On Fri, 10 Nov 2023, Andrew Cooper wrote: > On 09/11/2023 11:59 pm, Stefano Stabellini wrote: > > On Thu, 9 Nov 2023, Jan Beulich wrote: > >> On 08.11.2023 15:37, Andrew Cooper wrote: > >>> These 3 Kconfig docs were imported from Linux erroneously. They are > >>> GPL-2.0-only in Linux, but have no SPDX tag and were placed in such a way > >>> to > >>> be included by the blanket statement saying that all RST files are > >>> CC-BY-4.0. > >>> > >>> We should not be carrying a shadow copy of these docs. They aren't even > >>> wired > >>> into our Sphinx docs, and anyone wanting to refer to Kconfig docs is > >>> going to > >>> look at the Linux docs anyway. These, and more docs can be found at: > >>> > >>> https://www.kernel.org/doc/html/latest/kbuild/ > >>> > >>> which also have corrections vs the snapshot we took. > >> Imo this reference ... > >> > >>> Fixes: f80fe2b34f08 ("xen: Update Kconfig to Linux v5.4") > >>> Signed-off-by: Andrew Cooper <andrew.coop...@citrix.com> > >>> --- > >>> CC: George Dunlap <george.dun...@citrix.com> > >>> CC: Jan Beulich <jbeul...@suse.com> > >>> CC: Stefano Stabellini <sstabell...@kernel.org> > >>> CC: Wei Liu <w...@xen.org> > >>> CC: Julien Grall <jul...@xen.org> > >>> CC: Henry Wang <henry.w...@arm.com> > >>> --- > >>> docs/misc/kconfig-language.rst | 701 --------------------------- > >>> docs/misc/kconfig-macro-language.rst | 247 ---------- > >>> docs/misc/kconfig.rst | 304 ------------ > >>> 3 files changed, 1252 deletions(-) > >>> delete mode 100644 docs/misc/kconfig-language.rst > >>> delete mode 100644 docs/misc/kconfig-macro-language.rst > >>> delete mode 100644 docs/misc/kconfig.rst > >> ... wants putting into, say, the last of these three files you delete, as > >> a replacement. I can't spot any other place where we would have such a > >> reference. > >> > >> One problem I see with deleting our shadow copy is that by referring to > >> Linux'es doc, the wrong impression may arise that whatever new features > >> they invent we also support. Thoughts? (If nothing else, I'd expect this > >> aspect to be mentioned / justified in the description.) > > I think the ideal solution would be to replace the shadow copies with a > > link to the Linux docs of a specific Linux tag (v5.4), instead of > > generic Linux master. I am not sure where to place the links though. > > I don't personally think we need to keep any other reference around. > They're not interesting, because they're not going to be found by anyone > except those who already know they're there, and won't need to refer to > them for the kind of content they provide. > > Kconfig isn't a fast-moving target, and there's nothing new in Linux vs > what we've got here. The only interesting difference between us and > Linux is the fact we don't use modules, and we didn't even strip that > out of the shadow copy. > > We do have xen/tools/kconfig/README.source which states where it came from. > > I could be persuaded to add the following hunk. What we have isn't > precisely v5.4 anyway - we've got some reasonable differences in the > makefile side of things.
The below is good enough in my opinion. Ack. > diff --git a/xen/tools/kconfig/README.source > b/xen/tools/kconfig/README.source > index 44631f68e8..ac394106b9 100644 > --- a/xen/tools/kconfig/README.source > +++ b/xen/tools/kconfig/README.source > @@ -5,5 +5,7 @@ in this part of the Xen source tree. > > xen/tools/kconfig > ----------------- > -The kconfig directory was originally imported from the linux kernel > -git tree at kernel/git/torvalds/linux.git, path: scripts/kconfig > +The kconfig directory was originally imported from the Linux kernel > +git tree at kernel/git/torvalds/linux.git, path: scripts/kconfig of > +roughly v5.4. Linux's documentation can be found at: > +https://www.kernel.org/doc/html/latest/kbuild/ >
