BeagleBone kernel hung when powered via USB

GoogleGroups
1

Hi,
I have a problem when start the BB via USB, or when attach USB for debug via console.
The kernel, version 3.2.32-psp25 of Ubuntu 12.04.1, hungs writing this log:

Jan 1 01:21:26 scatoloarm kernel: [ 240.638078] INFO: task init:1 blocked for more than 120 seconds. Jan 1 01:21:26 scatoloarm kernel: [ 240.644413] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. Jan 1 01:21:26 scatoloarm kernel: [ 240.652679] init D c04e1384 0 1 0 0x00000000 Jan 1 01:21:26 scatoloarm kernel: [ 240.652707] Backtrace: Jan 1 01:21:26 scatoloarm kernel: [ 240.652769] [] (__schedule+0x0/0x43c) from [] (schedule+0x78/0x7c) Jan 1 01:21:26 scatoloarm kernel: [ 240.652807] [] (schedule+0x0/0x7c) from [] (schedule_timeout+0x20/0x16c) Jan 1 01:21:26 scatoloarm kernel: [ 240.652845] [] (schedule_timeout+0x0/0x16c) from [] (wait_for_common+0xc8/0x154) Jan 1 01:21:26 scatoloarm kernel: [ 240.652866] r6:00000002 r5:cf837abc r4:7fffffff Jan 1 01:21:26 scatoloarm kernel: [ 240.652906] [] (wait_for_common+0x0/0x154) from [] (wait_for_completion+0x18/0x1c) Jan 1 01:21:26 scatoloarm kernel: [ 240.652947] [] (wait_for_completion+0x0/0x1c) from [] (flush_work+0x30/0x3c) Jan 1 01:21:26 scatoloarm kernel: [ 240.652987] [] (flush_work+0x0/0x3c) from [] (tty_flush_to_ldisc+0x14/0x18) Jan 1 01:21:26 scatoloarm kernel: [ 240.653020] [] (tty_flush_to_ldisc+0x0/0x18) from [] (n_tty_poll+0x7c/0x180) Jan 1 01:21:26 scatoloarm kernel: [ 240.653069] [] (n_tty_poll+0x0/0x180) from [] (tty_poll+0x68/0x88) Jan 1 01:21:26 scatoloarm kernel: [ 240.653088] r7:00000000 r6:cef6db40 r5:cee09400 r4:cf36cb40 Jan 1 01:21:26 scatoloarm kernel: [ 240.653137] [] (tty_poll+0x0/0x88) from [] (do_select+0x2b8/0x460) Jan 1 01:21:26 scatoloarm kernel: [ 240.653156] r7:00000000 r6:00000002 r5:00000200 r4:00000400 Jan 1 01:21:26 scatoloarm kernel: [ 240.653197] [] (do_select+0x0/0x460) from [] (core_sys_select+0x244/0x308) Jan 1 01:21:26 scatoloarm kernel: [ 240.653231] [] (core_sys_select+0x0/0x308) from [] (sys_select+0xdc/0x10c) Jan 1 01:21:26 scatoloarm kernel: [ 240.653271] [] (sys_select+0x0/0x10c) from [] (ret_fast_syscall+0x0/0x30)

The message is repeated until I disconnect USB cable and the boot is completed.

Someone have this issue?

Bye

Luigi

2

At least upgrade that old v3.2.x kernel to the 'last' v3.2.x:

wget http://rcn-ee.net/deb/precise-armhf/v3.2.42-psp27/install-me.sh
sudo /bin/bash install-me.sh

(use the 5volt plug when doing it.)

But you should really look to switching to the v3.8.x tree..

Regards,

3

I think that it’s some hardware problem.
The last supplying, same revision (A6a) and same PCB revision (C2), with the same image start correctly.

Regards

4

Can I please get more detail on the specific HW issue?

Gerald

5

gigigui, the BBB can use up to around 460mA under full load according to the SRM. A typical USB 2.0 port can supply up to 500mA . . . assuming it works correctly.

LIke Gerald says you need to give more details. Otherwise, the sky is blue and . . .

6

Ok.

First supplying (december 2012 / january 2013):

  • demo image with Ubuntu 12.04 and kernel 3.2.32-psp25
  • powered from my notebook USB port
  • the boot hungs with the log that I reported
  • tested with about 10 boards

Second supplying (june 2013):

  • same board revision and PCB revision
  • same demo image
  • same notebook and same USB port
  • the boot process is ok and I can use the board withouth external power
  • tested with 3 board

I do not have a card of the old supply so I cannot remake the test.

So I think it’s some hw problem in that boards.

7

Why so many boards? What are you using these boards for?

Gerald

8

I was going to say with so many boards, and presumably using the same power supply, it is very unlikely a BB hw issue . . .