**** BEGIN LOGGING AT Mon Jan 29 03:00:02 2018 **** BEGIN LOGGING AT Mon Jan 29 03:03:43 2018 Jan 29 10:43:24 hi all Jan 29 10:44:16 * uw_fluxus_lurker waves. Jan 29 10:55:50 does anyonw know where fo tind a manual on how to alter images? (unpack, change a file, repack) I have a documented approach but that results in an image that doesn't boot. Jan 29 11:09:21 With regard to beagle-images, I have no clue. If it's a linux image, it might be doable by reading the image like a file system, perhaps convert it to a read-write filesystem first and then convert back after changing the file. But all of that might actually mess with the image's boot sector. You might need to re-assert that boot sector or do whatever uefi does. Jan 29 11:23:27 thanks, will look further Jan 29 11:23:33 via the forum Jan 29 11:37:07 Hi guys, I have been working ob Beaglebone blue and it was working fine till this morning Jan 29 11:37:40 the beaglebone boots and then just LED0 flashes and the board is not getting detected by usb Jan 29 11:37:52 also the octavo chip is getting hotter Jan 29 11:38:16 any ideas on what could have happened? Jan 29 11:53:19 if the beagleboard has booted properly which userled should we see at the end? is it the User led 0 which should keep on flashing once the beaglebone as booted up? Jan 29 11:56:01 Daksh: do you have the debug UART hooked up? Jan 29 11:59:22 I have just got a micro usb cable connected Jan 29 11:59:57 I have just got a micro usb cable connected Jan 29 12:00:50 then get yourself a USB-UART adapter and hook it up Jan 29 12:01:10 also obviously don't leave the BBBlue on if it's malfunctioning Jan 29 12:01:36 that is the thing I don't really know if it's malfunctioning Jan 29 12:02:12 well, also if you don't know. unless you deliberately want to risk destroying the device. Jan 29 12:02:44 it follows the boot sequence and all 4 leds light up... and then all leds turn off and just led0 keeps blinking Jan 29 12:03:09 and the board is not read by the usb Jan 29 12:03:22 it was working fine till this morning Jan 29 12:03:23 get a USB-UART adapter, connect it to the device debug UART Jan 29 12:03:55 what should I be looking for after connecting up the UART? Jan 29 12:08:45 you should see the power up sequence, SPL, U-Boot, Kernel Jan 29 12:18:43 hello again Jan 29 12:22:41 I'm trying to bitbang SPI1 SS pin while in spi communication Jan 29 12:23:27 I think rc_gpio_set_value(gpio, value) function vill work Jan 29 12:23:47 but I couln't find gpio number of that pin Jan 29 12:23:54 does anybody know it? Jan 29 12:24:49 what is the gpio number it should have a number like GP1_17 Jan 29 12:25:23 you can find it on the schematics Jan 29 12:25:42 I dont know that, it is slave select pin on SPI port, S1.1 socket named Jan 29 12:28:12 did you have a look at the strawson user manual? Jan 29 12:28:26 schematics says H18 Jan 29 12:28:45 yes Jan 29 12:30:17 int rc_manual_select_spi_slave(int slave); int rc_manual_deselect_spi_slave(int slave); If you configure a slave to be operated in manual select mode, it's necessary to use the above two select and deselect functions whenever talking to the slave. Selecting a slave results in driving the corresponding slave select line voltage to ground. It also automatically deselects the other slave if it was left selected by accident, but we do not Jan 29 12:30:36 did you try these two library functions? Jan 29 12:31:04 yes I'm using them Jan 29 12:31:13 well what I'm trying to do is Jan 29 12:32:20 I2m trying to control chip select pin, and these functions automatically controls that pin so I thought I need to bit bang that pin Jan 29 12:33:43 okay I found the gpio number for H18... you are using beaglebone blue right? Jan 29 12:33:59 for beaglebone blue it is GPIO0_29 Jan 29 12:34:34 so the exact pin number would be (32*0)+29 = 29 Jan 29 12:34:42 for example, if there are consequent read commands in code, chip