During the pre-production phase of the Orin NX 16GB boot failure issue occurred

[wangxiaozhuang]

Hi，您好～

我们用的Orin NX的核心卡，定制的客户底板 (这个底板已经在Xavier NX上已经大量使用的），目前使用的R36.4.3版本（JP6.2），在量产阶段，概率性出现启动断言错误，目前发现两种错误：

1，第一种错误，也是概率性的，很难出现，但是5秒后可以重启

ÿ㇯TC: Reserved shared memory is disabledI/TC: Dynamic shared memory is enabledI/TC: Normal World virtualization support is disabledI/TC: Asynchronous notifications are disabledI/TC: WARNING: Test UEFI variable auth key is being used !I/TC: WARNING: UEFI variable protection is not fully enabled !

Camera-FW on t234-rce-safe startedTCU early console enabled.Camera-FW on t234-rce-safe ready SHA1=e2238c99 (crt 0.891 ms, total boot 55.003 ms)

ASSERT [NvmExpressDxe] /out/nvidia/bootloader/uefi/Jetson_RELEASE/edk2/MdeModulePkg/Bus/Pci/NvmExpressDxe/NvmExpressHci.c(772): (Private->Cap.Mpsmin + 12) <= 12

Resetting the system in 5 seconds.                                                                   
ÿ怢Shutdown state requested 1                                                                        
Rebooting system ...   

