[Openembedded-architecture] Changes that switching from smart to dnf will cause

[Alexander Kanavin]

On 02/15/2017 11:16 AM, Anders Darander wrote:
> Could you add some strong wording in the documentation of
> pkg_postinst_ontarget4{PN}(), that it's only to be used as a _last_
> resort; when something really needs access to the real HW?

Sure, yes. Right now the 'exit 1' firstboot deferral can be abused in 
exactly same way, and probably is somewhere in 3rd party layers. It's 
only a bit more awkward to write:

pkg_postinst_${PN}() {
     # make sure running on the target
     if [ x"$D" != "x" ]; then
         exit 1
     fi
     /bin/mknod -m 0622 /dev/grsec c 1 13
}

(from meta-oe/recipes-support/gradm/gradm_3.1.bb)

(I'm not saying that gradm is doing this needlessly, maybe it has to, 
it's just an example of a recipe that is incompatible with read only 
rootfs).

Alex