summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)Author
2012-04-20manager: split transaction.[ch]Michal Schmidt
manager.c takes care of the main loop, unit management, signal handling, ... transaction.c computes transactions. After split: manager.c: 65 KB transaction.c: 40 KB
2012-04-20manager: Transaction as an objectMichal Schmidt
This makes it obvious that transactions are short-lived. They are created in manager_add_job() and destroyed after the application of jobs. It also prepares for a split of the transaction code to a new source.
2012-04-20job: job_uninstall()Michal Schmidt
Split the uninstallation of the job from job_free() into a separate function. Adjust the callers. job_free() now only works on unlinked and uninstalled jobs. This enforces clear thinking about job lifetimes.
2012-04-20manager: simplify transaction_abort()Michal Schmidt
This is equivalent.
2012-04-20job: allow job_free() only on already unlinked jobsMichal Schmidt
job_free() is IMO too helpful when it unlinks the job from the transaction. The callers should ensure the job is already unlinked before freeing. The added assertions check if anyone gets it wrong.
2012-04-20manager: fix commentMichal Schmidt
2012-04-20tmpfiles: fix error messageMichal Schmidt
2012-04-20log: fix LOG_TARGET_JOURNAL_OR_KMSGLennart Poettering
2012-04-20mount-setup: don't log with LOG_ERROR if a mount that doesn't matter failsLennart Poettering
2012-04-20main: log to the journal in container mode, by defaultLennart Poettering
2012-04-20docs: hook-up gtk-doc to 'make check'Kay Sievers
2012-04-19fix typo in src/shared/install.cSven Anders
2012-04-18remove MS_* which can not be combined with current kernel codeKay Sievers
MS_BIND|MS_MOVE can not be combined: do_mount() else if (flags & MS_BIND) do_loopback(&path, dev_name, flags & MS_REC); [...] else if (flags & MS_MOVE) do_move_mount(&path, dev_name); MS_REMOUNT|MS_UNBINDABLE can not be combined: do_mount() if (flags & MS_REMOUNT) do_remount(&path, flags & ~MS_REMOUNT, mnt_flags, data_page); [...] else if (flags & (MS_SHARED | MS_PRIVATE | MS_SLAVE | MS_UNBINDABLE)) do_change_type(&path, flags);
2012-04-17udev: unify /dev static symlink setupKay Sievers
2012-04-17silence a bunch of gcc warningsKay Sievers
2012-04-17selinux: unify systemd and udev codeKay Sievers
2012-04-17mkdir: do not use alloca() in a loopKay Sievers
2012-04-17udev: kill udev_selinux_setfscreateconat()Kay Sievers
2012-04-17udev: replace util_create_path() with mkdir_parents()Kay Sievers
2012-04-16udev: use startswith() and streq()Kay Sievers
2012-04-16udev: remove configuration options for /dev, /sys, /run directoriesKay Sievers
2012-04-16udev: replace ARRAY_SIZE() with ELEMENTSOF()Kay Sievers
2012-04-16logind: remove redundant entries from logind's default controller lists tooLennart Poettering
2012-04-16systemctl: show main and control PID explicitly in cgroup-showLennart Poettering
In some cases the main/control PID of a service can be outside of the services cgroups (for example, if logind readjusts the processes' cgroup). In order to clarify this for the user show the main/control PID in the cgroup tree nonetheless, but mark them specially.
2012-04-16util: introduce memdup()Lennart Poettering
2012-04-16cgls: don't show empty cgroups by defaultLennart Poettering
2012-04-16logind: hook up inhibit logic with idle hint logicLennart Poettering
2012-04-16logind: add shutdown/suspend/idle inhibition frameworkLennart Poettering
2012-04-16manager: remove unavailable/redundant entries from default controllers listLennart Poettering
2012-04-16cgroup: if a controller is not available don't try to create cgroups in its ↵Lennart Poettering
hierarchy
2012-04-15udev: udev_monitor_new_from_socket() - log error and remove from public headerKay Sievers
2012-04-15udev: update some rulesKay Sievers
2012-04-15udev: cleanup the udev cgroup when the daemon enters the idle stateKay Sievers
2012-04-14udev: move all unconditionally installed rules to rules/Kay Sievers
2012-04-14udev: keymap - move src/udev/force-release-maps/ to keymaps-force-release/Kay Sievers
2012-04-14udev: test - move test/ and src/test/Kay Sievers
2012-04-13service: place control command in subcgroup control/Lennart Poettering
Previously, we were brutally and onconditionally killing all processes in a service's cgroup before starting the service anew, in order to ensure that StartPre lines cannot be misused to spawn long-running processes. On logind-less systems this has the effect that restarting sshd necessarily calls all active ssh sessions, which is usually not desirable. With this patch control processes for a service are placed in a sub-cgroup called "control/". When starting a service anew we simply kill this cgroup, but not the main cgroup, in order to avoid killing any long-running non-control processes from previous runs. https://bugzilla.redhat.com/show_bug.cgi?id=805942
2012-04-13update .gitignoreKay Sievers
2012-04-13polkit: spawn agent in --fallback modeLennart Poettering
2012-04-13update .gitignoreKay Sievers
2012-04-13build-sys: add stub makefiles to all subdirs to ease development with emacsLennart Poettering
2012-04-13nspawn: add missing include linesLennart Poettering
2012-04-13watchdog: make watchdog dbus properties writableLennart Poettering
2012-04-13dbus: automatically send out changed events for properties written toLennart Poettering
2012-04-13udev: gudev - silence introspection scanner warningKay Sievers
<kay> walters: you happen to know how to silence this? gudev.h:24: Warning: GUdev: symbol='_GUDEV_INSIDE_GUDEV_H': Unknown namespace for symbol 'GUDEV_INSIDE_GUDEV_H' <walters> kay, probably: <walters> -#define _GUDEV_INSIDE_GUDEV_H 1 <walters> +#define _GUDEV_INSIDE_GUDEV_H <walters> kay, if the scanner sees a define with a value it assumes it's a constant for public consumption <walters> kay, patch in https://bugzilla.gnome.org/show_bug.cgi?id=674072 fwiw <kay> walters: cool, thanks! <kay> walters: your workaround removing the "1" works. nice!
2012-04-13dbus: expose whether we have a hardware watchdog on the busLennart Poettering
2012-04-13unit: signal explicitly if a condition failed in unit_start()Lennart Poettering
We shouldn't print a status message on the console if we skipped a unit due to a condition. Hence make unit_start() return -ENOEXEC in such a case which is mapped to JOB_SKIPPED which results in no console message.
2012-04-13udev: move libudev, gudev to src/; move gudev/docs/, udev/docs/ to to docs/Kay Sievers
2012-04-13logind: explicitly check for /dev/tty0Lennart Poettering
2012-04-13main: unset some more env varsLennart Poettering
generated by cgit v1.2.3-54-g00ecf (git 2.45.2) at 2024-10-25 15:26:40 +0000