Hello all,
This one is a bit of a linux noob question, but sometimes we all have 'em.
So, I'm running on a custom imx6 board that has 8 GB of space on its MMC. To initially flash the firmware, I essentially treated the board as a mass storage device and then did the required partitioning and writing with fdisk
and dd
.
When I do sysupgrades, the rootfs is shrunk back down to whatever I set TARGET_ROOTFS_PARTSIZE
to. This makes sense, as I would think there would be some argument whenever it's mounting root telling it the size (although I could not find where it might be).
What I don't get is that if I increase TARGET_ROOTFS_PARTSIZE
and make
, the size of the produced binary is larger, but not the size of the size of the produced rootfs file.
Here is my device profile if that reveals anything
define Device/mine
DEVICE_TITLE := manufacturer my-device-name
SUPPORTED_DEVICES := my-device
DEVICE_DTS := my-custom-device-tree
BOOT_SCRIPT := ha-i-dont-even-use-but-i-copy-pastad-and-didnt-remove-yet
UBOOT := unused
KERNEL := kernel-bin | append-uImage-fakehdr filesystem
KERNEL_IMG := kernel.img
FILESYSTEMS := ext4
IMAGES := sysupgrade.bin also-my-custom-dtb.dtb
IMAGE/also-my-custom-dtb.dtb := install-dtb
IMAGE/combined.bin := append-rootfs | pad-extra 128k | imx6-sdcard
IMAGE/sysupgrade.bin := sysupgrade-tar | append-metadata
endef
Lastly, is there some way I could programmatically change the size of rootfs on first boot after upgrade?
I know that this could be done by booting into another partition and then modifying, but I was hoping for something a bit cleaner than that.
Thanks for any insight!