3.12 and 3.11 kernel reboots sporadically on BBB

This forum is for supported devices using an ARMv7 Texas Instruments (TI) SoC.

3.12 and 3.11 kernel reboots sporadically on BBB

Postby rsainio » Fri Jan 17, 2014 11:39 am

I have updated to some recent kernels in 3.11 and 3.12 versions and noticed that after a while BBB reboots. This happens most of the time within 11 - 20 hours. I have several BBB-boards and they act in a similar way. I first suspected my home-made 5 V powersupply based on LM 2596, but after having linux-am33x-legacy 3.8.13-13 running for extended period stable I tend to think that it is the kernel.
On serial console one sees only that the bootloader starts loading the kernel again. No error messages from the linux kernel before the bootloader takes over.
I am happy to switch some additional debugging on if needed and try to reproduce the problem.
rsainio
 
Posts: 2
Joined: Fri Jan 17, 2014 11:24 am

Re: 3.12 and 3.11 kernel reboots sporadically on BBB

Postby kmihelich » Fri Jan 17, 2014 1:42 pm

I haven't seen this problem at all on two BBBs. Power supply to the boards is critical, I would start there seeing as it's homemade.
Arch Linux ARM exists and continues to grow through community support, please donate today!
kmihelich
Developer
 
Posts: 1133
Joined: Tue Jul 20, 2010 6:55 am
Location: aka leming #archlinuxarm

Re: 3.12 and 3.11 kernel reboots sporadically on BBB

Postby rsainio » Fri Jan 17, 2014 2:12 pm

The power supply delivers 3A @5V and I used the same design for 3 RasPi-boards as well - they run 24x7. Besides both BBB boards being tested run properly with 3.8 kernel. Was there a change in way the internal power handling is done in 3.11 and 3.12 kernels ?
rsainio
 
Posts: 2
Joined: Fri Jan 17, 2014 11:24 am

Re: 3.12 and 3.11 kernel reboots sporadically on BBB

Postby N3rdle » Mon Feb 17, 2014 8:15 pm

so, ive updated to the 3.13 kernel which has corrected a number of other device tree mapping issues which is excellent, but i too notice something similar to what you've reported. when i do a *reasonable* amount of io to the emmc device and do a "sync", I've got about a 50 percent chance of some sort of kernel panic(assumed because the box just seems to disappear).

anyone else seeing this? im guessing your crashes every 11 hours or so might be caused by something doing io on the mmc, such as an indexing operation, etc.

cheers

Peter
N3rdle
 
Posts: 1
Joined: Mon Feb 17, 2014 8:09 pm


Return to Texas Instruments (TI)

Who is online

Users browsing this forum: No registered users and 3 guests