**** BEGIN LOGGING AT Thu Oct 25 03:00:01 2012 Oct 25 06:28:43 <_av500_> west coast trolls all asleep? Oct 25 06:30:57 _av500_: on the east coast we troll. Oct 25 11:00:25 gm all Oct 25 11:00:39 gm one Oct 25 11:01:10 hey jkridner Oct 25 11:02:38 mourning Oct 25 11:14:28 koen: pls try to push the two meta-initramfs patches, thx. I'd have also a new kexecboot 3.4 kernel but yocto-tiny_3.4 is not in 1.3 so I'd wait Oct 25 11:27:50 hiho Oct 25 11:29:08 haha funny lance armstrong bug Oct 25 11:55:58 koen: trying to recall where I got last night... Oct 25 11:56:05 got some HDMI patches installed. Oct 25 11:56:23 haven't added the DT support for them. Oct 25 11:56:32 er, not installed... applied. Oct 25 11:56:41 pushed it to my fork. Oct 25 11:57:23 worked around EEPROM not being programmed by hacking the EEPROM dependencies out of u-boot. Oct 25 11:57:41 hmmm... Oct 25 11:57:49 you made him angry Oct 25 11:57:58 apparently. Oct 25 11:58:01 _av500_, what was the user ranting about Oct 25 11:59:05 http://www.youtube.com/watch?v=oolHYPGPTmM Oct 25 11:59:19 Crofton|work: badly worded error messages Oct 25 11:59:34 in "hob", whatever that is Oct 25 11:59:59 the fact that the error msg said "you need X" when the real error was "you need version Y.Z of X" Oct 25 12:00:25 apparently a hob is something about an English stove Oct 25 12:03:22 Crofton|work: it's the top cooking surface on a stove Oct 25 12:03:45 'stovetop' in US english Oct 25 12:04:11 jkridner, programming the eeprom would have been another option Oct 25 12:04:40 mdp: couldn't figure easier way to do that... Oct 25 12:04:53 I figured easiest way was to hack u-boot and get into Linux to program the EEPROM... Oct 25 12:05:04 of course, now I have less motivation to complete the programming step. Oct 25 12:05:19 koen: did I scare you off? Oct 25 12:05:22 the u-boot i2c commands worked for me Oct 25 12:05:41 mdp: yeah... those require hacking u-boot. Oct 25 12:05:48 which is all I did. Oct 25 12:05:49 but whatever works is always a good way to do things ;) Oct 25 12:07:05 arg, where did the riotig post gop Oct 25 12:07:55 koen: does 'make dtbs' do the trick of appending the DT? Oct 25 12:08:18 * mdp covers his ears Oct 25 12:08:37 lalalalala Oct 25 12:08:59 or, do I need to 'make uImage-dtb' Oct 25 12:09:07 av500, nobody wants a repeat of yesterday's rants ;) Oct 25 12:09:08 * jkridner feels like he just asked a FAQ Oct 25 12:09:22 mdp: its all in my scroll buffer Oct 25 12:09:26 k. Oct 25 12:09:29 should I resend? Oct 25 12:09:52 BeagleBot can tell me. Oct 25 12:11:54 jkridner, in summary, you have the option to simply load the dtb separately in u-boot Oct 25 12:12:03 it's smart and will pass it Oct 25 12:12:03 k, so the rant was just about if it is a good idea.... I agree it is not a good idea to ship an appended image. Oct 25 12:12:19 but, it is handy. Oct 25 12:12:37 I don't want to spend my bring-up time trying to debug passing in my DT. Oct 25 12:12:46 it's a standard feature Oct 25 12:13:00 jkridner: check the README.md in kernel.git Oct 25 12:13:12 it has something very close to exact steps Oct 25 12:13:18 my devel set up either loads from /tftpboot or mmc..pretty trivial Oct 25 12:13:51 fine. you got me lazy with the uImage-dtb yesterday Oct 25 12:15:09 koen: I'm using vanilla u-boot. anything keeping Angstrom from using it vanilla except for boot-time