I added a sample rootfs into the “Linux_for_Tegra/rootfs/” directory
then run sudo ./apply_binaries.sh (from Linux_for_Tegra/)
then flash,sh
dowload Tegra210_Linux_R32.2.1_aarch64.tbz2 and egra_Linux_Sample-Root-Filesystem_R32.2.1_aarch64.tbz2 (https://developer.nvidia.com/embedded/linux-tegra)
The basic procedure you’ve used seems correct.
Was the driver package (the part which creates the “Linux_for_Tegra/” directory) from the same release as the sample rootfs? Also, a lot of disk space is used, and it must be type ext4, so what do you see from the “Linux_for_Tegra/” directory on the host PC if you run this command (the “.” is important):
df -H -T .
If there was insufficient space, then the image would truncate. If the filesystem type was not ext4, then flash wouldn’t know about this, and would happily install the wrong filesystem type. If there is a difference in release used for the driver package versus the sample rootfs, then this too would cause an issue. If all of those are ruled out, then it would be useful to see the complete serial console boot log (the error shown is definitely a kernel issue, but it doesn’t say at what stage and gives no idea what lead up to the issue).
thank you very much
I command
df -H -T .
result:
/dev/sda1 ext4 105G 93G 6.9G 94% /
…
…
Not long ago I showed the following error in debugging the kernel.
Unable to handle kernel NULL pointer dereference at virtual address 00000178
[ 4.089150] pgd = ffffffc0015fa000
[ 4.089155] [00000178] *pgd=000000017a5d8003, *pud=000000017a5d8003, *pmd=000000017a5d9003, *pte=00e8000050041707
[ 4.089158] Internal error: Oops: 96000005 [#1] PREEMPT SMP
[ 4.089160] Modules linked in:
[ 4.089164] CPU: 1 PID: 1 Comm: swapper/0 Not tainted 4.4.159 #1
[ 4.089165] Hardware name: jetson_tx1 (DT)
[ 4.089167] task: ffffffc0fa6f0000 ti: ffffffc0fa6a4000 task.ti: ffffffc0fa6a4000
[ 4.089174] PC is at find_matching_color_fmt+0x10/0xa4
[ 4.089177] LR is at camera_common_enum_mbus_code+0x44/0xb4
[ 4.089179] pc : [<ffffffc0007b6d40>] lr : [<ffffffc0007b6e18>] pstate: 20000045
[ 4.089180] sp : ffffffc0fa6a7830
[ 4.089182] x29: ffffffc0fa6a7840 x28: 0000000000000018
[ 4.089185] x27: ffffffc0010f1fa0 x26: ffffffc0010f1fa8
[ 4.089187] x25: ffffffc0f8bb2050 x24: ffffffc0f8bb2050
[ 4.089189] x23: ffffffc07d35dbe8 x22: ffffffc07d35d018
[ 4.089191] x21: ffffffc0fa6a7908 x20: ffffffc0f9094c20
[ 4.089193] x19: ffffffc07d35d018 x18: 0000000000000000
[ 4.089195] x17: 0000000000000000 x16: 0000000000000000
[ 4.089197] x15: 0000000000000000 x14: 0ffffffffffffffe
[ 4.089200] x13: 0000000000000020 x12: 0101010101010101
I am using the command "make menuconfig "now.
And set “Compile the kernel with debug info” save and Recompile the kernel
Then I find vmlinux in the kernel_out directory
command “GDB vmlinux”
~/TX1283/kernel_sc/kernel_out$ sudo gdb vmlinux
GNU gdb (Ubuntu 7.11.1-0ubuntu1~16.5) 7.11.1
Copyright (C) 2016 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law. Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from vmlinux...done.
(gdb) l *find_matching_color_fmt+0x10/0xa4
(gdb)
But there is no error code.Where is it wrong?
At least part of the problem is that you seem to be out of disk space (“out of space” is a relative thing…a flash probably needs 34GB 20GB (TX1 has a smaller image than TX2…the previously mentioned 34GB was for TX2 for safe extra space) or more of empty space). A result of flashing when out of space would be that the flash appears to run normally, but the image on the installed filesystem would be truncated (and this is unpredictable as to what would happen). You only have 6.9GB. You could delete these two files, and measure again with “df -H” to see if spare disk space is at least 32GB 20GB (I’d actually recommend more):
cd "...wherever.../Linux_for_Tegra/bootloader/
sudo rm system.img.*
df -H -T .
Note that not all flash operations need that file space, but if you flash the rootfs without space, then it is guaranteed you will have unexpected issues. Your most recent command would include flashing the rootfs. It is possible the issues you are running into are caused by this, but it isn’t certain.
The actual error is related to camera color, but I couldn’t say specifically…someone knowing more about camera drivers would need to answer, but there is no way to answer unless we know the installed image is valid. You could verify at least that some of the NVIDIA-specific drivers are in place, but beware this is just a tiny amount of content (and for certain your camera software is not checked by this command):
sha1sum -c /etc/nv_tegra_release