Error flashing Drive Pegasus

Hi Nvidia,

I am having trouble flashing my drive pegasus system using SDK manager. Please see the attached screen shots for more information. I have been using the same method to flash my drive xavier platform and never had the same problem. Please advice what to do.

Thank you
Ian.

13:21:16 INFO : Flash Xavier A+B in parallel : -------------------------------------
13:21:16 INFO : Flash Xavier A+B in parallel :
13:21:16 INFO : Flash Xavier A+B in parallel : Tool OutPut:
13:21:16 INFO : Flash Xavier A+B in parallel : USB communication failed.Check if device is in recovery
13:21:16 INFO : Flash Xavier A+B in parallel : Tool OutPut to stderr:
13:21:16 INFO : Flash Xavier A+B in parallel : USB communication failed.Check if device is in recovery
13:21:16 INFO : Flash Xavier A+B in parallel : error-tool-tegrarcm
13:21:16 INFO : Flash Xavier A+B in parallel : bootburn flashing failed! error code = 50
13:21:16 INFO : Flash Xavier A+B in parallel : exit status 50
13:21:16 INFO : Flash Xavier A+B in parallel : [ Component Install Finished with Error ]
13:21:16 ERROR : Flash Xavier A+B in parallel : NV_FLASH_XAVIER_PDKFLASH_PARALLEL_COMP command ./pdk_flash.sh -b e3550b03-t194-es finished with error
13:21:16 INFO : Flash Xavier A+B in parallel :
13:21:16 INFO : Flash Xavier A+B in parallel : [ 604.00 KB used. Disk Avail: 303.35 GB ]
13:21:16 INFO : Flash Xavier A+B in parallel : [ NV_FLASH_XAVIER_PDKFLASH_PARALLEL_COMP Install took 33s ]
13:21:16 INFO : Flash Xavier A+B in parallel : command ./pdk_flash.sh -b e3550b03-t194-es finished with error
13:21:16 ERROR : Flash Xavier A+B in parallel : command terminated with error
13:21:16 ERROR : Flash Xavier A+B in parallel : install 'Flash Xavier A+B in parallel' failure, command <

Dear ian.liu,

Did you connect Host PC and Drive Xavier(USB Debug port) with USB 2.0 Type A cable?
If yes, could you please help to check the platform status via minicom tool?
Please refer to DRIVE OS → Utilitis → Minicom Terminal Emulation in DRIVE™ Software Documentation (ZIP).

Hi Steve,

I am able to loginto Tegra-A console via mincom tool. How do I check the platform status?

when I tried to flash the platform again while minicom is connected. Here was the console log:

nvidia@tegra-ubuntu:~$ ��                                                       
[0007.837] C> RATCHET: MB1 binary ratchet level 1 is less than ratchet level 2 f
rom HW fuses.                                                                   
[0007.845] E> task 4 failed (err: 0x5b5b0108)                                   
[0007.849] E> top caller module: RATCHET, error module: RATCHET, reason: 0x08, a
ux_info: 0x01                                                                   
[0007.857] I> MB1(00.24.180726-t194-41334769-23fb1f28) BIT boot status dump :   
11111000000000000000000000000000000000000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000000000000000000000000000000000000
0000000000000000                                                                
[0007.888] I> Reset to recovery mode                                            
[000

Thanks
Ian

Dear ian.liu,

Did you connect the USB cable when flashing with SDK manager?
If not, please connect the USB cable and try flashing with SDK manager.

13:21:16 INFO : Flash Xavier A+B in parallel : USB communication failed.Check if device is in recovery
According to the log you uploaded, USB cable seemed to be disconnected.

Hi Steve,

I am 100% sure that I have the USB cable connected. I was able to get the console print out in my host computer thru minicom. As I mentioned, I used the same method to flash Drive Xaiver board and I did not encounter the flashing problem.

