• State
  • Replies 3 replies
  • Subscribers 99 subscribers
  • Views 49 views
  • Users 0 members are here
Support feedback
Options
Options
Related

AM3352: The system sometimes fails to startup

zhang lumin
Prodigy 30 points
Part Number: AM3352

Tool/software:

Our CPU is used in HMI products，We are currently encountering several devices using the vxworks system, When Uboot started to load the kernel, the system did not jump successfully and the startup froze，the CPU is AM3352.The specific phenomenon is as follows

1)  The devices that have encountered problems initially did not start successfully, but after repeated power on and off restarts, they may start normally. If the startup is successful, it can still be successfully started by powering off and then powering on again. However, if the power is turned off and restarted after waiting for a few minutes, the startup will still fail.

2）We tried using other identical devices with the same version but were unable to reproduce the issue. Currently, only five devices will consistently reproduce this issue;

3)  We have taken X-rays of the CPU and DDR and have not found any issues with virtual soldering so far；

4）We are currently trying to replace the DDR chip for comparative testing，we use ddr3 now（This requires a relatively long time and has not been completed yet）；

5）We also tested the main power and clock of the CPU and DDR, and did not find any problems, but we are not sure if there are any other parts that need to be tested that we did not detect；

6）In addition, we have discovered a strange phenomenon where we switch the boot mode sysboot0 to 3 through hardware, and after power off and power on, we can  restart it again to start normally，we have checked evenif porting the code operations performed on this special mode in uboot to normal mode and then restarting the device still fails to start successfully， So I don't know what other controls the CPU has for normal mode and special mode with sysboot0=3；

Although this phenomenon seems to be an individual hardware issue, we are unsure if there is any defect on hardware or software as we encounter the same phenomenon for all five devices，

Do you have any further analysis suggestions ?

thanks

  • 0
    TI__Guru 68600 points

    1/. have we performed the HW A/B swap test? for example, swap the SoC on the non-working board with the SoC on the working board, and vice versa...
    2/. How many boards were tested?
    Best,
    -Hong

  • 0 in reply to Hong Guan64
    Prodigy 30 points

    Hi
    1、Because we cannot directly swap SOC or DDR in the laboratory, we need help from the factory to replace it. Therefore, we are arranging for the factory to operate it, but it has not been completed yet；

    2、About a few dozen pieces will happen once, with a relatively high probability；

    Because the probability is relatively high and the phenomena are consistent, we suspect that there may be some software or design flaws

    Is there any information that can be exported for comparison and judgment? Such as some configuration or DDR information, etc，

  • 0 in reply to zhang lumin
    TI__Guru 68600 points

    It'll be very helpful to perform the HW A-B-A swap test to isolate the HW issue on the board.

    zhang lumin said:
    6）In addition, we have discovered a strange phenomenon where we switch the boot mode sysboot0 to 3 through hardware, and after power off and power on, we can  restart it again to start normally，we have checked evenif porting the code operations performed on this special mode in uboot to normal mode and then restarting the device still fails to start successfully， So I don't know what other controls the CPU has for normal mode and special mode with sysboot0=3；

    I guess you were reconfiguring SYSBOOT[15:0], will you share more details on the changes, and test result.

    Best,
    -Hong