**** BEGIN LOGGING AT Sun Jun 07 02:59:56 2020 Jun 07 03:02:30 The portion of ram 0x402F0000 0x402F03FF, is reserved for some particular use? Jun 07 03:35:51 that linker script is gold for learning, e.g. HIDDEN( __init_array_start = . ); Jun 07 08:22:08 what happens when the 7th seal opens? Jun 07 09:33:24 Hi all, Jun 07 09:33:57 I've just bought the beaglebone black wireless and having some troubles with it becomming unavailble when leaving it turn on for 3+ hours Jun 07 09:36:23 Do anyone know if it have some sort of sleep mode enabled or something? I experience that when it's working USR0 is blinking heartbeat and USR2 is blinking a bit, when it is working I can ping it at 192.168.1.133. When it fails USR0 and USR2 are permanently on and i cannot ping it anymore. My router doesn't see the device as connected to the Jun 07 09:36:24 wireless, eventhough WL led is on. Do anyone know how i can debug this ? Jun 07 09:42:20 Munke: the best way is to attach to it's debug UART port Jun 07 09:42:37 also it will depend on what distribution you are running on the device Jun 07 09:51:02 Ok, I'm running Debian Buster IoT Image 2020-04-06, which I have manually slimmed by removing bonescript and many other of the standard applications installed. Jun 07 09:52:13 @tbr The debugging through UART port, i have a raspberry pi, might that work as a debugger? And how would I go forward with this, do you have some reference page for debugging ? Jun 07 09:54:06 @tbr essentially you just want to see if I can log in through serial, or what have you thought about ? Jun 07 10:01:29 @Munke Did you observe this everytime ? and is it happening exactly at the same inervals? Jun 07 10:03:48 I don't know exactly how long, all i know is that I can leave it running, come home and it is no longer "working" Jun 07 10:04:01 I will then press the reset button and it will start up Jun 07 10:04:19 it should be as easy as interlinking ground and crossing RX-TX TX-RX Jun 07 10:04:19 To debug I've enabled persistent journalctl, which show the following: Jun 07 10:04:51 I'd expect the image to have kernel output enabled on UART, if not you'd need to turn that on Jun 07 10:04:55 Jun 06 14:57:44 beaglebone systemd[1]: Started Cleanup of Temporary Directories.-- Subject: A start job for unit systemd-tmpfiles-clean.service has finished successfully-- Defined-By: systemd-- Support: https://www.debian.org/support-- -- A start job for unit systemd-tmpfiles-clean.service has finished successfully.-- -- The job identifier is Jun 07 10:04:55 407.Jun 06 15:17:01 beaglebone CRON[1747]: pam_unix(cron:session): session opened for user root by (uid=0)Jun 06 15:17:01 beaglebone CRON[1748]: (root) CMD ( cd / && run-parts --report /etc/cron.hourly)Jun 06 15:17:01 beaglebone CRON[1747]: pam_unix(cron:session): session closed for user rootJun 06 15:35:57 beaglebone wpa_supplicant[811]: wlan0: Jun 07 10:04:56 WPA: Group rekeying completed with 48:f8:b3:b9:85:1f [GTK=CCMP]Jun 06 16:17:01 beaglebone CRON[1772]: pam_unix(cron:session): session opened for user root by (uid=0)Jun 06 16:17:01 beaglebone CRON[1773]: (root) CMD ( cd / && run-parts --report /etc/cron.hourly)Jun 06 16:17:01 beaglebone CRON[1772]: pam_unix(cron:session): session closed for user Jun 07 10:04:56 rootJun 06 16:35:54 beaglebone wpa_supplicant[811]: wlan0: WPA: Group rekeying completed with 48:f8:b3:b9:85:1f [GTK=CCMP]Jun 06 17:17:01 beaglebone CRON[1798]: pam_unix(cron:session): session opened for user root by (uid=0)Jun 06 17:17:01 beaglebone CRON[1799]: (root) CMD ( cd / && run-parts --report /etc/cron.hourly)Jun 06 17:17:01 beaglebone Jun 07 10:04:57 CRON[1798]: pam_unix(cron:session): session closed for user rootJun 06 17:35:52 beaglebone wpa_supplicant[811]: wlan0: WPA: Group rekeying completed with 48:f8:b3:b9:85:1f [GTK=CCMP]-- Reboot --Jun 07 09:28:13 beaglebone kernel: Booting Linux on physical CPU 0x0Jun 07 09:28:13 beaglebone kernel: Linux version 4.19.94-ti-r43 Jun 07 10:04:57 (voodoo@w8-imx6q-wandboard-2gb) (gcc version 8.3.0 (Debian 8.3.0-6)) #1buster SMP PREEMP Jun 07 10:06:00 please don't paste multi-line output here, use a pastebin / gist / snippet and give us the URL Jun 07 10:06:58 tbr Ok, will remember that, sorry. I will try what you suggested and see what happens. But do you have any suggestions as to what you expect is happening ? Jun 07 10:07:09 @Munke can you look at /etc/systemd/sleep.conf ? Jun 07 10:07:45 aswin all is commented out Jun 07 10:08:53 @Muke Okay! Then try the log method what tbr suggesting Jun 07 10:10:15 tbr aswin Thank you, I'll set it up and get back to you. Thank you for your time, but now I'll need to