**** BEGIN LOGGING AT Sun Sep 24 03:00:01 2017 Sep 24 03:00:11 same issue. I think the MRAA and UPM modules I could not get working have something to do with it. Sep 24 03:00:53 The UPM module is what software I am using to get this hardware to work. UPM relies on MRAA to work. Sep 24 03:01:14 wait, GPRS... didn't you say it was a GPS module ? Sep 24 03:01:22 Yea. Sep 24 03:01:24 GPS. Sep 24 03:01:34 GPRS and GPS are totally different things Sep 24 03:01:49 Sim28! Sep 24 03:01:56 Oh? Sep 24 03:02:14 I have been reading the wrong material. Sorry. Sep 24 03:03:12 Um...it is a GPS module but there are no GPS software, i.e. only gprs. Sep 24 03:03:26 I will get my act together soon. Please hold. Sep 24 03:03:30 GPRS has absolutely nothing to do with GPS Sep 24 03:03:45 they are not related in any way other than their names being somewhat similar Sep 24 03:04:00 I understand. Sep 24 03:04:12 Wrong software. Sep 24 03:04:32 there should be plenty of software for gps modules Sep 24 03:06:14 https://pastebin.com/xzKkFw1X Sep 24 03:06:58 okay? Sep 24 03:09:04 zmatt: I am sorry. I keep trying and failing. It is not your fault. I appreciate your help. Sep 24 03:10:09 you didn't ask any question about your pastebin link Sep 24 03:10:21 Please hold. Sep 24 03:11:49 Let me stick this module in place and see if this puppy works now. Sep 24 03:12:57 It works now. Sep 24 03:13:06 This is odd. That software did not work yesterday. Sep 24 03:13:25 I hope you don't mean to say you're hotplugging the module into a powered system, since that's a recipe for damaging it Sep 24 03:13:50 I am using UART2 like you told me to do. Sep 24 03:13:57 But not like you just mentioned, I lie. Sep 24 03:14:04 ? Sep 24 03:14:08 No excuse. I got excited. Sep 24 03:14:19 I hotplugged it! Sep 24 03:14:22 I did it. Sep 24 03:15:14 So zmatt: Why would that software work now? Is it b/c of the configure-pin cmd on the terminal? Sep 24 03:16:21 if the two pins were previously not configured for uart functionality, then yes Sep 24 03:16:29 But zmatt: You were right. I should have not hotplugged it at all. I got excited after I remembered that the configure-pin cmd works. Sep 24 03:16:38 ... Sep 24 03:16:55 That stuff you did with pearl is nice, though. Sep 24 03:17:35 I'm making a variant for the green wireless right now (with correct pin descriptions) Sep 24 03:17:49 zmatt: I know I cannot make up what I put you through this night. I am sorry. Sep 24 03:18:29 Thank you. Sep 24 03:20:10 though, the longer I look at the beaglebone green wireless, the more I'm amazed at how incompetent its design is Sep 24 03:21:47 lol, they're also driving 3.3v signals directly into 1.8v inputs (WLAN_EN and BT_EN) of the wireless chipset... nicely done Sep 24 03:23:21 Oops! Sep 24 03:24:18 Hey zmatt: Can you direct me to who has the power to change the Cloud9 Autorun program? Sep 24 03:24:40 I know absolutely nothing about cloud9 Sep 24 03:25:04 Okay...no issue. Sep 24 03:25:06 though I'm pretty sure you can configure autorun stuff yourself Sep 24 03:25:13 Oh? Sep 24 03:25:24 I will look into it. Sep 24 03:26:28 that stuff you wrote in perl and config-pin come in handy with our BBB. Sep 24 03:26:30 Thank you. Sep 24 03:27:01 a green-wireless branch should be on github fairly soon Sep 24 03:27:11 And to think I was complaining about the cellular tech. while I needed to be investing in GPS. Sep 24 03:27:12 Okay. Sep 24 03:27:15 Thank you. Sep 24 03:28:27 oh god the misdesign just goes on and on... /o\ Sep 24 03:29:05 Stick with USA, please BBB! Sep 24 03:29:24 ^ nearly impossible... Sep 24 03:29:30 ? Sep 24 03:30:08 zmatt: I thought the components were all outsourced for the BBGW? Sep 24 03:30:54 I'm not talking about components or production, I'm talking about the design Sep 24 03:31:00 Oh. Sep 24 03:31:07 basically seeed took the BBB schematic and made a bunch of changes Sep 24 03:31:45 and pretty much everything they touched was done in a stupid way Sep 24 03:31:45 Oh, I was unaware. Sep 24 03:32:54 Maybe, they have their reasons. For instance, they have ideas of math that cannot be interchanged into English. No interpretations and no translations. Sep 24 03:33:08 one of the consequences