My TX2 can’t started system. after updata jetpack4.2 on myself PCB.carrier plate
Hi 358271128,
Is your xavier able to boot up with devkit + module?
Yes , the xavier can boot up with devkit + module .
Could you provide the boot log to us?
Is your carrier board able to boot up when using rel-28 based release?
using rel-28 on my carrier board can able to boot up, and no found boot.log in my system.
What do you mean find no boot log in your system…?
Can’t you see the kernel boot up log? Do you have UART debugger?
I can’t find /var/log/boot.log in my system.
During my xavier startup, the system did not generate a new kernel Log & Boot Log & system log.
I suspect that the power settings did not start TX2. Has JetPack 4.2 been modified in this regard?
Of course you cannot find anything in /var/log when your system cannot boot up…
We always debug such issue with UART debugger. Do you have one??
I’m sorry. We don’t set up UART on the PCB board.
Is there any other way to test this?
How do you tell the device does not boot up? Do you judge it by using a HDMI monitor?
Are you able to see nvidia logo when boot up?
1.Redirect to the file to see if the system is started by setting the self-startup script execution file.
2.By viewing network connections.I set up the IP on the devkit + module and install TX2 on the PSB carrier plate, but Ping didn’t work.
I found some more documents in the
…/64_TX2/Linux_for_Tegra/kernel/dtb
Compared with previous versions.
Are there any changes in the device tree that make it impossible to boot?
358271128,
Currently I would say you may need to find out a way to enable UART debugger. There is nothing we can do for you now.
The kernel version is different between rel-28(k4.4) and rel-32.1 (k4.9) so it is normal the dtb is different.
my UART log
log.txt (125 KB)
Many thanks.
Just found this kernel error.
[ 11.208884] edid invalid
[ 11.212206] tegradc 15210000.nvdisplay: hdmi: tmds rate:25200K prod-setting:prod_c_hdmi_0m_54m
[ 11.222214] tegradc 15210000.nvdisplay: hdmi: get RGB quant from EDID.
[ 11.228764] Unable to handle kernel read from unreadable memory at virtual address 00000063
I guess the issue comes from your device cannot read EDID from HDMI monitor and cause following error.
Also, there is i2c error in your log. Is your carrier board able to read HDMI edid through i2c when rel-28/jetpck3.3? This kind of issue should be same on both rel-28 or rel-32.
It just hangs because we have more to check on rel-32.
Thank you for your reply!
We did not set up the interface of HDMI on my carrier board. The reason for I2C error may be that we used COM1 as the serial port of connection, and COM1 may be the reuse of HDMI and COM.
If we want to solve this problem, what is a good way?
Got it. Then please set the status to disable for nvdisplay@15210000 in your device tree.
Thanks very much
Can you tell me which file this nvdisplay@15210000 is in? In /public_sources
Add this node to “tegra186-quill-p3310-1000-a00-00-base.dts”