wait some hours before it will happen again :D Jun 07 10:10:29 a USB-UART adapter is always a very handy tool to have when working with embedded devices Jun 07 10:11:05 I have a drawer full of them. I bought some extremely simple ones for <1$/€ a piece and just attach them permanently to some projects Jun 07 10:11:13 yea, we had them at school, but now that I'm done I find myself missing some tools :) Jun 07 10:11:13 @Muke What exactly your BBB is performing in your setup? Jun 07 10:11:40 @as Jun 07 10:11:54 aswin it is doing nothing Jun 07 10:12:03 currently just turned on Jun 07 10:12:49 aswin it seemed to me, to be an out of memory error, so I added some swap and installed everything running but it still happens. Jun 07 10:12:50 Get cp2303 or FTDI232 based USB-Serial bridges Jun 07 10:13:05 I think cp2303 is cheaper Jun 07 10:13:13 but FTDI is better ? Jun 07 10:13:52 FTDI is good as a primary tool, but the cheap ones usually work just fine Jun 07 10:14:03 (if you get a genuine FTDI!) Jun 07 10:14:04 I dont think that cp2303 is bad, I dont know about that much in detail. Jun 07 10:14:34 @tbr yeh thats the problem. market is flooded with chinese copys Jun 07 10:15:02 you'll need to do fairly demanding things to hit problems. Longer wires, very high baud rates, RF-noisy/EMF environments, etc Jun 07 10:15:04 ok, I'd rather purchase whatever works stable for a long time. Original FTDI would be the better choice then ? Jun 07 10:15:23 CP2303 is from Silicon Labs and they are trusted manufacture Jun 07 10:16:57 get one proper FTDI and a bag of these: https://www.aliexpress.com/item/32903090600.html Jun 07 10:18:00 @Munke For now, If you have an arduino that will do the job for you :) Jun 07 10:19:57 Cool, thanks :) Jun 07 10:21:28 oh, one thing is the BBB 5V logic i serial communications ? Jun 07 10:22:07 it uses 3.3V Vref, but the debug UART *should* be 5V tolerant Jun 07 10:22:14 cause the pi is 3.3V, so i might need to make some protection setup for it Jun 07 10:22:18 oh then it is fine Jun 07 10:23:10 if you run into issues with serial, remember to do "loop tests" on both sides, by just shorting RX to TX Jun 07 10:29:25 I'm not familiar with that, what should happen when I short them ? Jun 07 10:30:06 @Munke You will be sending signals to the same UART Jun 07 10:30:28 btw, https://www.amazon.co.uk/DSD-TECH-SH-U09C5-Converter-Support/dp/B07WX2DSVB/ref=sr_1_16?dchild=1&keywords=FTDI232&qid=1591525606&s=electronics&sr=1-16 <-- would that be a good choice for the FDTI solution ? Jun 07 10:30:58 of cause :D .. so see if i send an 'a' I would get that back Jun 07 10:31:09 Yes Jun 07 10:31:22 makes sense :) Jun 07 10:31:23 kinda echo Jun 07 11:12:37 tbr: the bbb's serial console port is 5V tolerant yeah (thanks to an isolation buffer it has), but 5V usb-serial converters may not be able to properly receive the 3.3V signal from the bbb Jun 07 11:17:15 CoffeeBreakfast: 0x402F0000-0x402F03FF is reserved for secure world and access to it is blocked by a firewall. I'm pretty sure it is not actually used (on GP devices), but I think the size of the firewalled region could not be set to zero (because it is configured with an inclusive endpoint rather than an exclusive endpoint) Jun 07 11:24:36 amusingly the secure-world MMU translation table is located outside this protected region, though it is normally never accessed because the only entry ever needed once public-world execution starts (the section covering secrom) is locked into iTLB Jun 07 11:26:37 (you can unlock and manipulate it, but you'll just make the SSM angry and it will trigger a SoC reset) Jun 07 11:56:40 aswin tbr Now I'm setup, but i struggle to get login screen connecting the raspberry pi to BBBWs debug serial port. I've made loop test on both sides, ensuring that they are working. BBBW does however fail at this sometimes, where i won't get anything returned. I'm using screen like this: screen /dev/ttyO0 115200screen /dev/ttyO0 115200 Jun 07 11:56:56 oh double pasted.. screen /dev/ttyO0 115200 Jun 07 11:58:23 I think one thing to beware of is that when agetty detects a break (e.g. disconnecting and reconnecting the cable) it will cycle through its list of baudrates Jun 07 12:00:11 nevermind, I saw a video where he had tagged the pins as you should connect them, not what they were... Jun 07 13:01:20 CoffeeBreakfast: "your asm-demo works.. but some of StarterWare examples don't.." ... it's entirely possible that's due to StarterWare being a steaming pile of pig manure Jun 07 13:09:02 also known as "enterprise grade" Jun 07 13:09:12 CoffeeBreakfast: for example, its uart driver is really bad and many