Flash nvme on orin nano（8G）

Robotics & Edge Computing Jetson & Embedded Systems Jetson Orin Nano
, ,
1

我們自己設計的底板，cpu核心處理器使用的是orin nano（8G）
官網下載工程為：Jetson_Linux_R35.3.1_aarch64.tbz2
當使用燒寫指令“sudo ./tools/kernel_flash/l4t_initrd_flash.sh --external-device nvme0n1p1 -c tools/kernel_flash/flash_l4t_external.xml -p“-c bootloader/t186ref/cfg/flash_t234_qspi.xml” --showlogs --network usb0 p3509-a02+p3767-0000 internal ”來燒寫nvme硬盤時，燒寫提示失敗，失敗日志如下：
Copied 400960 bytes from /qspi/internal/adsp-fw_sigheader.bin.encrypt to address 0x03b30000 in flash
[ 220]: l4t_flash_from_kernel: Warning: skip writing B_reserved_on_boot partition as no image is specified
[ 220]: l4t_flash_from_kernel: Warning: skip writing uefi_variables partition as no image is specified
[ 220]: l4t_flash_from_kernel: Warning: skip writing uefi_ftw partition as no image is specified
[ 220]: l4t_flash_from_kernel: Warning: skip writing worm partition as no image is specified
Writing bct_backup.img (parittion: BCT-boot-chain_backup) into /dev/mtd0
Sha1 checksum matched for /qspi/internal/bct_backup.img
Writing /qspi/internal/bct_backup.img (32768 bytes) into /dev/mtd0:66715648
Copied 32768 bytes from /qspi/internal/bct_backup.img to address 0x03fa0000 in flash
[ 220]: l4t_flash_from_kernel: Warning: skip writing reserved_partition partition as no image is specified
Writing gpt_secondary_3_0.bin (parittion: secondary_gpt_backup) into /dev/mtd0
Sha1 checksum matched for /qspi/internal/gpt_secondary_3_0.bin
Writing /qspi/internal/gpt_secondary_3_0.bin (16896 bytes) into /dev/mtd0:66846720
Copied 16896 bytes from /qspi/internal/gpt_secondary_3_0.bin to address 0x03fc0000 in flash
Writing qspi_bootblob_ver.txt (parittion: B_VER) into /dev/mtd0
Sha1 checksum matched for /qspi/internal/qspi_bootblob_ver.txt
Writing /qspi/internal/qspi_bootblob_ver.txt (109 bytes) into /dev/mtd0:66912256
Copied 109 bytes from /qspi/internal/qspi_bootblob_ver.txt to address 0x03fd0000 in flash
Writing qspi_bootblob_ver.txt (parittion: A_VER) into /dev/mtd0
Sha1 checksum matched for /qspi/internal/qspi_bootblob_ver.txt
Writing /qspi/internal/qspi_bootblob_ver.txt (109 bytes) into /dev/mtd0:66977792
Copied 109 bytes from /qspi/internal/qspi_bootblob_ver.txt to address 0x03fe0000 in flash
Writing gpt_secondary_3_0.bin (parittion: secondary_gpt) into /dev/mtd0
Sha1 checksum matched for /qspi/internal/gpt_secondary_3_0.bin
Writing /qspi/internal/gpt_secondary_3_0.bin (16896 bytes) into /dev/mtd0:67091968
Copied 16896 bytes from /qspi/internal/gpt_secondary_3_0.bin to address 0x03ffbe00 in flash
[ 220]: l4t_flash_from_kernel: Successfully flash the qspi
[ 220]: l4t_flash_from_kernel: Flashing success
Error flashing non-qspi storage

Cleaning up…

然後重新上電，查看串口調試信息，出現“Busy Pin”的打印提示，串口打印提示如下：
I> Task: I2C register (0x50001e44)
I> Task: Map CCPLEX_INTERWORLD_SHMEM carveout (0x50001e2c)
I> Task: Program CBB PCIE AMAP regions (0x50019b64)
I> Task: Boot device init (0x50001d74)
I> Boot_device: QSPI_FLASH instance: 0
I> Qspi clock source : pllc_out0
I> QSPI-0l initialized successfully
E> Error in command_complete 18001 int_status
E> Sending CMD_SD_SEND_IF_COND failed
E> Error opening sdcard-0
E> Failed to initialize device 6-0
C> Storage init failed
C> Task 0x0 failed (err: 0x39390706)
E> Top caller module: SDMMC, error module: SDMMC, reason: 0x06, aux_info: 0x07
I> Busy Spin

3

Hi,

請問你們對自己設計底板之後要對BSP做哪些修改有概念嗎？

p3509-a02+p3767-0000

這個config file是給Orin NX插在Xavier NX DevKit的底板上時用的，除非你們的底板設計和Xavier NX的DevKit一模一樣（當然如果一樣的話你們也沒有自己設計的必要），否則pinmux和device tree都有可能不一樣的情況下，直接刷刷不過也是很正常的。

4

感謝回復，但把參數p3509-a02+p3767-0000改成jetson-orin-nano-devkit燒寫，能燒寫成功，然後系統能正常啓動。我們按照Orin Nano Developer Kit開發辦來設計的我們自己的底板！然後看官網鏈接orin Nano和Orin NX用的參數一樣的，參考鏈接： Quick Start — Jetson Linux Developer Guide documentation (nvidia.com)

5

那就是你們一開始的時候用錯config file了而已…
你可以發現jetson-orin-nano-devkit.conf實際上是link到p3768-0000+p3767-0000.conf，裡面指定的kernel dtb是tegra234-p3767-0000-p3768-0000-a0.dtb，也和p3509-a02+p3767-0000.conf裡用的不一樣。

6

感謝解答！p3509-a02+p3767-000參數只適用於ORIN NX！

7

你的結論搞錯了 p3509-a02+p3767-0000 的重點在你的底板設計要跟xavier nx devkit一樣, 這個config可以用在orin nano 跟orin nx這兩個所有款的module.

