I have a system built into an enclosure that is exhibiting behavior very similar to issue [1723265] that was fixed in L4T version 24.1 where the board does not like to cold boot when the temperature is slightly warm (the bug was above 44 degrees, see 24.1 release notes). The board is currently built into an enclosure so I don’t have very good access to it.
I am unsure if the issue is caused by the [1723265] bug or if there is something else wrong (such as my COTS carrier board not supplying the correct power on sequencing, or power up pulse etc). If I boot the system and run some power intensive processing, I will have to wait some time for the system to cool before it cold boots again. Warm boots seem to work correctly.
I believe this board has been flashed with an image created from another board which had 24.1. Cloning procedure can be found here: http://elinux.org/Jetson/TX1_Cloning
Running “uname -r” on the console shows “3.10.96-tegra” as expected with a L4T 24.1.
My question: where was the fix applied? Is the fix for [1723265] compiled into the kernel such that seeing the correct kernel version is sufficient to ensure that I have the fix for [1723265] or is something else required (dtb file, etc?). I have not compiled my own kernels outside of installing new Jetpack releases.
If the fix is not contained in the kernel, is it possible that cloning a board would not include the patch?
Cloning will give you exactly what was on the other rootfs. So kernel and all configurations, even UUID of partition, will remain the same on the new board.
Cloning includes only what was cloned. It could include everything, it could be a subset, so on. The information I’ve seen on cloning the JTK1 includes cloning “all” partitions or just the rootfs (APP), but I think the JTX1 clone information is for only one partition at a time. The default in the information for JTX1 is the “APP” partition, which is the rootfs only.
However, the other partitions don’t have the kernel or firmware, I doubt they are related to the fix you mentioned. I won’t guarantee it, as perhaps it could be a fix in the bootloader binary, but odds are low for that. I don’t know what fix “[1723265]” was or what it changed, so I can’t give a definite answer.
Ya, the only information I have is the release notes for L4T 24.1 which states:
1.4 TOP ISSUES FIXED SINCE LAST RELEASE
The following issues are resolved in this release.
[1723265] Hard resetting the TX1 board while it is above 44°C will cause boot failure due
to default system shut down value is set to 44°C. The system will boot up once cooled
down to below 44°C. This issue does not occur upon warm reset.
[1723265] Hard resetting the TX1 board while it is above 44°C will cause boot failure due
to default system shut down value is set to 44°C
Isn’t clear to me where the change was made and if this would have been included in my clone procedure.
If it was in kernel or firmware your clone should have worked. If it is part of the bootloader it would have been missed. Anyone know if this fix was in kernel, firmware, or bootloader?