Adrian Schmutzler <m...@adrianschmutzler.de> [2019-11-28 15:40:32]: > have mixed kernel releases again
This is not an option, is it? > or would have to throw out a lot of targets for each of the stable branches > (which might be always the same set of less-cared targets)? If the target is obsolete and/or orphaned, what could we do instead? I mean, if there is no interest to rebase the patches to the next kernel version and at least compile test them (not even dare to run test it), whats the point of claiming "support" for this device? It would be two step process, source-only in one release (users notice this very quickly, perhaps someone manage to rebase it), then remove it in the next release. > This might also have negative impact on the "slow" targets, as they might > quickly end up with no supported stable branch anymore (unless there are > mixed-kernel releases again). Flash size is limiting factor, not the speed. Take a look at 4/32 state, it's still somehow doable to support some of those devices even with 4.19 and maybe 5.4 kernels, but there's no interest to solve that in maintainable and acceptable way. In other words, obsolete/orphaned devices. -- ynezz _______________________________________________ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/mailman/listinfo/openwrt-devel