platform/upstream/systemd.git
7 years agoMerge pull request #4131 from intelfx/update-done-timestamps-precision
Zbigniew Jędrzejewski-Szmek [Fri, 16 Sep 2016 02:53:00 +0000 (22:53 -0400)]
Merge pull request #4131 from intelfx/update-done-timestamps-precision

condition: ignore nanoseconds in timestamps for ConditionNeedsUpdate=

Fixes #4130.

7 years agologind: fix /run/user/$UID creation in apparmor-confined containers (#4154)
Tomáš Janoušek [Thu, 15 Sep 2016 23:26:31 +0000 (01:26 +0200)]
logind: fix /run/user/$UID creation in apparmor-confined containers (#4154)

When a docker container is confined with AppArmor [1] and happens to run
on top of a kernel that supports mount mediation [2], e.g. any Ubuntu
kernel, mount(2) returns EACCES instead of EPERM.  This then leads to:

    systemd-logind[33]: Failed to mount per-user tmpfs directory /run/user/1000: Permission denied
    login[42]: pam_systemd(login:session): Failed to create session: Access denied

and user sessions don't start.

This also applies to selinux that too returns EACCES on mount denial.

[1] https://github.com/docker/docker/blob/master/docs/security/apparmor.md#understand-the-policies
[2] http://bazaar.launchpad.net/~apparmor-dev/apparmor/master/view/head:/kernel-patches/4.7/0025-UBUNTU-SAUCE-apparmor-Add-the-ability-to-mediate-mou.patch

7 years agohwdb: Update database of Bluetooth company identifiers
Marcel Holtmann [Thu, 15 Sep 2016 20:09:35 +0000 (22:09 +0200)]
hwdb: Update database of Bluetooth company identifiers

7 years agotest-execute: fix %n typo (#4153)
Zbigniew Jędrzejewski-Szmek [Thu, 15 Sep 2016 12:21:42 +0000 (08:21 -0400)]
test-execute: fix %n typo (#4153)

7 years agoMerge pull request #4150 from ssahani/net1
Martin Pitt [Thu, 15 Sep 2016 10:20:09 +0000 (12:20 +0200)]
Merge pull request #4150 from ssahani/net1

networkd: trivial fixes

7 years agoUpdate systemctl.xml (#4151)
kristbaum [Thu, 15 Sep 2016 07:11:33 +0000 (09:11 +0200)]
Update systemctl.xml (#4151)

7 years agohwdb: add Lenovo *40 series resolution fixes (#4149)
Peter Hutterer [Thu, 15 Sep 2016 06:16:05 +0000 (16:16 +1000)]
hwdb: add Lenovo *40 series resolution fixes (#4149)

7 years agonetworkd: network fix log message
Susant Sahani [Thu, 15 Sep 2016 04:49:26 +0000 (10:19 +0530)]
networkd: network fix log message

7 years agonetworkd: netdev fixup copy paste error
Susant Sahani [Thu, 15 Sep 2016 04:48:59 +0000 (10:18 +0530)]
networkd: netdev fixup copy paste error

7 years agoTODO: update networkd TODO
Susant Sahani [Thu, 15 Sep 2016 04:48:22 +0000 (10:18 +0530)]
TODO: update networkd TODO

7 years agoupdate-done, condition: write the timestamp to the file as well and use it to prevent...
Ivan Shapovalov [Tue, 13 Sep 2016 00:04:35 +0000 (03:04 +0300)]
update-done, condition: write the timestamp to the file as well and use it to prevent false-positives

This fixes https://bugs.freedesktop.org/show_bug.cgi?id=90192 and #4130
for real. Also, remove timestamp check in update-done.c altogether since
the whole operation is idempotent.

7 years agotime-util: export timespec_load_nsec()
Ivan Shapovalov [Thu, 15 Sep 2016 02:19:31 +0000 (05:19 +0300)]
time-util: export timespec_load_nsec()

7 years agoshell-completion: add --wait to systemd-run completions (#4140)
Davide Cavalca [Wed, 14 Sep 2016 20:38:53 +0000 (13:38 -0700)]
shell-completion: add --wait to systemd-run completions (#4140)

7 years agogitignore: ignore image.raw from mkosi (#4141)
Davide Cavalca [Wed, 14 Sep 2016 17:15:21 +0000 (10:15 -0700)]
gitignore: ignore image.raw from mkosi (#4141)

7 years agonetworkd: add support to configure virtual CAN device (#4139)
Susant Sahani [Wed, 14 Sep 2016 16:15:16 +0000 (21:45 +0530)]
networkd: add support to configure virtual CAN device (#4139)

1. add support for kind vcan
2. fixup indention netlink-types.c, networkd-netdev.c

7 years agoMerge pull request #4133 from keszybz/strerror-removal
Martin Pitt [Wed, 14 Sep 2016 09:17:58 +0000 (11:17 +0200)]
Merge pull request #4133 from keszybz/strerror-removal

Strerror removal and other janitorial cleanups

7 years agokernel-install: Add KERNEL_INSTALL_NOOP (#4103)
Colin Walters [Wed, 14 Sep 2016 05:57:43 +0000 (01:57 -0400)]
kernel-install: Add KERNEL_INSTALL_NOOP (#4103)

Will be used by rpm-ostree (and likely lorax) to suppress
RPM->kernel->%posttrans->dracut runs, and basically everything
else this script is doing.