being that a lot of expansion header pins are reserved for wifi/bt (which is totally unnecessary, the beaglebone black wireless doesn't have that problem) Sep 24 03:33:19 Oh. Sep 24 03:33:22 how so? Sep 24 03:33:36 just really stupid usage of pins Sep 24 03:33:42 One chip? Sep 24 03:33:50 SoC? Sep 24 03:34:07 they took expansion header pins for wifi/bt while leaving other pins of the SoC completely unused Sep 24 03:34:19 Oh. Sep 24 03:34:26 even though the pins they left unused contain exactly the interfaces you need for the wifi/bt chipset Sep 24 03:35:00 They have this weird, online set-up for WiFi. It was never working, i.e. so I went with connmanctl cmd. Sep 24 03:35:47 I think you are the person that told me about the conmanctl first. Sep 24 03:36:28 ... Sep 24 03:36:28 that's software though and probably not seeed's fault Sep 24 03:36:46 Nope...someone wrote that stuff to be fun for children. I am pretty sure of it. Sep 24 03:37:04 ehm Sep 24 03:37:17 Everyone wants to have premade connections and easy adapters. Sep 24 03:38:37 For instance zmatt: I want a simple software to run. When this simple software does not run, I cannot move on until I get this software to run. I would like to learn more but without basics working, I am stuck doing repetitive moves on the BBB. Sep 24 03:39:19 A new book would be nice, i.e. with extended text. Sep 24 03:40:20 For example: That old guy in the office that talks slow and has been working day in and day out. That guy probably knows a bit too much and must have some way to have a release (a book). Sep 24 03:40:43 ... Sep 24 03:41:24 the problem with books is that they're always outdated Sep 24 03:42:47 Yep. What if someone recorded what and when and then info. informed people could write text? Sep 24 03:42:58 Yep...sounds honesdt. Sep 24 03:43:25 I have absolutely no idea what it is you're trying to communicate to me Sep 24 03:43:34 Write a book! Sep 24 03:43:56 I proof read. Sep 24 03:44:07 <<<< tester set_! Sep 24 03:44:27 I absolutely suck at writing docs Sep 24 03:44:30 Fine. Sep 24 03:44:51 Me too! I bet we need a doc writing class on the BBB! Sep 24 03:45:40 also, writing a book about the bbb seems like a soul-crushing activity since the audience is relatively limited and its content likely to be obsolete by the time it gets published Sep 24 03:45:51 For instance...I go to the BBB.io online and to the Google + page at random. I see key note speakers online talking about the BBB. Sep 24 03:46:23 Okay...fine. I just know that I really enjoyed all those people who produced the books that I got to read. Sep 24 03:46:30 There is so much more, though. Sep 24 03:47:08 Some of those classes are expensive and people are cheap-skates. They want now and cheap. Sep 24 03:47:51 I looked online at LinuxFoundation and they have classes. Some are exciting and relevant. Others, not so much. I want more embedded classes to see and learn from. Sep 24 03:48:03 ... Sep 24 03:51:41 I pressured you. I apologize. Enough with my banter. Sep 24 03:51:48 I am on another break. Sep 24 03:52:04 at no time did I feel pressured Sep 24 03:52:16 confused maybe, but not pressured Sep 24 03:54:04 I just pushed a green-wireless branch -> https://github.com/mvduin/bbb-pin-utils/tree/green-wireless Sep 24 03:54:26 haven't really tested it or checked the pins very thoroughly, but it's probably not too wrong Sep 24 03:55:19 but I'm afraid that if I stare at this idiocy much longer my brain would start to dribble out of my ear Sep 24 03:55:30 I lied again. I am back and I never really left. Sep 24 03:55:31 Ha! Sep 24 03:55:42 Oh...Okay. Thank you zmatt. Sep 24 03:55:44 so it'll have to do Sep 24 03:56:33 For now! Ha, Ha, Ha, Ha! Sep 24 03:58:49 there, patched the README too Sep 24 04:01:38 Cool beans, mate! Sep 24 04:01:46 I will check it out...NOW! Sep 24 04:30:44 is config-pin just for one go and I am guessing that it does not stick after a reboot? Sep 24 04:50:13 Well, that damn GPS tinker toy works. Now, onto better and more robust "thangs." I owe it all to that person (what was the name). Oh yea, zmatt! Sep 24 04:58:43 https://www.seeed.cc/BBGW-starter-tutorial%235-Where-are-you%3F-p-1648.html is where I got that darn