In some part of the system, we determine whether the rootfs is read-only by checking the fstab or /proc/mounts, in other part of the system, we use the value of ROOTFS_READ_ONLY in /etc/default/rcS as a criteria. This is confusing and makes system inconsistent. We should drop the use of ROOTFS_READ_ONLY and use a uniform way to determine whether rootfs is read-only. On target, we should use /proc/mounts to check whether / is read-only; on host, we should use $ROOT_DIR/etc/fstab to check whether the rootfs is going to be mounted as read-only or not. //Chen Qi
A question is it important to know if the rootfs is READ-ONLY during build or install time? If so, the rootfs and thus ${ROOT_DIR} will not be available yet.
(In reply to comment #1) > A question is it important to know if the rootfs is READ-ONLY during build > or install time? The answer would be 'No' with current approach. //Chen Qi
The first solution spawns too many processes at boot time. We need a better solution.
Considering the overhead brought in by checking the /proc/mounts or /etc/fstab in each boot script, the READ_ONLY_ROOTFS value in /etc/default/rcS is still useful. However, we should at least add a check for conflicting configurations between /etc/fstab and /etc/default/rcS. //Chen Qi
merged into poky/master: e6e44cf9c10116b9bd2260eb344031d9b897f669 http://git.openembedded.org/?p=openembedded-core.git&a=commit;h=1565a0c5a3f245703e280ca90cf11d3f9374788a