19:28:40 jdoles: sorry I miss it, for what? Jun 06 19:28:44 jdoles: Um...as far as I have seen, the architecutre itself tends to be different as well. Jun 06 19:29:02 agmlego: how do you mean the 'architecture'? Jun 06 19:29:05 agmlego: the internals? Jun 06 19:29:07 For example, PLCs rarely if ever have any concept of interrupts, because interrupts break determinism. Jun 06 19:29:15 Yes, at a silicon level. Jun 06 19:29:24 agmlego: right, that's why I mentioned the AND blocks. Jun 06 19:29:44 agmlego: the outputs of the programming language are connected to physical outputs. Jun 06 19:29:45 I was under the impression that a plc generally runs a fairly simple loop of code at a fixed rate Jun 06 19:30:01 read inputs, calculate some things, set outputs Jun 06 19:30:03 repeat Jun 06 19:30:12 or did they get more complex since last time I looked? Jun 06 19:30:25 mru: I think you can make them arbitrarily complex. Jun 06 19:30:30 jdoles: So, what you get out of something like the beagle, other than it being wholly unsuited for the rigors of an industrial environment, is *soft* realtime, since even with an RTOS, the underlying sarchitecture still can cuase non-deterministic behavior. Jun 06 19:30:35 mru: thats usually the case Jun 06 19:30:37 mru: as long as you meet the hard-real time constraint. Jun 06 19:30:50 mru: If you really want toi simplify that far, yes, that is true. Jun 06 19:30:54 mru, you just describe the PRU Jun 06 19:30:56 cause that's why they're done Jun 06 19:31:09 But that "simple" loop tends to be hundreds of rungs of logic. Jun 06 19:31:27 agmlego: yes, but not millions of lines of code Jun 06 19:31:31 Yeah, with thousand of inputs/outputs. Jun 06 19:31:31 like a full os Jun 06 19:31:46 mru: Sure, but PLCs are not designed to be general-purpose computers. Jun 06 19:31:54 And general-purpose computers are not PLCs. Jun 06 19:31:57 PLCs typically control much more I/O than a PC. Jun 06 19:32:02 mranostay thinks I'm a hardware peoples! yey! :) Jun 06 19:32:04 agmlego: that's what I thought Jun 06 19:32:13 my plan is working! Jun 06 19:32:17 I'm pretty sure PLC tasks are handled by standard boxes that just run specialized software Jun 06 19:32:21 bradfa: we all know you're mr blurrycam Jun 06 19:32:27 no matter what lies mranostay spreads Jun 06 19:32:28 at least the most non-time critical parts Jun 06 19:32:28 mru, that, too Jun 06 19:32:34 agmlego: aren't the pauses documented? Jun 06 19:32:42 agmlego: even Intel does that. Jun 06 19:32:48 hi bradfa Jun 06 19:32:52 hi woglinde Jun 06 19:32:55 jdoles: Intel does what? Jun 06 19:33:02 I haven't been too impressed with most PLCs I've worked with. Yes they are quite deterministic but the reality is that an embedded linux system running preempt rt can always react to inputs and send controls quicker than the scan time of the PLC which is often 50ms or so Jun 06 19:33:11 panto: I am pretty sure you are wrong, actually. Jun 06 19:33:23 I know for sure that Rockwell uses a custom processor architecture. Jun 06 19:33:24 georgem: if 50ms is enough, who cares? Jun 06 19:33:26 agmlego: specify upperbounds on the time it takes for instructions to execute. Jun 06 19:33:29 agmlego, not all PLCs are that critical Jun 06 19:33:29 PLC: perfectly logical crapshoot? Jun 06 19:33:29 it's a hard guarantee Jun 06 19:33:33 most are pretty mundane Jun 06 19:33:41 what if the airplane falls out of the sky if it takes 100ms just once? Jun 06 19:33:47 ok, that would be a rather poor plane Jun 06 19:33:48 jdoles: Yes, but they rarely specify lower bounds... Jun 06 19:33:49 but still Jun 06 19:33:51 mru, and quite fast Jun 06 19:33:56 jdoles: Hard real time requires both. Jun 06 19:34:06 or quite low flying Jun 06 19:34:10 agmlego: it does? Jun 06 19:34:13 I bet those centrifuges PLC controller that got sabotaged with a virus didn't run on custom hardware Jun 06 19:34:16 I believe we just bricked one of our beaglebones because we were just unplugging it .. what is the best way to power cycle or shutdown the bone? Jun 06 19:34:30 jdoles: Uh, yeah. That is kind of the definition. Jun 06 19:34:32 panto: I doubt the virus ran _on_ the plc Jun 06 19:34:33 clh, "shutdown -h now"? Jun 06 19:34:42 mru, it ran on the plc controller Jun 06 19:34:44 agmlego: all definitions I know say that it should be done before the deadline. Jun 06 19:34:45 Hard real time requires you to know *exactly* when something will occur. Jun 06 19:34:47 more likely it ran on the workstation controlling the plc Jun 06 19:34:52 bradfa: the german airforce parked one of their tornado fighters in one of the swiss mountains a couple of years back... Jun 06 19:34:58 and loaded a dodgy program Jun 06 19:35:04 KotH, fast mountains! Jun 06 19:35:06 bradfa: tx Jun 06 19:35:14 has anyone bricked a bone? Jun 06 19:35:14 bradfa: mach 2! Jun 06 19:35:24 agmlego: you can just run a loop when you are done waiting until the right time. Jun 06 19:35:33 mru: I'm not saying PLCs are bad I'm just saying that the argument that you can't use an embedded linux machine running preempt RT doesn't really hold water because it the kernel is done right it will NEVER take 50ms to react Jun 06 19:35:40 jdoles: That is the broader concept of soft real time. Hard real time needs to know the earliest something could happen too--if you are too early, it is as bad as being too late. Jun 06 19:35:57 clh, apply 120 V AC to pins, brick! Jun 06 19:36:13 georgem: I like how your argument is predicated on "if the kernel is done right". Jun 06 19:36:15 there's no such thing as soft realtime Jun 06 19:36:25 mru: Uh, yeah, there is. Jun 06 19:36:27 either it's realtime, or it's not Jun 06 19:36:35 agmlego: soft real time means that statistically you sometimes don't meet the hard real time bound. Jun 06 19:36:36 either you meet the deadline, or you don't Jun 06 19:36:41 This is my line of work, and nearly everything we produce is soft real time. Jun 06 19:36:52 that's a marketing term Jun 06 19:37:15 jdoles: No. Soft real time means there is a deadline by which time results are needed, but there is no reason you cannot have results before that point. Jun 06 19:37:16 I'm with mru here Jun 06 19:37:31 And how many of you actually work in industrial automation? Jun 06 19:37:42 agmlego, I used to Jun 06 19:37:42 An example of a soft real-time system is trading on the stock market. Jun 06 19:37:47 I am not Jun 06 19:37:47 does the occational industrial project count? Jun 06 19:37:57 If you do it too slowly, you go bankrupt, but not immediately. Jun 06 19:37:57 occasional Jun 06 19:38:00 I am only working with canopen 447 soon Jun 06 19:38:13 jdoles: then it's not realtime Jun 06 19:38:14 KotH: Sure, if it means you understand that soft vs. hard real time is not a marketing thing. Jun 