patches? Oct 25 12:15:33 tftpboot ${loadaddr} uImage.bone2;tftpboot ${fdtaddr} am335x-bone2.dtb;bootm ${loadaddr} - ${fdtaddr} Oct 25 12:15:59 mdp: works if we ship each bone with a PC :) Oct 25 12:16:14 jkridner: jsut the boottime patches Oct 25 12:16:21 jkridner: http://dominion.thruhere.net/koen/angstrom/beaglebone/dt/ Oct 25 12:16:30 av500, we have the chromebook running chrbuntu now! Oct 25 12:16:32 jkridner: for regular bone, that is vanilla uboot + xip patch Oct 25 12:16:34 are you moving to meta-beagleboard anytime soon? Oct 25 12:16:44 jkridner: this xip approach was shot down, but it does save us ~1s Oct 25 12:16:51 av500, s/tftpboot/fatload/ :P Oct 25 12:16:55 jkridner: give me 48 hours in a day first Oct 25 12:17:15 a second saved is a second earned Oct 25 12:17:22 no worries if it doesn't happen any time soon... Oct 25 12:17:25 just don't reboot Oct 25 12:17:27 just wondering if I should be on-watch for it. Oct 25 12:17:43 jkridner: and bring me the head of the TI person that recommended using an aptina sensor for the QL cape Oct 25 12:18:05 koen, I'm hoping my latest news means I get to dodge the bullet ;) Oct 25 12:18:18 koen: firsrt I get the head of the guy that proposed the sensor on the blaze Oct 25 12:18:34 the very same sensor that was in the new ipad Oct 25 12:18:40 that you just could not buy... Oct 25 12:19:48 koen, it was probably the same decision process that led to the nxp hdmi transmitter selection Oct 25 12:20:22 "our agent in shenzhen found an entire van-full of these parts for cheap!!!" Oct 25 12:21:11 where do I find things like the name for the .dtb file that gets loaded? Oct 25 12:21:30 ${devtree} doesn't tell me much. Oct 25 12:21:34 mdp: unlikey Oct 25 12:21:38 unlikely Oct 25 12:21:40 I'll start looking through u-boot source. :-/ Oct 25 12:21:43 jkridner: type 'printenv' in the uboot console Oct 25 12:22:23 jkridner, u-boot inherently knows the dtb name now? Oct 25 12:22:35 jkridner: or you know, read the README.md in kernel.git, it's stated there Oct 25 12:22:49 I wonder why I called it README Oct 25 12:22:56 README only gives 1 name. Oct 25 12:23:06 you called it README.md not README Oct 25 12:23:22 jkridner: ext2load mmc ${mmcdev}:2 ${fdtaddr} ${devtree} Oct 25 12:23:26 jkridner: that's in README.md Oct 25 12:23:42 yeah, I could see from that ${devtree} Oct 25 12:24:13 jkridner: one line up: devtree=/boot/am335x-bone.dtb Oct 25 12:24:31 so now you know the name of the file that gets loaded Oct 25 12:24:39 well, there you go. that isn't in the README.md Oct 25 12:24:51 * mdp files an IR Oct 25 12:25:01 jkridner: ehm Oct 25 12:25:06 jkridner: I urge you to check again Oct 25 12:25:52 jkridner: line 28 and line 36 Oct 25 12:25:56 can you two take that to a private channel? Oct 25 12:26:04 exact steps! Oct 25 12:26:21 I'm not seeing on https://github.com/beagleboard/kernel/blob/3.7/README.md Oct 25 12:26:31 anyway, I see. Oct 25 12:26:50 93019025 (Koen Kooi 2012-10-17 08:42:32 +0200 28) ```devtree=/boot/am335x-bone.dtb Oct 25 12:28:06 ah Oct 25 12:28:09 markdown fail Oct 25 12:54:48 * mdp hugs the dt sram driver Oct 25 12:54:55 mdp why? Oct 25 12:55:09 I needed it...and it's so elegant Oct 25 12:55:17 makes me want to cry Oct 25 12:55:25 timing Oct 25 12:55:26 s Oct 25 12:55:53 it means one less hack Oct 25 13:12:35 I ordered circuit boards for my four-channel zigbee 120V christmast