• State
  • Locked Locked
  • Replies 7 replies
  • Subscribers 99 subscribers
  • Views 780 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

USB hangs on Begale bone, Debian 6.0.5 Linux 3.2.28-psp21.1 armv7l

Gustav Bergquist

Hi!

We use beagle bone in an embedded application and we have problem with the USB-host. 

We need 3 ports and thus we use a  hub with external power supply (we have tried different brands of usb hub and currently a d-link hub is used PN: EUBH4EB......C1E).

In total the power supply for beagle bone and usb-hub is 5V 3A.

USB hub port device:

1. SanDisk Cruzer Blade 4GB memstick

2 .FT4232

3. ATMEL at32uc3c1512 (while in bootloader, when in application no usb device is active on port)

4. Not connected

The problem we have is that sometimes the kernel reports

ti81xx_interrupt 956: Please issue long reset to make usb functional !!


The only soulution we have so far is to do a soft reboot of the beagle bone. In our application this is not acceptable and we hope you can help us to find a proper solution to this problem. 

  • 0
    TI__Mastermind 25235 points

    Hi Gustav,

    I'm sorry, but we don't support Debian OS. The print output you have pasted is not available in the TI supported AM335x PSP and is most probably Debian OS specific. Perhaps you will find support on http://beagleboard.org/bone or by following the links posted here. There is a mailing list, as well as an IRC channel.

    Best regards,
    Miroslav

  • 0 in reply to Miroslav Kiradzhiyski XID

    Hi,

    When looking at the source code to the kernel it's the ti81xx_ driver that at least detects this problem. 
    This is written by TI and i thought you'd support that.

    I'm just curious if there's anyway to reset the usb through software? This could be acceptable if it can be done quick enough. 

  • 0 in reply to Gustav Bergquist
    TI__Mastermind 25235 points

    Hi,

    The driver in the Debian distribution seems to differ compared to the one we have in the Arago distribution. There is no such print:

    Please issue long reset to make usb functional !!

    Perhaps the driver has been modified?

    Are there any other helpful kernel log messages?

    I believe you can power off a USB port with:

    echo suspend >/sys/bus/usb/devices/usb42/power/level
    echo disabled >/sys/bus/usb/devices/usb42/power/wakeup

    Best regards,
    Miroslav

  • 0 in reply to Miroslav Kiradzhiyski XID
    TI__Guru**** 168211 points

    Miroslav,

    That message is printed from ti81xx.c due to Babble interrupt happening, but from an older version of PSP. The latest driver has been modified to control Babble condition which is available since PG2.x.

    Gustav,

    The current Beaglebone still uses AM335x PG1.0 which has no s/w control on Babble condition. But PG2.1 has s/w control, and the latest PSP already support it. The recently released Beablebone Black has PG2.x on it.

    When the Babble interrupt happens in your setup? while connecting the hub without any device? or while connecting any of the devices?

  • 0 in reply to Bin Liu
    TI__Guru**** 168211 points

    Is the Beaglebone also powered from the HUB through the USB port? If so, could you please try to power it from a dedicated power supply?

    To rule out any s/w variables, could you please try to use the prebuilt binaries from SDK release? including uboot, uimage, and root filesystem? because that is what has been validated by TI.

    You can download the SDK prebuilt package from: am335x-evm-sdk-bin-05.07.00.00.tar.gz

  • 0 in reply to Bin Liu

    The HUB is powered from a dedicated source. 

    The babble condition happens randomly. The beagle bone is the user frontend for a 5kW inverter and there are significant noise when emergency condition is issued, however this have been tested thoroughly and the EMC-noise does not seem to affect the USB. 

    We use BeagleBone rev A6, does this support PSP 2.x ? Provided we use the right software

    Thank you for all your help

  • 0 in reply to Gustav Bergquist

    We will try the attached file. 

    The condition occours, as stated in previous message, randomly. 

    Constant factors are: Powered up and running for some time ( more than 30 min) and all USB-devices are connected from power up.

    They are never physically disconnected or attached while power is applied to the system.