06 19:38:14 agmlego: yes. for instance on one I did I had a timer loop generating a GPIO output. I had a logic analyzer on it to see if more than 2ms elapsed between transitions. I did find drivers that were faulty which caused locking longer than that, but I fixed them and made sure it didn't happen again. Jun 06 19:38:22 agmlego: I have *never* ever heard of that definition. Jun 06 19:38:35 agmlego: for me they are Jun 06 19:38:37 agmlego: point at a definition in some source that's not written by you. Jun 06 19:38:38 jdoles: Now you have. Welcome to automation. Jun 06 19:38:46 agmlego: I have books saying otherwise. Jun 06 19:38:51 agmlego: for me, the only thing that counts is what the spec says Jun 06 19:39:10 for me counts if you get stuxnet or not Jun 06 19:39:12 haha Jun 06 19:39:22 * panto watches #beagle heading for another troll-fest Jun 06 19:39:27 * panto fetches popcorn Jun 06 19:39:29 jdoles: https://en.wikipedia.org/wiki/Real-time_computing#Criteria_for_real-time_computing Jun 06 19:39:49 KotH: Sure, we always design to spec. But there are general terms which are useful. Jun 06 19:40:03 needs some "citation needed" sprinkled around me thinks Jun 06 19:40:34 agmlego: unfortunately, most of these terms are so vague and infested with different definitions, that we as engineers have to nail the customer down with those specs Jun 06 19:40:48 KotH: I do not disagree with you. Jun 06 19:40:52 agmlego: it doesn't say that. Jun 06 19:40:54 * KotH sprinkles some "citation needed" around bradfa Jun 06 19:41:01 jdoles: It does, actually. Jun 06 19:41:15 jdoles: Or, rather, my definition fits into that broader definition. Jun 06 19:41:25 * bradfa dances in the citations! Jun 06 19:42:07 agmlego: you can always choose 0 to be a lower bound. Jun 06 19:42:26 It all depends on when your deadline is. if you're deadline is 10ms in the future you have a lot different of a problem than if its 10ns. Jun 06 19:42:43 if your deadline is 10 ns, run! Jun 06 19:42:57 agmlego: so, I maintain that you only need an upperbound to be called a real-time system. Jun 06 19:43:07 if your deadline is a piano wire around your neck... stand still! Jun 06 19:43:17 jdoles: As a broad umbrella, I do not disagree. Jun 06 19:43:33 jdoles: Hard and soft are subsets within the overal term "real time". Jun 06 19:43:42 I'd like to enforce a real-time constraint: a beer must be brought to me no earlier than 5pm but no later than 6pm. Is this hard or soft real time? Jun 06 19:44:06 bradfa: soft beer time Jun 06 19:44:10 agmlego: because you cannot guarantee any exact time anyway. Jun 06 19:44:20 bradfa: Hard. If the beer is early, you are not there to enjoy it and it gets warm. If it is late, you got fed up with no beer, and you have already gone. Jun 06 19:44:21 agmlego: you will always miss some theoretical time by some amount. Jun 06 19:44:25 bradfa: if it were 10 bottles, it would be hard beer time :) Jun 06 19:44:34 agmlego: perhaps a few picoseconds, or whatever. Jun 06 19:44:44 jdoles: Now you are picking nits. Jun 06 19:44:49 * _av500_ fetches beer too Jun 06 19:44:57 so, wait, if I spend 5 years in jail, is that "doing hard time" or "doing soft time"? Jun 06 19:45:04 When a scan time of milliseconds is enough, no one cares about picoseconds. Jun 06 19:45:17 bradfa: Neither. Bring lube. Jun 06 19:45:22 agmlego: these are definitions, so I can go all crazy with mathematical arguments. Jun 06 19:45:31 picoseconds are the shrinking violets of real time Jun 06 19:45:33 salut prpplague Jun 06 19:45:39 prpplague, just in time! Jun 06 19:45:43 prpplague: welcome to another useless troll discussion :) Jun 06 19:45:43 hi prpplague Jun 06 19:45:46 prpplague, get popcorn Jun 06 19:45:50 greetings all Jun 06 19:45:54 oh lord Jun 06 19:45:58 i can't handle any more issues today Jun 06 19:46:04 no issues, just fun Jun 06 19:46:10 bradfa: bah! Jun 06 19:46:11 prpplague, I think I've fixed hdmi Jun 06 19:46:12 * mru wrote a blag post about trolls today Jun 06 19:46:15 bradfa: i was just about to write that! Jun 06 19:46:16 agmlego: how do you actually program those real-time devices? Jun 06 19:46:19 + audio till 1080p@24 Jun 06 19:46:22 there isn't even any meat to this discussion Jun 06 19:46:31 agmlego: do you have some kind of compiler which assures that it always works? Jun 06 19:46:35 panto: lovely Jun 06 19:46:41 vegan discussion? Jun 06 19:46:44 jdoles: yes, the hex editor Jun 06 19:46:49 agmlego: or is it just testing and hoping for the best? Jun 06 19:46:55 mru: yeah, right. Jun 06 19:46:55 enter the codes on keypad, blind Jun 06 19:47:02 like real PLC does it Jun 06 19:47:18 * panto expect the discussion to get up to godel's theorem Jun 06 19:47:19 oh, and my tty ran out of paper again... Jun 06 19:47:19 jdoles: you'd be surprised how poorly some PLCs are designed Jun 06 19:47:20 <_av500_> isnt the code on punch cards? Jun 06 19:47:34 jdoles: Ladder logic in the case of PLCs, a variety of things in the case of our soft realtime vision PCs. Jun 06 19:47:34 georgem: not really, because they come from hardware people who don't know software. Jun 06 19:47:48 agmlego: oh, my. Jun 06 19:48:01 I thought that was a bad idea 10+ years ago. Jun 06 19:48:16 * bradfa goes back to telling windows programmers why floating point is bad while they don't listen Jun 06 19:48:18 jdoles: According to who? Jun 06 19:48:19 bradfa: I'm trying to get my BBB connectiviy through USB with setupgateway_wlan-eth1.sh Jun 06 19:48:38 mastiff_, ok, I've never tried that with USB or a black, but what's up? Jun 06 19:48:50 bradfa: what happens if you tell them while they listen? Jun 06 19:48:54 ran the script on the host, but BBB isn't online so just wanted to ask what I should do next Jun 06 19:49:09 mru, they don't listen, they're still windows programmers, they haven't gotten an update from MS yet on how to listen Jun 06 19:49:10 agmlego: it's not exactly compositional, etc. Jun 06 19:49:15 We support IEC 61131 ladder, etc on our linux based control systems for non critical PLC uses in addition to its other functions. Jun 06 19:49:30 Hello Jun 06 19:49:35 jdoles: What does "compositional" mean in this context? Jun 06 19:49:36 mastiff_, I use that script to setup a Debian Squeeze box that has eth1 as a physical Ethernet interface to route over my wlan connection Jun 06 19:49:45 jdoles: And why do you think it matters? Jun 06 19:49:47 if it doesn't work for some other use case, I'm sorry but I've not tried any other setup Jun 06 19:49:52 agmlego: a modern system would perhaps compile to ladder logic and would allow one to prove that the business objective is met before loading it onto the device. Jun 06 19:50:06 agmlego: I assume you still need testing to see whether it works. Jun 06 19:50:23 