light controller last night. ChristmasBone here we come! Oct 25 13:15:19 \o/ Oct 25 13:17:00 picture or it's fake Oct 25 13:17:57 silicalgel with javafx gui? Oct 25 13:51:35 hmm... I'm getting a "data abort" error from u-boot now. Oct 25 13:54:17 well, better then nothing Oct 25 13:54:32 jkridner: regular bone or ddr3 one? Oct 25 13:55:35 ddr3 one... I figured that I was using a wrong bootcmd, because kloadaddr isn't defined under my vanilla u-boot. Oct 25 13:56:32 README.md has a sample uenvt.text for vanilla uboot as well Oct 25 13:57:23 jkridner: I get the same data abort, I use an appended DT as a stopgap for that Oct 25 13:57:43 jkridner: only on the ddr3 board, though Oct 25 13:58:42 I figured out the missing var, but now I'm getting a "data error" after kernel decompression. Oct 25 14:01:37 * av500 productizes koen 's stopgap and proves a point :) Oct 25 14:02:40 :( Oct 25 14:19:12 hmm. :( my kernel still gets the data abort error even with appended DT Oct 25 14:28:31 ship it Oct 25 14:29:47 mdp: ah, finally you see the light of real world customers solutions :) Oct 25 14:31:57 sings "welcome to the real world and the show" Oct 25 14:32:20 http://www.youtube.com/watch?v=wuKwBwYKvKs Oct 25 14:35:13 leto running man? Oct 25 14:35:29 with schwarzenegger? Oct 25 14:43:36 woglinde: nah, more like standard german metal. Oct 25 14:44:49 woglinde: running man is a classic, though. a pretty good descendent is death race (/w jason statham) in the long cut. :) Oct 25 14:45:55 av500, data abort is about equal to most consumer devices shipped ;) Oct 25 14:52:51 Can anyone point me to a specification of the BeagleBone for standard power draw under full load without and attached capes? Oct 25 14:53:11 define "full load" Oct 25 14:53:28 100% cpu load let's say Oct 25 14:53:38 or for that matter just idle Oct 25 14:53:41 why get fancy? Oct 25 14:56:09 death race was shit, running man is awesome :) Oct 25 14:57:14 no Oct 25 14:57:20 running man features 2 US governers as well Oct 25 14:57:25 just like predator Oct 25 15:01:30 okay, I found it- in the manual. Imagine that :-/ Oct 25 15:01:36 derp Oct 25 15:39:18 https://www.circuitlab.com/editor/ Oct 25 15:39:23 http://www.circuits.io/circuits/new Oct 25 15:52:21 I have some problems with my beagle board setup - and I was hoping that this has been solved.. Oct 25 15:52:59 seems like nobody still has solved the lack of explicit questions Oct 25 15:52:59 I have a board that is a Beagle-based design (a copy of the Cx ref design) Oct 25 15:53:35 and when the CPU gets heavily loaded I occasionaly get "illegal instruction" errors Oct 25 15:53:49 these can ahppen at any time on any program Oct 25 15:54:06 for example, ls will sometimes give an illegal instruction Oct 25 15:54:20 or top will "segmentation fault" Oct 25 15:54:30 This is some weird stuff.. Oct 25 15:54:47 it feels like a hardware error Oct 25 15:55:01 Russ: how is your mite project coming along? Oct 25 15:55:14 I have added a ton of bypass caps on the power supply rails - and they look good,, but are at strange voltages.. Oct 25 15:55:56 louloulou_: sounds like your voltage scaling is not working Oct 25 15:56:12 louloulou_: or that is a possible cause of that sort of thing Oct 25 15:56:26 like 1.3 volts on the vcore 1.3 and 1.0 vols on vcore2 Oct 25 15:59:49 how do i adjust the voltage scaling? Oct 25 16:00:08 it is