Wondering what options left after unable to see xavier agx with sdkmanager

I just opened the Xavier AGX and powered it up. I tried to put into recovery mode (power, recovery hold, press release reset, release recovery button), but only found xavier one time in the sdkmanager software (1.3.7110). I have not been able to repeat it.

During boot, I see output on the serial monitor, but it eventually stops. On the screen connected to a monitor over hdmi, it sort of keeps blinking (the screen) and it is stuck on 'Started update UTMP about system RUNLEVEL changes…"
Is there some way to recover or maybe get it back to factory state? Or what is the best option?
I am currently trying to get the files to load onto USB, but not sure if that is a good next step?
Thanks!

[0010.595] I> Continuing with default option: 1
[0010.595] I> Loading kernel sig file from rootfs ...
[0010.595] I> rootfs path: /sdmmc_user/boot/Image.sig
[0010.614] I> Loading kernel binary from rootfs ...
[0010.614] I> rootfs path: /sdmmc_user/boot/Image
[0010.834] I> Validate kernel ...
[0010.834] I> T19x: Authenticate kernel (bin_type: 37), max size 0x5000000
[0011.194] I> No kernel-dtb binary path
[0011.194] I> Loading kernel-dtb binary from partition ...
[0011.194] W> No valid slot number is found in scratch register
[0011.195] W> Return default slot: _a
[0011.195] I> A/B: bin_type (38) slot 0
[0011.195] I> Loading partition kernel-dtb at 0x91000000 from device(0x1)
[0011.206] I> Validate kernel-dtb ...
[0011.206] I> T19x: Authenticate kernel-dtb (bin_type: 38), max size 0x400000
[0011.212] I> Loading ramdisk from rootfs ...
[0011.212] I> rootfs path: /sdmmc_user/boot/initrd
[0011.265] I> Kernel hdr @0xa4ab0000
[0011.266] I> Kernel dtb @0x90000000
[0011.266] I> decompressor handler not found
[0011.266] I> Copying kernel image (34191376 bytes) from 0xa4ab0000 to 0x800800e
[0011.285] I> Updated bpmp info to DTB
[0011.286] I> Ramdisk: Base: 0x92000000; Size: 0x54eadd
[0011.287] I> Updated initrd info to DTB
[0011.287] W> WARN: Fail to override "console=none" in commandline
[0011.287] E> tegrabl_linuxboot_add_disp_param, du 1 failed to get display paras
[0011.288] E> tegrabl_linuxboot_add_disp_param, du 1 failed to get display paras
[0011.293] W> No valid slot number is found in scratch register
[0011.299] W> Return default slot: _a
[0011.302] I> Active slot suffix: 
[0011.306] I> add_boot_slot_suffix: slot_suffix = 
[0011.310] I> Linux Cmdline: root=/dev/mmcblk0p1 rw rootwait rootfstype=ext4 co 
[0011.349] I> Updated bootarg info to DTB
[0011.353] W> MAC addr invalid!
[0011.356] E> Failed to get WIFI MAC address
[0011.360] W> MAC addr invalid!
[0011.362] E> Failed to get Bluetooth MAC address
[0011.367] I> eeprom_get_mac_addr: MAC (type: 2): 48:b0:2d:2b:80:e4
[0011.374] E> "plugin-manager" doesn't exist, creating
[0011.378] I> Adding /chosen/plugin-manager/cvm
[0011.382] E> "chip-id" doesn't exist, creating
[0011.386] I> Adding /chosen/plugin-manager/chip-id
[0011.391] E> "configs" doesn't exist, creating
[0011.395] I> Adding /chosen/plugin-manager/configs
[0011.400] E> "ids" doesn't exist, creating
[0011.404] I> Adding /chosen/plugin-manager/ids
[0011.409] E> "odm-data" doesn't exist, creating
[0011.413] I> Adding /chosen/plugin-manager/odm-data
[0011.421] E> "memory" doesn't exist, creating
[0011.422] I> [0] START: 0x80000000, END: 0xac000000
[0011.426] I> [1] START: 0xac004000, END: 0xf09d0000
[0011.431] I> [2] START: 0xf09dc000, END: 0xf09e0000
[0011.436] I> dram_block larger than 80000000
[0011.440] I> [3] START: 0x100000000, END: 0x880000000
[0011.445] I> added [base:0x80000000, size:0x2c000000] to /memory
[0011.451] I> added [base:0xac200000, size:0x44600000] to /memory
[0011.456] I> added [base:0x100000000, size:0x780000000] to /memory
[0011.464] I> Updated memory info to DTB
[0011.466] E> add_disp_param: failed to get display params for du=1
[0011.473] E> "reset" doesn't exist, creating
[0011.477] I> NVG: Logical CPU: 0; MPIDR: 0x80000000
[0011.481] I> Enabled cpu-0 (mpidr: 0x0) node in FDT
[0011.486] I> NVG: Logical CPU: 1; MPIDR: 0x80000001
[0011.491] I> Enabled cpu-1 (mpidr: 0x1) node in FDT
[0011.495] I> NVG: Logical CPU: 2; MPIDR: 0x80000100
[0011.500] I> Enabled cpu-2 (mpidr: 0x100) node in FDT
[0011.505] I> NVG: Logical CPU: 3; MPIDR: 0x80000101
[0011.510] I> Enabled cpu-3 (mpidr: 0x101) node in FDT
[0011.514] I> NVG: Logical CPU: 4; MPIDR: 0x80000200
[0011.519] I> Enabled cpu-4 (mpidr: 0x200) node in FDT
[0011.524] I> NVG: Logical CPU: 5; MPIDR: 0x80000201
[0011.529] I> Enabled cpu-5 (mpidr: 0x201) node in FDT
[0011.534] I> NVG: Logical CPU: 6; MPIDR: 0x80000300
[0011.538] I> Enabled cpu-6 (mpidr: 0x300) node in FDT
[0011.543] I> NVG: Logical CPU: 7; MPIDR: 0x80000301
[0011.548] I> Enabled cpu-7 (mpidr: 0x301) node in FDT
[0011.554] E> "misc-data" doesn't exist, creating
[0011.558] I> Boot-device: eMMC
[0011.560] I> Add boot-sdmmc to plugin-manager/misc-data
[0011.566] I> Add storage-sdmmc to plugin-manager/misc-data
[0011.571] W> Unknown storage device
[0011.574] I> Add serial number:1423620010089 as DT property
[0011.581] I> Plugin-manager override starting
[0011.584] I> node /plugin-manager/fragement-tegra-wdt-en matches
[0011.591] I> node /plugin-manager/fragement-soft-wdt matches
[0011.599] I> node /plugin-manager/fragment-pcie-c5-rp matches
[0011.603] I> node /plugin-manager/fragment-tegra-ufs-lane10 matches
[0011.614] I> Disable plugin-manager status in FDT
[0011.615] I> Plugin-manager override finished successfully
[0011.617] I> tegrabl_load_kernel_and_dtb: Done
[0011.646] I> Kernel EP: 0x80080000, DTB: 0x90000000
[    0.000000] Booting Linux on physical CPU 0x0
[    0.000000] Linux version 4.9.140-tegra (buildbrain@mobile-u64-2390-d3000) (9
[    0.000000] Boot CPU: AArch64 Processor [4e0f0040]
[    0.000000] OF: fdt:memory scan node memory, reg size 48,
[    0.000000] OF: fdt: - 80000000 ,  2c000000
[    0.000000] OF: fdt: - ac200000 ,  44600000
[    0.000000] OF: fdt: - 100000000 ,  780000000
[    0.000000] earlycon: tegra_comb_uart0 at MMIO32 0x000000000c168000 (options)
[    0.000000] bootconsole [tegra_comb_uart0] enabled
[    0.000000] Found tegra_fbmem: 00800000@a069b000
[    0.000000] Found lut_mem: 00002008@a0696000
<hit enter to activate fiq debugger>
��WARNING: pll_d3 has no dyn ramp
��[    6.157061] cgroup: cgroup2: unknown option "nsdelegate"
[    7.443828] random: crng init done
[    7.443964] random: 7 urandom warning(s) missed due to ratelimiting
[    7.479171] using random self ethernet address
[    7.479340] using random host ethernet address
[    9.326910] CPU4: shutdown
<hit enter to activate fiq debugger>
��WARNING: pll_d3 has no dyn ramp
��[    6.157061] cgroup: cgroup2: unknown option "nsdelegate"
[    7.443828] random: crng init done
[    7.443964] random: 7 urandom warning(s) missed due to ratelimiting
[    7.479171] using random self ethernet address
[    7.479340] using random host ethernet address
[    9.326910] CPU4: shutdown
[    9.392214] CPU5: shutdown
[    9.450996] CPU6: shutdown
[    9.516020] CPU7: shutdown