I'll also likely change the `kernel.spec` to respect this as well.

7 years agoNEWS: add a bunch of stuff for the 232 release (#4132)
Zbigniew Jędrzejewski-Szmek [Wed, 14 Sep 2016 05:40:02 +0000 (01:40 -0400)]
NEWS: add a bunch of stuff for the 232 release (#4132)

This does not include the description of the mixed v1/v2 mode, but
everything important apart from that should be covered.

7 years agoTODO: remove duplicated item
Zbigniew Jędrzejewski-Szmek [Mon, 12 Sep 2016 19:20:33 +0000 (20:20 +0100)]
TODO: remove duplicated item

7 years agoAlways use unicode ellipsis when ellipsizing
Zbigniew Jędrzejewski-Szmek [Mon, 12 Sep 2016 19:09:36 +0000 (20:09 +0100)]
Always use unicode ellipsis when ellipsizing

We were already unconditionally using the unicode character when the
input string was not pure ASCII, leading to different behaviour in
depending on the input string.

systemd[1]: Starting printit.service.
python3[19962]: foooooooooooooooooooooooooooooooooooo…oooo
python3[19964]: fooąęoooooooooooooooooooooooooooooooo…oooo
python3[19966]: fooąęoooooooooooooooooooooooooooooooo…ąęąę
python3[19968]: fooąęoooooooooooooooooąęąęąęąęąęąęąęą…ąęąę
systemd[1]: Started printit.service.

7 years agoTODO: remove strerror entry
Zbigniew Jędrzejewski-Szmek [Mon, 12 Sep 2016 18:56:50 +0000 (19:56 +0100)]
TODO: remove strerror entry

I believe the remaining call sites are legitimate uses which cannot be
easily replaced with %m.

7 years agotests: get rid of strerror
Zbigniew Jędrzejewski-Szmek [Mon, 12 Sep 2016 18:55:34 +0000 (19:55 +0100)]
tests: get rid of strerror

7 years agotree-wide: use %m in calls to sd_bus_error_set_errnof
Zbigniew Jędrzejewski-Szmek [Mon, 12 Sep 2016 18:54:35 +0000 (19:54 +0100)]
tree-wide: use %m in calls to sd_bus_error_set_errnof

sd_bus_error_set_errnof supports %m, so there's no need to call
strerror manually.

7 years agojournal-verify: get rid of strerror
Zbigniew Jędrzejewski-Szmek [Mon, 12 Sep 2016 18:41:33 +0000 (19:41 +0100)]
journal-verify: get rid of strerror

7 years agomicrohttpd-util: add the trailing newline automatically
Zbigniew Jędrzejewski-Szmek [Mon, 12 Sep 2016 18:33:37 +0000 (19:33 +0100)]
microhttpd-util: add the trailing newline automatically

It's prone to error and annoying to have to add it manually. It was
missing from a few places.

7 years agojournal-remote: implement %m support in mhd_respondf
Zbigniew Jędrzejewski-Szmek [Mon, 12 Sep 2016 18:24:45 +0000 (19:24 +0100)]
journal-remote: implement %m support in mhd_respondf

errno value is not protected (it is undefined after this function returns).
Various mhd_* functions are not documented to protect errno, so this could not
guaranteed anyway.

7 years agoudev: use get_proc_cmdline_key instead of FOREACH_WORD_QUOTED
Zbigniew Jędrzejewski-Szmek [Mon, 12 Sep 2016 18:03:35 +0000 (19:03 +0100)]
udev: use get_proc_cmdline_key instead of FOREACH_WORD_QUOTED

7 years agoman: "disabled on" does not sound right
Zbigniew Jędrzejewski-Szmek [Mon, 12 Sep 2016 17:46:19 +0000 (18:46 +0100)]
man: "disabled on" does not sound right

7 years agofileio: simplify mkostemp_safe() (#4090)
Topi Miettinen [Tue, 13 Sep 2016 06:20:38 +0000 (06:20 +0000)]
fileio: simplify mkostemp_safe() (#4090)

According to its manual page, flags given to mkostemp(3) shouldn't include
O_RDWR, O_CREAT or O_EXCL flags as these are always included. Beyond
those, the only flag that all callers (except a few tests where it
probably doesn't matter) use is O_CLOEXEC, so set that unconditionally.

7 years agoRelicense hwdb/parse_hwdb.py as MIT (#4129)
Zbigniew Jędrzejewski-Szmek [Tue, 13 Sep 2016 06:18:13 +0000 (07:18 +0100)]
Relicense hwdb/parse_hwdb.py as MIT (#4129)

This parser will also be used in libinput, which uses the MIT license, so
relicense this file to the more permissive license to make bidirectional code
flow easier. parse_hwdb.py is only useful during building of the project, and
is not part of the installation, so effectively both licenses are very similar.
In particular, the licensing of binary packages produced by systemd is not
influenced in any way, because the MIT licensed part is not installed.