select pin goes down for once and not going high again until all reads done Jan 29 12:34:58 oh Jan 29 12:35:06 thank you very much Jan 29 12:35:39 no worries... I hope it works for future pin reference use this table https://github.com/beagleboard/beaglebone-blue/blob/master/BeagleBone_Blue_Pin_Table.csv Jan 29 12:48:49 that csv contains errors Jan 29 12:49:05 quite a few errors I noticed Jan 29 12:50:53 zmatt (IRC): would be great to get a bug report of any. Same info tries to be replicated at https://github.com/BeagleBoard/capes. Jan 29 12:54:42 unfortunately I didn't keep track of it, I just noticed it because I started working on a pins spreadsheet for the blue (see "BBBlue" sheet of , work in progress) and first merged the data from the CSV into it and then I checked the schematic to cover the rest of the pins Jan 29 12:56:49 hold on, it shouldn't be hard to make a sheet that compares the two Jan 29 12:57:35 there are also weird things in the csv like the words "green" or "yellow" appearing in random places Jan 29 13:07:41 jkridner[m]: see 'tmp' tab of the same spreadsheet doc Jan 29 13:11:21 some mismatches are just due to spelling, but definitely not all, e.g. SPI1_SS1/2, WLAN_IRQ, DCAN1_SILENT Jan 29 13:11:51 DCAN in general Jan 29 13:12:22 btw, did you see my comment earlier about bluetooth a2dp ? Jan 29 13:17:18 hi there can someone tell me about ttl to rs232r cable Jan 29 13:17:26 can be used in bbb Jan 29 13:20:47 pls someone to tell me can be used in bbb or can damage the board Jan 29 13:24:28 svitanov: any 3.3v ttl serial cable should be fine Jan 29 13:25:33 5v cables won't damage the board (the bbb has 5v-tolerant buffers for the console uart port), but may not work reliably since some 5v cables don't understand the 3.3v signal from the bbb Jan 29 13:25:58 svitanov: some cable suggestions can be found here: https://elinux.org/Beagleboard:BeagleBone_Black_Serial Jan 29 13:29:23 I need to use in bbb for flashing onboard 6 pin be done or I donk know Jan 29 13:31:13 and how to flash bbb onboard with new operatin system Jan 29 13:44:28 someone to help me pls to install getminera OS on bbb Jan 29 14:02:22 "I need to use in bbb for flashing onboard 6 pin be done or I donk know" <--- I have absolutely no idea what this word salad could possibly mean Jan 29 14:03:33 instructions for reflashing a bbb with a new OS are below the download links at bbb.io/latest-images Jan 29 14:04:06 I quickly googled what "getminera" is, but it isn't an OS, it's just a piece of software Jan 29 14:04:33 presumably they have installation instructions themselves Jan 29 15:22:44 Hi all Jan 29 15:31:48 what the actual fuck Jan 29 15:32:07 pardon? Jan 29 15:32:11 address space layout randomization is fucking ignoring the alignment specification of ELF sections Jan 29 15:32:40 oh. that's not clever Jan 29 15:32:56 I have a 32KB-aligned object in .bss Jan 29 15:33:07 objdump on the executable confirms .bss has the required alignment: Jan 29 15:33:10 22 .bss 00010000 00018000 00018000 00001008 2**15 Jan 29 15:33:44 I print the pointer to that object... 0x4be000 Jan 29 16:28:05 Just got BeagleBoneBlack, none of the LED examples work. Says "error: Unable to find LED: usr0". Ideas? Jan 29 16:33:22 Thomas__ what image are you using? Can you read /etc/dogtag? Jan 29 16:34:18 If you typed it in with an older BoneScript library, it is a good possibility it is case sensitive. Jan 29 16:34:28 Try USR0. Jan 29 16:35:13 It's uppercase in the source code Jan 29 16:36:48 What command can I use to read dogtag? Jan 29 16:37:03 It's the example that shipped on the board Jan 29 17:43:19 jkridner[m]: let me know when you're done with my 'tmp' spreadsheet showing the differences between the csv and the schematic (so I can remove it) Jan 29 18:05:50 I'm experimenting with Python scripts that draw to a framebuffer. With the Beaglebone connected to my PC via USB cable is it possible