Ubuntu 18.04.3 LTS linux ttyTCU0

Hi,

Just try the same steps again and it will enter recovery mode again.
During recovery mode, the device has no OS running so there would be nothing on monitor. You can only check the board is in recovery mode by command “lsusb” on your host. There should be a “Nvidia Corp” device on it.

2 Likes

which recovery mode works best?

  • Press and release power button
  • press and hold recovery button
  • press and release reset
  • release recovery after 2 seconds
    ?

so, this pattern works on other machines, but i can’t seem to get this one into recovery mode. It doesn’t seem bricked, but it doesn’t like recovery mode.
Any suggestions?

The instructions in SDK Manager is well and good.

Press and hold the middle button then press power button release both.

I think the problem you face is due to wrong port, connect the Host PC to the USB-C on the board where LED indicator is. You problem should be solved.

Enjoy booting😉😉😉 @wegunterjr

1 Like

I wish it was the wrong port, it is in that exact port you mentioned.
I am all happy with a different device and am making progress on that one by following the instructions. Not sure what’s up with the other. I will revisit it and update if I have success.
Thanks for the suggestions and encouragement!

The recovery mode in AGX Xavier is simple.

Connect the Host PC and Xavier with USB-C cable.
Make sure you connect it where LED indicator is.

Plug the power cable

  1. press and hold the middle button
  2. along with it press power button
  3. Release both

Now in Host PC terminal check $ lsusb
It will show NVIDIA CORP

Then continue flashing SDK Manager will do the job😉

1 Like

I think there is a easier way to understand when to press the power button.

If you board just plug out and in the power supply cable, your board is now still not powered on.
In such case,

  1. press the recovery button
  2. press the power button (power LED should be ON now)
  3. release the power button
  4. release the recovery button

If your board is already in powered-on state, then it is pointless to press a power key again because it will not trigger anything. In such case, do the above 4 steps again but replace the power button to reset button in step 2,3.

So the timing to use either power key or reset key depends on your power status of your board.

1 Like

valid comments, but i think my device is bad. All my other devices work just fine with this suggested method, but not this one. It doesn’t seem to want to work.
I appreciate the help. It gave me confidence that it wasn’t me, and I was able to get another device working.

1 Like