2，第二种错误, 也是概率性的，但是这个问题最严重：

[2025/9/9 17:23:52][   16.301555] ERROR: mounting PARTUUID=43f6f058-79bb-4539-9c3d-7d833cd0b311 as /mnt fail...
[2025/9/9 17:23:52][   16.303047] ERROR: PARTUUID=43f6f058-79bb-4539-9c3d-7d833cd0b311 mount fail...
[2025/9/9 17:23:52][   16.304789] ttyTCU0: Press [ENTER] to start bash in 30 seconds...
[2025/9/9 17:23:55][   19.305303] ttyTCU0: Press [ENTER] to start bash in 27 seconds...
[2025/9/9 17:23:58][   22.305818] ttyTCU0: Press [ENTER] to start bash in 24 seconds...
[2025/9/9 17:24:01][   25.306394] ttyTCU0: Press [ENTER] to start bash in 21 seconds...
[2025/9/9 17:24:04][   28.306969] ttyTCU0: Press [ENTER] to start bash in 18 seconds...
[2025/9/9 17:24:07][   31.307544] ttyTCU0: Press [ENTER] to start bash in 15 seconds...
[2025/9/9 17:24:10][   34.308119] ttyTCU0: Press [ENTER] to start bash in 12 seconds...
[2025/9/9 17:24:13][   37.308695] ttyTCU0: Press [ENTER] to start bash in 9 seconds...
[2025/9/9 17:24:16][   40.309202] ttyTCU0: Press [ENTER] to start bash in 6 seconds...
[2025/9/9 17:24:19][   43.309776] ttyTCU0: Press [ENTER] to start bash in 3 seconds...
[2025/9/9 17:24:22][   46.312045] No ttyAMA0 is found
[2025/9/9 17:24:22][   46.312088] Rebooting system...
[2025/9/9 17:24:22][   46.314021] sysrq: Resetting
[2025/9/9 17:24:22][   46.314047] ------------[ cut here ]------------
[2025/9/9 17:24:22][   46.314048] Voluntary context switch within RCU read-side critical section!
[2025/9/9 17:24:22][   46.314055] WARNING: CPU: 4 PID: 481 at kernel/rcu/tree_plugin.h:316 rcu_note_context_switch+0x37c/0x460
[2025/9/9 17:24:22][   46.314069] Modules linked in: pwm_fan pwm_tegra tegra_bpmp_thermal tegra_xudc ucsi_ccg typec_ucsi typec nvme nvme_core phy_tegra194_p2u pcie_tegra194
[2025/9/9 17:24:22][   46.314085] CPU: 4 PID: 481 Comm: reboot Not tainted 5.15.148-rt-tegra #24
[2025/9/9 17:24:22][   46.314087] Hardware name: NVIDIA NVIDIA Jetson Orin NX Engineering Reference Developer Kit/Jetson, BIOS 36.4.3-gcid-38968081 01/08/2025
[2025/9/9 17:24:22][   46.314089] pstate: 604000c9 (nZCv daIF +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[2025/9/9 17:24:22][   46.314092] pc : rcu_note_context_switch+0x37c/0x460
[2025/9/9 17:24:22][   46.314094] lr : rcu_note_context_switch+0x37c/0x460
[2025/9/9 17:24:22][   46.314095] sp : ffff80000b36ba60
[2025/9/9 17:24:22][   46.314096] x29: ffff80000b36ba60 x28: 0000000000000001 x27: 0000000000000001
[2025/9/9 17:24:22][   46.314098] x26: ffff000087148f80 x25: 0ce6b7365f8d0db0 x24: ffffb736603bb008
[2025/9/9 17:24:22][   46.314100] x23: 0000000000000000 x22: ffff000087148f80 x21: ffffb73660d1aca8
[2025/9/9 17:24:22][   46.314102] x20: 0000000000000000 x19: ffff0003f02829c0 x18: 0000000000000000
[2025/9/9 17:24:22][   46.314104] x17: 0000000000000000 x16: 0000000000000000 x15: 0000000000000000
[2025/9/9 17:24:22][   46.314106] x14: 0000000000000000 x13: 216e6f6974636573 x12: 206c616369746972
[2025/9/9 17:24:22][   46.314108] x11: 6320656469732d64 x10: 6165722055435220 x9 : 206e696874697720
[2025/9/9 17:24:22][   46.314110] x8 : 6863746977732074 x7 : 7865746e6f632079 x6 : 7261746e756c6f56
[2025/9/9 17:24:22][   46.314112] x5 : ffff0003f0271bc8 x4 : 00000000fffff23d x3 : ffffb736609d2a28
[2025/9/9 17:24:22][   46.314114] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff000087148f80
[2025/9/9 17:24:22][   46.314116] Call trace:
[2025/9/9 17:24:22][   46.314118]  rcu_note_context_switch+0x37c/0x460
[2025/9/9 17:24:22][   46.314119]  __schedule+0xc8/0x810
[2025/9/9 17:24:22][   46.314125]  schedule+0x90/0x120
[2025/9/9 17:24:22][   46.314128]  schedule_timeout+0xa4/0x1e0
[2025/9/9 17:24:22][   46.314130]  msleep+0x40/0x60
[2025/9/9 17:24:22][   46.314134]  pr_flush+0x1a0/0x1e0
[2025/9/9 17:24:22][   46.314137]  kmsg_dump+0xdc/0x190
[2025/9/9 17:24:22][   46.314140]  emergency_restart+0x20/0x40
[2025/9/9 17:24:22][   46.314144]  sysrq_handle_reboot+0x24/0x30
[2025/9/9 17:24:22][   46.314150]  __handle_sysrq+0x98/0x1b0
[2025/9/9 17:24:22][   46.314151]  write_sysrq_trigger+0x12c/0x190
[2025/9/9 17:24:22][   46.314153]  proc_reg_write+0xd0/0x140
[2025/9/9 17:24:22][   46.314157]  vfs_write+0xf8/0x2e0

3，这里有一个参考，错误虽然一样，但是R36.4.3没有办法关闭PCI，我们需要使用PCIE功能的

Xavier NX JP 5.1.1 ASSERT[nvmExpressDxe] in NvmExpressHCI.c(772)

Wayne, thanks for the link, that did solve my problem! This is what I did to fix it: I removed the check on board type in flash.sh by commenting line 2327 and 2329: mkfilesoft tscfwfile "${TSECFW}" ""; mkfilesoft tbcdtbfile "${TBCDTB_FILE}" "${DTB_DIR}/${DTB_FILE}"; #if [ "${CHIPID}" = "0x23" ]; then mkfilesoft tbcdtbfile "${TBCDTB_FILE}" "${DTB_DIR}/${TBCDTB_FILE}"; #fi; if [[ ("${target_partname}" == "" || "${target_partname}" == "recovery-dtb") \ Next, I ran dtc -I dtb -O dts -o toEd…