info. on the GPRS instead of GPS. Sep 24 05:56:25 config-pin is indeed not persistent Sep 24 05:57:24 lol, yeah, nice that they never bothered to test their tutorial code Sep 24 05:57:41 lesson learned: don't assume seeed is competent Sep 24 07:24:12 Hi All, according to this : https://github.com/RobertCNelson/bb-kernel/issues/49 usb gadget serial isn't supported with windows 7 without proper inf dile, doesn anyone has a clue on how to make one ? Sep 24 07:38:55 cyberguy_: https://www.kernel.org/doc/Documentation/usb/linux-cdc-acm.inf Sep 24 07:58:26 zmatt : how do i use it? i need to add the USB PID/VID ? Sep 24 07:58:50 zmatt: how do i know what is the PID, VID? Sep 24 08:02:04 given that this inf file is included with the linux kernel source tree, I'd assume those are already set correctly Sep 24 08:15:23 zmatt: s you mean it is already there? i just need to pick the inf file in device manager and put it as the CDC Serial updated driver? Sep 24 08:16:16 that would be the obvious thing to try yes Sep 24 08:16:56 too obvious, can not do, must ask further questions instead... Sep 24 08:28:47 zmatt :doesn't work : windows was unable to install CDC Serial ... Windows could not find driver software for your device.... Sep 24 08:29:31 zmatt: i also tried to do so with the Gadget Seial COM entry , but it says that Windows detecmined the driver software is up to date Sep 24 08:31:16 zmatt : btw, the date for the driver is from 2007 .. are you sure they put on the BBB VID/PID in there...?!?!? Sep 24 08:36:45 the BBB uses a non-standard vid/pid ? Sep 24 08:38:03 note: I don't actually know how any of this insane Windows stupidify works, I just googled to find the inf file Sep 24 08:47:53 zmatt: I'm too lazy to get my BBB and boot it with a recent image, but apparently *something* changed, probably that thing doesn't apply anymore. Sep 24 08:50:18 zmatt :each vendor has it's own PID/VID i think, so am335x should have one Sep 24 08:52:30 iirc normally the g_serial gadget uses a vid/pid allocated to the linux foundation Sep 24 08:53:38 but, like I said, I don't actually know how the windows inf file works, nor do I actually know how current images configure the usb gadget since I don't use any of that Sep 24 08:59:20 zmatt: how do i find out how to make this works... .. ? Sep 24 08:59:36 try magic Sep 24 09:01:09 zmatt : magic! Sep 24 09:01:10 !?1 Sep 24 09:04:31 zmatt :what do you mean..? Sep 24 09:54:05 https://gist.github.com/dm8tbr/252927df3b5c8f559e1638fce0f625c3 - bone-debian-8.9-iot-armhf-2017-09-01-4gb.img as seen over USB Sep 24 09:54:49 this still uses CDC-acm, no idea why that would be a problem... Sep 24 10:03:33 on windows 10 everything except something called CDC-ECM works fine. For the latter there is no driver, but the rndis network works just fine Sep 24 10:04:12 I guess that might be a secondary network inferface to improve OSX compatibility Sep 24 10:15:43 tbr: you mean this version should work for me?!? Sep 24 10:18:50 cyberguy_: what I'm saying is that it should work just the same as previous images. At least with my limited understanding Sep 24 12:57:51 Up, Up and OTay! Sep 24 13:03:18 tbr: i tried debian-8.9-iot-armhf same cdc serial annoying issue Sep 24 13:11:15 cyberguy_: did this work before? Sep 24 13:11:31 only with version <7.9 Sep 24 13:12:17 tbr: RCN wrote me they changed the gadget serial machnism so you won't need a special driver (and it works with windows 8/10) but with windows 7 you need to create inf file . Sep 24 13:13:35 tbr: see here https://github.com/RobertCNelson/bb-kernel/issues/49 Sep 24 13:15:11 cyberguy_: my recommendation is to wait until RCN gets that figured out Sep 24 13:15:57 as nobody in here is good at windows things, don't expect help here Sep 24 13:16:35 aside from that I'd use a proper USB serial device to plug into the debug port anyway Sep 24 13:29:07 any recommandation to someone who is fimilar with that Sep 24 15:21:14 tbr: do you know how can i get the VID/PID of the BBB ? Sep 24 15:23:25 "lsusb" on linux will do that Sep 25 00:59:21 Is it still required to activate the SPIDEV0 virtual cape at boot or can it be done at runtime with an echo to the capemgr with the 4.4 kernel? Sep 