jdoles: So, you mean you want a simulator. Congratulations, every PLC I have worked with has a simulator in the programming environment. Jun 06 19:50:27 There is no need for testing if you setup your system engineering process correctly. Jun 06 19:50:32 agmlego: no, not a simulator. Jun 06 19:50:33 bradfa: the BBB shows up on the host as eth1 and I'm using wlan0 on the host to connect to wifi router Jun 06 19:50:41 agmlego: a simulator only tests. Jun 06 19:50:46 Does anyone know why hciconfig might stop returning data on a Beaglebone? The only thing I did was update the udev package Jun 06 19:50:54 agmlego: I am talking about making it work in general. Jun 06 19:51:10 mastiff_, on DEbian Squeeze? Jun 06 19:51:15 agmlego: anyway, perhaps this is overkill Jun 06 19:51:20 bradfa: do I need to set the gateway on the BBB to the eth1 address? Jun 06 19:51:21 yes Jun 06 19:51:26 agmlego: thanks for the talk anyway. Jun 06 19:51:32 debian on host, angstrom on bbb Jun 06 19:51:40 mastiff_, how's the BBB getting an IP? Jun 06 19:51:46 static? Jun 06 19:51:49 From the ether! Jun 06 19:51:53 Some of our control systems are used by breweries. Now thats some time critical stuff. Better be in my fridge by time I get home from work :) Jun 06 19:51:55 jdoles, ether doesn't give IPs Jun 06 19:52:02 static yeah, 192.168.7.2 Jun 06 19:52:12 mastiff_, and what's your host's eth1 IP? Jun 06 19:52:29 192.168.7.1 Jun 06 19:52:57 mastiff_, I dunno, have you tried the script using real networking interfaces? sometimes the USB networking is wonky Jun 06 19:53:01 Isn't it ipv6 enabled?! Jun 06 19:53:56 who needs ipv6 anyways? Jun 06 19:54:03 It works fine if I plug the BBB into my router with eth Jun 06 19:54:21 mastiff_, but then you're not using the script, right? Jun 06 19:54:23 it's only the japanese who run out of ips... and they are a threat to our economy anyways! Jun 06 19:54:31 ha Jun 06 19:54:33 Right right Jun 06 19:54:40 Don't need to then Jun 06 19:55:05 <_av500_> KotH: they dont need ip, they have imode Jun 06 19:55:08 Just trying to figure out how to share internet from laptop or desktop when the BBB is connected with USB Jun 06 19:55:08 mastiff_, sorry, I can't help much here, I've not used that script in a while and no longer have a debian host to test it on Jun 06 19:55:25 maybe someone else can help Jun 06 19:55:30 _av500_: i'm a kddi customer :) Jun 06 19:56:13 <_av500_> KotH: k-mode then? Jun 06 19:56:15 maybe I just need to set the 192.168.7.1 address for the BBB gateway and it will work. The script looks like it set things up on the host. Jun 06 19:56:39 * mru has a /29 ipv4 subnet Jun 06 19:56:44 what's this talk of running out? Jun 06 19:56:45 jdoles thinks you do not need to test if your system is designed right... i now know an entire company whose products i will avoid on my life. Jun 06 19:57:09 because...shit, man. Jun 06 19:57:10 mru: same here... still 6 free ips Jun 06 19:57:21 agmlego, which company? Jun 06 19:57:30 bradfa: jdoles' Jun 06 19:57:37 that's not very helpful to me Jun 06 19:57:52 /whois jdoles doesn't help me much Jun 06 19:57:53 bradfa: nor to me, actually. Jun 06 19:58:09 safest not to use _any_ products Jun 06 19:58:09 * bradfa needs to get some ip addresses before they run out! Jun 06 19:58:12 jdoles: who do you work for, so we can boycott it? Jun 06 19:58:19 lol Jun 06 19:58:20 haha Jun 06 19:58:24 agmlego: you need to test to verify the hardware only. Jun 06 19:58:31 rotfl Jun 06 19:58:33 agmlego: the software would be proven to be correct. Jun 06 19:58:36 no, you are wrong. Jun 06 19:58:38 hahahahahahah Jun 06 19:58:45 jdoles: are you up for the piano test? Jun 06 19:58:47 hoohoohoohoo Jun 06 19:58:52 agmlego: and that's much better than your software testing. Jun 06 19:58:54 hopelessly, criminally wrong. Jun 06 19:59:04 http://blog.regehr.org/archives/364 Jun 06 19:59:06 ^^ the piano test Jun 06 19:59:12 agmlego: Airbus also uses these techniques. Jun 06 19:59:20 agmlego: Intel does for certain instructions. Jun 06 19:59:28 no. Jun 06 19:59:32 agmlego: go and never fly with Airbus again. Jun 06 19:59:32 just...no. Jun 06 19:59:36 Maybe if you write your code in lisp you can write a mathematical proof to prove its correct. I'll stick with C Jun 06 19:59:46 C is correct Jun 06 19:59:51 just like on SAT exams Jun 06 19:59:51 georgem: your logic is wrong, yet again. Jun 06 20:00:08 georgem: I can prove C to be correct just as easily. Jun 06 20:00:20 jdoles: i do a lot of work with airbus. i can tell you that you are wrong on that. Jun 06 20:00:20 mru: nice! Jun 06 20:00:35 agmlego: they use abstract interpretation. Jun 06 20:00:42 agmlego: which is a less powerful form of proof. Jun 06 20:00:45 intel has shown that they sometimes get things wrong. Jun 06 20:00:50 jdoles: my logic is wrong? Jun 06 20:00:59 see the pentium division thing. Jun 06 20:01:00 jdoles, can you prove if eglibc or glibc are correct? I need to pick which one to use Jun 06 20:01:03 bradfa: they let you near a compiler? Jun 06 20:01:11 agmlego: they fixed the pentium division thing with formal proof. Jun 06 20:01:19 agmlego: because they never wanted it to happen again. Jun 06 20:01:25 agmlego: AMD also does that. Jun 06 20:01:46 agmlego: they use a rather simplistic kind of formal proof which can easily be automated, however. Jun 06 20:01:57 bradfa: sure, I could do that. Jun 06 20:01:57 AARGH MY EYES! Jun 06 20:01:58 mranostay, I wrote some exactsteps on how to build canadien compiler! Jun 06 20:01:59 jdoles: I honestly have no idea what you are even arguing anymore. I am not even sure *you* even know what you are arguing anymore. You should just stop. Jun 06 20:02:07 * panto waits until godel's theorem makes it's appearance Jun 06 20:02:22 panto: you probably don't even understand the theorem. Jun 06 20:02:38 jdoles, try to explain it to me then Jun 06 20:02:47 agmlego: this is like a walk in the park for me. Jun 06 20:02:47 like I'm a complete idiot Jun 06 20:02:49 regarding airbus control systems: http://aviation-safety.net/database/record.php?id=20081007-0 Jun 06 20:03:12 panto: Godel Jun 06 20:03:32 jdoles, panto is not a complete idiot, but pretend Jun 06 20:03:39 panto: 's theorems mostly relate to how most interesting proof systems can only be proven to be relatively consistent. Jun 06 20:03:45 * KotH hands panto an ö Jun 06 20:03:52 Hilbert was a nub!!1! Jun 06 20:03:57 * bradfa searches for that key on keyboard, fails to find it... Jun 06 20:03:59 <_av500_> Godöl Jun 06 20:04:00 panto: and that's a completely basic result. Jun 06 20:04:02 jdoles: What, rambling on and on about wrong things? I believe it. Jun 06 20:04:31 a C program can't be proven correct since there is no formal model of the C spec Jun 06 20:04:38 Godel is funny