* Re: [OE-core] [PATCH v2] bitbake.conf: Extend distro feature usrmerge to merge bin+sbin [not found] <18C52E3640016A00.1442561@lists.openembedded.org> @ 2026-07-24 9:05 ` Jörg Sommer 2026-07-24 20:51 ` Randolph Sapp 0 siblings, 1 reply; 2+ messages in thread From: Jörg Sommer @ 2026-07-24 9:05 UTC (permalink / raw) To: openembedded-core, richard.purdie, ross.burton, mathieu.dubois-briand, rs [-- Attachment #1: Type: text/plain, Size: 1777 bytes --] Jörg Sommer via lists.openembedded.org schrieb am Fr 24. Jul, 11:00 (+0200): > From: Randolph Sapp <rs@ti.com> > > The split of bin and sbin became more artificial over time, and it is no > longer clear why programmes goes into the one or the other directory; e.g. > forever normal users have used sendmail in /usr/sbin, and fdisk and mkfs > learned to operate on files to create images long ago. > > Because systemd drives this movement to have only /usr/bin and symlink /bin, > /sbin and /usr/sbin to there, the usrmerge feature (started by systemd too) > should be extended to include the binmerge. For a system with split bin/sbin > `systemctl status` reports the system as tainted for *unmerged-bin.* > > Override most sbin variables to point to their bin equivalent when > usermerge is enabled, with the exception being base_sbindir_native and > sbindir_native. There is no guarantee the host machine is using this > filesystem scheme. > > This check was added to systemd back in 2024 [2]. > > [1] https://gitlab.archlinux.org/archlinux/packaging/packages/filesystem/-/blob/main/PKGBUILD > [2] https://github.com/systemd/systemd/commit/844863c61e7b501097da84a1e4d1e4a6aa6d9f0d > > Signed-off-by: Randolph Sapp <rs@ti.com> > Signed-off-by: Jörg Sommer <joerg.sommer@navimatix.de> > --- > > I took Randolph's version and reworked it a bit. But the idea stays the same. This is a follow-up to <20260408195024.1874335-2-rs@ti.com> https://lists.openembedded.org/g/openembedded-core/message/234858 -- Navimatix GmbH T: 03641 - 327 99 0 Tatzendpromenade 2 F: 03641 - 526 306 07745 Jena www.navimatix.de Geschäftsführer: Steffen Späthe, Jan Rommeley Registergericht: Amtsgericht Jena, HRB 501480 [-- Attachment #2: smime.p7s --] [-- Type: application/pkcs7-signature, Size: 6684 bytes --] ^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [OE-core] [PATCH v2] bitbake.conf: Extend distro feature usrmerge to merge bin+sbin 2026-07-24 9:05 ` [OE-core] [PATCH v2] bitbake.conf: Extend distro feature usrmerge to merge bin+sbin Jörg Sommer @ 2026-07-24 20:51 ` Randolph Sapp 0 siblings, 0 replies; 2+ messages in thread From: Randolph Sapp @ 2026-07-24 20:51 UTC (permalink / raw) To: Jörg Sommer, openembedded-core, richard.purdie, ross.burton, mathieu.dubois-briand, rs On Fri Jul 24, 2026 at 4:05 AM CDT, Jörg Sommer wrote: > Jörg Sommer via lists.openembedded.org schrieb am Fr 24. Jul, 11:00 (+0200): >> From: Randolph Sapp <rs@ti.com> >> >> The split of bin and sbin became more artificial over time, and it is no >> longer clear why programmes goes into the one or the other directory; e.g. >> forever normal users have used sendmail in /usr/sbin, and fdisk and mkfs >> learned to operate on files to create images long ago. >> >> Because systemd drives this movement to have only /usr/bin and symlink /bin, >> /sbin and /usr/sbin to there, the usrmerge feature (started by systemd too) >> should be extended to include the binmerge. For a system with split bin/sbin >> `systemctl status` reports the system as tainted for *unmerged-bin.* >> >> Override most sbin variables to point to their bin equivalent when >> usermerge is enabled, with the exception being base_sbindir_native and >> sbindir_native. There is no guarantee the host machine is using this >> filesystem scheme. >> >> This check was added to systemd back in 2024 [2]. >> >> [1] https://gitlab.archlinux.org/archlinux/packaging/packages/filesystem/-/blob/main/PKGBUILD >> [2] https://github.com/systemd/systemd/commit/844863c61e7b501097da84a1e4d1e4a6aa6d9f0d >> >> Signed-off-by: Randolph Sapp <rs@ti.com> >> Signed-off-by: Jörg Sommer <joerg.sommer@navimatix.de> >> --- >> >> I took Randolph's version and reworked it a bit. But the idea stays the same. > > This is a follow-up to <20260408195024.1874335-2-rs@ti.com> > https://lists.openembedded.org/g/openembedded-core/message/234858 Thanks for taking this forward Jörg. My main build machine is too slow to look at Yocto right now. Trying to requisition more hardware. ^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-07-24 20:51 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <18C52E3640016A00.1442561@lists.openembedded.org>
2026-07-24 9:05 ` [OE-core] [PATCH v2] bitbake.conf: Extend distro feature usrmerge to merge bin+sbin Jörg Sommer
2026-07-24 20:51 ` Randolph Sapp
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.