25 01:12:11 Yep... Sep 25 01:13:00 I got that...did I get that? Yep. I set up a service on my BBGW to run what you, zmatt, taught me about those config-pin cmd. Sep 25 01:13:20 Now... Sep 25 01:13:45 I can run my software without having to enable pins each time. Sep 25 01:15:51 Cool beans! Sep 25 01:30:01 ahh I see now I should be using config-pin instead. Sep 25 01:30:40 What are you doing? Sep 25 01:30:56 trying to reflash an spi eeprom via flashrom Sep 25 01:31:05 Oh. Sep 25 01:31:18 Dang...I know nothing about that stuff. Sep 25 01:31:29 Sorry. I hope things go well. Sep 25 01:32:34 well show-pins at least says the pins are configed for spi so now I guess it is just a matter of figuring out if my issue is the wiring or something on the BBB is still configured incorrectly Sep 25 01:33:41 Pastebin? Sep 25 01:35:37 Jonimus: use config-pin to select pin function Sep 25 01:36:12 zmatt: yeah I did that Sep 25 01:36:16 ok :) Sep 25 01:36:20 https://github.com/adafruit/adafruit-beaglebone-io-python/blob/master/doc/run_config-pin_during_startup.md to set up services. Sep 25 01:36:31 SPI0 should be /dev/spidev1.0 correct? Sep 25 01:36:56 no, 0.0 Sep 25 01:37:50 well there is only /dev/spidev1.0 and /dev/spidev2.0 so I'm not sure what else it'd be? Sep 25 01:38:06 oh, 1-based? weird Sep 25 01:39:08 but I can't find anything that actually confirms that or would show me that Sep 25 01:39:17 oh right, I patched that in my kernel build Sep 25 01:40:54 I thought rcn merged that patch though Sep 25 01:40:58 but maybe not in 4.4 Sep 25 01:42:11 yeah bingo, he applied my patches in the 4.9 series Sep 25 01:42:52 zmatt: is there a way to verify I have the bone setup correct so I can say my issue is for sure the wireing to the chip? Sep 25 01:44:18 so on a 4.9 kernel spi0 would be spidev0. on a 4.4 kernel spi devices are numbered 1-based sequentially in the order the kernel probes them... which is hopefully { spi0, spi1 } but I'm not even sure if that's guaranteed :) Sep 25 01:44:58 can you pastebin the relevant output lines from show-pins ? Sep 25 01:46:29 (with "devices" I meant buses actually... the devices are named spidev$bus.$cs ) Sep 25 01:47:05 spidev1.0, 1.1, 2.0, 2.1! Sep 25 01:47:07 https://gist.github.com/Jonimoose/e2600fa40288927cc31107f6ef39628c Sep 25 01:47:22 But that is from the 4.9.x. Sep 25 01:48:55 set_: huh, you have spidev1/2 instead of 0/1 on a 4.9 kernel? Sep 25 01:49:18 Yep! Sep 25 01:49:23 that is odd, the patches are most definitely in rcn's 4.9 tree Sep 25 01:49:44 ohh, wait, that's his 4.9-bone tree... maybe he didn't apply them to 4.9-ti ... that would be confusing though Sep 25 01:50:16 Jonimus: yeah, looks okay to me Sep 25 01:50:59 of course that doesn't mean the issue is necessarily in wiring, it could be in how software configures the spi port Sep 25 01:51:08 (e.g. spi mode) Sep 25 01:51:21 How many modes do we have? Sep 25 01:51:50 4 Sep 25 01:52:06 Damn Lucy! That is one, two, three, four! Sep 25 01:52:14 set_: maybe you're running an *old* 4.9-ti kernel ? Sep 25 01:52:26 I am...I have a July kernel. Sep 25 01:52:51 I can update but I do not want to right now. Sep 25 01:53:02 the patches seem to have been included in 4.9-ti 11 months ago Sep 25 01:54:14 Damn! I wonder. Oh, I got these images from http://elinux.org/Beagleboard:BeagleBoneBlack_Debian#2017-09-17_-_Debian_9_.28Stretch.29_-_Weekly. Sep 25 01:54:22 weeird Sep 25 01:54:36 Hmmm...well. I guess I can update later. Sep 25 01:54:41 well I think I'm gonna have to wait for my chipclip to get here because the odds of my wireing hell being the issue are fairly likely. Sep 25 01:55:04 https://twitter.com/Jonimus/status/912129887112171521 for pics Sep 25 01:55:04 set_: an image from this year should definitely include this 11-month old patch Sep 25 01:55:24 I know. Let me check to make sure. Sep 25 01:55:56 BeagleBoard.org Debian Image 2017-07-02 Sep 25 01:56:31 Jonimus: you're trying to access a flash chip in-situ ? that feels like a recipe for trouble Sep 25 01:57:17 <<< on break from being a non-smoker. Sep 25 01:59:09 zmatt: yeah, but that header is explicitly labled JSPI so I assume that is its intended purpose. Sep 25 01:59:31 I just don't have the