• State Resolved
  • Locked Locked
  • Replies 14 replies
  • Answers 1 answer
  • Subscribers 97 subscribers
  • Views 968 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.

SK-AM62A-LP: Unable to boot up

Maurice Lu
Prodigy 160 points

Part Number: SK-AM62A-LP
Other Parts Discussed in Thread: AM62A7, SK-AM62B-P1, , SK-AM64B, SYSCONFIG

Hi Sir,

We are facing problem with boot up EVM.

We have downloaded file and create MicroSD card image through Quick Start Guide, but the UART log did not allow me to enter "root".

There are 3 different logs:

Log 1:

[ 218.276979] rcu: INFO: rcu_preempt detected expedited stalls on CPUs/tasks: { 0-.... } 52881 jiffies s: 93 root: 0x1/.
[ 218.287696] rcu: blocking rcu_node structures (internal RCU debug):
[ 218.293961] Task dump for CPU 0:
[ 218.297190] task:swapper/0 state:R running task stack:0 pid:0 ppid:0 flags:0x0000000a
[ 218.307101] Call trace:
[ 218.309548] __switch_to+0xd4/0x130
[ 218.313049] 0xffee280c
[ *** ] A start job is running for D-Bus Sy＃ Message Bus (3min 35s / 4min 33s)
[ ***] A start job is running for D-Bus Sy＃ Message Bus (3min 35s / 4min 33s)

Log 2:

[ OK ] Reached target Basic System.
[ 27.645029] rcu: INFO: rcu_preempt self-detected stall on CPU
[ 27.650781] rcu: 0-....: (5249 ticks this GP) idle=433c/1/0x4000000000000000 softirq=1387/1393 fqs=2602
[ 27.660247] (t=5253 jiffies g=345 q=23085 ncpus=4)
[ 27.665117] CPU: 0 PID: 227 Comm: systemd-udevd Tainted: G O 6.1.46-g247b2535b2 #1
[ 27.674058] Hardware name: Texas Instruments AM62A7 SK (DT)
[ 27.679615] pstate: 40000005 (nZcv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[ 27.686561] pc : _stext+0xa4/0x2a4
[ 27.689961] lr : _stext+0x6c/0x2a4
[ 27.693353] sp : ffff800008003f60
[ 27.696656] x29: ffff800008003f60 x28: ffff000801788000 x27: 0000000000000280
[ 27.703782] x26: 0000000000000000 x25: 0000000000000000 x24: ffff8000090bd000
[ 27.710906] x23: 0000000060000005 x22: ffff8000090c3000 x21: ffff8000090bd9c8
[ 27.718031] x20: ffff8000092260c0 x19: ffff000801788000 x18: 0000000000000000
[ 27.725156] x17: ffff800876775000 x16: ffff800008000000 x15: 00009875fe97ce38
[ 27.732280] x14: 0000000000000312 x13: 0000000000000001 x12: 0000000000000000
[ 27.739404] x11: ffff00087f8391c0 x10: ffff80000923fcb0 x9 : ffff800009229d38
[ 27.746528] x8 : 000000018091b500 x7 : 7fffffffffffffff x6 : 00000000462690a3
[ 27.753652] x5 : 03ffffffffffffff x4 : 0000000000000015 x3 : ffff000801788000
[ 27.760777] x2 : ffff800876775000 x1 : 00000000000000e0 x0 : ffff8000090c3b00
[ 27.767902] Call trace:
[ 27.770338] _stext+0xa4/0x2a4
[ 27.773384] ____do_softirq+0x10/0x20
[ 27.777038] call_on_irq_stack+0x24/0x4c
[ 27.780952] do_softirq_own_stack+0x1c/0x30
[ 27.785125] __irq_exit_rcu+0xcc/0xf4
[ 27.788778] irq_exit_rcu+0x10/0x20
[ 27.792258] el1_interrupt+0x38/0x70
[ 27.795829] el1h_64_irq_handler+0x18/0x2c
[ 27.799914] el1h_64_irq+0x64/0x68
[ 27.803306] finish_task_switch.isra.0+0x80/0x230
[ 27.808002] __schedule+0x234/0x664
[ 27.811483] schedule+0x5c/0xcc
[ 27.814616] do_nanosleep.constprop.0+0x5c/0x160
[ 27.819226] hrtimer_nanosleep+0xa0/0x120
[ 27.823228] common_nsleep+0x44/0x70
[ 27.826797] __arm64_sys_clock_nanosleep+0xcc/0x150
[ 27.831663] invoke_syscall+0x48/0x114
[ 27.835404] el0_svc_common.constprop.0+0xd4/0xfc
[ 27.840098] do_el0_svc+0x20/0x30
[ 27.843405] el0_svc+0x28/0xa0
[ 27.846453] el0t_64_sync_handler+0xbc/0x140
[ 27.850712] el0t_64_sync+0x18c/0x190
[ 27.854378] rcu: INFO: rcu_preempt detected expedited stalls on CPUs/tasks: { 0-.... } 5287 jiffies s: 113 root: 0x1/.
[ 27.865071] rcu: blocking rcu_node structures (internal RCU debug):
[ 27.871331] Task dump for CPU 0:
[ 27.874555] task:systemd-udevd state:R running task stack:0 pid:227 ppid:198 flags:0x00000a0e
[ 27.884468] Call trace:
[ 27.886912] __switch_to+0xd4/0x130
[ 27.890401] 0xffff000801788000

