The answer was simple: the rules for the stable tree do not allow the inclusion of any patch that has not already been merged, in some form, into the mainline.
This is not quite correct. If a patch goes into a stable branch and causes a regression there, but we know it doesn't cause that regression in mainline, then it is OK to revert it on the stable branch only (though in some cases it's better to cherry-pick the missed dependencies of the original patch from mainline). Also, a patch applied to the stable branch may sometimes look quite different from the corresponding patch(es) in mainline, but this is quite rare.
Copyright © 2017, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds