summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2013-03-26readahead: cleanupsAuke Kok
- check for OOM - no need to use floats and round()
2013-03-26readahead: chunk on spinning mediaAuke Kok
Readahead has all sorts of bad side effects depending on your storage media. On rotating disks, it may be degrading startup performance if enough requests are queued spanning linearly over all blocks early at boot, and mount, blkid and friends want to insert reads to the start of these block devices after. The end result is that on spinning disks with ext3/4 that udev and mounts take a very long time, and nothing really happens until readahead is completely finished. This has the net effect that the CPU is almost entirely idle for the entire period that readahead is working. We could have finished starting up quite a lot of services in this time if we were smarter at how we do readahead. This patch sorts all requests into 2 second "chunks" and sub-sorts each chunk by block. This adds a single cross-drive seek per "chunk" but has the benefit that we will have a lot of the blocks we need early on in the boot sequence loaded into memory faster. For a comparison of how before/after bootcharts look (ext4 on a mobile 5400rpm 250GB drive) please look at: http://foo-projects.org/~sofar/blocked-tests/ There are bootcharts in the "before" and "after" folders where you should be able to see that many low-level services finish 5-7 seconds earlier with the patch applied (after).
2013-03-26NEWS: tweak grammarZbigniew Jędrzejewski-Szmek
2013-03-26NEWS: typo fixsystemd/v199Lennart Poettering
2013-03-26build-sys: bump version and .so revisionsLennart Poettering
2013-03-26cryptsetup: when prompting for password use GPT partition labelLennart Poettering
If there's a GPT partition label set for a LUKS partition, then it's nicer to show that than the model number, when asking for a passphrase.
2013-03-26man: document that the passive units cannot be pulled in mnaullyLennart Poettering
2013-03-26fstab-generator: drop x-initrd.rootfs mount optionHarald Hoyer
x-initrd.mount now has different meanings, if fstab-generator is called in the initramfs. initrd:/etc/fstab and x-initrd.mount defines mounts for the initrd-root-fs.target initrd:/sysroot/etc/fstab and x-initrd.mount defines mounts for the initrd-fs.target
2013-03-26units: disallow manual starting of passive unitsLennart Poettering
As passive units only are useful for ordering things within the initial transaction there is no point in ever activating them manually, hence refuse it.
2013-03-26fstab-generator: degrade the message about missing "root=" to log_debugHarald Hoyer
Some installation media (fedora at least) does not have and need a "root=" argument on the kernel command line.
2013-03-26Drop src/login/uaccess.c, dead codeMartin Pitt
This moved to src/udev/udev-builtin-uaccess.c a while ago.
2013-03-26bus: fix missing variable initializationLennart Poettering
2013-03-26modules-load: there's really no point in mentioning that a certain modules ↵Lennart Poettering
is already loaded After all, this runs in parallel to udev, so there's quite a chance it already is....
2013-03-26build-sys: make gcc shut upLennart Poettering
2013-03-26build-sys: ship missing unit fileLennart Poettering
2013-03-26units: downgrade dependency on sockets.target/timers.target/paths.target by ↵Lennart Poettering
basic.target There isn't really any need to require any targets but sysinit.target from basic.target, so downgrade sockets.target, paths.target, timers.target.
2013-03-26bus: when we are talking to a bus, SCM_CREDS/SCM_SECLABEL are not very usefulLennart Poettering
2013-03-26bus: automatically generate minimal introspection data to find installed objectsLennart Poettering
2013-03-26bus: rename send_hello flag to bus_cientLennart Poettering
This way we can hide more than just the hello logic behind this flag, for example, later on automatic match management.
2013-03-26build-sys: create kernel/install.d directoriesKay Sievers
2013-03-26NEWS: updateKay Sievers
2013-03-25cryptsetup-generator: let's be a bit more efficient with strv_extend()Lennart Poettering
2013-03-25cryptsetup-generator: add a missing OOM checkLennart Poettering
2013-03-25update NEWSLennart Poettering
2013-03-25prepare NEWS for 199Lennart Poettering
2013-03-25man: properly document the system targets that are also available for the ↵Lennart Poettering
user instance
2013-03-25man: rearrange systemd.special(7) to clarify which units are passive and ↵Lennart Poettering
which ones aren't This also adds a short explanation paragraph for this.
2013-03-25units: there is no point in pulling in ordering 'provides'-style targetsLennart Poettering
Units such as nss-lookup.target, nss-user-lookup.target, remote-fs-pre.target, local-fs-pre.target, time-sync.target, rpcbind.target are to be pulled in by the implementing services, and that's there only purpose. They should not have any 'active component' otherwise, so let's drop all further deps from these units.
