On 31/10/2019, Bjørn Mork <bj...@mork.no> wrote: > John Crispin <j...@phrozen.org> writes: > >> Hi, >> should we use v5.4 as our next kernel ? > > Hello John! > > I am still struggling to understand how the project makes release > decisions. I don't think your question makes sense without considering > target release dates, at least in a yearly resolution, given the massive > amount of work involved with switching kernel versions. > > So my question is: Is it realistic to expect an OpenWrt release with > v5.4 next year? > > Because if it isn't, then v5.4 is really a waste of time IMHO. You > should target the next longterm kernel after v5.4 instead for the first > 2021 release. > > And of course, if you think a 2020 release with v5.4 is realistic, then > there's an open question on what to do with v4.19. Are two OpenWrt > releases next year likely/possible? If not, then v5.4 should probably > replace v4.19 as the current target kernel as soon as possible. > > Yes, I know what your meeting minutes says. But two releases in 2019 is > obviously not going to happen. And personally I don't see how that's > suddenly going to change for 2020 unless you do something drastic to > simplify the release process. Yes, that means lowering the standards. > Maybe making releases that won't be maintained, and therefore don't need > to be "future proof"? > > > Bjørn > > > _______________________________________________ > openwrt-devel mailing list > openwrt-devel@lists.openwrt.org > https://lists.openwrt.org/mailman/listinfo/openwrt-devel >
Yes, pretty strong arguments to skip 5.4 and just use 4.19 for the 20.xx release (which should not be 20.01 or 20.07 but more likely 20.10) _______________________________________________ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/mailman/listinfo/openwrt-devel
