**** BEGIN LOGGING AT Sat Aug 10 03:00:09 2019 Aug 10 03:20:24 take notes! Aug 10 03:21:11 I just went in on #pocoo to discuss what could be the reason on my errors. It was the old bonescript.socket service running. Aug 10 03:21:14 I keep forgetting. Aug 10 03:25:17 Oh and the tmp36 sensor needs to have ---> celusius = (mV - 500) / 10 instead of just ---> celsius = mV / 10 when mV = ADC.read(Px.xx) * 1800. Aug 10 03:25:38 This was my mistake for years until today. Aug 10 03:26:41 Hell, I even changed programming languages for not figuring this key fact out. Aug 10 03:44:08 Blah! Aug 10 10:09:33 how to setup can module Aug 10 17:04:43 hi,everyone. i'm setting the can bus with p9.24 and p9.26, but it says not found Aug 10 17:04:48 https://pastebin.com/EQsRghxq Aug 10 17:05:17 are you booting from sd card or eMMC ? Aug 10 17:05:27 from sd card Aug 10 17:06:08 after the update it doesn't work Aug 10 17:07:24 flashing to eMMC is recommended, but if for whatever reason you don't want to, you should either erase eMMC (or at least the bootloader on it) or bypass the bootloader on eMMC by powering on with the S2 button held down (you can (and should) let go once the power led turns on. this bypass remains in effect across reboot/reset, it is only sampled at power-up) Aug 10 17:08:19 since by default the processor prefers using the bootloader on eMMC over the one on SD card. if you have an old bootloader on eMMC, it will usually still manage to load the system from sd card, but it won't do so quite right, resulting in weird problems like the one you have Aug 10 17:09:48 that means i should flasch it with "Debian 9.5 2018-10-07 4GB eMMC IoT Flasher "? Aug 10 17:10:51 yeah, or you can turn the sd card you already have into a flasher by uncommenting one line in /boot/uEnv.txt Aug 10 17:11:35 (I think it's the last line... it's something like "cmdline=..yada yada init=/opt/something-with-eMMC-flasher-or-whatever) Aug 10 17:11:54 and then reboot, and it'll proceed to wipe and reflash eMMC Aug 10 17:14:00 and after update i can't get access to PC with 192.168.7.2 Aug 10 17:14:17 but with putty is ok Aug 10 17:14:32 so what then do you mean with "access" web browser" ? Aug 10 17:15:04 yes, i mean with web browser Aug 10 17:15:30 odd, if you can connect to it via ssh you should also be able to connect to it using the web browser Aug 10 17:15:48 unless somehow the incorrect bootloader is causing problems with that as well, but I don't see how it could Aug 10 17:18:55 i will try the new flasch, see if it works on it,thanks a lot Aug 10 19:48:56 Are any ADC pins available on the BBGW? I will test but if you know, please let me know. I have tested the p9.40 pin so far only. Aug 10 19:50:02 Oh and if any ADC pins exist on the BBGW, what do I need to comment in the uEnv.txt file to alter my task at hand, e.g. eMMC or HDMI? Aug 10 19:53:35 Nevertheless...I had disable ADC uncommented out. This is probably the issue. Reboot! Aug 10 20:02:56 TMP36 = too hot to handle? Aug 10 20:03:11 poweroff! Aug 10 20:03:37 No smoke but my finger is burnt. Aug 10 20:09:37 uboot_overlay_adc does not work on the BBGW! Notice my finger! Aug 10 20:16:51 It was reading in at 200+ degrees. Yikes! Aug 10 20:17:06 GPIO = too hot! Aug 10 20:18:16 Or whatever that pin was that is nonconfigurable. Aug 10 20:21:01 Ut oh. Turning off now. Gosh. Aug 10 20:29:11 Off to the BBB for this sensor. Aug 10 20:58:20 Where is the ADC file system on our BBBs w/ the 4.19.x kernel? Aug 10 21:09:19 Is it under Indust. I/O? Aug 10 21:13:16 set_: ADC pins are exactly the same on the BBGW as they are on the BBB Aug 10 21:13:18 zero difference Aug 10 21:13:37 Oh. Aug 10 21:13:57 Okay. I received an error when attempting to config. those pins w/ config-pin. Aug 10 21:14:09 ADC pins are non-configurable Aug 10 21:14:12 Oh. Aug 10 21:14:14 Nice! Aug 10 21:14:17 they're always ADC pins, they have no other function Aug 10 21:14:22 I was trying to put in work. Aug 10 21:14:25 Okay. Thank you. Aug 10 21:15:00 Hey @zmatt: Where should I start to look for using a driver on the ADC subsystem for 4.19.x? Aug 10 21:15:05 (which also means there's normally no reason to disable them. the option only exists in /boot/uEnv.txt in case you want to enable a custom overlay that configures the ADC differently, e.g. as touchscreen controller) Aug 10 21:15:15 Oh. Aug 10 21:15:16 Okay. Aug 10 21:15:57 I looked online. It seems like nothing