There was a new finding today. I can flash the Drive Pegasus Board successfully if I log into the partner account (NVONLINE) in SDK manager, but I can’t flash the board (same problem as mentioned above) if I log into the developer account instead. Please let me know your suggestions and comments.

Thanks
Ian

Dear ian.liu,

Could you please let me know what is the DRIVE build the you are trying to flash with both accounts?
When flashing failed, could you export and attach the full sdkm log?

Hi Steve,

it seems like I am facing the same problem. I am also able to connect via minicom, but during flash process the usb connection cannot be established.

I exported the debug logs. You can find them attached. Feel free to contact me if you need further info

Kind regards,
Florian
SDKM_logs_04022019_1137.zip (814 KB)

Dear florian.haselbeck,

We reviewed the log, it looks like there is a process that has a hold on the tty for the target:
+++++++++++++++++++++++++++++++++++++++++++++++++
38575 2019-04-02 11:29:56.512 - info: Aurix Port = /dev/ttyUSB3
38576 2019-04-02 11:29:56.512 - info: Target rootfs will be flashed into emmc
38577 2019-04-02 11:29:56.512 - info: ./bootburn.sh -b e3550b03-t194 -B qspi -x /dev/ttyUSB3
38578 2019-04-02 11:29:56.513 - info: Execute Xavier Script
38579 2019-04-02 11:29:56.544 - info: Successfully acquired lock over /var/lock/LCK…bootburn
38580 2019-04-02 11:29:56.545 - info: Successfully acquired lock over /var/lock/LCK…ttyUSB3
38581 2019-04-02 11:29:56.563 - info: Read skuinfo from InfoRom…
38582 2019-04-02 11:30:00.624 - info: Disabling SIGINT <Ctrl+C> temporarily
38583 2019-04-02 11:30:04.514 - info: Setting Tegra-A on hold… Done
38584 2019-04-02 11:30:07.493 - info: Setting Tegra-B on hold… Done
38585 2019-04-02 11:30:10.087 - info: Setting Tegra-A in recovery… Process CMD (16547) still running. Trying to kill after 5 seconds
38586 2019-04-02 11:30:18.435 - info: Done
38587 2019-04-02 11:30:24.095 - info: Setting Tegra-B in recovery… Process CMD (16907) still running. Trying to kill after 5 seconds
38588 2019-04-02 11:30:29.583 - info: Done
38589 2019-04-02 11:30:29.818 - info: Enabling SIGINT <Ctrl+C>
+++++++++++++++++++++++++++++++++++++++++++++++++
We are a little unclear on the exact process going on here, but could you check if the port is has a user on it?

$ sudo fuser /dev/ttyUSB3

Hi Steve,

I do not get a user on it and I also tried to get the processes on the port via lsof (see below).

I also tried to kill all processes on the port, but the result when I try to flash the board is still the same. Do you have more ideas?

