Dear nvidia team:
We use the mksparse tool to generate system images. The system.img converted from the same SSD is 19G and 26G. Why is there such a difference? APP partitions are all the same size.The command is “mksparse -v --fillpattern=0 goldenimage.raw system.img”
Hi,
Can you please elaborate the issue more?
You mean you have two different systems, and get different image sizes after backup?
Size of partitions is not the point; it’s how much space actually being taken up matters.
I don’t know what you are talking about.
What do you have with df- h when getting the 26GB image?
How did you backup the raw image?
dd if=/dev/nvme0n1p1 | dd of=Public/goldenimage.raw
It’s the same as this topic:
mksparse behaving incorrectly - Jetson & Embedded Systems / Jetson TX2 - NVIDIA Developer Forums
I don’t feel like it’s the correct way doing it.
Please see if you get the same result with this:
https://docs.nvidia.com/jetson/archives/r35.4.1/DeveloperGuide/text/SD/FlashingSupport.html#cloning-rootfs-with-initrd
I don’t think the initrd includes any driver for manipulating NTFS partitions; use ext4 instead.
I follow the way you told me and get the system.img size is 19GB and not 26G.So I will release the version in the way you said. If the image I get is still much larger than the actual occupancy size, this way may be problematic and unstable.
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.