is "Popular" right now. Aug 10 21:16:04 I'm not aware of anything having changed w.r.t. the ADC in 4.19.x, but I also haven't really checked. I'm using 4.14 Aug 10 21:16:13 oh. Aug 10 21:16:14 Okay. Aug 10 21:16:28 I should try to revert if things are going to take too much time on my part. Aug 10 21:16:38 pretty sure 4.14 is also still the default on current images Aug 10 21:16:44 It is. Aug 10 21:17:04 I'd still expect the ADC to be an IIO device, like it was in 4.14 Aug 10 21:19:16 Okay. Aug 10 21:19:31 I found it but I was unsure as to how to alter things on the command line. Aug 10 21:19:42 it is an iio device. Aug 10 21:19:51 "alter things" ? Aug 10 21:20:04 Yea. Turn on something, i.e. TMP36 sensor. Aug 10 21:20:16 W/out programming anything in Python or whatever langauge. Aug 10 21:20:40 I will look through my books again. Aug 10 21:21:41 I have no clue what you mean, there's no "turning on" of anything involved... your sensor is always on Aug 10 21:22:23 Oh. Aug 10 21:22:32 (unless it's not connected) Aug 10 21:22:47 So, I can just go to /trigger and create a readout? Aug 10 21:23:23 you can just read the value via sysfs... I know there's more to the IIO interface but I don't really know anything about that Aug 10 21:23:30 no idea what trigger is Aug 10 21:24:16 Okay. No issue. I am reading in the kernel source and they seem to have very little on these devices. Odd days. Anyway, thank you. I will read about it all later. I need a break. Aug 10 21:24:50 https://www.kernel.org/doc/html/latest/iio/iio_configfs.html Aug 10 21:24:53 generally speaking you can just directly read the adc value from a sysfs attribute Aug 10 21:25:01 you definitely don't need anything in configfs Aug 10 21:25:25 Okay. Right. That is what i read. I am in the wrong file/dir. Aug 10 21:25:29 that iio has stuff that can be configured via configfs is complete news to me Aug 10 21:25:38 Oh. Aug 10 21:25:47 Yea. Me too. I am just learning more as time passes. Aug 10 21:26:17 I wonder if it is located on the 4.19 kernel. Off to look real quickly. Aug 10 21:26:34 is it not just in the same place as on 4.14 ? Aug 10 21:26:54 https://pastebin.com/gh9dKnj3 Aug 10 21:27:57 Nope. I think it is located somewhere else. Aug 10 21:27:59 Please hold. Aug 10 21:28:51 is it /sys/bus/iio? Aug 10 21:29:16 There is no platform. Aug 10 21:29:27 there's more than one way to reach it Aug 10 21:29:38 I have devices drivers and some other files. Aug 10 21:29:43 there will 100% absolutely be platform Aug 10 21:30:21 under /sys/class or /sys/bus = no Aug 10 21:30:35 oops. Aug 10 21:30:37 I see it now. Aug 10 21:30:52 Oh. Aug 10 21:31:29 have you tried just literally trying the commands from that paste? Aug 10 21:31:43 No, not yet. Aug 10 21:32:27 Let me go and try it. Aug 10 21:33:07 no such file or dir. Aug 10 21:33:18 Both lines as one? Aug 10 21:33:33 ?? Aug 10 21:33:59 I tried the first line. "no such file or directory" was the error. Aug 10 21:34:30 I'm installing the latest 4.19-ti series kernel currently to check Aug 10 21:34:48 Okay. Aug 10 21:34:50 brb Aug 10 21:35:03 I am literally walking outdoors to check something. Aug 10 21:35:07 brb Aug 10 21:39:44 4.19.31-ti-r19 is my kernel. Aug 10 21:40:00 ok, I had hoped that sysfs path would be somewhat stable, but I guess not Aug 10 21:40:17 hardcoding sysfs paths is a bad idea in general anyway, for the most part they have no guarantee of stability across kernel versions Aug 10 21:40:30 Understood. Aug 10 21:40:44 Things change, people have other ideas. Aug 10 21:41:18 cd /sys/bus/platform/devices/*adc*/iio:* Aug 10 21:41:24 that at least works :P Aug 10 21:41:32 Okay. Aug 10 21:41:39 Let me go and check. Aug 10 21:41:48 of course that's a pretty disgusting way to locate a device in sysfs Aug 10 21:42:29 I cannot get into any adc stuff at all under devices. Aug 10 21:43:03 Oh. Aug 10 21:43:08 I got it now. Aug 10 21:43:23 aw Aug 10 21:44:55 Do I use uevent? Aug 10 21:45:30 I have a TMP36 sensor on my P9.40 pin. Aug 10 21:46:05 you just read the value from one of the in_voltage*_raw files (there's one for each channel) Aug 10 21:46:48 Okay. I am new to this idea. Seriously. so, would this be okay? read in_voltage7_raw? Aug 10 21:47:25 I have no idea what you're asking. the concept of reading a file is new to you? Aug 10 21:47:36 On the command line, yes. Aug 10 21:47:54 I do not think I have ever read about reading files on the command