proper conenector hence the hook clips. Sep 25 02:00:39 fair enough... I just wonder how they deal with that exactly w.r.t. power and avoiding drive conflict Sep 25 02:01:50 spi isn't exactly a multimaster capable bus Sep 25 02:03:53 well the board is completley unpowerred and power is supposed to be provided via the header, I have another atx powersupply hot wired to give me the 3.3v and supposedly this should work Sep 25 02:04:48 if not I do have more correct wireing on the way and it should be here tomorrow assuming everything goes as planned. Sep 25 02:05:28 If I can't get the BBB working I have a bus pirate comming but I'd rather be able to say I've used this Beagle for at least one thing since I purchased it years ago. ;) Sep 25 02:06:10 be very very careful when using a 3.3v supply that's not the one from the bbb... having the chip is connected to the BBB while one is powered and the other is not, there is risk of damaging hardware Sep 25 02:08:52 good to know, from what I've seen though the BBB's 3.3v rail doesn't cut it for this situation. Sep 25 02:09:12 that suggests a lot more than the flash chip is being powered Sep 25 02:10:12 it might also be a good idea to use 100 ohm resistors or so for the spi lines instead of wires, to limit current in case of any drive conflict Sep 25 02:11:39 kk Sep 25 02:14:59 Do all of the smaller LEDs have a 2V current draw and if not, how would I be able to tell? Sep 25 02:17:07 I am asking because this book says that red LEDs run 1.8v and clear white runs 2.4v. Sep 25 02:17:27 voltage drop varies somewhat per led Sep 25 02:17:35 note, voltage drop != current draw Sep 25 02:18:10 the voltage drop also varies somewhat depending on the current (but not remotely the linear relationship that a resistor has) Sep 25 02:19:41 Is voltage drop a term for applied light over time? Sep 25 02:20:05 so, this is the datasheet of some random led: http://www.cree.com/led-components/media/documents/C503B-RAS-RAN-AAS-AAN-RBS-RBN-ABS-ABN-RCS-RCN-ACS-ACN-1079.pdf if you look at the first diagram on page 8, it shows the relationship between voltage and current for two different leds Sep 25 02:20:37 Okay. Sep 25 02:21:09 these particular red leds evidently have a higher voltage drop (aka "forward voltage") than 1.8v Sep 25 02:21:43 more like 2.1v Sep 25 02:22:46 I was going to do some math on LEDs with the BBB. Sep 25 02:23:11 My LEDs do not have any mfg. name. Sep 25 02:24:02 basically you can think of a diode as a path where electrons fall down some ledge... since they can't get back up, current can only flow in one direction. the height of the drop is the voltage difference between the two terminals when current is flowing, and when the electrons fall down and smack into the floor, that energy gets turned into heat (regular diodes) and/or light (leds) Sep 25 02:25:48 I am listening to this book on the BBB, "The BeagleBone Black Primer." Sep 25 02:26:21 The book has a section dealing with LEDs, like most books on the subject, but this one is specific with more text. Sep 25 02:26:47 it has math, explanations, and ideas/theory. Sep 25 02:27:10 I think not knowing anything about my LEDs is an issue. Sep 25 02:27:17 Dang! Sep 25 02:28:21 Oh and zmatt: This book states that there is a 4mA supply on our BBBs. Sep 25 02:29:17 dealing with leds is pretty simple when efficiency isn't a concern: you choose how much current you want to be flowing through the led (this is directly proportional to light output), look up or guess the voltage drop across the led for that current, and then the series resistor value should be (supply voltage - led voltage drop)/current Sep 25 02:30:19 Okay. Sep 25 02:30:24 I think the digital outputs are specified at 4 mA. you should preferably avoid driving leds directly with gpios, they're not designed to handle much current Sep 25 02:31:57 Alright...I will keep on reading. Sep 25 02:32:05 the blue leds on the beaglebone for example all use a transistor to drive the led Sep 25 02:32:43 I have not gotten to transistors yet. Sep 25 02:33:10 (or actually they use some slightly weird transistor with integrated series resistors) Sep 25 02:33:37 Oh. **** ENDING LOGGING AT Mon Sep 25 03:00:00 2017