**** BEGIN LOGGING AT Thu Sep 13 10:04:13 2018 Sep 13 10:15:37 thinkfat: cool i'll have a look. Sep 13 10:18:41 thinkfat: i have to mention that i am not using overlays though :) Sep 13 10:19:10 I don't think that makes a difference Sep 13 10:19:48 if you look into zmatt's overlay-utils, it's just a dts with some dressing Sep 13 10:25:08 ok :) i somehow felt the need to mention it :D Sep 13 10:34:01 thinkfat: i tried the snippet but no luck yet. Sep 13 10:41:52 Duality: what'cha mean? Sep 13 11:03:37 thinkfat: well i copied over the clock settings from the overlay, and loaded it in but the frame rate is still wrong (which means mcasp is using the internal oscillator instead of my external one. Sep 13 11:04:28 Duality: there's also the pinmux which might be interesting... but it should lead to a different result if they're wrong Sep 13 11:04:47 this is what my sound setup in the device tree looks like: https://pastebin.com/tx6xuMef Sep 13 11:04:50 thinkfat: good point Sep 13 11:05:56 the clk_mcasp0... Sep 13 11:16:01 Duality: you're feeding your own clock into gpio3_21 right? Sep 13 11:16:21 yes Sep 13 11:16:48 Duality: I think that snippet is wrong anyway. The thing is, on the BBB there's an oscillator that supplied the audio clock, and you need to disable it if you feed in your own clock Sep 13 11:17:44 Duality: and the snippet you pasted is sort of wrong there, you copied the clk_mcasp0 node from the HDMI overlay, right? Sep 13 11:18:16 yes Sep 13 11:18:26 Duality: if you want to feed your own clock, you need to make sure the internal oscillator is off. Sep 13 11:18:27 then edited to add my gpio Sep 13 11:19:04 Duality: without understanding what you did ;) Sep 13 11:20:44 you don't really need a "clk_mcasp0 as a go-between. This node just describes the gpio that switches the internal oscillator on and off Sep 13 11:21:48 what you need to do however is make sure that gpio1_27 is low to disable the internal oscillator Sep 13 11:21:57 check page 3 of the BBB schematics Sep 13 11:23:04 as otherwise you'll probably feed your own clock into the output of the on-board oscillator Sep 13 11:24:23 this is all new to me :) i made an assumption and it was wrong :D Sep 13 11:24:47 doh Sep 13 11:24:51 can you not read schematics? Sep 13 11:28:38 i am working on a custom board, that schematic does not mention such a thing. Sep 13 11:30:38 oh, ok, you mean, not actually a beaglebone black? Sep 13 11:31:16 thinkfat: no i am sorry if that should have been clear :) Sep 13 11:32:23 in that case I guess, just drop the mcasp0_clk node and replace all references to it by mcasp0_clk_fixed Sep 13 11:36:57 i am missing something and i do not know what, the samplerate is still 46.8 instead of 44.1 Sep 13 12:00:25 Duality: take a frequency counter to your clock ;) Sep 13 12:00:54 Duality: which frequency are you feeding? Sep 13 12:02:51 Duality: 22.579200 MHz, from your devicetree Sep 13 12:03:12 the ration between the actual clock and what you want suggests that the actual clock is 24MHz Sep 13 12:03:26 e.g. the one you're really feeding into the mcasp0 Sep 13 12:04:09 from whatever source Sep 13 12:06:31 the sample rate should be 46.875kHz Sep 13 12:11:54 thinkfat: my logic analyzer says it's the 22Mhz clock Sep 13 12:12:39 thinkfat: yes 24Mhz is the internal clock Sep 13 12:12:45 which i don't want. Sep 13 12:13:01 https://github.com/beagleboard/bb.org-overlays/blob/master/src/arm/BB-BONE-AUDI-02-00A0.dts Sep 13 12:13:02 ? Sep 13 12:14:20 Duality: note how this snippet has the clk_mcasp in the sound_master node, not in the sound_cpu node Sep 13 12:15:20 thinkfat: your right i had not noticed that Sep 13 12:15:54 Duality: I'm just fishing in the dark, though. I have as much of an idea as you have ;) Sep 13 12:17:54 Duality: thing is, the audio configuration for the bbb is really all with external clock, mainly because the internal clock is not able to generate 44.1 khz with a fixed divider Sep 13 12:18:15 you'd need a pll or another fractional divider to get it Sep 13 12:18:36 hence the external oscillator on the bbb Sep 13 12:19:22 so, all of the examples you will find should all work for your design, except that you don't need to gate the internal oscillator if you want to feed your own clock Sep 13 12:26:09 thinkfat: i changed it to be