Thanks and kind regards,
Florian

mlpf@rosbart:~$ sudo fuser /dev/ttyUSB3
mlpf@rosbart:~$ sudo fuser -u /dev/ttyUSB3
mlpf@rosbart:~$ sudo fuser -m /dev/ttyUSB3
/dev/ttyUSB3:            1   212   272   625   667   703   838   840   843   847   853   855   859   861   862   864   901   903   904   907   908   968  1010  1012  1023  1083  1110  1113m  1129  1199  1303  1373  1426  1450  1521  1547  1558  1588  2172  2199  2861  2863  2870  2872  2946  2950  2962  2993  2998  3004  3005  3013  3018  3022  3028  3030  3047  3048  3058  3112  3155  3165  3409  3411  3416  3422  3431  3434  3476  3494m  3505  3529  3531  3533  3535  3539  3541  3543  3544  3562  3594  3595  3625  3723  3742  3761  3767  3779  3780  3784  3785  3796  3840  3852  3900  3946  3948  3952  3966  3976  3984  3990  3995  4004  4013  4072  4081  4136  4244  4269  4270  5492  5508  5512  5520  6004  7368  7456  7525  7596  7901 10049 10320 10598 10919 11993 12916 14074 14078 14092 14113m 14119
mlpf@rosbart:~$ lsof /dev/ttyUSB3
lsof: WARNING: can't stat() overlay file system /var/lib/docker/overlay2/6b3666df24ef65d257b978f24b590a53087400a34e78075beb750bbfa9376e6a/merged
      Output information may be incomplete.
lsof: WARNING: can't stat() tmpfs file system /var/lib/docker/containers/3b58f6dc589ce219cb20c24c40ec3ac9b68831d3254331e943867a483b769834/mounts/shm
      Output information may be incomplete.
lsof: WARNING: can't stat() nsfs file system /run/docker/netns/5a1b7e57c901
      Output information may be incomplete.
lsof: WARNING: can't stat() overlay file system /var/lib/docker/overlay2/120df9ad26c5fc4f9c33d8e97a225fb1d67193b7cb07c1ffdcf7c14901884493/merged
      Output information may be incomplete.
lsof: WARNING: can't stat() tmpfs file system /var/lib/docker/containers/e2388ed347efff199c953a4e9dd2de514f365a9ad44654bb1b13d6df6a9a6b6d/mounts/shm
      Output information may be incomplete.
lsof: WARNING: can't stat() nsfs file system /run/docker/netns/897bc0f15938
      Output information may be incomplete.

Hi Steve,

I think I am in the same situation as ian.liu has reported before:

  1. I cannot flash “Drive software 8.0” on Pegasus via Developer account. Got the same error message on minicom:
[0007.587] C> RATCHET: MB1 binary ratchet level 1 is less than ratchet level 2 from HW fuses.
[0007.595] E> task 4 failed (err: 0x5b5b0108)
[0007.599] E> top caller module: RATCHET, error module: RATCHET, reason: 0x08, aux_info: 0x01
[0007.607] I> MB1(00.24.180726-t194-41334769-23fb1f28) BIT boot status dump :
1111100000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
[0007.637] I> Reset to recovery mode
[000
  1. I can flash “Drive OS 5.1.0.0” via Partner account successfully, but DriveWorks is not installed by default.

By talking with the Nvidia representative, I understand the discrete GPU is not supported by “Drive Software 8.0” yet, will have to wait for next release.

In this situation, If assume I don’t use the discrete GPU features on Pegasus for now, is it possible to emulate the “Drive-Xavier” configurations with “Drive-Pegasus” hardware? Or is there any other way to run some of the “Drive Networks” apps on Pegasus?

Thanks.

Dear ian.liu,

Regarding below log, could you let me know whether you updated any files after installing the package with SDKM?

nvidia@tegra-ubuntu:~$ ��                                                       
[0007.837] C> RATCHET: MB1 binary ratchet level 1 is less than ratchet level 2 f
rom HW fuses.                                                                   
[0007.845] E> task 4 failed (err: 0x5b5b0108)                                   
[0007.849] E> top caller module: RATCHET, error module: RATCHET, reason: 0x08, a
ux_info: 0x01                                                                   
[0007.857] I> MB1(00.24.180726-t194-41334769-23fb1f28) BIT boot status dump :   
11111000000000000000000000000000000000000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000000000000000000000000000000000000
0000000000000000                                                                
[0007.888] I> Reset to recovery mode                                            
[000

Dear ian.liu,

One more thing, could you please let me know what kernel version you are using using “uname -a”? Thanks.