all under i2c control Oct 25 16:00:15 its under i2c control Oct 25 16:00:30 is there a deriver or a config file? Oct 25 16:00:40 comes out of the omap on the Smart Reflex I2C port and goes into the PMIC on a dedicated port Oct 25 16:00:59 did someone write code for it/ Oct 25 16:01:01 ? Oct 25 16:01:02 there is config somewhere in arch/arm/mach-omap2 Oct 25 16:01:11 its working on beagle Oct 25 16:01:53 I do not care at all about power savings- can i just put the voltages to max? Oct 25 16:02:01 yes you can Oct 25 16:02:07 ok- Oct 25 16:02:12 let me look in that dir Oct 25 16:02:29 its easier to do it in u-boot then disable Smart Reflex driver Oct 25 16:02:46 the mach-omap2 stuff gives me a headache Oct 25 16:03:00 where is the /arch directory? Oct 25 16:03:19 mkay... Oct 25 16:03:19 of course it could be some other issue which means not enough power gets to CPU Oct 25 16:03:32 arch/.... in the linux kerne] Oct 25 16:03:51 xora overheating maybee too? Oct 25 16:04:01 ok -- (i am an idiot) so you cant adjust the voltages with a file mod - you have to recomple kernel? Oct 25 16:04:06 or do it in uboot? Oct 25 16:04:31 woglinde: I always assumed hardware guys check that :-D Oct 25 16:04:41 xora haha Oct 25 16:04:47 the die temp is about 160F Oct 25 16:05:02 (I wrote code to read the die temp sensor) Oct 25 16:05:17 louloulou_: default beagle config AFAIK enables voltage scaling depending on CPU frequency Oct 25 16:05:26 till later Oct 25 16:05:31 louloulou_: you should see vcore1/2 vary with load Oct 25 16:06:16 they are not varying - but i think smart reflex is disabled Oct 25 16:06:35 so i would assume that they would go to max Oct 25 16:07:13 Im not sure if they stay at u-boot levels or go to max Oct 25 16:07:45 or if u-boot levels are max or some lower value Oct 25 16:07:57 how do i do it in uboot? Oct 25 16:08:01 i am in it now? Oct 25 16:08:17 louloulou_: read the code, I cant remember off top of head but it should be fairly obvious Oct 25 16:08:26 * XorA is more used to omap4 Oct 25 16:08:44 anohter stupid question- is there a memory test i can do in uboot? Oct 25 16:08:52 Like the old PC bios mem test Oct 25 16:17:49 koen, with everything going on, I haven't had time to mess with it other than the opportunity to use the one I have Oct 25 16:18:08 koen, at some point, I'll probably do a respin so I don't have to bit bang i2c Oct 25 16:30:08 ok I am an idiot. Where can I find out how to change the CPU voltage and freq via the /sys/devices/ files? Oct 25 16:31:23 the linux whiz says that buy cat 5000000 > /sys/devices/system/cpu/cpu0/cpufreq/scaling_setspeed I can put the cpu at 500 mhz Oct 25 16:31:46 my question is how the hell did he know this, and where can I go to learn it? Oct 25 16:31:56 read the source luke Oct 25 16:32:25 OK i respect that, but you are telling me to look in the phone book -- is ther an index? Oct 25 16:32:42 (where do i start?) Oct 25 16:33:41 there are many nice books on the linux kernel and embedded linux Oct 25 16:37:11 dm8tbr and with DT they are all obsolete Oct 25 16:42:30 woglinde: job security, new books want to be written, printed, sold ... Oct 25 16:55:00 dm8btr sure but not that he wonders why the structure in the books are not there anymore Oct 25 16:58:26 man I am knee deep in u-boot and I dont see anything regarding smart-reflex -- looking in /u-boot/board/ti/beagle/beagle.c