2013-03-25update TODOLennart Poettering
2013-03-25units: introduce remote-fs-setup.target to pull in dependencies from remote ↵Lennart Poettering
mounts This introduces remote-fs-setup.target independently of remote-fs-pre.target. The former is only for pulling things in, the latter only for ordering. The new semantics: remote-fs-setup.target: is pulled in automatically by all remote mounts. Shall be used to pull in other units that want to run when at least one remote mount is set up. Is not ordered against the actual mount units, in order to allow activation of its dependencies even 'a posteriori', i.e. when a mount is established outside of systemd and is only picked up by it. remote-fs-pre.target: needs to be pulled in automatically by the implementing service, is otherwise not part of the initial transaction. This is ordered before all remote mount units. A service that wants to be pulled in and run before all remote mounts should hence have: a) WantedBy=remote-fs-setup.target -- so that it is pulled in b) Wants=remote-fs-pre.target + Before=remote-fs-pre.target -- so that it is ordered before the mount point, normally.
2013-03-25Revert "remote-fs.target: want remote-fs-pre.target"Lennart Poettering
This reverts commit 6bde0b3220e95a08cefb70846f73b2cf24b7734a. We should not pull in remote-fs-pre.target unconditionally. It's supposed to be pulled in by the implementors of it, rather then its users.
2013-03-25units: and also, order all early-boot sockets before sockets.targetLennart Poettering
2013-03-25units: order all udev services before sysinit.target, tooLennart Poettering
Not that it would matter much, but let's make things a bit more systematic: early boot services shall order themselves before sysinit.target, and nothing else.
2013-03-25update TODOLennart Poettering
2013-03-25units: introduce new timers.target and paths.target to hook timer/path units ↵Lennart Poettering
into for boot
2013-03-25fstab-generator: add missing strempty() callsMantas Mikulėnas
2013-03-25build-sys: add missing sed substitution for DEBUGTTYKay Sievers
2013-03-25build-sys: configurable debug shell tty pathUmut Tezduyar
2013-03-25fstab-generator: fix minor memory leak on error pathLennart Poettering
2013-03-25fstab-generator: add missing OOM checkLennart Poettering
2013-03-25fstab-generator: rename x-initrd-rootfs.mount to x-initrd.rootfsLennart Poettering
This changes the fstab mount option x-initrd-rootfs.mount to x-initrd.rootfs, in order to only use a single namespace "x-initrd." for all mount options of the initrd.
2013-03-25update TODOLennart Poettering
2013-03-25journal: Add sync timer to journal serverOleksii Shevchuk
Add option to force journal sync with fsync. Default timeout is 5min. Interval configured via SyncIntervalSec option at journal.conf. Synced journal files will be marked as OFFLINE. Manual sync can be performed via sending SIGUSR1.
2013-03-25update TODOLennart Poettering
2013-03-25core: ensure LSB Provides are handled correctlyFrederic Crozat
Let's say you have two initscripts, A and B: A contains in its LSB header: Required-Start: C and B contains in its LSB header: Provides: C When systemd is parsing /etc/rc.d/, depending on the file order, you can end up with either: - B is parsed first. An unit "C.service" will be "created" and will be added as additional name to B.service, with unit_add_name. No bug. - A is parsed first. An unit "C.service" is created for the "Required-Start" dependency (it will have no file attached, since nothing provides this dependency yet). Then B is parsed and when trying to handle "Provides: C", unit_add_name is called but will fail, because "C.service" already exists in manager->units. Therefore, a merge should occur for that case.
2013-03-25udevd.c: set udev children_max according to CPU countHarald Hoyer
Setting children_max according to RAM leads to too much concurrent I/O.
2013-03-25timer: downgrade time change message to debugMichal Schmidt
The manager already prints "Time has been changed" at level info. It seems too verbose to print the time change message additionally for every waiting timer unit. Downgrade the per-unit message to debug.
2013-03-25bus: split socket related code into bus-socket.[ch], to prepare for kdbus ↵Lennart Poettering
backend
2013-03-25bus: implement 'unixexec:' protocolLennart Poettering
2013-03-25bus: make optional whether unix socket passing is negotiated and whether ↵Lennart Poettering
hello is sent This alos gets rid of explicit sd_open_fd() and sd_open_address() constructors in favour of sd_new() + sd_new_start() where the negotiation parameters may be set it in between.