of its functions (including UARTIntEnable/UARTIntDisable) will cause corruption if called while a byte is being transmitted or received, as a result of pointlessly setting and restoring EFR.4 instead of just setting it once during initialization Jun 07 13:10:18 and the uartEcho example will compile into a deadloop if compiled with any level of optimization other than -O0 Jun 07 15:15:35 A "hello world" code worked now (Using StarterWare).. with this .lds https://pastebin.com/m0EANMKf (it doesn't mean that script is entirely right) Jun 07 15:19:22 does starterwave not include its own linker scripts? Jun 07 15:19:27 *starterware Jun 07 15:23:53 this is the linker script I used for my own baremetal C++ experiments: https://pastebin.com/raw/7KZrR2ja Jun 07 15:24:52 (it is also an option to treat a8ram and ocmc as a single contiguous memory segment) Jun 07 15:26:58 (also, it is not meant to be complete... I just added stuff as I needed them) Jun 07 15:27:43 One that tries to load the image to 0x8000000 (with the help of U-Boot?), but peripheral boot loads it to 0x402F04000. Jun 07 15:28:16 ah, they assume you're using u-boot Jun 07 15:28:25 which makes sense I guess Jun 07 15:49:22 what's the difference between `} >a8ram :image` and `} :image` ? Jun 07 15:52:28 that the latter doesn't specify to which memory it should go (which matters when you have two memories defined, as this linker script does) Jun 07 15:53:29 it would make more sense if you could do so for the entire segment ("image"), but you can't Jun 07 16:07:33 PHDRS is not necessary when using .bin instead of .elf ? Jun 07 19:44:50 ERROR: open() for /sys/devices/platform/ocp/ocp:P9_12_pinmux/state failed, No such file or directory <<< How can I overcome this idea on my BBGW w/ config-pin? Jun 07 19:45:49 I tried echo 60 > /sys/class/gpio/export Jun 07 19:46:12 But...there gave me the error. Jun 07 19:46:41 I guess I can try another pin. Forget it. Jun 07 21:39:56 P9.12 is occupied on the BBGW and therefore can't be reconfigured: https://pastebin.com/Fi8Vh0n5 Jun 07 21:42:47 Thank you. Jun 07 21:43:12 I used another GPIO. Jun 07 21:43:27 But, thank you for that paste. Jun 07 21:43:51 I keep forgetting about the GPIO pin usage on the BBGW compared to the BBB/BBBW. Jun 07 21:44:20 this is something you should have known, I've mentioned often enough that the BBGW has a bunch of pins occupied for wifi/bt and I think I might actually have made this paste originally for you. also, you could have used (the green-wireless branch of) show-pins to check what's up with that pin Jun 07 21:44:43 Okay. Jun 07 21:45:22 I figured it was P8 pins that were used. I am going to keep that paste if that is okay? Jun 07 21:45:34 Oh. Jun 07 21:45:43 show-pins works on the BBGW too. Nice. Jun 07 21:46:30 there's even a variant whose labels are updated for the green-wireless: https://github.com/mvduin/bbb-pin-utils/tree/green-wireless#show-pins Jun 07 21:46:57 @zmatt: Could I just cancel the Bluetooth stuff for getting that specific pin for use? Jun 07 21:47:31 I see the branches now. Jun 07 21:47:33 no. it might be freed up if you disable wifi+bluetooth entirely Jun 07 21:47:37 You have many. Jun 07 21:47:38 Oh. Jun 07 21:47:39 Okay. Jun 07 21:47:51 (but then what's the point of using the BBGW in the first place) Jun 07 21:47:58 I am going to leave the Wifi and Bluetooth alone then. Jun 07 21:48:01 Right. Jun 07 21:48:45 I am learning about Python again, matey. I am watching some edX videos on Recursion and Iteration (again). Jun 07 21:49:05 This way, I can show you something neat (finally). Jun 07 21:49:23 os, re, intertools, and so on... Jun 07 21:50:20 I am still trying to learn all about the import modules from Python (built-in). Jun 07 21:50:48 as you say, "Pure Python." Jun 07 21:51:41 trying to learn about those topics sounds pointless at your current level of understanding. any time I've seen you attempt to write or modify python code it was clear you had not even remotely any sort of mental model of what was going on Jun 07 21:51:56 Right! Jun 07 21:52:07 But, things in time take time. Jun 07 21:52:20 It is not easy to just jump on board and have all ideas. Jun 07 21:52:36 It takes a while. You know this, man. Jun 07 21:53:27 sure, until you have a basic understanding of programming it's probably a waste of time to watch some tedX videos "on Recursion and Iteration (again)" Jun 07 21:53:58 Fine. If you say so. Do you recommend a book or some other ideas of understanding? Jun 07 21:54:26 you've asked that before, you know I don't have any Jun 07 21:54:38 Okay. No issue. **** ENDING LOGGING AT Mon Jun 08 02:59:58 2020