line. Aug 10 21:48:05 I will go and look it up. Aug 10 21:48:08 read it to what? if you just want to see its contents you'd generally use "cat filename" Aug 10 21:48:18 Oh. Aug 10 21:48:19 Okay. Aug 10 21:48:22 cat Aug 10 21:48:41 note that in_voltage7_raw is not one of the 7 adc inputs on the expansion header, it's an internal one, hence unlikely to be of any use to you Aug 10 21:48:51 Got it. Aug 10 21:49:05 P9.40 has a specific adc input or not? Aug 10 21:50:04 you can find an overview of the beaglebone's expansion header pins in my pins spreadsheet ( https://goo.gl/Jkcg0w .. the P9 tab would be more useful in this case) and numerous other places on the internet Aug 10 21:50:46 given how long you've been using beaglebones now, I really feel like that's not a question you should still need to be asking Aug 10 21:51:50 I did not understand the comparison of header pins and file systems well enough at the beginning and then I jumped into Python for testing. My bad. Aug 10 21:52:19 note btw that the value in these files is the raw measurement (as indicated by the filename)... they're a number in range 0..4095 corresponding to 0V .. 1.8V Aug 10 21:52:35 Right. That is what I noticed. Aug 10 21:52:50 math time? Aug 10 21:53:54 echo $(( $(cat in_voltage7_raw) * 1800 / 4095 )) Aug 10 21:54:01 that would show the result in milivolt Aug 10 21:54:17 having said that, doing math in bash scripting quickly gets pretty ugly Aug 10 21:54:25 That is a nice google doc. there. Aug 10 21:54:27 Okay. Aug 10 21:54:53 have you seriously never seen it before? I link to it like all the time here Aug 10 21:55:15 I think I may have reviewed it but not for any other reason outside of previewing. Aug 10 21:55:34 Did you make that google doc? Aug 10 21:55:41 Time! Aug 10 21:55:47 yes Aug 10 21:55:50 Nice! Aug 10 21:56:08 Can I use it now? Aug 10 21:56:20 It shows which ball and everything!@ Aug 10 21:56:52 I very much doubt the ball is of any interest to you :P Aug 10 21:56:59 I should have stuck w/ C/C++. Python libraries are falling apart everywhere. Aug 10 21:57:02 also i recommended the P9 tab, which definitely doesn't show balls Aug 10 21:57:09 No Balls! Aug 10 21:57:17 what python libraries? Aug 10 21:57:26 BBIO and PyBBIO. Aug 10 21:57:41 I do not think they work w/ the newest kernels. Aug 10 21:57:56 from what I've seen, adafruit's python library suck pretty hard Aug 10 21:58:02 I cannot even use pip3 to install their libraries any longer. Aug 10 21:58:49 Yea but i have not found other libraries. W/ my inexperience of making libraries, I am up "Poop's" creek. Aug 10 21:58:49 they also seem mostly redundant... for the most part they don't do anything you can't already do with more generic libraries (or often none is needed at all, other than what's included with python already) Aug 10 21:59:09 I will consider learning more, I guess. Aug 10 21:59:45 I have been more of a user, i.e. outside of understanding how to program chips in C/C++ w/ help. Aug 10 22:00:29 @zmatt: I do not know as much as you give me credit of holding. Aug 10 22:00:40 <---- learning still! Aug 10 22:01:06 don't worry, I have very little expectations of you Aug 10 22:01:13 Thank you! Aug 10 22:01:38 apart from enthausism... you seem to have a limitless supply of that Aug 10 22:02:04 I will keep trying but outside of school, I have no real studies taking place. Oh and I am done w/ school, too. Aug 10 22:02:14 No! Aug 10 22:02:25 I will try something one day. No rush. Aug 10 22:06:48 But... Aug 10 22:06:53 Got to go. gtg? Aug 10 22:07:15 Thank you again. No matter how much you seem to have no affect on me, you are helping me tend to move in specific directions. Aug 11 00:30:18 @zmatt: I tried that command you gave earlier ---> echo $ ((cat in_voltage0_raw * 1800 / 4095)) Aug 11 00:30:49 Something about a missing parenthesis or one extra too many. Aug 11 00:36:17 -bash: syntax error near unexpected token '(' Aug 11 00:36:20 That is the error. Aug 11 00:39:36 Do not fret. I will try some other ideas before going into this excursion of ideas. Please forget I ever said anything about the adc subsystem "stuff." Aug 11 00:42:31 4.14.x, here I come! Aug 11 02:01:58 I got that damn TMP36 working finally. Did that take all day, yes! Does it finally work on my console w/ the correct temp, yes! Aug 11 02:28:08 Has anyone used matplotlib on the AI yet (for testing)? **** ENDING LOGGING AT Sun Aug 11 02:59:57 2019