Log 3:

Arago Project am62axx-evm -

Arago 2023.10 am62axx-evm -

am62axx-evm login: ***************************************************************
***************************************************************
NOTICE: This file system contains the following GPL-3.0 packages:
[ 20.617041] omap_i2c 20010000.i2c: controller timed out

Thanks for your help.

Maurice

  • 0
    TI__Guru**** 181996 points

    Hi Maurice,

    How do you power the EVM? Please ensure to use a type-C power supply which is minimum 5V 3A.

    Maurice Lu said:
    We have downloaded file and create MicroSD card image through Quick Start Guide, but the UART log did not allow me to enter "root".

    Please ensure UART flow control is disabled in the terminal program.

  • 0 in reply to Bin Liu
    Prodigy 160 points

    Hi Bin,

    Yes I'm using NB 45W adapter on EVM.

    We have tried teraterm and CCS and cannot find UART flow control in program.

    Somehow we can enter "root" cammand in random boot up operation but still cannot have display output.

    Log as below:

    [ OK ] Finished Print notice about GPLv3 packages.

    _____ _____ _ _
    | _ |___ ___ ___ ___ | _ |___ ___ |_|___ ___| |_
    | | _| .'| . | . | | __| _| . | | | -_| _| _|
    |__|__|_| |__,|_ |___| |__| |_| |___|_| |___|___|_|
    |___| |___|

    Arago Project am62axx-evm -

    Arago 2023.10 am62axx-evm -

    am62axx-evm login: root
    [ 21.052582] audit: type=1006 audit(1651167763.592:10): pid=1013 uid=0 old-auid=4294967295 auid=0 tty=(none) old-ses=4294967295 ses=3 res=1
    [ 21.065167] audit: type=1300 audit(1651167763.592:10): arch=c00000b7 syscall=64 success=yes exit=1 a0=8 a1=ffffefcc3738 a2=1 a3=ffff9ddb8020 items=0 ppid=1 pid=1013 auid=0 uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0 sgid=0 fsgid=0 tty=(none) ses=3 comm="(systemd)" exe="/lib/systemd/systemd" key=(null)
    [ 21.091601] audit: type=1327 audit(1651167763.592:10): proctitle="(systemd)"
    [ 21.098690] audit: type=1334 audit(1651167763.616:11): prog-id=11 op=LOAD
    [ 21.105538] audit: type=1300 audit(1651167763.616:11): arch=c00000b7 syscall=280 success=yes exit=8 a0=5 a1=ffffc40a0950 a2=78 a3=0 items=0 ppid=1 pid=1013 auid=0 uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0 sgid=0 fsgid=0 tty=(none) ses=3 comm="systemd" exe="/lib/systemd/systemd" key=(null)
    [ 21.131012] audit: type=1327 audit(1651167763.616:11): proctitle="(systemd)"
    [ 21.138110] audit: type=1334 audit(1651167763.632:12): prog-id=11 op=UNLOAD
    [ 21.145119] audit: type=1334 audit(1651167763.632:13): prog-id=12 op=LOAD
    [ 21.151948] audit: type=1300 audit(1651167763.632:13): arch=c00000b7 syscall=280 success=yes exit=8 a0=5 a1=ffffc40a09f0 a2=78 a3=0 items=0 ppid=1 pid=1013 auid=0 uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0 sgid=0 fsgid=0 tty=(none) ses=3 comm="systemd" exe="/lib/systemd/systemd" key=(null)
    [ 21.177425] audit: type=1327 audit(1651167763.632:13): proctitle="(systemd)"

    Is there any suggestion?

    Or can I send the complete UART log to you by email?

    (The internet policy in our company do not allow us to upload files through web page.)

    Thanks. B,R.

    Maurice

  • 0 in reply to Maurice Lu
    TI__Guru**** 181996 points

    Hi Maurice,

    Maurice Lu said:
    We have tried teraterm and CCS and cannot find UART flow control in program.

    I only use Linux and am not familiar with Teraterm. flow control should be in somwhere in the program settings. Have you searched on the internet?

    Maurice Lu said:
    Somehow we can enter "root" cammand in random boot up operation but still cannot have display output.

    Do you mean the display on the HDMI port? If so, we can leave it for later, my main concern is the rcu_preempt related messages in your initial post above? Do you still see them in every Linux boot?

  • 0 in reply to Bin Liu
    Prodigy 160 points

    Hi Bin,

    We've tried flow control "none" & "hardware" & "XON/XOFF" in Teraterm, all seems no improvement.