8

我重新解釋一下

Orin nano 跟Orin NX module有兩種底板可以使用

  1. Orin nano devkit底板 (只有dp port)
  2. Xavier NX devkit底板 (p3509) (只有hdmi port)

當你使用(1)的底板的時候 不論你是哪種module, 都要用 jetson-orin-nano-devkit.conf燒錄
當你用(2)的時候, 都要用 p3509-a02+p3767-0000 燒錄

所以不管哪種module都可以透過這兩個config燒 差別是在底板的設計.

9

針對你nvme燒不過這件事情 麻煩附上完整的log. Host端跟UART都附上來

10

然後重新上電，查看串口調試信息，出現“Busy Pin”的打印提示，串口打印提示如下：
I> Task: I2C register (0x50001e44)
I> Task: Map CCPLEX_INTERWORLD_SHMEM carveout (0x50001e2c)
I> Task: Program CBB PCIE AMAP regions (0x50019b64)
I> Task: Boot device init (0x50001d74)
I> Boot_device: QSPI_FLASH instance: 0
I> Qspi clock source : pllc_out0
I> QSPI-0l initialized successfully
E> Error in command_complete 18001 int_status
E> Sending CMD_SD_SEND_IF_COND failed
E> Error opening sdcard-0
E> Failed to initialize device 6-0
C> Storage init failed
C> Task 0x0 failed (err: 0x39390706)
E> Top caller module: SDMMC, error module: SDMMC, reason: 0x06, aux_info: 0x07
I> Busy Spin

另外 這邊你要抓的log應該是燒錄當下碰到的log. 而不是已經燒失敗之後你才把機器重新上電抓log.

11

燒寫log：
flash_nvme.log (267.9 KB)
燒寫時的串口打印log：
uart_log.log (90.2 KB)

12

請問你們的nvme用pcie哪一個controller?

13

pcie2控制器， 部分引脚如下圖：

20230706-172610
20230706-1726101345×628 128 KB

14

PCIe controller C7 對應到的是pcie@141e0000

現在這個問題的原因是Orin nano devkit的配置上這個controller是有打開的. 但是p3509上沒開
PCIe controller沒開, nvme自然會找不到

所以麻煩去device tree把設定打開.

https://docs.nvidia.com/jetson/archives/r35.3.1/DeveloperGuide/text/HR/JetsonModuleAdaptationAndBringUp/JetsonOrinNxNanoSeries.html#enabling-pcie-in-a-customer-carrier-board-design

15

您好，改完device tree仍然有問題，這個pcie的設備樹改完之後，改完之後的操作：
（1）./apply_binaries.sh
（2）./tools/l4t_flash_prerequisites.sh
（3）./tools/kernel_flash/l4t_initrd_flash.sh --external-device nvme0n1p1 -c tools/kernel_flash/flash_l4t_external.xml -p“-c bootloader/t186ref/cfg/flash_t234_qspi.xml” --showlogs --network usb0 p3509-a02+p3767-0000 internal > flash_nvme.log
其中設備樹如下：
tegra234-soc-pcie.dtsi (50.6 KB)

燒寫日志：
flash_nvme.log (267.4 KB)
燒寫時候的串口打印日志：
uart_log.log (94.9 KB)

16

Hi,

Device tree部份還是沒有改對. 看你要不要直接把orin nano devkit的設定直接複製過來…
還有, 141e0000.pcie_ep這個不用打開. 這個是給pcie endpoint mode用的

1734 [    4.944283] tegra194-pcie 141e0000.pcie_ep: Failed to find PHY entries: -22^M
1735 [    4.945713] tegra194-pcie 141e0000.pcie_ep: Failed to parse device tree: -22^M
1736 [    4.952801] tegra194-pcie: probe of 141e0000.pcie_ep failed with error -22^M
1737 [    4.959881] tegra194-pcie 14100000.pcie: Adding to iommu group 10^M
1738 [    4.972183] tegra194-pcie 14100000.pcie: Using GICv2m MSI allocator^M
1739 [    4.973872] tegra194-pcie 14160000.pcie: Adding to iommu group 11^M
1740 [    4.984352] tegra194-pcie 14160000.pcie: Using GICv2m MSI allocator^M
1741 [    4.985091] tegra194-pcie 141e0000.pcie: Adding to iommu group 9^M
1742 [    4.990380] tegra194-pcie 141e0000.pcie: Failed to find PHY entries: -22^M
1743 [    4.996908] tegra194-pcie 141e0000.pcie: Failed to parse device tree: -22^M
1744 [    5.003757] tegra194-pcie: probe of 141e0000.pcie failed with error -22^M
17

orin nano devit 的pcie設備樹是哪個文件，在jetson-orin-nano-devkit.conf文件裏面找到的設備樹並沒有pcie相關的配置

18

請問你是第一次碰device tree相關的東西嗎?

19

orin nano的產品第一次用，配置過其他廠商cpu處理器的設備樹

20

在jetson-orin-nano-devkit.conf文件裏面找到的設備樹並沒有pcie相關的配置

這件事情一定是錯的, 因為你用這個conf燒錄之後那個對應的pcie有開.
有幾種可能

  1. 你找錯dtb/dts
  2. 你找對dts, 但你需要trace code找到pcie相關的設定在哪.
22

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.