looking in those glasses Jun 06 20:04:43 doesn't seem like robert nelson's 5-29 ubuntu and debian images want to boot on the bbb Jun 06 20:04:44 mru: compcert. Jun 06 20:04:49 wrong Jun 06 20:04:52 <_av500_> bradfa: Göögel glasses Jun 06 20:04:57 mru: compcert is incomplete. Jun 06 20:04:58 touche! Jun 06 20:04:58 mru: maybe the machine code could be? Jun 06 20:05:09 nobody has proven that what compcert compiles is C code Jun 06 20:05:10 mru: but complete enough for a real system. Jun 06 20:05:12 it's something close Jun 06 20:05:15 but might be different Jun 06 20:05:19 and it's buggy Jun 06 20:05:20 * bradfa still can't find utf-8 keyboard buttons... Jun 06 20:05:27 mru: and that makes it impossible? Jun 06 20:05:29 <_av500_> bradfa: try wtf-8 Jun 06 20:05:34 k, brb Jun 06 20:05:37 You are all way too practical. Jun 06 20:05:48 to formally prove something you first need a formal model Jun 06 20:05:54 and there isn't one for C Jun 06 20:05:56 haha Jun 06 20:05:56 jdoles, yes and for good reason Jun 06 20:05:58 mru: so, write one. Jun 06 20:06:02 nor for any common processor architecture Jun 06 20:06:10 jdoles: and how do you prove _that_ correct? Jun 06 20:06:36 mru: if you followed PhD. programs you would know what people are already doing that. Jun 06 20:06:57 and the cpus are designed in verilog using non-verified compilers too Jun 06 20:06:58 mru: formalization of common CPU instruction sets are popular PhD. assignments. Jun 06 20:07:01 etc etc Jun 06 20:07:02 jdoles, if you followed common sense you'd know that people are idiots Jun 06 20:07:21 mru: you are stretching it now. Jun 06 20:07:21 I prefer to work in the real world Jun 06 20:07:31 troll, troll, troll :) Jun 06 20:07:33 mru: ah men to that Jun 06 20:07:35 mru: in the real world where you have to hope that what you create works, you mean? Jun 06 20:07:43 jdoles: do you know how many cpu bugs I have _personally_ found? Jun 06 20:07:49 my bowels work, can it be proven? Jun 06 20:07:57 mru: in what kind of CPUs? Jun 06 20:08:01 mru: AMD/Intel like? Jun 06 20:08:03 arm for instance Jun 06 20:08:03 georgem + mru holding hands? trolling brings us all together brothers and sisters Jun 06 20:08:21 heh Jun 06 20:08:32 mru: I don't count ARM for my convenience. Jun 06 20:08:43 panto: forgett it! i will never hold your hand! you are a greek! Jun 06 20:08:52 ...and with that, jdoles is now on /ignore for being a useless person. Jun 06 20:08:53 mru, I like the piano test Jun 06 20:08:56 jdoles: that's academia talking right there Jun 06 20:09:04 mru: it's easy to find bugs in systems which are not used very often. Jun 06 20:09:12 mru: I have also found lots of compiler bugs. Jun 06 20:09:15 cortex-a8 isn't used often? Jun 06 20:09:19 mru: except in gcc I only found one. Jun 06 20:09:21 haha ARM isn't used often?! Jun 06 20:09:24 * bleeter falls off chair Jun 06 20:09:29 jdoles: I find bugs in _all_ compilers Jun 06 20:09:29 bleeter, don't hurt yourself! Jun 06 20:09:31 it's my job Jun 06 20:09:32 <_av500_> arm is a passing fad Jun 06 20:09:42 leg is next big processor arch Jun 06 20:09:47 heh Jun 06 20:09:47 way longer and stronger than arm Jun 06 20:09:49 mru: I don't think your job is finding bugs in compilers. Jun 06 20:09:52 how is lower powered CPUs a passing fad? Jun 06 20:09:54 jdoles: it is Jun 06 20:09:59 are* Jun 06 20:10:01 that's part of my job Jun 06 20:10:08 I'm allowed to fix them too Jun 06 20:10:11 mru: sure, it's part of every software engineer's job. Jun 06 20:10:23 mru: ok, well, that sort of is special. Jun 06 20:10:28 jdoles, mru really does make compilers suck less Jun 06 20:10:31 jdoles: I'm in the linaro compiler team ffs Jun 06 20:10:34 jdoles: you may not be aware that arm replaced the 6800 family as the most used and most sold processor family a couple of years ago Jun 06 20:10:56 mru: ok, well, thanks for that then. Jun 06 20:10:57 KotH, but how do the punch cards fit in such small chips? Jun 06 20:10:57 jdoles: with sold chips per year way over what intel would ever dream to acheive Jun 06 20:11:00 more arm cores are sold per year than intel has sold ever Jun 06 20:11:15 bradfa: realy really tiny punch cards Jun 06 20:11:22 ah, micro punch cards Jun 06 20:11:27 Intel just doesn't care about low-profit products. Jun 06 20:11:32 Just like IBM. Jun 06 20:11:34 maybe pico punch cards are next? Call them PPC? Jun 06 20:11:48 mranostay, run! we're dissing intel! Jun 06 20:11:58 <_av500_> mranostay: shut up you are drunk! Jun 06 20:12:04 Then why did Intel just invest in creating a new lower powered mobile architecture? Jun 06 20:12:10 * KotH hands mranostay a beer Jun 06 20:12:16 KotH: enabler! Jun 06 20:12:18 trev-hub_: they were strongarmed into it :) Jun 06 20:12:22 <_av500_> arm is a play to bring down intel prices Jun 06 20:12:28 <_av500_> so far it works Jun 06 20:12:41 mru: +1 Jun 06 20:12:46 heh Jun 06 20:12:55 lol Jun 06 20:12:55 mru, +1 Jun 06 20:13:01 +1 for everyone! Jun 06 20:13:04 av500 lol Jun 06 20:13:07 hahaha Jun 06 20:13:10 YAY! a +1 for me! Jun 06 20:13:10 strongarmed Jun 06 20:13:16 Me looks at the IXP 425 in the corner of his desk Jun 06 20:13:17 av500 it do not works Jun 06 20:13:20 trev-hub_: to make the ARM people feel important of course! Jun 06 20:13:34 x86 all THE THINGS! Jun 06 20:13:38 <_av500_> +1 Jun 06 20:13:41 still intel sells more than most of the other together Jun 06 20:13:55 woglinde, if by other you mean AMD Jun 06 20:13:55 georgem, only the iaxp was a bigger disaster than the ixp Jun 06 20:13:58 trev-hub_: no, the real reason is that they might see a demand for such products now. Jun 06 20:14:01 <_av500_> and on intel I can run dbase2!!!!! Jun 06 20:14:02 nobody denied that intel is successful Jun 06 20:14:09 trev-hub_: low power laptops for example is one such product. Jun 06 20:14:19 <_av500_> yes, suddelny in 2013 they are needed Jun 06 20:14:19 mru, intel is not successful! Jun 06 20:14:26 If they see people buying ARM laptops, they might think this would eat into their profits. Jun 06 20:14:37 * _av500_ sees no such people Jun 06 20:14:38 bradfa: they make money, no? Jun 06 20:14:45 _av500_: I do. Jun 06 20:14:45 bradfa look at the figures Jun 06 20:14:52 mru, but think how much more they could make if they made more! Jun 06 20:15:00 mru they are making a shit lot of money Jun 06 20:15:01 Hence, the only logical thing to do is to beat ARM at its own game. Jun 06 20:15:11 <_av500_> it takes an apple to make people buy arm laptops Jun 06 20:15:17 AMD is going bankrupt in 2-3 years if they continue like this. Jun 06 20:15:24 <_av500_> like the past 20ys Jun 06 20:15:27 jdoles, AMD has been going bankrupt for the past few decades Jun 06 20:15:30 <_av500_> they are used to it Jun 06 20:15:31 just like freescale Jun 