Oct 25 17:02:56 I thought the smart reflex stuff was all proprietary in an i2c channel that's otherwise inaccessible Oct 25 17:11:33 louloulou_: I might have lead you down wrong path, maybe what I know is only valid for omap4 :-( Oct 25 17:16:14 louloulou_: its possible the pmic on beagle is already at the correct V for booting Oct 25 17:16:35 _av500_: no #LeavesOnThursday ? Oct 25 17:34:37 <_av500_> ? Oct 25 17:35:42 lol Oct 25 17:48:45 mdp, I found my Ethernet LED problem - am335x PRM_DEVICE.RSTTIME1 register sets RSTTIME1 bitfield to 06 and that is too short of a time to reset the PHY, the PHY effectively debounces the reset signal, increasing the time to 0xFF causes the PHY to warm reset properly Oct 25 17:56:24 what version of u-boot should I run for beaglebone? I just tried compiling mainline and it won't load for me. Oct 25 17:56:53 devfunks, 2012.10 mainline works well Oct 25 17:57:06 ok- maybe i did something wrong then Oct 25 17:57:19 are you using am335x_evm_config ? Oct 25 17:57:39 no i used omap3_beagle_config Oct 25 17:57:54 am335x_evm_config is for beaglebone Oct 25 17:57:59 thanks :) Oct 25 17:58:00 omap3_beagle_config is for beagle and xM Oct 25 17:58:05 gotcha Oct 25 17:58:18 if things were clear then it wouldn't be fun :) Oct 25 17:58:39 haha, definitely wouldn't learn anything if it was plug-n-play Oct 25 17:59:14 why do people insist on using a canned config at all? Oct 25 18:01:37 mru: because for for some, it's nice to at least start with something that works. For beginners, it's nearly mandatory. Oct 25 18:01:47 I like to start with what works, and then proceed to break it from there. Oct 25 18:02:10 alan_o: I've never seen a canned config that worked for me Oct 25 18:02:26 and that's definitely a problem. Oct 25 18:02:44 They get out of date unfortunately. Oct 25 18:03:04 The ones in koen's 3.6 and 3.7 are good. Those are what I started with. Oct 25 18:03:51 they're not just out of date Oct 25 18:03:57 they're plain wrong for what I want to do Oct 25 18:04:32 I'm sure patches would be welcomed by those maintainers. Oct 25 18:04:47 Especially from someone who knows what he's doing. Oct 25 18:05:03 mru, do you suggest not starting with a canned config for u-boot? Oct 25 18:06:26 as a starting point they can be useful Oct 25 18:06:27 bradfa: where can I find the am335x_evm_config? Oct 25 18:06:34 to get the basic requirements set Oct 25 18:06:47 devfunks, if you look in boards.cfg the target is called "am335x_evm" Oct 25 18:07:16 the config is in include/configs/am335x_evm.h Oct 25 18:15:57 i must have checked out the wrong branch or something because I can't find it anywhere Oct 25 18:16:31 master is generally the default Oct 25 18:16:48 i'm on 3.7 Oct 25 18:17:01 bradfa: fascinating Oct 25 18:17:27 devfunks: I thought you were still talking about u-boot..that's what bradfa is referencing above Oct 25 18:19:04 oh sorry, I am.. I couldn't find it in mainline u-boot so I was looking in my beagleboard checkout Oct 25 18:19:45 devfunks, if you checkout u-boot tag 2012.10 it's there, believe me :) Oct 25 18:20:18 mdp, now I'm trying to find somewhere in u-boot setup to stick a write to that register to set it Oct 25 18:20:29 k Oct 25 18:20:38 i basically followed http://elinux.org/BeagleBoard#U-Boot Oct 25 18:20:41 it's even in master too Oct 25 18:21:04 well, ok, if you insist, yes, it's in