• State Answer Suggested
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 100 subscribers
  • Views 589 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

Linux/AM5718: Problem with Linux booting on AM5718

user4759774
Intellectual 745 points
Part Number: AM5718
Other Parts Discussed in Thread: AM5728

Tool/software: Linux

Dear All,

I do have quite strange problem - I have one carrier board and two SoC modules (with AM5718 and AM5728)

In u-boot both works correctly.

However, problem starts in Linux.

In Linux only AM5728 works (even with tunned dts file for AM5718).

From my  "blinking LED" debugging it seems like the SoC is going to sleep (kernel idle thread) without waking up. Also the console (which was working in u-boot) is dead.

I took  linux-4.4.19+gitAUTOINC+db0b54cdad-gdb0b54cdad file arch/arm/boot/dts/am571x-idk.dts as a template.

What is strange - AM5728 works with this dts prepared for AM5718.

One more issue:

- According to 

http://www.ti.com/lit/an/sprabx8c/sprabx8c.pdf

There are some ROM modification for AM5718. Is this important to my problem?

And one more remark - have anybody was able to force DEBUG_LL and EARLY_PRINTK to work on recent (v4.9) linux kernel? Moreover, is it working on reference TI's linux 4.4 kernel?

Thanks in advance,

Łukasz

  • 0
    Intellectual 745 points
    Hi all,

    The problem was not related to any ROM modifications, but differences with default (out of reset) configurations regarding AM572x and AM571x.

    In my case - I cannot reuse PCIe code directly since PCIe2 pins are by default on AM5718 muxed to USB3.0. However, on AM5728 they are dedicated and muxed to PCIe2.

    With misconfigured pins I do experience system hanging on AM571x.

    And one more remark - in my case DEBUG_LL and EARLY_PRINTK works with ICEDCC enabled - DCC console via JTAG.

    BR,
    Łukasz
  • 0 in reply to user4759774
    TI__Guru**** 170625 points
    Hi Lukasz,

    It seems that you have fix it, correct? Thanks for sharing that info with the community.

    Regards,
    Pavel
  • 0 in reply to Pavel Botev
    Intellectual 745 points
    Hi Pavel,

    I've found the reason of Linux hang, but didn't (yet) come up with a proper solution (I've started another thread on this).

    BR,
    Łukasz