06 20:15:31 apples are useful, if you're making cider Jun 06 20:15:34 jdoles they always doing it Jun 06 20:15:41 * bradfa loves cider... Jun 06 20:15:42 bradfa: who is giving them all the money then? Jun 06 20:15:44 heh well intel is going to have to stop using some/all of the x86 architecture to beat ARM at its own game. Jun 06 20:15:44 has AMD never not been bankrupt?! Jun 06 20:15:50 bradfa: just new stock? Jun 06 20:15:57 <_av500_> what has x86 got to do with anything? Jun 06 20:16:02 bleeter hm one year it was near Jun 06 20:16:02 probably intel, so that they anti-monopoly laws won't be applied to them Jun 06 20:16:03 jdoles, I don't understand your questions Jun 06 20:16:10 my favorite is when i get my dickens cider Jun 06 20:16:13 _av500_, what's love got to do with it? Jun 06 20:16:15 bradfa: how is AMD not going to go bankrupt? Jun 06 20:16:19 <_av500_> woglinde: i guess when they do, intel gives them money Jun 06 20:16:24 bradfa: they lose 400 million/year Jun 06 20:16:26 bradfa: love is in the air Jun 06 20:16:29 bradfa: they have 1 billion. Jun 06 20:16:33 jdoles, they've been losing money like that for ever Jun 06 20:16:43 bradfa: you cannot lose money forever. Jun 06 20:16:48 jdoles, watch AMD Jun 06 20:16:53 they're experts by now Jun 06 20:16:54 bradfa: Didn't hurt myself. Nor did I break my current 'never argue with an idiot on the internet' protection circuit :D Jun 06 20:17:04 av500 yes often intel did it Jun 06 20:17:07 bleeter, I need one of those, wher'ed you buy yours? Jun 06 20:17:07 bradfa: so, what happened the last few times? Jun 06 20:17:17 jdoles, lmgtfy Jun 06 20:17:18 bradfa: who gave them money? Jun 06 20:17:20 jdoles: i know a company, 30 people, that made >1mioCHF loss per year... for 15 years Jun 06 20:17:24 These instructions worked great for getting BBB internet access over USB. http://travis-lab.blogspot.com/2012/02/getting-on-beaglebone-on-internet-using.html Jun 06 20:17:36 jdoles: sorry..10mioCHF loss per year Jun 06 20:17:37 <_av500_> offtopic please! Jun 06 20:17:50 KotH: sure, but someone must have funded them. Jun 06 20:17:54 mastiff_, does that work for you? Jun 06 20:18:09 <_av500_> when AMD is likely to go bust, intel will miss a stepping Jun 06 20:18:11 jdoles: nope...lived on borrowed money Jun 06 20:18:14 yes, it worked, but I'm not sure if it will be persistent after reboot Jun 06 20:18:21 mastiff_ aeh that topic is not bbb specific Jun 06 20:18:22 no it won't persist Jun 06 20:18:26 In the case of AMD, they either must take on new loans, or give out new stock. Jun 06 20:18:29 usbnet works on other machines too Jun 06 20:18:32 <_av500_> AMD is a playto make intel look nicer Jun 06 20:18:35 bradfa: iuno. all I know is it was somewhere, sometime, last century. I do break it sometimes, but at least there are user servicable parts inside Jun 06 20:18:35 I suspect AMD is still making $ off of their wafer fabs aka GlobalFoundries Jun 06 20:18:42 <_av500_> nah Jun 06 20:18:49 trev-hub they closed Jun 06 20:18:54 yeah offtopic sorry hehe Jun 06 20:18:55 trev-hub_, amd got out of global foundries mostly Jun 06 20:18:57 GlobalFoundries is wanky too Jun 06 20:18:59 at the moment Jun 06 20:19:04 trev-hub_, it was too successful for them Jun 06 20:19:09 http://online.wsj.com/article/SB10001424127887323844804578526871014195516.html ? Jun 06 20:19:10 bradfa: lol Jun 06 20:19:10 I meant the closed some stuff Jun 06 20:19:14 did they split off? Jun 06 20:19:21 <_av500_> GF was a play to pump money into east germany Jun 06 20:19:36 _av500_, girl friend? Jun 06 20:19:38 http://arstechnica.com/business/2013/04/amd-on-ropes-from-the-top-of-the-mountain-to-the-deepest-valleys/ Jun 06 20:19:43 <_av500_> panto: globalfloundires Jun 06 20:19:50 av500 into saxxonia Jun 06 20:19:54 only Jun 06 20:19:55 <_av500_> woglinde: yeah Jun 06 20:19:55 <_av500_> right Jun 06 20:20:00 <_av500_> Saxxon ftw! Jun 06 20:20:01 well..boys, trolls.. gtg and get some sleep Jun 06 20:20:06 I know, but an east germany gf sounded more interesting Jun 06 20:20:11 jdoles: do me a favor and come back on a rainy weekend! Jun 06 20:20:13 <_av500_> panto: Jun 06 20:20:14 jdoles: you're fun! Jun 06 20:20:14 KotH: did you watch that video? Jun 06 20:20:18 KotH: good (ugt) night, jimbob! Jun 06 20:20:22 mru: oh..dman.. Jun 06 20:20:24 mru: iw ill now Jun 06 20:20:31 KotH, it's raining here Jun 06 20:20:37 what other images are out there for the BBB other than nelson's and armhf's? Jun 06 20:20:54 cyronin_, angstrom? Jun 06 20:20:55 android ! Jun 06 20:21:02 but not angstrom? Jun 06 20:21:05 * bradfa has to go pick up car now, troll on troll brothers! Jun 06 20:21:07 <_av500_> biuldroot Jun 06 20:21:12 <_av500_> build even Jun 06 20:21:14 QNX Jun 06 20:21:17 <_av500_> WinCE Jun 06 20:21:25 freedos Jun 06 20:21:25 hannah montana linux Jun 06 20:21:30 <_av500_> lol Jun 06 20:21:34 im just looking for something that I can actually install a built kernel onto and actually boot Jun 06 20:21:37 +1 for bleeter Jun 06 20:21:48 cyronin_, angstrom can do that =) Jun 06 20:22:05 bleeter: thats's for the biebelboard Jun 06 20:22:09 bradfa: sunshowers at least? Jun 06 20:22:14 weirdos Jun 06 20:22:18 mru: I'd buy that for a dollar! Jun 06 20:22:19 mru beiberboard ?! Jun 06 20:22:32 * m_billybob cringes Jun 06 20:22:33 <_av500_> mranostay: shut up and get drunk Jun 06 20:22:41 mranostay, no, god's not beating his wife today Jun 06 20:23:07 bradfa: i never heard of that term before you posted it Jun 06 20:23:07 so are the drops her tears? Jun 06 20:23:17 mru: o_0 Jun 06 20:23:21 mranostay, me neither Jun 06 20:23:24 but now I really gtg Jun 06 20:23:26 cya Jun 06 20:23:51 KotH: tom waits is awesome Jun 06 20:24:09 <_av500_> +10 Jun 06 20:24:13 <_av500_> heck, +11 Jun 06 20:24:18 he goes to 11? Jun 06 20:24:23 <_av500_> at least Jun 06 20:25:01 <_av500_> mru: do you have that life recording with all the shit he tells the audience? Jun 06 20:25:03 If the BBC News has a video of him, he goes to 11 (for their website video volume control does this) Jun 06 20:25:12 <_av500_> film at 11 Jun 06 20:25:15 _av500_: I have spotify Jun 06 20:25:41 <_av500_> good luck Jun 06 20:26:19 weird stuff Jun 06 20:26:27 must watch more Jun 06 20:26:31 mru: thanks! Jun 06 20:27:37 http://bitsavers.informatik.uni-stuttgart.de/pdf/motorola/68000/ Jun 06 20:27:41 omg memories Jun 06 20:27:50 <_av500_> quick, get him a chair Jun 06 20:27:59 <_av500_> sit, granpa Jun 06 20:28:09 * panto waves cane Jun 06 20:29:06 panto: i still have an HC11 manual at home... on paper! Jun 06 20:29:08 * mranostay plays loud music Jun 06 20:29:33 anyways.. off to sleep Jun 06 20:29:35 me too Jun 06 20:29:36 night boys! Jun 06 20:29:43 nite koth Jun 06 20:29:45 g'night Jun 06 20:29:47 dont let the