master too :) Oct 25 18:21:21 I hope you didn't use "Steve's Beagle U-Boot" repo ;) Oct 25 18:22:06 fun stuff...x-loader references..lots of things that aren't useful any longer Oct 25 18:22:14 https://github.com/bradfa/u-boot/blob/mainline/include/configs/am335x_evm.h <- there it is! :) Oct 25 18:22:17 TI should really update their docs Oct 25 18:22:23 TI has docs? Oct 25 18:22:46 mdp, is it in sprzbarf89.pdf? Oct 25 18:22:51 :) Oct 25 18:23:23 docs? This is a community board. You'll read the source and like it Oct 25 18:24:01 and follow the traces to understand :P Oct 25 18:24:10 beagleboard.org points to elinux wiki as the canonical source of info Oct 25 18:24:30 not to be confused with Canonical! :P Oct 25 18:24:39 mranostay: I went to #rpi and found no trace of understanding Oct 25 18:25:07 heh no nodejs for powerpc. Oct 25 18:25:13 bradfa, nm, that wiki is only for "BeagleBoard and BeagleBoard-xM wiki" Oct 25 18:25:15 thanks bradfa Oct 25 18:25:50 there's a wiki for the wiki? Oct 25 18:26:13 mru, I'm so confused! Oct 25 18:26:58 * mranostay is probably going to hear not being able to run nodejs on powerpc as a "limitation" today :) Oct 25 18:27:07 mru: can i borrow the cluebat? Oct 25 18:27:16 Hmmmmmm imx6 Hmmmmmm Oct 25 18:27:38 * mdp reminded of cubieboard. Oct 25 18:27:51 anybody here order a cubieboard? Oct 25 18:27:51 clubieboard? Oct 25 18:28:14 mranostay: https://lh4.googleusercontent.com/-H8_IN0R8q5s/Ti1ck1Lj7ZI/AAAAAAAAAD0/7f9URI0Rq7w/s0/ Oct 25 18:28:25 I was considering going hands on with our future SoC overlords Oct 25 18:29:21 amazon was it? Oct 25 18:30:23 mdp: ddr3 that is nice Oct 25 18:31:19 mru, not that overlord, the allwinner overlords Oct 25 18:32:03 I just can't make myself take anything with such a name seriously Oct 25 18:32:54 wonder if it is pincompatible with the 33x Oct 25 18:32:57 I suppose having awesome WM running on turbolinux on the allwinner a10 would be the naming trifecta win Oct 25 18:33:54 mru, the name does betray the geographical origin now doesn't it? Oct 25 18:35:24 mranostay, jkridner was working on a ddr3 board earlier...I wonder what that means? Oct 25 18:36:05 worst keep secret board? :) Oct 25 18:36:16 that's sekrit to you, sir Oct 25 18:36:18 *kept Oct 25 18:36:44 well sorry i see the u-boot commits. :) Oct 25 18:38:57 I've recommended we terminate this FOSS experiment because of that Oct 25 18:43:32 <_av500_> allwinner is awesome Oct 25 18:43:47 <_av500_> get it quick before the company names changes again Oct 25 18:54:52 mranostay: a limitation of node.js? I'd agree.... Oct 25 18:55:25 at your day job, what would you say it is that you do? Oct 25 18:56:58 I'm writing an application for the Beagleboard xM, I use the framebuffer for graphics and a USB touchscreen in /dev/input/event1 for input. How do I deal with the calibration of the touchscreen, is that supposed to be done at the application layer or the kernel layer ? Oct 25 18:57:33 kroon I would use qt Oct 25 18:58:03 alan_o: well i know i'm going to hear limitation of powerpc :) Oct 25 18:58:27 mranostay, there are no limitations for powerpc Oct 25 18:58:41 hmm Oct 25 18:58:42 Send mdp into your meeting. Oct 25 18:59:01 slideware and some handwaving fixes things in meetings Oct 25 18:59:29 Is that really a thing though? Node won't compile/run on ppc linux? Oct 25 18:59:38 alan_o: v8 doesn't Oct 25 18:59:53 * mranostay wonders why magic V8 is really doing Oct 25 19:00:03 endiannes? Oct 25 19:00:25 I have to say, even the node website is funny Oct 25 19:00:39 All the things from that parody video are basically above-the-fold Oct 25 19:00:51 Then the microsoft quote. yikes..... Oct 25 19:01:12 Then on the download page, if you download the source, the mime type is screwed up and it wants to open in gedit Oct 25 19:01:23 mranostay, when discussing powerpc related things, you must drop the phrase "boundedly undefined" once in a while and people will nod their heads Oct 25 19:01:33 pretty good webserver setup for a site that's peddling a webserver..... Oct 25 19:02:12 mdp: "why did't we switch to a better platform like x86" Oct 25 19:02:21 i heard that one :).......... Oct 25 19:02:37 probably because tlbmiss is there and won't allow a switch away from powerpc Oct 25 19:03:29 mranostay: So it sounds like this place is a significant step down from your previous employer Oct 25 19:03:43 alan_o, that is not possible. Oct 25 19:04:19 hehe no comment in any case Oct 25 19:04:31 mdp: really? I only know one guy from the previous place. Maybe I don't have enough data points. Oct 25 19:04:35 mranostay, you know what I say is true. ;) Oct 25 19:04:57 alan_o: was he/she stuck in Mobile, AL? :) Oct 25 19:05:08 hmm..... Oct 25 19:05:17 mranostay, it was bad enough working remote for them Oct 25 19:05:28 mranostay: not sure... I can check. Oct 25 19:05:42 alan_o: russian guy? :) Oct 25 19:05:50 alan_o, I know a ton of people since we all went in the acquisition ;) Oct 25 19:06:31 I don't really know him very well. Just met at ELC. He gave off competence. I decided I like him. Oct 25 19:06:35 lots of happy people Oct 25 19:07:05 that sounds like a linkedin skill Oct 25 19:07:13 "emanates competence" Oct 25 19:07:25 It true. It should be. Oct 25 19:07:47 "feigns competence" Oct 25 19:07:51 You know how when you meet someone and you can tell pretty quickly whether they know what they're doing or whether they're full of BS? Oct 25 19:08:40 Paul Graham says good people can sniff one another out. That's either true, or I'm not as good as I think I am :) Oct 25 19:09:08 how about bad people sniffing bad people out? Oct 25 19:10:30 heh "Graham's Hierarchy of Disagreement" Oct 25 19:12:14 mranostay: I hadn't thought about bad people sniffing one another out. You'd have to know you're bad first. Is that like 75% of people think they're better than average drivers? (no, I don't want to talk about about medians vs means) Oct 25 19:12:53 It's probably higher than 75% Oct 25 19:13:16 If I was a pollster, man we'd have some entertaining polls. Oct 25 19:13:38 anyway, so mdp, since you don't like mranostay's previous employer...... Who do you think are the top Embedded Linux companies? Oct 25 19:14:00 are there any left? :) Oct 25 19:14:01 I don't really know anything about any of them. Oct 25 19:14:12 I don't think there are any top embedded linux companies Oct 25 19:15:02 well, depends on how you define "top" Oct 25 19:15:08 :) Oct 25 19:15:10 Slimlogic obviously :-) Oct 25 19:15:37 alan_o: if good people can sniff one another out, so can stupid people Oct 25 19:15:39 In one method of categorizing, I would say linuxtronix, denx, pengutronix Oct 25 19:16:20 Maybe (clearly) I just don't know. I always had envisioned those companies being staffed