to get the remote framebuffer to display on my PC screen? Jan 29 18:12:25 with "a framebuffer" you mean "the (legacy) framebuffer" (/dev/fb0) presumably? Jan 29 18:13:45 zmatt: that's a nice spreadsheet. Jan 29 18:14:16 do you highlight any errors on min? Jan 29 18:14:19 er, mine Jan 29 18:15:11 jkridner: yellow = matches "Use on Blue" rather than "Blue Name", red = matches neither Jan 29 18:15:44 (see conditional formatting rules) Jan 29 18:16:45 in a few cases where the schematic net was actually not named at all (e.g. those for the sd card) I just adopted the "Blue Name" as net name Jan 29 18:17:36 or in a few cases left it blank Jan 29 18:18:37 I don't see those columns... only "BB Blue", "" and "Name". Jan 29 18:18:50 are you looking at the 'tmp' sheet? Jan 29 18:19:06 oh. sorry. I was on BBBlue sheet. Jan 29 18:21:34 note I inserted a leading zero for the row number of balls in rows 1-9, for consistency with the rest of my sheet. other than that columns A-D are straight from the csv Jan 29 18:22:17 and thanks :) Jan 29 18:22:30 had you never seen my pin spreadsheet for the bbb before? Jan 29 18:25:58 zmatt: yeah Jan 29 18:29:05 this sounds like it might do what you want -> https://github.com/ponty/framebuffer-vncserver Jan 29 18:48:29 zmatt looks like there might be a real delta in there, but most are just adjustments of the naming style. Jan 29 18:48:51 some are cosmetic, but definitely not all Jan 29 18:49:10 can we identify which arent? Jan 29 18:49:37 some are also flagged due to being deltas from the Robotics Cape. Jan 29 18:50:10 the highlighting rules I added don't look at the robotics cape column Jan 29 18:50:29 the EQEP vs. QEP naming is probably one of the most confusing items. Jan 29 18:54:53 I figured you could determine which differences were significant or not yourself :P but to help out I've highlighted the ones that look like real differences to me Jan 29 18:55:17 btw, did you see my remark earlier about a2dp ? Jan 29 19:44:48 hi all Jan 29 19:46:27 i got a question about yocto and beaglebone. i want to build an image for my bbgw. can i use the bbb meta-layer and just add the bbgw device-tree? Jan 29 19:55:26 probably Jan 29 20:16:24 Hi all, do you know if this is possible to use BeagleBoard Blue with machinekit and use stepper motor driver (leadshine dm556) ? Jan 29 21:07:47 Alkabal_: according to the datasheet the dm556 logic inputs are 5V, so some level-shifting will be required since the beaglebone has 3.3v outputs Jan 30 01:18:47 zmatt: no, I missed the A2DP remark. We ended up never using the I2S interface to the WL1835. Jan 30 01:20:55 jkridner: here's the log of what I said: https://pastebin.com/qPNsWTYN Jan 30 01:21:41 zmatt: ah, yes, the level shifters. Jan 30 01:21:56 I think the BSP package didn't make it possible, but the BSM does. Jan 30 01:22:15 we've got a draft Blue design with the BSM that removes the level shifters. Jan 30 01:22:18 yeah that's what I discovered too Jan 30 01:22:37 I don't see any "errors" with the list... just naming "deltas" Jan 30 01:22:41 anyway, if you do connect a2dp, be sure to use fsx/clkx and not fsr/clkr Jan 30 01:23:15 in the "Use on Blue" column I've marked cells in purple that seem to mismatch the current schematic Jan 30 01:23:28 maybe MDIR_2B wasn't updated right. Jan 30 01:24:33 and a bunch more Jan 30 01:27:07 btw, I'm a bit concerned about the multitude of power domains on the blue... right now it looks to me like it would be a very bad idea to shut the beaglebone down while either V_BATT or VCC is supplied Jan 30 01:29:34 in that situation there are voltages being put on ADC pins (AIN5/6), and a 5V supply remains available in various places where they very much invite the user to drive a signal into the unpowered am335x Jan 30 01:29:55 e.g. connecting a gsm module to the GSM connector would result in exactly that Jan 30 01:33:13 (also, a minor note, you may want to