**** BEGIN LOGGING AT Fri Nov 06 02:59:58 2015 Nov 06 03:28:35 you shouldn't point at people, least so I've been told. Nov 06 03:40:51 yeah, hence I skipped it Nov 06 13:12:11 hi, i've a strange issue with a Beaglebone black, i cannot boot from SD to flash the firmware, i've tried with samsung evo sdhc (orange and gray version) 16 GB, i've checked the boot flag on first partition; the strange thing is that i can see the sd card from eMMC installed OS only if extract then reinsert it. Nov 06 13:12:56 i use a 3000 mA power supply Nov 06 13:24:43 barsigor: make sure to hold down S2 before powering up the board Nov 06 13:54:09 i hold s2, but nothing... Nov 06 13:54:22 i'ved tried now with another sd Nov 06 14:59:01 anyone have experience setting up and using the linux watchdog subsystem? I'm looking to protect myself from crashes that right now require a manual reset via the reset button. Nov 06 16:24:10 can I boot from the USB port on the BeagleBone Black? Nov 06 16:26:01 GT_: if you mean by connecting a USB storage device, no. Nov 06 16:26:21 GT_: if you mean by connecting it to a computer and feeding it data through the USB port, yes. Nov 06 16:27:01 what you can do though is to have U-boot and kernel on eMMC or SD and rootfs on a USB storage device Nov 06 16:33:00 How do you load it from a PC? Nov 06 16:33:57 and do you know if the limitation ( SD yes, USB no ), is BBB specific? Nov 06 16:34:45 I want to be able to give users the ability to reflash the board, but not using the SD card. Nov 06 16:35:08 and without knowing anything about u-boot. Nov 06 16:38:21 GT_: there is a project for flashing over usb. lfbb or such, but it's not so straight forward AFAIU Nov 06 17:31:03 May I ask a question? Nov 06 17:40:56 kaak: I use the daemon in the 'watchdog' package Nov 06 17:44:16 with this config file http://pastebin.com/KUbBc9NN Nov 06 17:44:53 (you can reduce the timeout of course) Nov 06 19:55:24 rcn-ee: heh, I've been working on adding iosets to my spreadsheet... some of them really make no sense Nov 06 19:56:28 like I started typing over those for the uarts before I realized... wait, all of these signals are either loosely coupled or completely asynchronous to each other, why on earth have iosets been defined for them? Nov 06 19:57:56 zmatt, that bad! ;) Nov 06 20:00:05 defining iosets for peripherals with 1 pin (eCAP) is also interesting Nov 06 20:16:08 time to catch train, bbl Nov 06 20:55:59 how can I invoke the uBoot prompt while booting the BBB? Nov 06 21:00:39 GT_, "any key" over serial debug terminal when you reboot/boot.. Nov 06 21:01:39 rcn-ee, hmm do you know why then my Putty serial conn would not take input from my keyboard? Nov 06 21:02:15 rcn-ee, I also tried Tera Term but the console does not seem to know that I am pressing a key Nov 06 21:02:44 no idea... Nov 06 21:05:42 Anyone here have ideas on this https://www.mail-archive.com/linux-btrfs@vger.kernel.org/msg48104.html (or previously run into something similar)? Nov 06 21:07:16 rcn-ee, nevermind I had a loose cable Nov 06 21:12:51 flow control is the other fav. Nov 06 21:51:25 * zmatt is still annoyed flow control wasn't included in the console header but left not-connected Nov 06 21:55:04 jmesmon: that instruction decodes to ldrex r3, [r9] Nov 06 21:55:18 and r9 isn't aligned Nov 06 21:55:37 also, r9 isn't a valid address even if it were aligned Nov 06 21:55:40 zmatt: I've been digging, and it looks like it's really a "almost NULL" deref that just happens to be caught via the alignment exception Nov 06 21:55:42 yep Nov 06 21:55:57 btrfs_root == 1, for some reason >.> Nov 06 21:56:19 (btw, why is the kernel compiled in ARM mode instead of thumb mode?) Nov 06 21:56:31 should it be compiled in thumb mode? Nov 06 21:56:37 * jmesmon has no idea Nov 06 21:56:47 saves space, increases performance Nov 06 21:57:02 why doesn't everything use thumb mode then? Nov 06 21:57:03 (mostly by saving space, hence reducing pressure on the instruction cache) Nov 06 21:57:18 legacy Nov 06 21:57:42 you can fully use an ARMv7 processor without ever executing a single instruction in ARM mode if you want to Nov 06 21:57:52 I'll try flicking that switch on, but I'm doubtful it will help with this particular bug Nov 06 21:58:13 no it was just a random observation Nov 06 21:58:31 yep, and I appreciate it (I'm in the middle of trying to shrink the kernel) Nov 06 21:58:47 and upgrade it to a new version, hence my fight with this bug Nov 06 22:00:32 just for fun, click "next message" Nov 06 22:00:52 the description of that patch makes me wonder if btrfs ought to be considered production quality :P Nov 06 22:05:50 well, I'm not sure I'm too supprised. It's amazing how many bugs code we run everyday has. Nov 06 22:06:03 https://medium.com/message/everything-is-broken-81e5f33a24e1 Nov 06 22:06:03 :P Nov 06 22:45:25 btrfs has always been regarded as beta I thought .. Nov 06 22:45:28 on all I've read anyhow Nov 06 22:57:41 is there a way to tell what of the kernels for the debian BBB release has PRU enabled ? Nov 06 22:58:46 i just checked and the image that i flashed has CONFIG_UIO_PRUSS unset Nov 06 22:58:56 or do i need to recompile the kernel **** ENDING LOGGING AT Sat Nov 07 02:59:59 2015