trolls bite you! Jun 06 20:29:47 time to go too Jun 06 20:29:49 and the quick reference card Jun 06 20:29:51 KotH: http://youtu.be/nMqxNPsfN50 Jun 06 20:29:57 I miss paper manuals Jun 06 20:30:00 mru: tomorrow ;) Jun 06 20:30:06 and the 6808 manual Jun 06 20:30:07 dammit I don't want to print them Jun 06 20:30:08 panto: printers? Jun 06 20:30:17 er, 6800 Jun 06 20:31:01 Hey all, can anyone point me to a BBB image where the Cloud9 IDE is not broken for Analog Read? Appreciate Help Jun 06 20:31:57 ha. tom waits kicks ass Jun 06 20:32:13 I heard he was big in japan Jun 06 20:32:32 at least he says so in a song Jun 06 20:33:39 http://theoatmeal.com/comics/bacon_flavor Jun 06 20:35:32 weather is nice today. planning on going to the park and setting some hunks o'pig a flame. Jun 06 20:37:03 Protip: Jun 06 20:37:07 Kill the pig first. Jun 06 20:37:17 heh Jun 06 20:37:29 <_av500_> the fire will do that Jun 06 21:19:57 Hey, I'd like to set up ICS for my beaglebone black on my linux laptop via usb connection, but I have no idea how to go about it. Jun 06 21:20:01 if I take a .dtbo file from angstrom, and load it via debian is there any reason it shouldn't work? Jun 06 21:21:46 loaded question bakers. Jun 06 21:22:17 may not work because its comflicting with other dt overlays Jun 06 21:22:22 conflicting too Jun 06 21:23:09 I took the bone-iio (for analog) .dtbo from angstrom and put it in my Debian install Jun 06 21:23:19 It loads properly, but the analog pins don't show up Jun 06 21:24:49 yeah sorry i dont know why Jun 06 21:24:57 still new to this meself Jun 06 21:25:25 I dunno if there is more than just the .dtbo file I need Jun 06 21:25:49 im thinking there is . . . or i believe ive read somethign to that effect, but for the life of me can not remember what Jun 06 21:25:57 maybe panto will chime in ? ;) Jun 06 21:31:19 panto is sleeping Jun 06 21:32:32 nite Jun 06 21:33:09 Crofton: did the germans take all your beer? Jun 06 21:34:27 not all of it Jun 06 21:34:32 thy lft today Jun 06 21:34:34 I am sad Jun 06 21:35:18 you needed to sprinkle Bud Light around the good beer Jun 06 21:35:41 like salt to vampires.. germans can't cross a stream of bad beer Jun 06 21:35:50 haha Jun 06 21:35:54 +1 mranostay Jun 06 21:36:07 s/Bud Light/Generic American Beer Jun 06 21:36:16 s/germans/canadians Jun 06 21:36:31 amirite? Jun 06 21:38:25 mranostay: don't think that works with koen Jun 06 21:38:41 koen will drink anything Jun 06 21:38:44 even solids Jun 06 21:41:29 ds2: he is immune to our crappy "beers" Jun 06 21:44:52 wonder if he will go for bottled vinger Jun 06 21:45:04 try and find out Jun 06 21:45:05 hmm need to test that out Jun 06 21:45:42 * mranostay fully expect koen to move to Texas someday :P Jun 06 21:46:22 what a weird thing to do Jun 06 21:46:54 so is drinking a bud light Jun 06 21:49:31 Hello Everybody. Has anyone an BBB running with the LDVS Cape 7" bundle from chalkboard ? Jun 06 21:50:53 I just got my BBB and the cape today and "plug and play" doesn´t seem to work ;) Jun 06 21:52:11 A lot of display based capes have issues and the LDVS isn't listed on the BBB cape page Jun 06 21:54:21 hmmm this is bad Jun 06 21:54:23 prpplague: I notice that the BBB cape page on elinux (http://www.elinux.org/BeagleBone_Black_Capes) lists TiWi-BLE cape as being hardware incompatible. I guess this means that the idea to disable the eMMC isn't viable? Jun 06 21:55:17 As I see it the Calkboard cape isn´t in the list Jun 06 21:56:51 The problem is I don´t have an micro HDMI cable right now so i can´t even check if the cape gets recognized or not. Jun 06 21:58:18 i own this bundle http://www.chalk-elec.com/?page_id=1280#!/~/product/category=3094859&id=13727570 Jun 06 21:59:41 Berun: serial console Jun 06 22:01:28 hmm i have no adapter or something to attach it so and the cape was made for the BBW so the serial console isn´t connected to the cape. no way to get to it while the cape is on Jun 06 22:03:18 ubuntu doesn´t do the trick out of the box. I will try to set the resolution parameters by hand Jun 06 22:12:08 build ur own LCD board Jun 06 22:14:28 "to build an lcd board, you must first create the universe" -bill crosby Jun 06 22:18:09 hmm Jun 06 22:19:46 the cape should be compatible to the BBB even if it was created for the BBW, because it uses only the same ports and pins which havent cahnged from BBW to BBB Jun 06 22:24:22 I am trying to make a little project and had just come to the conclusion that my mini2440 with 7" display was not fast enough and thought that the BBB will do the trick and now i am also stuck here Jun 06 22:24:32 berun, you can't ssh in? Jun 06 22:25:35 i am trying but the ubuntu kernel needs an login o.O Jun 06 22:25:43 root root doesn´t work ;) Jun 06 22:25:53 if you don't know the login, you're screwed in all sorts of ways :P Jun 06 22:26:01 should be root without password Jun 06 22:26:09 i will try Jun 06 22:26:44 and, you should be able to open a browser to the beaglebone and read teh intro guide, which would tell you that. :P Jun 06 22:26:53 root is incorrect as username Jun 06 22:26:57 err, "getting started" guide. Jun 06 22:27:40 there is none ;) when i run the start html it sends me on a page with ayax.html Jun 06 22:27:57 oh crap, you said ubuntu. my bad. Jun 06 22:27:58 AJAX_terminal.html Jun 06 22:28:03 i have no idea. Jun 06 22:28:14 what the login is. and the getting started guide wouldn't be there. Jun 06 22:28:33 but, without a login, you're board is a little usless, wouldn't you say? Jun 06 22:28:37 or am i missing something? Jun 06 22:28:39 HEHE Jun 06 22:29:23 yes you are right just wanted to try this kernel and see if it was cape(a)ble to use the cape ;) Jun 06 22:30:06 i see. Jun 06 22:30:40 if i were to guess without any insight whatsoever, i would guess display is probably going to hdmi. Jun 06 22:30:43 i have it froom here http://elinux.org/BeagleBoardUbuntu Jun 06 22:30:55 any clue what they normaly use as login ? Jun 06 22:32:04 Default user: ubuntu pass: temppwd Jun 06 22:32:19 at least that's what that link you sent says :P Jun 06 22:35:02 Login incorrect arm login: Jun 06 22:35:15 grr Jun 06 22:35:28 nice. Jun 06 22:40:21 do you know any ububtu image for the BBB where you know the login data ;) Jun 06 22:45:53 try ubuntu as a username? Jun 06 22:46:22 or, mount the filesystem somewhere else and look at it there Jun 06 22:46:26 did already :( Jun 06 22:46:36 what does that mean? Jun 06 22:46:56 sorry, i tryed already ubunti as username Jun 06 22:47:09 you try ubuntu ubuntu Jun 06 22:47:10 u Jun 06 22:47:14 yes Jun 06 22:47:27 vi /mnt/etc/shadow or something and change the password field to blank Jun 06 22:47:57 where did Berun find this image? Jun 06 22:48:07 caps lock on? :P Jun 06 22:48:16 