    Do you mean the messages like:

    [ 27.612959] rcu: INFO: rcu_preempt detected stalls on CPUs/tasks:
    [ 27.619052] rcu: 0-....: (14 GPs behind) idle=b31c/0/0x3 softirq=939/940 fqs=2578

    Yes it shows up in every Linux boot, and every time it looks a little bit different.

    Also, it only did not show up in the situation I can enter "root" command.

    Thanks.

    Maurice

  • 0 in reply to Maurice Lu
    TI__Guru**** 181996 points
    Maurice Lu said:
    We've tried flow control "none" & "hardware" & "XON/XOFF" in Teraterm, all seems no improvement.

    it should be set to "none" for the console UART.

    Maurice Lu said:

    Do you mean the messages like:

    [ 27.612959] rcu: INFO: rcu_preempt detected stalls on CPUs/tasks:

    Yes.

    Maurice Lu said:

    Yes it shows up in every Linux boot, and every time it looks a little bit different.

    Also, it only did not show up in the situation I can enter "root" command.

    What is your SK EVM version? The board has a silkprint like "PROCXXXX...", what is this string?

    Which Processor SDK version do you use?

    Please attach the full console boot log including the u-boot messages.

    Can you please try different brand of SD cards to see if the issue also happens?

  • 0 in reply to Bin Liu
    Prodigy 160 points

    Hi Bin,

    EVM Version: PROC135E3

    SDK Version: 09.01.00.07 (18 Dec 2023)

    We've tried 16G and 32G SD cards with different brand and both fail.

    Can you provide email or cloud drive link so that we can send files to you directly?

    We are unable to upload files from our side (or even post the logs here).

    Thanks.

    Maurice

  • 0 in reply to Maurice Lu
    TI__Guru**** 181996 points

    Hi Maurice,

    I have send your a friendship request which should includes my email address.

    Do you have another EVM to check if this is a board issue?

  • 0 in reply to Bin Liu
    Prodigy 160 points

    Hi Bin,

    We only have one SK-AM62A-LP, another EVM is SK-AM62B-P1.

    SK-AM62B-P1 is unbootable either, I'll send logs to you once I get the email address.

    I've reached out to you through private message, please check.

    Thanks.

    Maurice

  • 0 in reply to Maurice Lu
    TI__Guru**** 181996 points

    Hi Maurice,

    I have provided my email address, waiting for the boot logs for both boards if the failure doesn't look like the same on both boards.

  • 0 in reply to Bin Liu
    Prodigy 160 points

    Hi Bin,

    We can boot both EVMs into image now.

    SK-AM62A-LP use SDK 9.0.1 and another monitor with pure HDMI cable.

    SK-AM62B-P1 use another monitor with pure HDMI cable.

    But I have another question for the 3rd EVM SK-AM64B:

    How should we make sure it has already booted into image while there is no HDMI port onboard?

    Thanks for your great help.

    Maurice

  • 0 in reply to Maurice Lu
    TI__Guru**** 181996 points

    HI Maurice,

    Maurice Lu said:

    We can boot both EVMs into image now.

    SK-AM62A-LP use SDK 9.0.1 and another monitor with pure HDMI cable.

    SK-AM62B-P1 use another monitor with pure HDMI cable.

    Glad the issue is resolved. But I am curious to know what caused the issues in your initial post, such as unable to type in "root" for login and kernel log shows "rcu_preempt self-detected stall" etc?

    Maurice Lu said:

    But I have another question for the 3rd EVM SK-AM64B:

    How should we make sure it has already booted into image while there is no HDMI port onboard?

    You can connect the Linux console UART to a PC and check the Linux boot log.

  • 0 in reply to Bin Liu
    Prodigy 160 points

    Hi Bin,

    Thanks for the reply.

    We'd like to focus on two bootable EVM first, I will start a new thread for SK-AM64B later.

    We want to change DDR density and other configuration through SysConfig.

    How should we implement new setting into demo image?

    Thanks.

    Maurice

  • 0 in reply to Maurice Lu
    TI__Guru**** 181996 points

    Hi Maurice,

    Maurice Lu said:

    We want to change DDR density and other configuration through SysConfig.

    How should we implement new setting into demo image?

    Please create a new e2e thread for this question so that our DDR expert can provide support.

    Maurice Lu said:
    We'd like to focus on two bootable EVM first, I will start a new thread for SK-AM64B later.

    Can you please summarize how the boot issue ("rcu_preempt self-detected stall") is resolved? by hardware changes, software changes, or something else?

  • 0 in reply to Bin Liu
    Prodigy 160 points

    Hi Bin,

    Sure, will create a new thread.

    Issue "rcu_preempt self-detected stall" -> Change SDK image to 9.0.1 version.

    Issue "Cannot output video" -> Use the monitor with pure HDMI cable, do not use HDMI-VGA dongle.

    Thanks a lot.

    Maurice