**** BEGIN LOGGING AT Wed Jul 19 03:00:03 2017 Jul 19 08:39:56 hi i wants to implement can protocol Jul 19 10:03:00 * zmatt . o O ( ... okay? ) Jul 19 13:18:29 hello everyone Jul 19 13:18:46 i need help with using uart with BBB Jul 19 13:19:15 ttyS0 works Jul 19 13:19:33 but ttys1/s2 doesnt Jul 19 13:20:24 any ideas? Jul 19 13:26:39 Do you enable them with your device tree? Jul 19 13:26:58 afaik they're defaulting to GPIO unless you load an overlay Jul 19 13:37:02 or if using cape-universal (which is the default), select pinmux with the config-pin utility Jul 19 13:53:09 zmatt: I recently got a git tree for the kernel from another ARM vendor. Jul 19 13:53:33 zmatt: It contained a whole commit, claimed to be 4.1.15, and only diffed 450k lines vs. upstream 4.1.15. Jul 19 13:53:51 Vendors... Jul 19 14:04:59 lol Jul 19 14:26:54 what was the commit message? "stuff" ? :D Jul 19 14:30:23 zmatt: I don't rightly remember. I think it was "support for board" or something like that. Jul 19 14:30:34 zmatt: Also, all the files had the same timestamp. As well. Jul 19 14:41:38 git doesn't record timestamps Jul 19 14:48:02 Hi all.. I have a BBB with an odd issue. The USB seems to only be scanned on boot. If I have a device plugged in on boot, it's detected and shows up in lsusb. However if I unplug the device, nothing happens in syslog or dmesg, and it doesn't disappear. Jul 19 14:48:32 If I boot with nothing attached and plug in a keyboard or something, it doesn't appear, isn't probed, etc. Jul 19 14:48:53 Just pulled the image yesterday, 3.8.13-bone47 Jul 19 14:49:36 wow that's a really ancient kernel Jul 19 14:50:09 you downloaded an image yesterday? what did you download? Jul 19 14:50:51 current images use a 4.4 kernel Jul 19 14:51:34 Jessie for Beaglebone, beagleboard: https://beagleboard.org/latest-images Jul 19 14:51:58 Debian 8.7 2017-03-19 4GB SD LXQT Jul 19 14:52:23 uhh, there's no way that comes with a 3.8 kernel Jul 19 14:52:59 actually I know 100% sure it doesn't Jul 19 14:54:40 try powering on with the S2 button held down (the button closest to the card slot) Jul 19 14:55:19 (to ensure you aren't unintentionally booting from eMMC) Jul 19 15:29:57 hello all Jul 19 15:35:51 hello Jul 19 15:36:33 you're testing the acoustics of the room? :) Jul 19 15:37:03 just seeing if anyone is online Jul 19 15:37:11 also switched clients Jul 19 15:37:34 most people are not likely to respond to "hello" here (I'm generally not either) Jul 19 15:37:45 ah gottcha Jul 19 15:37:51 ah gotcha Jul 19 15:38:05 wow its been too long since ive used IRC Jul 19 15:38:52 anyway the reason I joined was to see if anyone could answer a question i have regarding the hardware on the beaglebone Jul 19 15:40:07 I am designing a new board and I see that the beaglebone black wireless uses the Octavo OSD335x chip instead of the AM335x and its supporting chips Jul 19 15:41:04 my qestion is, what is the longevity like on the OSD335x chip? Like if I make my board with that chip, will it still be availible for the next 5-10 years? Jul 19 15:41:40 or would I be better off using the TI AM335x along with its supporting chips? Jul 19 15:41:44 that's really a question you'd need to ask octavo Jul 19 15:42:14 alright, thats probably a good start Jul 19 15:42:33 my opinion would be: if it is within your abilities to use the am335x directly, then go for that Jul 19 15:43:52 both options are within my abilities, Its just that the Octavo chip is desirable due to the level of integration it has which makes board layout much quicker and simpler Jul 19 15:45:52 thanks for the advise Jul 19 15:45:59 sure, that's the reason it exists... of course there's always downsides, like the inflexibility of being tied to their design and component choices Jul 19 15:46:38 thats true too Jul 19 15:46:58 I think for my application, it has all of the requirements and then some Jul 19 15:47:57 something I'd be concerned about is that they adopted the BBB's power supply design, including the 3v3 regulator issue Jul 19 15:48:28 what is the 3.3v issue? Jul 19 15:48:43 does it burn out? or