http://elinux.org/BeagleBoardUbuntu Jun 06 22:48:30 no sorry that would be to easy Jun 06 22:49:00 http://rcn-ee.net/deb/flasher/raring/BBB-eMMC-flasher-ubuntu-13.04-2013-05-29.img.xz Jun 06 22:52:40 Berun: mount the flasher image on your PC (assuming a linux box), look in /build Jun 06 22:52:53 or rather /media//build Jun 06 22:53:00 i run windows :( Jun 06 22:53:14 step 1, fix that shit Jun 06 22:53:17 but my laptop is ubuntu Jun 06 22:53:18 mom Jun 06 22:53:23 HIHI Jun 06 22:54:06 so, plug it into your laptop Jun 06 22:57:05 ok it is in Jun 06 22:57:13 berun, virtualbox is your friend. Jun 06 22:57:17 now i see the rootfs Jun 06 22:57:31 i had it but its not the same Jun 06 22:58:01 the laptop is only for linux and i must say it runs it fine Jun 06 22:58:25 so where do i find the login now ? Jun 06 22:58:51 * russell-- presumes this is the flasher image you used to program the BBB? Jun 06 22:59:44 yes right now i have pluged in the micro sd from which i have installed ubuntu on the BBB Jun 06 22:59:59 look in "build" Jun 06 23:00:22 in the main directory is no build Jun 06 23:00:35 bin, boot, dev , ... Jun 06 23:00:53 of your flasher media or your laptop?> Jun 06 23:01:17 on the sd card Jun 06 23:01:26 mom only one way to make sure Jun 06 23:02:31 when i plug in the sdcard i get two partitions one boot and one rootfs Jun 06 23:03:05 in boot : backup, debug, Docs, Drivers, dtbs, tools, ans some files Jun 06 23:03:20 ok, dome some more tests Jun 06 23:03:29 anyone running debian-7.0.0-console-armhf-2013-05-29.tar.xz on a beaglebone black ? Jun 06 23:04:01 in angstrom land, there is build directory in the rootfs that holds the tarball and scripts to install to onboard flash Jun 06 23:04:36 in rootfs: bin, boot, dev, etc, home, lib, lost+foun(secured), media, mnt, opt, proc, root, run, sbin, selinux, srv, sys, tmp, usr, var Jun 06 23:04:57 by right now i am lost in ubuntuland Jun 06 23:05:14 i have used BBB-eMMC-flasher-ubuntu-13.04-2013-05-29.img.xz Jun 06 23:05:26 * russell-- doesn't know, suggests you ask the person who created the image Jun 06 23:06:35 i don´t care which image i use as long as i get the cape running and the display too Jun 06 23:06:49 angstrom is also fine to me Jun 06 23:06:56 or debian Jun 06 23:07:00 or or ro Jun 06 23:07:02 or Jun 06 23:07:26 * russell-- doesn't have your cape or a display Jun 06 23:07:33 serial++ Jun 06 23:07:40 i just want to use OWFS and an webserver Jun 06 23:07:56 ehh browser Jun 06 23:08:32 hmm i though these chalkboard thing where ehhh popular Jun 06 23:08:37 s Jun 06 23:10:41 i assue it is not possible to use an BBW image on an BBB ? Jun 06 23:11:46 russel i will send you one if you get it running ;) Jun 06 23:14:58 berun, someone got the 3.2 working on the black, but not with emmc or hdmi working. Jun 06 23:15:19 check the newsgroup Jun 06 23:15:43 hmm but chalkboard doesnt have a 3.2 LCD Jun 06 23:15:57 or are all LCDS capes the same ? Jun 06 23:18:21 hmm i thought i had started an thread on the forum, but i can´t find it Jun 06 23:18:57 https://github.com/RickKimball/beaglebone/tree/master/ws2811_spi_test Jun 06 23:19:13 can someone with DTS knowledge comment if i have anything bogus there Jun 06 23:19:42 found it Jun 06 23:22:07 anything is possible Jun 06 23:32:48 ds2: except the impossible Jun 06 23:39:39 the impossible is possible Jun 06 23:39:51 PERL? Jun 06 23:41:38 what about perl? Jun 06 23:42:21 just the common tag line "the impossible is possible" Jun 06 23:42:37 don't let me hijack the thread Jun 06 23:50:08 that is all what we do here Jun 06 23:51:24 anychance someone with a BBB can try something for me (needs gadget or serial access) Jun 06 23:52:38 ifconfig eth0 down && ifconfig eth0 netmask Jun 06 23:52:46 and see if eth0 still actually works Jun 06 23:58:25 it does Jun 06 23:58:30 bah Jun 06 23:58:34 don't even have to test it Jun 06 23:58:44 mranostay: oh, it doesn't for me Jun 06 23:59:09 https://gist.github.com/cornet/9813ac80cbde8e99fe10#file-gistfile1-txt Jun 06 23:59:38 i appear to get similar results with 2 different debian images and angstrom Jun 07 00:06:39 https://gist.github.com/cornet/b5a444b7a17f46295df9 # here is Angstrom log Jun 07 00:11:11 Hello, does anyone have a solution for changing PinModes via the command line? Jun 07 00:11:21 for the beaglebone black that is. Jun 07 00:18:01 since sys/kernel/debug/omap_mux/ is not available I am not sure how to change the PinMode properly Jun 07 00:18:24 that is never the proper way to change it Jun 07 00:19:45 Well that is good to know, any suggestions to get me steered into the proper way? I have also been trying to achieve changing the pinmode in c. Yet, I think something is wrong with my C code and wanted to check via the command line. Jun 07 00:20:31 you can see the pin modes with: cat /sys/kernel/debug/pinctrl/44e10800.pinmux/pins Jun 07 00:24:17 mastiff thanks this will help some Jun 07 00:31:37 so will writing to the file /sys/kernel/debug/pinctrl/44e10800.pinmux/pins or pinmux-pins change the pin mode? I am new to memory mapping and am a little lost with this. Jun 07 00:33:32 Try both and let us know if one way works. Jun 07 00:39:03 mastiff It isn't very happy with me writing to this file, even after changing the file permissions. So, I am going to say it doesn't work? My aplogies for being a little ignorant Jun 07 00:40:46 I've had similar problems, can't change LED brightness even though file permissions are ok Jun 07 00:41:21 not for turning led on or off, but changing the brightness value Jun 07 00:49:17 I'm sure you guys get 1000 questions a day about PWMs. :( Jun 07 00:50:28 Is PWM functional yet with 3.8/3.9 or do I need to be hunting an old image? Jun 07 00:59:31 so guys due to it is very urgent that i get the beaglebone black running with a touch screen, do you think there will be an solution soon to the BBW LDVS cape comparability to the BBB Jun 07 01:00:44 email the creators of it. Jun 07 01:03:00 they don´t seem to be very ehhh interested in giving me an solution and gave me links to old artikels in which they explain how to set up the screen for the BBW. Jun 07 01:03:45 PimpSmurf: Here's this one. I'm about to try it a few minutes also. http://hipstercircuits.com/enable-pwm-on-beaglebone-with-device-tree-overlays/ Jun 07 01:04:03 but as someone mentioned in the forums the BBB uses a different grafics "driver" so that these instructiuon should not work Jun 07 01:04:25 hi Jun 07 01:04:56 Berun: Aye your using OWFS on BBW? I've not tried it, but I've been using 1-wire on Linux PCs with a USB-1-wire dongle. Jun 07 01:05:11 I am new Jun 07 01:05:25 I am just planning to buy one board Jun 07 01:05:26 hey new. Jun 07 01:05:29 for study purpose Jun 07 01:05:41 I kind of panicked a bit and now have buyed an