with really smart people. Maybe that's just fantasy. Maybe only a small handful are really good, and those are the ones they send to ELC. :) Oct 25 19:16:53 So the code on SlimLogic's front page wouldn't make it into the kernel. Oct 25 19:17:09 mdp, free-electrons make your list? Oct 25 19:17:21 I like their online content. Oct 25 19:17:25 heh is this a backslapping thread? Oct 25 19:17:26 you guys are forgetting sakoman Oct 25 19:17:43 can i throw my fake company in? :P Oct 25 19:17:59 mranostay, only if you continue to break into fire alarms Oct 25 19:18:08 alan_o: I must declare a vested interest in Slimlogic! Oct 25 19:18:18 mranostay: do you also set fire to burglar alarms? Oct 25 19:18:26 XorA: did you write the code on the front page :) Oct 25 19:18:45 alan_o: what code? Oct 25 19:18:58 in the picture Oct 25 19:19:00 bottom left Oct 25 19:19:06 slimlogic.co.uk Oct 25 19:19:10 that one? Oct 25 19:19:11 I suspect that is from alsalib Oct 25 19:19:16 or libscenario Oct 25 19:19:45 ah, I just know I'd have gotten told to return from both the if and the else. Maybe it's just the maintainers I submit to Oct 25 19:19:56 bradfa, sure, I just don't know how many qualify as "the top" Oct 25 19:20:00 It doesn't bother me a bit, but others have preferences :) Oct 25 19:20:14 alan_o: its not kernel as its got fgets in it :-D Oct 25 19:20:30 ugh, french spacing Oct 25 19:20:32 oh hey, there's a blurry fgets up there Oct 25 19:20:57 it also might be some random bit of code the GFX artist nabbed Oct 25 19:21:12 mdp, how would you even quantify "the top"? Like good places to work or good at contributing to the upstreams? Or good at implementing for customers? Oct 25 19:21:13 like I said, I'm just busting chops. Looks fine to me. As long as it's not javascript :) Oct 25 19:21:22 is it fgets or fgetc? Oct 25 19:22:19 alan_o: unfortuneately I spend too much time coding and upstreaming to keep webpage flowing Oct 25 19:23:08 bradfa, exactly Oct 25 19:23:41 XorA: hehe. no doubt. It's far better than my website :) I don't have any stones to throw in that department. Oct 25 19:25:33 mru: everybody always forgets me! what are we talking about anyway? I haven't been paying attention ;-) Oct 25 19:25:51 sakoman: best embedded linux company Oct 25 19:26:12 sakoman: you are on the top embedded linux companies list! Oct 25 19:26:16 congrats Oct 25 19:26:23 heh, that is funny Oct 25 19:26:32 we are making the list now Oct 25 19:26:41 stay tuned for a Phoronix article on it Oct 25 19:26:55 sakoman: can we have your bank account number to wire you the prize money :P Oct 25 19:29:03 mranostay: cash works much better ;-) Oct 25 19:31:19 how about payment in beer? Oct 25 19:32:07 haha Oct 25 19:32:10 mranostay: beer works too! Oct 25 19:53:53 TIL powerpc is obsolete to x86.... Oct 25 19:54:12 some 10 years Oct 25 19:54:48 you mean x86 is obsolete, right? :) Oct 25 20:01:11 is the beaglebone serial connection ttyO0? Oct 25 20:01:24 through USB* Oct 25 20:01:49 we need to think of a reason to introduce ttyl1 as well Oct 25 20:04:12 ttfn? Oct 25 20:04:23 mru: and ttyI1 Oct 25 20:05:26 devfunks: ttyO0 is connected to the FTDI USB chip Oct 25 20:05:43 That's the default console for u-boot and the kernel. Oct 25 20:05:50 thanks- just wanted to verify Oct 25 20:06:18 kinda stuck on what to do... currently hangs on "Starting Kernel ..." Oct 25 20:06:30