s6-linux-init/src/init/s6-linux-init-maker.c, branch v0.2.0.0 Helpers to build an s6-based init program https://git.skarnet.org/cgit/s6-linux-init/atom/src/init/s6-linux-init-maker.c?h=v0.2.0.0 2016-09-10T13:43:57Z Change initial_path to include /usr/sbin and /sbin, because stage 3 2016-09-10T13:43:57Z Laurent Bercot ska-skaware@skarnet.org 2016-09-10T13:43:57Z urn:sha1:2e56f6dd37022646f1c44c8cc9365f0da249a39a Make s6-linux-init a real init package. 2016-09-10T13:29:56Z Laurent Bercot ska-skaware@skarnet.org 2016-09-10T13:29:56Z urn:sha1:dfa16e4a89158f3d3198db71e2eba3bda9330078 This means: - removing leaky options to s6-linux-init-maker. Default initial_path is now always /usr/bin:/bin; the uid and gid of the catch-all logger (used at boot time) can now be given numerically instead of relying on the (run-time) user db mapping. - moving s6-halt, s6-poweroff and s6-reboot over here, from s6-linux-utils - clarifying on the documentation a bit. - make s6-linux-init-maker write stage1 to dir/init instead of stdout 2016-04-14T13:21:19Z Laurent Bercot ska-skaware@skarnet.org 2016-04-14T13:21:19Z urn:sha1:4448842bb141430e23498d5a969b5cefe8786e57 - version: 0.1.0.0 Fix update/global-links dependencies 2016-03-14T22:26:07Z Laurent Bercot ska-skaware@skarnet.org 2016-03-14T22:26:07Z urn:sha1:1e6f50c7c7f9202114be4e300807623ddbf77255 Empty env in stage 1, add option to dump it somewhere better configure, better strip, small init-maker bugfix 2015-11-19T21:47:45Z Laurent Bercot ska-skaware@skarnet.org 2015-11-19T21:47:45Z urn:sha1:73bea8ab25859ca8a3e43c008087dd34a9d0b945 - added stage2_finish support 2015-10-13T16:05:05Z Laurent Bercot ska-skaware@skarnet.org 2015-10-13T16:05:05Z urn:sha1:e00d2c4d88c90ccc6d091bde07d412936f8886d3 - s6 dep bumped to 2.2.2.0 Add s6-setsid -qb invocation to stage 1 (thanks Dr. Willy) 2015-09-08T12:12:25Z Laurent Bercot ska-skaware@skarnet.org 2015-09-08T12:12:25Z urn:sha1:34d555e6010052d84904b8005f51f4ad909ebdd7 Minor, cosmetic fixes to s6-linux-init-maker 2015-06-25T23:04:31Z Laurent Bercot ska-skaware@skarnet.org 2015-06-25T23:04:31Z urn:sha1:cac0c78d8a99be335eca80ef0bb3392d274b1058 This was still racy as fuck, because redirfd -wnb is a dirty hack and 2015-06-25T22:35:06Z Laurent Bercot ska-skaware@skarnet.org 2015-06-25T22:35:06Z urn:sha1:60bd7b90a21690178c17b1a3daa671076fbed9d4 actually opens the fifo for reading, long enough for pid 2 to unblock! So it must be called *before* the fork. Rewrote the init script logic; found a nice way to factorize the redirect_stage2 impact. This one's good... I think. replace if with foreground in finish script 2015-06-25T21:07:20Z Laurent Bercot ska-skaware@skarnet.org 2015-06-25T21:07:20Z urn:sha1:21cb73b507d27bcb101a1f3ae95312d22952988b