On 25/02/16 07:23 AM, Ben Armstrong wrote: > > All of that being said, some people have reported success with at > least the Stretch live-boot (and possibly also live-config?) being > used directly in Jessie builds, provided that a jessie-backports > kernel is also used. If that use case continues to work for people, we > do not actually need to put through a proposed-updates version of > either live-boot or live-config to support it. We would only need > proposed-updates uploads if we also want to support pure Jessie builds > (i.e. no backports kernel) with fixes introduced in Stretch. For > example, if the UEFI patch currently under discussion gets accepted, > and we continue to use live-build for future Jessie point-releases, > and we want that UEFI patch to be included in those point-releases, it > would be best if we could get that patch backported for the live-build > currently in Jessie and then accepted through proposed-updates.
To clarify, my UEFI example doesn't really match the topic introduced in this paragraph, since UEFI support would need to go into live-build, not the other two packages. There are also important fixes in live-boot, and possibly also live-config, that have been made arlready in stretch and yet are also applicable to Jessie (autologin via systemd being broken in the Jessie standard image falls in this category) and therefore, proposed-updates uploads of those packages are probably also a good idea. I did the initial work for the autologin patch backport, but have lacked the time and incentive since then to get the patch through proposed-updates, as it is lower down on my list of priorities than other things in my life at this time. Ben
