On Wed, Mar 18, 2020 at 3:02 AM Otavio Salvador <otavio.salva...@ossystems.com.br> wrote: > > On Tue, Mar 17, 2020 at 12:29 PM Alex Kiernan <alex.kier...@gmail.com> wrote: > > udev is packaged before systemd so any wildcard inclusions in FILES will > > override later specifics. List all udev rules explicitly so that the > > systemd specific rules, packaged alongside systemd, appear in the > > correct package. > > > > Signed-off-by: Alex Kiernan <alex.kier...@gmail.com> > > I believe udev ought to be on PACKAGE_BEFORE_PN so it is still handled > with the wildcard and allow for easier maintenance in future. >
It's done with an =+ at the moment, replacing it with PACKAGE_BEFORE_PN would certainly make it clearer (and avoid holes where dev/dbg stuff could leak through). But I don't think that helps with the greedy wildcard problem - we have: FILES_udev += "... ${rootlibexecdir}/udev/rules.d/*.rules \ and FILES_${PN} = " ... ${nonarch_base_libdir}/udev/rules.d/70-uaccess.rules \ ${nonarch_base_libdir}/udev/rules.d/71-seat.rules \ ${nonarch_base_libdir}/udev/rules.d/73-seat-late.rules \ ${nonarch_base_libdir}/udev/rules.d/99-systemd.rules \ Where the FILES_udev consumes all the rules before ${PN} gets a look in, unless there's some magic I'm overlooking? The inconsistent path usage wants fixing too... I'd not noticed that before! -- Alex Kiernan -- _______________________________________________ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core