change the 100K pull-up/down resistors on sysboot pins to something stronger like 10K to reduce the risk that stuff connected to those pins accidently interferes with sysboot config. for example, sysboot 2/3/14 have 100K pull-ups but also connect to the motor driver which has 120-600 K pull-downs, which actually means those pins are not guaranteed to be sampled high) Jan 30 01:34:33 cut the ADC traces and put in a buffer powered by VDD_ADC Jan 30 01:36:18 that would fix that particular problem yes. (do include a series resistor on the output of the buffer to avoid running afoul of the am335x adc power-up erratum) Jan 30 01:37:16 or use a low enough powered op-amp and power it via a GPIO ;) Jan 30 01:37:32 ew Jan 30 01:37:42 what is your objection? noise? Jan 30 01:38:03 I just don't get why you would do that instead of using vdd_adc Jan 30 01:38:22 software control of the power Jan 30 01:38:30 why would you need that? Jan 30 01:38:35 save on quiescent current Jan 30 01:40:20 the voltage dividers currently used can't be turned off either :P Jan 30 01:49:53 usb host power should probably have a much larger output cap than the 10 uF it currently has (220 uF recommended by tps2105 datasheet "for 100-mA loads") Jan 30 02:04:00 Hi all, I'm trying to use Adafruit's python GPIO library to set a pin high and I'm only getting ~500 mV instead of 3.3V. This feels like I'm doing something simple and stupid wrong, so what is it? Jan 30 02:04:06 If it's relevant, the pin is P8_27 and I've disabled the hdmi virtual cape. P8_29 being used in as far as I can tell identical fashion also works just fine, which is the really odd thing. Jan 30 02:04:49 that sounds odd Jan 30 02:05:01 this is without any load connected to the pin? Jan 30 02:08:21 yup Jan 30 02:08:50 Odder: I just switched up my .dts file and python code to use P8_29 and P8_31, 29 continues to work as expected and 31 now behaves just like 27 was. Jan 30 02:08:58 High state is 500 mv or so, low state is 0 as it should be Jan 30 02:10:15 weird as hell Jan 30 02:12:17 dts file for the 8_27 version, in case you see anything I don't: https://pastebin.com/rDn9w21e Jan 30 02:12:44 it doesn't sound good to me... the only reasons I can imagine for measuring 500 mV on a gpio that is driven high (assuming it's not somehow a measurement error) is if there is a very heavy load to ground or if the output driver is broken (with the former probably leading swiftly to the latter) Jan 30 02:12:55 although even as a failure mode it sounds odd to me Jan 30 02:13:16 you can also double-check pin configuration with my show-pins script -> https://github.com/mvduin/bbb-pin-utils/ Jan 30 02:13:46 but if you can see the pin toggle based on the value you're setting, it kinda has to be configured right Jan 30 02:15:28 Do 8_27 and 8_31 share a driver but not 8_29? Jan 30 02:15:38 nice script btw, been wanting something like that. Jan 30 02:15:55 It shows 29 and 31 (haven't rebooted back to 27 29 yet) with exactly identical configs though Jan 30 02:16:58 P8.27 and P8.29 are basically about as similar as pins can get Jan 30 02:19:15 good lord... I'm a bit sick right now and haven't been getting much food down, so I guess I'm feeling a bit spacey Jan 30 02:19:35 The two channels on my scope were scaled different. Both are actually at 3.3v... Jan 30 02:20:02 I _swear_ it really was at 500 mV and properly scaled at some point, but it isn't now so it probably never was. Jan 30 02:20:02 I'm glad I included the "assuming it's not somehow a measurement error" caveat in my statement earlier :) Jan 30 02:20:27 heh, well thanks for the help anyway...! Jan 30 02:20:40 at least I got a handy new script out of this wasted ~hour Jan 30 02:20:45 :) Jan 30 02:21:00 let me throw in a spreadsheet as bonus -> https://goo.gl/Jkcg0w Jan 30 02:21:14 I personally find the P9/P8 tabs to be handy overviews **** ENDING LOGGING AT Tue Jan 30 03:00:01 2018