beagleboard-xm and dispaly just to make shure i have something that runs Jun 07 01:06:02 I wanted to know few features of Beagel Bone Black Jun 07 01:06:12 Berun: I'm still calling BBB beaglebone-beta at this point. :P Jun 07 01:06:28 Zafar: did you read the reference manual available on the website? Jun 07 01:07:17 yes But I am not Sure How to flash my own imag into it as it have eMMC in it Jun 07 01:07:29 ...well that's not what you asked. Jun 07 01:07:38 pru you may be right at this point, hopefully it will get better in some weeks Jun 07 01:08:07 Berun: it really is getting better every week! so much so, that I'm probably going to pull mine out of its box finally. Jun 07 01:08:13 soon. Jun 07 01:08:18 hehe Jun 07 01:08:24 (not kidding :-|) Jun 07 01:08:31 I have zero knowledge about device tree. I'm just trying to access the PWM controls in the /sys interface. I have to recompile a kernel for that? Everytime I want to change it? Jun 07 01:08:38 ich will have to present something in a week Jun 07 01:08:57 no kernel recompile. that's the whole POINT of device tree, so you don't have to recompile the kernel. Jun 07 01:09:10 Berun: why not just buy a white and use an old 3.2 image? Jun 07 01:09:14 (assuming it worked on white) Jun 07 01:09:17 Zafar: Features are at http://beagleboard.org/Products/BeagleBone%20Black and http://circuitco.com/support/index.php?title=BeagleBoneBlack and http://elinux.org/BeagleBone Jun 07 01:09:32 the white is out of stock everywhere in germany Jun 07 01:09:41 the xm was also Jun 07 01:09:48 they have whites on the shelf at microcenter near me :P Jun 07 01:09:50 but i found one on ebay Jun 07 01:09:53 I want to get a white but they're spendy Jun 07 01:09:58 as direct buy Jun 07 01:10:07 I suspect they're just renewing BBW stock while BBB is all the rage Jun 07 01:10:20 not* renewing Jun 07 01:10:24 Berun, since you're in a bind, i'll sell you mine. how d oes $400 usd sound? Jun 07 01:10:36 i'll ship it priority air. Jun 07 01:10:50 anyone know offhand what BBB mac addresses tend to start with? Jun 07 01:10:52 :P Jun 07 01:10:53 of cause for extra money right Jun 07 01:11:12 ssi: i would absolutely hope and assume it would be unique, as all mac addresses should. Jun 07 01:11:42 ssi: link/ether c8:a0:30:af Jun 07 01:11:45 Then I'm confused. I'll try to find an intro to device tree or something. The link above doesn't explain what to do with all that device tree code to enable pwm's Jun 07 01:11:52 yep, that's more like it Jun 07 01:11:54 thanks fiola :D Jun 07 01:11:55 fiola: it's seriously static!?!? Jun 07 01:12:11 pru_evtout_: they're unique, but the particular phys that bbb uses probably allare in a range Jun 07 01:12:12 ssi: c8:a0:30 is the manufacture part i think Jun 07 01:12:13 ok guys i will go to sleep now. its 3 o´clock in the morning here in germany Jun 07 01:12:24 see you later Jun 07 01:12:33 I was trying to find it in my arp table :P Jun 07 01:12:42 I don't know if it's static or not, but it's always initialized to a commercially allocated prefix per manufacturer. Jun 07 01:12:47 oh..ok. worried me. pretty sure my switch wouldn't like me attaching 20 or so with identical mac addresses. Jun 07 01:13:00 no, not identical Jun 07 01:13:07 but the first three or four bytes are probably all the same :) Jun 07 01:14:34 Never heard of a any manufacturer of any equipment sending out units with the same MAC address. That would be awesomely dumb, lol Jun 07 01:15:28 It may be allowed though. As long as they use their allocated prefix, it's probably compliant. Jun 07 01:15:39 yeah. angstrom doesn't have the "dtc" command Jun 07 01:19:13 PimpSmurf: my bbb, burned with one of the regular images, has it. :-\ Jun 07 01:19:27 you running something old? Jun 07 01:29:35 anyone using wifi on their BBB? Jun 07 01:32:37 hello Jun 07 01:32:55 how a can do a backut an trasfer for FTP Jun 07 01:32:55 ? Jun 07 01:32:58 in beaglebone black Jun 07 01:40:56 same way you do it in linux :D Jun 07 01:45:51 I was running the very latest angstrom and it didn't have it. Jun 07 01:46:23 I'm flashing arch's new version now. I prefer arch normally, and It has a package for dtc Jun 07 01:46:44 maybe I can get it working in arch afterall Jun 07 01:46:58 * m_billybob just cross compiled dts's Jun 07 01:47:14 they're platform agnostic as i understand it Jun 07 01:59:23 Yeah I have DTC again! :) Jun 07 02:00:53 m_billybob: hi-o sailor. Jun 07 02:25:13 Jeez I can't find anything that works on linux 3.8.13 to get the PWM working Jun 07 02:25:27 there are instructions to use it 2-3 different ways, 10 different places around the internet Jun 07 02:25:45 but I can't get period_ns and duty_ns to actually do anything when I "echo 1 > run" Jun 07 02:29:20 is it muxed correctly? Jun 07 02:31:32 I would guess not. I'm trying to use P8_13 Jun 07 02:32:58 It appears none of them are mux'ed to PWM Jun 07 02:47:55 emeb_mac: you know what i'm thinking? Jun 07 02:48:17 mranostay: ICE CREAM! Jun 07 02:48:45 no? Jun 07 02:48:56 FPGA Jun 07 02:49:40 oh, that. Jun 07 02:50:20 one of these days I need to build up my RF ADC board & test it out. Jun 07 02:52:31 emeb_mac: I was able to program my fpga via spi using your bcc_tool Jun 07 02:52:45 ssi: nice! Jun 07 02:52:48 emeb_mac: I haven't been able to program the flash though, and I'm not sure how to troubleshoot it Jun 07 02:53:09 did you add the -v option to see the error results? Jun 07 02:53:15 yeah Jun 07 02:53:18 hang on lemme try it again Jun 07 02:53:39 and did you first program the FPGA with the pass-thru bitfile? Jun 07 02:53:51 heh Long Ranger movie? Jun 07 02:54:18 yeah Jun 07 02:54:27 so I'm trying to do that again now, and I get intermittent issues Jun 07 02:54:33 bcc_cfg: INIT low during bitstream send Jun 07 02:54:44 I'm getting taht every time now... before I'd get that occasionally Jun 07 02:55:01 check that all your solder joints are good? Jun 07 02:55:09 it programmed that time Jun 07 02:55:20 ./bcc_tool -v ../flash_prog/flash_prog.bit Jun 07 02:55:51 so then I do: ./bcc_tool -v -p ../blinky/blinky.bit Jun 07 02:56:20 and I get: http://pastebin.com/6mfDWnXV Jun 07 02:56:52 btw anyone know if the NAND cape actually works? Jun 07 02:57:59 ssi: that's what I'd expect if there's a problem in the hookup from the FPGA to the SPI flash Jun 07 02:58:19 that's what I was afraid of Jun 07 02:58:21 how can I troubleshoot it? Jun 07 02:58:32 check that the SPI flash is soldered in correctly Jun 07 02:58:46 that's U302, correct? Jun 07 02:58:48 check continuity from pins on FPGA to pins on Flash Jun 07 02:59:23 yes - U302 Jun 07 02:59:34 I'm pretty sure it's oriented correctly Jun 07 02:59:37 ST logo toward U303 Jun 07 02:59:46 I'll try to ring out its pins **** ENDING LOGGING AT Fri Jun 07 02:59:58 2013