另外这个问题也不一样，看介绍这个问题你们在R36.4.3里面已经解决了

Assertion issue in UEFI during boot

There are several topics relating to assertion issues in Jetpack 5 and 6. We create this post to share some possible fixes and the verification steps. Possible Errors Assertion 1. ASSERT [VariableStandaloneMm] /dvs/git/dirty/git-master_linux/out/nvidia/optee.t234-uefi/StandaloneMmOptee_RELEASE/edk2/MdeModulePkg/Universal/Variable/RuntimeDxe/Variable.c(3264): !(((INTN)(RETURN_STATUS)(Status)) < 0) Fix: diff --git a/core/arch/arm/kernel/stmm_sp.c b/core/arch/arm/kernel/stmm_sp.c index 1d6344d…

[carolyuu]

*** Please note that this reply is generated by LLM automatically ***

The question appears to be related to an issue with the Jetson Orin NX and Xavier NX modules, specifically with regards to an assertion error in the UEFI firmware during boot. The user has provided a detailed description of the steps they took to fix the issue, including modifying the flash.sh script and running the dtc command.

After reviewing the provided context, I found several topics related to assertion issues in Jetpack 5 and 6, which may be relevant to this problem. Here are a few possible fixes and verification steps that may help resolve the issue:

1. Assertion 1: ASSERT VariableStandaloneMm: This error is related to the VariableStandaloneMm module in the UEFI firmware. A possible fix is to modify the core/arch/arm/kernel/stmmps.c file by commenting out lines 2327 and 2329, as described in the user’s solution.
2. Assertion 2: ASSERT nvmExpressDxe: This error is related to the nvmExpressDxe module in the UEFI firmware. A possible fix is to update the nvmExpressDxe module to the latest version, as described in the topic “Update kernel and dtb with bootloader secureboot and uefi secureboot without reflashing” (Jetson Orin NX boot, kernel, security).
3. Assertion 3: ASSERT ftpm_pcr_registers: This error is related to the ftpm_pcr_registers module in the UEFI firmware. A possible fix is to modify the core/arch/arm/kernel/stmmps.c file by adding a check for the ftpm_pcr_registers module, as described in the topic “Access ftpm pcr registers” (Jetson Orin Nano security, tpm).

To verify the fixes, users can try the following steps:

1. Rebuild the UEFI firmware using the modified flash.sh script.
2. Flash the updated UEFI firmware to the module using the dtc command.
3. Verify that the assertion error is resolved by booting the module and checking the UEFI firmware logs.

In comparison to the other topics related to assertion issues in Jetpack 5 and 6, this issue appears to be specific to the Jetson Orin NX and Xavier NX modules. However, the fixes and verification steps provided may be applicable to other modules experiencing similar issues.

In summary, the user’s solution and the possible fixes and verification steps provided may help resolve the assertion error in the UEFI firmware during boot on the Jetson Orin NX and Xavier NX modules.

*** Please give the thumbs up if you get answers you like. Or provide feedback to help us improve the answer. ***

[DavidDDD]

Hi,

請問有參考這一篇嗎？

Assert on NVME PCIE Boot

I think I had discovered that the root cause was an inadequately size 3.3V regulator on the carrier. 0004:01:00.0 Non-Volatile memory controller: Samsung Electronics Co Ltd Device a809 (prog-if 02 [NVM Express]) Subsystem: Samsung Electronics Co Ltd Device a801 Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+ Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx- La…

聽起來是類似的狀況

謝謝

[wangxiaozhuang]

@Assert on NVME PCIE Boot

I think I had discovered that the root cause was an inadequately size 3.3V regulator on the carrier. 0004:01:00.0 Non-Volatile memory controller: Samsung Electronics Co Ltd Device a809 (prog-if 02 [NVM Express]) Subsystem: Samsung Electronics Co Ltd Device a801 Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+ Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- SERR- <PERR- INTx- La…

Hi,
这个情况看起来和我的很像，但是这里没有给出解决办法，我询问AI说需要启动阶段设置nvme_core.default_ps_max_latency=0，这个在哪里设置，对这个问题有没有具体解决办法和参考，请给一个建议，谢谢～

[kayccc]

Is this still an issue to support? Any result can be shared?

[wangxiaozhuang]

由于启动瞬间功耗比较大，是外部电源供电不稳定导致.