**** BEGIN LOGGING AT Sat Jul 01 03:00:02 2017 Jul 01 08:41:11 :q! Jul 01 08:48:30 E37: No write since last change (add ! to override) Jul 01 08:52:12 I had a ! you dummy :P Jul 01 12:25:53 hi, I'm a bit confused about working with i2c devices. I have an accelerometer connected to i2c bus 2, when i ran i2cdetect -r 2 the first time, showed a device at address 0x19, then I ran i2cdump 2 0x19 and it printed half of the register values for my device (I think) then it got to address 0x5a and it started to print XX's Jul 01 12:27:02 after that, any time I run i2cdump it prints nothing but XX's and it seems to read each byte very slowly (timing out). Also, when I re-run i2cdetect, that also runs very slowly and it fails to detect my device Jul 01 12:27:10 why would these commands stop working? Jul 01 12:31:30 Hello all, I have a quick question; I have an original beaglebone (white), and would like to run Volumio on it. However, I can only find an image for BBB, and that does not run (4 LEDs solid green, then nothing). Jul 01 12:32:06 Is a BBB image backwards compatible with the original BeagleBone> Jul 01 12:32:09 ? Jul 01 15:05:59 spuz: sounds like you managed to lock up the bus Jul 01 15:06:13 or the accelerometer Jul 01 15:06:33 in general, neither i2cdetect nor i2cdump is a safe thing to run Jul 01 15:07:00 you should only perform register accesses documented in the datasheet of the accelerometer Jul 01 15:07:12 (or use the kernel driver if one is available for it) Jul 01 15:10:49 i2c is not a discoverable bus. i2cdetect attempts to detect the presence of a device in a way that hopefully doesn't mess it up, but that's definitely not guaranteed ("By default, the command used is the one believed to be the safest for each address." -- i2cdetect manpage) Jul 01 15:11:25 or, again quoting the i2cdetect manpage, "WARNING This program can confuse your I2C bus, cause data loss and worse!" Jul 01 19:35:00 zmatt: thanks, yes i figured that i2cdump was upsetting the bus some how. after a reset, and restricting i2cdump to the first 4 rows of data everything seems to be working correctly Jul 01 20:05:17 spuz: more specifically, upsetting the accelerometer Jul 01 20:05:57 :) Jul 01 20:06:07 maybe it crashed, hehe Jul 01 20:11:24 hmm if that was the case then i wouldn't expect i2cdetect to start behaving strangely Jul 01 20:20:41 why not? what did the kernel log say btw? Jul 01 20:21:55 it is easy to envision the i2c controller in the accelerometer keeping the i2c bus stalled while waiting for the core of the accelerometer to respond **** ENDING LOGGING AT Sun Jul 02 03:00:01 2017