@SteveNV,
I also suffer from this issue, please refer skdm.lo

2019-04-26 12:50:55.696 - info: +++++++++++++++++++++++++++++++++++++++++++++++++
2019-04-26 12:50:55.696 - info: Aurix Port = /dev/ttyUSB3
2019-04-26 12:50:55.696 - info: Target rootfs will be flashed into emmc
2019-04-26 12:50:55.697 - info: ./bootburn.sh -b e3550b03-t194a -B qspi -x /dev/ttyUSB3
2019-04-26 12:50:55.700 - info: Execute Xavier Script
2019-04-26 12:50:55.714 - info: Successfully acquired lock over /var/lock/LCK…bootburn
2019-04-26 12:50:55.714 - info: Successfully acquired lock over /var/lock/LCK…ttyUSB3
2019-04-26 12:50:55.718 - info: Read skuinfo from InfoRom…
2019-04-26 12:50:59.343 - info: Disabling SIGINT <Ctrl+C> temporarily
2019-04-26 12:51:02.053 - info: Setting Tegra-A on hold… Done
2019-04-26 12:51:04.766 - info: Setting Tegra-B on hold… Done
2019-04-26 12:51:10.262 - info: Setting Tegra-A in recovery… Done
2019-04-26 12:51:15.753 - info: Setting Tegra-B in recovery… Done
2019-04-26 12:51:15.815 - info: Enabling SIGINT <Ctrl+C>
2019-04-26 12:51:15.979 - info: Target BR_CID: 0x8802191164520784080000000c000200
2019-04-26 12:51:15.979 - info: Detected prod device type
2019-04-26 12:51:15.990 - info: ‘/home/garywang/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn_t19x/…/…/…/…//firmwares/bin/t19x/mb1/mb1_t194_prod.bin’ → ‘/home/garywang/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn_t19x/_temp_dump/_temp_dump_tvrTpihbDd/rcm-flash/mb1_recovery.bin’
2019-04-26 12:51:15.990 - info: ‘/home/garywang/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn_t19x/…/…/…/…//firmwares/bin/t19x/mb1/mb1_t194_prod.bin’ → ‘/home/garywang/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn_t19x/_temp_dump/_temp_dump_tvrTpihbDd/rcm-flash/mb1.bin’
2019-04-26 12:51:15.995 - info: ~/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn_t19x/_temp_dump/_temp_dump_tvrTpihbDd/rcm-flash/applet ~/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn_t19x/_temp_dump/_temp_dump_tvrTpihbDd/rcm-flash ~/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn_t19x ~/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn_t19x
2019-04-26 12:51:15.997 - info: f_FlashingMB2Applet = /home/garywang/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn_t19x/…/…/…/…//tools/host/flashtools/flash/t19x/rcmbl/nvtboot_applet.bin
2019-04-26 12:51:16.031 - info: ~/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn_t19x/_temp_dump/_temp_dump_tvrTpihbDd/rcm-flash ~/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn_t19x ~/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn_t19x
2019-04-26 12:51:16.100 - info: ~/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn_t19x/_temp_dump/_temp_dump_tvrTpihbDd/rcm-flash/applet ~/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn_t19x/_temp_dump/_temp_dump_tvrTpihbDd/rcm-flash ~/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn_t19x ~/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn_t19x
2019-04-26 12:51:16.101 - info: Sending MB1 & Applet
2019-04-26 12:51:17.132 - info:
2019-04-26 12:51:17.133 - info: ------------ Stack Trace ------------
2019-04-26 12:51:17.133 - info: stack frame 0 - 287 AbnormalTermination /home/garywang/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn_t19x/bootburn_lib.sh
2019-04-26 12:51:17.133 - info: stack frame 1 - 2082 SendAndBootAppletSigned /home/garywang/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn_t19x/bootburn_lib.sh
2019-04-26 12:51:17.133 - info: stack frame 2 - 2476 BootRCM /home/garywang/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn_t19x/bootburn_lib.sh
2019-04-26 12:51:17.133 - info: stack frame 3 - 362 source /home/garywang/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn_t19x/bootburn_active.sh
2019-04-26 12:51:17.133 - info: stack frame 4 - 1014 main /home/garywang/nvidia/nvidia_sdk/DRIVE/Linux/5.0.13.2/SW/DriveSDK/drive-t186ref-foundation/tools/host/flashtools/bootburn/…/bootburn_t19x//bootburn.sh
2019-04-26 12:51:17.133 - info: -------------------------------------
2019-04-26 12:51:17.134 - info:
2019-04-26 12:51:17.134 - info: Tool OutPut:
2019-04-26 12:51:17.134 - info: USB communication failed.Check if device is in recovery
2019-04-26 12:51:17.134 - info: Tool OutPut to stderr:
2019-04-26 12:51:17.134 - info: USB communication failed.Check if device is in recovery
2019-04-26 12:51:17.135 - info: error-tool-tegrarcm
2019-04-26 12:51:20.978 - info: Process CMD (31319) still running. Trying to kill after 5 seconds
2019-04-26 12:51:26.019 - info: bootburn flashing failed! error code = 50
2019-04-26 12:51:26.019 - info: exit status 50
2019-04-26 12:51:26.019 - info: [ Component Install Finished with Error ]
2019-04-26 12:51:26.019 - error: NV_FLASH_XAVIER_PDKFLASH_A_COMP command ./pdk_flash.sh -b e3550b03-t194a-es finished with error

