Xavier NX Low EMC Frequency

Robotics & Edge Computing Jetson & Embedded Systems Jetson Xavier NX
,
1

Hi,
I have an issue with low EMC frequency. MaxFreq=204000000 instead of 1866000000.
The tegra release is:
R32 (release), REVISION: 6.1, GCID: 27863751, BOARD: t186ref, EABI: aarch64, DATE: Mon Jul 26 19:36:31 UTC 2021
L4T version R32.6.1

When I try to change the POWER_MODEL to 8, I’m getting the following:
$ sudo /usr/sbin/nvpmodel -m 8

NVPM WARN: patching tpc_pg_mask: (0x1:0x4)
NVPM WARN: patched tpc_pg_mask: 0x4
  • And the clocks are as follows:

SOC family:tegra194 Machine:NVIDIA Jetson Xavier NX Developer Kit
Online CPUs: 0-5
cpu0: Online=1 Governor=schedutil MinFreq=1190400 MaxFreq=1420800 CurrentFreq=1420800 IdleStates: C1=1 c6=1
cpu1: Online=1 Governor=schedutil MinFreq=1190400 MaxFreq=1420800 CurrentFreq=1420800 IdleStates: C1=1 c6=1
cpu2: Online=1 Governor=schedutil MinFreq=1190400 MaxFreq=1420800 CurrentFreq=1420800 IdleStates: C1=1 c6=1
cpu3: Online=1 Governor=schedutil MinFreq=1190400 MaxFreq=1420800 CurrentFreq=1420800 IdleStates: C1=1 c6=1
cpu4: Online=1 Governor=schedutil MinFreq=1190400 MaxFreq=1420800 CurrentFreq=1420800 IdleStates: C1=1 c6=1
cpu5: Online=1 Governor=schedutil MinFreq=1190400 MaxFreq=1420800 CurrentFreq=1420800 IdleStates: C1=1 c6=1
GPU MinFreq=114750000 MaxFreq=1109250000 CurrentFreq=114750000
EMC MinFreq=204000000 MaxFreq=204000000 CurrentFreq=204000000 FreqOverride=0
Fan: PWM=130
NV Power Mode: MODE_20W_6CORE

I use custom board and the rootfs located on nvme drive. To make sure it’s not related to HW, I switched the Xavier NX module to another one and in the new module it’s fine without any issues. Both modules were flashed the same way.

Please help me to fix the issue
Thanks

3

please move back this module to devkit. Reflash with sdkamanger + latest jetpack version.

Run sudo jetson_clocks and see if the emc clock could be up or not.

4

Thanks WayneWWW,
After clean flash, the emc clock set to 1866000000
I need to flash the device using rootfs after modifications with application running on it and use nvme storage. I use the following command to do it:
sudo ROOTFS_AB=1 ./tools/kernel_flash/l4t_initrd_flash.sh --external-device nvme0n1 -c ./tools/kernel_flash/flash_l4t_nvme_rootfs_ab.xml -S 32GiB --showlogs --erase-all jetson-xavier-nx-devkit-emmc nvme0n1p1

During the boot I’ve got the following error message:

[   16.562186] podgov: can't create debugfs directory
[   16.562940] Kernel panic - not syncing: nvhost_scale_emc_debug_init
[   16.563201] CPU: 0 PID: 6126 Comm: das Tainted: G           O    4.9.253-tegra #1
[   16.563384] Hardware name: NVIDIA Jetson Xavier NX Developer Kit (DT)
[   16.563527] Call trace:
[   16.563649] [<ffffff800808ba40>] dump_backtrace+0x0/0x198
[   16.563777] [<ffffff800808c004>] show_stack+0x24/0x30
[   16.563920] [<ffffff8008f6f5fc>] dump_stack+0xa0/0xc4
[   16.564059] [<ffffff8008f6c6a0>] panic+0x12c/0x2a8
[   16.564190] [<ffffff8008ccf660>] nvhost_scale_emc_debug_init.isra.12+0x128/0x1a0
[   16.564359] [<ffffff8008ccfa0c>] nvhost_pod_event_handler+0x334/0x400
[   16.564507] [<ffffff8008ccc960>] devfreq_add_device+0x318/0x470
[   16.564640] [<ffffff8008cccb1c>] devm_devfreq_add_device+0x64/0xc0
[   16.565148] [<ffffff8000fefa48>] gk20a_scale_init+0xf0/0x190 [nvgpu]
[   16.565651] [<ffffff8000fe8cb8>] gk20a_pm_finalize_poweron+0x388/0x418 [nvgpu]
[   16.566166] [<ffffff8000fe8f00>] gk20a_pm_runtime_resume+0x58/0x70 [nvgpu]
[   16.566346] [<ffffff800878b204>] pm_generic_runtime_resume+0x3c/0x58
[   16.572450] [<ffffff800878d54c>] __rpm_callback+0x74/0xa0
[   16.578042] [<ffffff800878d5ac>] rpm_callback+0x34/0x98
[   16.582844] [<ffffff800878e010>] rpm_resume+0x470/0x710
[   16.587926] [<ffffff800878ee3c>] __pm_runtime_resume+0x4c/0x70
[   16.594124] [<ffffff8000fe8e1c>] gk20a_busy+0xd4/0x160 [nvgpu]
[   16.600319] [<ffffff8000fc9f8c>] gk20a_ctrl_dev_open+0x8c/0x168 [nvgpu]
[   16.606760] [<ffffff8008261534>] chrdev_open+0x94/0x198
[   16.611908] [<ffffff8008257dd8>] do_dentry_open+0x1d8/0x340
[   16.617680] [<ffffff8008259390>] vfs_open+0x58/0x88
[   16.622241] [<ffffff800826c9c0>] do_last+0x530/0xfa0
[   16.627384] [<ffffff800826d4c0>] path_openat+0x90/0x378
[   16.632203] [<ffffff800826ea30>] do_filp_open+0x70/0xe8
[   16.637448] [<ffffff8008259854>] do_sys_open+0x174/0x258
[   16.643060] [<ffffff80082599bc>] SyS_openat+0x3c/0x50
[   16.648229] [<ffffff8008083900>] el0_svc_naked+0x34/0x38
[   16.653474] SMP: stopping secondary CPUs
[   16.657711] Kernel Offset: disabled
[   16.661142] Memory Limit: none

I’ve followed the instructions lhoang gave in this post:

  1. sudo ./tools/kernel_flash/l4t_initrd_flash.sh --no-flash jetson-xavier-nx-devkit-qspi internal
  2. sudo ROOTFS_AB=1 ./tools/kernel_flash/l4t_initrd_flash.sh --no-flash --external-device nvme0n1p1 -S 32GiB -c ./tools/kernel_flash/flash_l4t_nvme_rootfs_ab.xml --showlogs --external-only --append jetson-xavier-nx-devkit-emmc external
  3. sudo ./tools/kernel_flash/l4t_initrd_flash.sh --flash-only

On the first boot I was able to check the clocks and it was fine but after reboot, I was getting the Kernel panic - not syncing: nvhost_scale_emc_debug_init and reboots in a loop without option to login.

What could be the reason for this behavior?
Thanks

6

Are there any updates on this topic since I’m facing a similar problem.

7

file your own topic please.