in the master_sound node but still no luck :S Sep 13 12:34:46 i got a samplerate of 44.09 now ! Sep 13 12:34:48 :) Sep 13 12:35:27 oh? Sep 13 12:35:32 all of a sudden? Sep 13 12:35:50 https://github.com/beagleboard/bb.org-overlays/blob/master/src/arm/BB-HDMI-TDA998x-00A0.dts Sep 13 12:36:14 see the line "simple-audio-card,bitclock-master"? Sep 13 12:43:06 thinkfat: no only after i made a mistake and put clk_mcasp0_fixed in both sound_master node and sound_cpu node. Sep 13 12:51:40 Duality: I hope you can backtrack Sep 13 12:54:42 Duality: https://www.kernel.org/doc/Documentation/devicetree/bindings/sound/simple-card.txt Sep 13 12:54:49 Duality: I think it's all explained therein Sep 13 12:55:11 Duality: you have a sound_master, which is the one supplying the clock, and in your case it's the codec, not the cpu Sep 13 12:56:35 Duality: and you have the simple-audio-card,bitclock-master and frame-master property which would reference the sound master Sep 13 13:11:08 thinkfat: i see, I had read simple-card.txt :) but did not get that info you said from it. Sep 13 13:48:38 thinkfat: i appreciate your help so far ! :) and that your willing to think along :) Sep 13 15:49:13 With a beaglebone black wireless acting as a wifi router, how does one go about directing the traffic of connected devices from bbbexample.com to BBB_IP_ADDRESS:PORT? Sep 13 15:55:47 any linux pros here? Sep 13 16:13:42 dns-question: i'm not a pro .. but i wonder if your question isn't a bit unclear Sep 13 16:51:04 thinkfat: the audio osc on the bbb is for an 48 kHz samplerate btw, not 44.1 Sep 13 19:17:20 gpio_leds { ... led0 {... gpios=<0x5 0x15 0x0>; what are these 3 values? Sep 13 20:01:39 Hi I need some help on analog read on BBB. Everything I find online is out dated for the Kernal I am using. I am using: Linux beaglebone 4.14.49-ti-r54 #1 SMP PREEMPT Fri Jun 15 22:14:13 UTC 2018 armv7l GNU/Linux Sep 13 20:01:59 Where should I start. Sep 13 20:02:05 ? Sep 13 20:09:09 Anyone can help me with the Analog read? Sep 13 20:20:25 I'm running one of latest debian 9 images on BBB, config-pin doesn't work for me. Universal cape is enabled in uboot. Logs: https://pastebin.ubuntu.com/p/2fXwSBKsDq/ Sep 13 20:20:40 any ideas what to check are appreciated Sep 13 20:25:22 https://github.com/cdsteinkuehler/beaglebone-universal-io tells about /sys/devices/ocp.* folder, while in my logs /sys/devices/platform/ocp/ is used. It's hard to understand what's going on and what docs to read **** ENDING LOGGING AT Thu Sep 13 20:58:25 2018 **** BEGIN LOGGING AT Thu Sep 13 23:44:37 2018 Sep 14 01:11:07 Okay, okay! Before I set this place on fire, should I plug in the battery first or the USB to test software w/ the Motor Cape? Sep 14 01:12:57 For future ideas, I would put a start button on the Cape. I can do this but I think that the Cape would be "nice" if the button would allow for power to supply the board. Sep 14 01:13:22 But this power should only be supplied when the user allows it to be so. Sep 14 01:14:58 Testing ahead! Sep 14 01:16:19 Okay! Sep 14 01:16:49 The battery does not start the board! It only allows for motor action. The board needs a separate power supply. Sep 14 01:23:14 Be careful set_ Sep 14 01:24:52 don't let the magic smoke escape Sep 14 01:25:04 OTay! Sep 14 01:25:21 What are the chips on the board? Sep 14 01:25:25 I cannot tell just yet. Sep 14 01:25:40 ST? Sep 14 01:25:57 I need a big magnifying glass. Sep 14 01:27:31 The schematic says "thermal pad." Sep 14 01:27:49 generic is fine, I guess. Sep 14 01:36:20 Is x1, the "Motor" header on the schematic? Sep 14 01:36:42 w/ the power in and GND? Sep 14 01:39:24 Forget that last question. Sep 14 01:39:29 Sorry. Sep 14 01:50:03 I think the two motors, one and four respectively, need to be on separate Thermal Pads. Sep 14 01:50:11 Right/ Sep 14 01:51:51 Or can I use the Motor Cape w/ M1-, M1+, M2-, and M2+ for one motor while in one piece of software? Sep 14 01:51:53 ... Sep 14 01:52:11 Back to it! Sep 14 01:58:54 So, since the each thermal pad sources "two" motors, say Thermal Pad One for example sources only Motor One and Motor Four, then the connections for a stepper must be for one and four instead of one and two. Sep 14 01:58:55 ... Sep 14 01:59:55 yatzee! **** ENDING LOGGING AT Fri Sep 14 03:00:10 2018