my uname listed below,

Linux tegra-ubuntu 4.9.131-rt93-tegra #1 SMP PREEMPT RT Mon Nov 26 10:10:24 PST 2018 aarch64 aarch64 aarch64 GNU/Linux

@SteveNV,
Could you please kindly help me to clarify if SDKManager v0.9.9.2.3.5.1 (or any newer version) is ready to flash Pegasus SKU2200? I cannot flash my two units of Pegasus so far…:( Thanks!

Gary

@SteveNV,
It seems my issue is not belonging to Chip ID Error while flashing a DRIVE AGX Pegasus, no “Failed to get UID of chip” from error-tool-tegrarcm and being similar to problem of ian.liu.

@ian.liu
Do you fix it and any suggestion for me? Thanks!

Gary

@SteveNV,
I have been investing on flashing Pegasus but saw comment from leowang,

"…I can flash “Drive OS 5.1.0.0” via Partner account successfully, but DriveWorks is not installed by default.

By talking with the Nvidia representative, I understand the discrete GPU is not supported by “Drive Software 8.0” yet, will have to wait for next release…"

Could you please help us to confirm it? Thanks!

Gary

@SteveNV,
I flash Pegasus by NVONLINE account (garywang@itri.org.tw) again but suffer this account is not have authorized to flash Pegasus. Any idea to help me?

I can create issue in nvidia online (e.g. #2485258 Gary Wang 1/13/2019 ITRI automotive [ITRI TW] Clock skew detected when booting finished).

Thanks!

Gary

Dear Gary,

Please refer to the following link.
https://devtalk.nvidia.com/default/topic/1050909/general/find-suitable-version-current-drive-sw-8-0-or-newer-of-driveworks-sdkmanager-for-pegasus-/

@SteveNV, got it. Thanks!

Because I only used built-in DRIVE OS (5.1.0.0-13431798) for testing and development so far.

For risk management in project schedule, I am afraid some situations of my Pegasus fail rebooting due to our testing or Ubuntu setting in DRIVE OS then needs flash mechanism to reset our DRIVE OS in Pegasus.

Could you please help to teach me how to flash or factory reset for DRIVE OS (5.1.0.0-13431798) on Pegasus.

Thanks!

ps: please refer my flash failed log.

Gary
NV_FLASH_XAVIER_PDKFLASH_A_COMP.log (129 KB)
NV_FLASH_XAVIER_PDKFLASH_PARALLEL_COMP.log (268 KB)

Dear Gary,

Did you connect USB cable between Pegasus and HostPC?
Because according to the log files you attached, USB cable seems to be disconnected.
“Make sure only one target is connected to USB”

.
.
.

Tool OutPut:
USB communication failed.Check if device is in recovery
Tool OutPut to stderr:
USB communication failed.Check if device is in recovery
error-tool-tegrarcm