is it just unstable? Jul 19 15:49:13 it shuts down later than it's supposed to according to power supply sequencing requirements Jul 19 15:49:24 ah Jul 19 15:49:30 which, depending on external electronics, may result in significant current injection Jul 19 15:50:12 https://lh3.googleusercontent.com/-_xHMJGdwpTA/VUKJFjC8koI/AAAAAAAAACc/VmSqofhSmXM/s1600/off-dc-noserial.png Jul 19 15:50:15 I think for 3.3v external devices, I will have my own regulator Jul 19 15:50:25 oh wow Jul 19 15:50:29 they include a separate regulator, that's actually the root cause of the problem Jul 19 15:50:44 because the regulator isn't controlled by the pmic Jul 19 15:50:56 ah ok that make sense Jul 19 15:51:51 you can find a description of what you're seeing there in this post I wrote: https://groups.google.com/d/msg/beagleboard/7sxPePT7wkM/V1Ft-xxh0agJ Jul 19 15:52:47 excellent! This will be quite helpful Jul 19 15:52:55 it was discovered in relation to people trying to use the battery terminals of the BBB (which hugely amplifies the problem) Jul 19 15:53:07 ah I see Jul 19 15:53:20 well my application wont be running off batteries Jul 19 15:53:28 also I plan to make my own power supplies Jul 19 15:53:41 the problem is somewhat mitigated because the PMIC switches to battery power unconditionally at the start of the powerdown sequence Jul 19 15:54:01 which causes VSYS (input to the LDOs) to fall during power-down sequencing Jul 19 15:54:22 which is quite idiotic in itself, but it does reduce the impact of the 3v3 bug Jul 19 15:54:23 interesting Jul 19 15:55:05 one big downside of the osd335x in my opinion is that you can't do much about the power supply scheme anymore Jul 19 15:55:21 true, its all integrated Jul 19 15:55:33 e.g. you can't run some or all of the I/O banks at 1.8V (which is a much more comfortable voltage for the processor than 3.3V) Jul 19 15:56:16 or swap out the not-very-well-designed pmic :) Jul 19 15:57:09 lol true Jul 19 15:57:40 awesome, thanks for the help. I think I have what I need to make a decision Jul 19 15:57:45 good luck Jul 19 15:57:49 thanks Jul 19 16:32:09 I am new to beaglebone. I have beaglebone blue. But I can't find much resources on how to use it's robotics functionality like I want to write the program to control brushless motor using ESC. But I don't know how to access the servo pin and what library to use. If anyone can help me or point me the right resource Jul 20 00:54:46 hey zmatt! do the newer beaglebone kernels (4.4, 4.9) have sgx driver support? Jul 20 00:56:00 sure, with a bit of effort (unless rcn has packages nowadays, haven't really checked) Jul 20 00:56:11 we have them working on a 4.9-bone kernel Jul 20 00:56:31 is there a guide to the effort i need to do to get them working on 4.9? Jul 20 00:56:56 trying to get hardware acceleration for that silly android board Jul 20 00:57:31 looks like rcn actually does have packages Jul 20 00:57:48 ah right android, so a debian package isn't that useful to you Jul 20 00:58:06 yeah... Jul 20 00:58:46 wait a sec - arent these proprietary blobs? Jul 20 00:59:01 the userspace libs are yes Jul 20 00:59:18 but with permissive redistribution license Jul 20 00:59:35 if i had a deb, couldnt i just rip them out and copypasta them into my android image? Jul 20 01:00:28 if you're using an unmodified rcn kernel for which there's a corresponding sgx package Jul 20 01:00:47 building the kernel drivers isn't hard though Jul 20 01:01:01 is there a repo i can clone? Jul 20 01:01:12 https://github.com/mvduin/omap5-sgx-ddk-linux this is my mirror of the kernel driver + a little crufty build script Jul 20 01:01:48 the build script will need tweaking since it contains hardcoded paths, and it builds a debian package currently Jul 20 01:17:07 for the userspace part you need https://github.com/mvduin/omap5-sgx-ddk-um-linux and https://github.com/mvduin/libgbm Jul 20 01:38:14 thanks for pointing me in the right direction, zmatt! Jul 20 01:38:42 and ensure there aren't conflicting mesa libs installed unless you enjoy headaches **** ENDING LOGGING AT Thu Jul 20 03:00:01 2017