7 years agohwdb: add Lenovo X1 Tablet pointing stick speed fix (#4128)
dwassenberg [Mon, 12 Sep 2016 12:29:55 +0000 (14:29 +0200)]
hwdb: add Lenovo X1 Tablet pointing stick speed fix (#4128)

Like many other recent thinkpads the factory default pointingstick
sensitivity on these devices is quite low, making the pointingstick
very slow in moving the cursor.

This extends the existing hwdb rules for tweaking the sensitivity to
also apply to the X1 Tablet models.

Signed-off-by: Dennis Wassenberg <dennis.wassenberg@secunet.com>
7 years agoshared/install: fix set-default with empty root (#4118)
Zbigniew Jędrzejewski-Szmek [Sat, 10 Sep 2016 11:07:51 +0000 (12:07 +0100)]
shared/install: fix set-default with empty root (#4118)

https://bugzilla.redhat.com/show_bug.cgi?id=1374371

When root was empty or equal to "/", chroot_symlinks_same was called with
root==NULL, and strjoina returned "", so the code thought both paths are equal
even if they were not. Fix that by always providing a non-null first argument
to strjoina.

7 years agotest-fs-util: also empty TEMP and TMP env vars (#4121)
Marc-Antoine Perennou [Sat, 10 Sep 2016 07:38:04 +0000 (09:38 +0200)]
test-fs-util: also empty TEMP and TMP env vars (#4121)

A follow-up for #3818 (992e8f2).

7 years agoMerge pull request #4119 from keszybz/drop-more-kdbus
Evgeny Vereshchagin [Sat, 10 Sep 2016 06:26:43 +0000 (09:26 +0300)]
Merge pull request #4119 from keszybz/drop-more-kdbus

Drop more kdbus functionality

7 years agoservice: fixup ExecStop for socket-activated shutdown (#4120)
Kyle Russell [Sat, 10 Sep 2016 05:55:36 +0000 (01:55 -0400)]
service: fixup ExecStop for socket-activated shutdown (#4120)

Previous fix didn't consider handling multiple ExecStop commands.

7 years agoshared: recognize DNS names with more than one trailing dot as invalid (#4111)
Martin Pitt [Fri, 9 Sep 2016 15:11:54 +0000 (17:11 +0200)]
shared: recognize DNS names with more than one trailing dot as invalid (#4111)

One trailing dot is valid, but more than one isn't. This also fixes glibc's
posix/tst-getaddrinfo5 test.

Fixes #3978.

7 years agounit: sent change signal before removing the unit if necessary (#4106)
Michael Olbrich [Fri, 9 Sep 2016 15:05:06 +0000 (17:05 +0200)]
unit: sent change signal before removing the unit if necessary (#4106)

If the unit is in the dbus queue when it is removed then the last change
signal is never sent. Fix this by checking the dbus queue and explicitly
send the change signal before sending the remove signal.

7 years agopid1: drop kdbus_fd and all associated logic
Zbigniew Jędrzejewski-Szmek [Fri, 9 Sep 2016 14:16:26 +0000 (15:16 +0100)]
pid1: drop kdbus_fd and all associated logic

7 years agoman: drop kdbus descriptions from sd_b_negotiate_fds(3)
Zbigniew Jędrzejewski-Szmek [Fri, 9 Sep 2016 14:06:06 +0000 (15:06 +0100)]
man: drop kdbus descriptions from sd_b_negotiate_fds(3)

7 years agobuild-sys: drop last reference to --have-kdbus
Zbigniew Jędrzejewski-Szmek [Fri, 9 Sep 2016 14:03:11 +0000 (15:03 +0100)]
build-sys: drop last reference to --have-kdbus

8 years agoservice: Continue shutdown on socket activated unit on termination (#4108)
Kyle Russell [Fri, 9 Sep 2016 02:34:43 +0000 (22:34 -0400)]
service: Continue shutdown on socket activated unit on termination (#4108)

ENOTCONN may be a legitimate return code if the endpoint disappeared,
but the service should still attempt to shutdown cleanly.

8 years agoman: add missing <constant> tag (#4109)
Lucas Werkmeister [Thu, 8 Sep 2016 08:18:16 +0000 (10:18 +0200)]
man: add missing <constant> tag (#4109)

8 years agoMerge pull request #4087 from fsateler/detect-seccomp-filter
Evgeny Vereshchagin [Wed, 7 Sep 2016 03:44:06 +0000 (06:44 +0300)]
Merge pull request #4087 from fsateler/detect-seccomp-filter

seccomp: also detect if seccomp filtering is available

8 years agoREADME: document that CONFIG_SECCOMP_FILTER is required for SECCOMP support
Felipe Sateler [Mon, 5 Sep 2016 22:16:13 +0000 (19:16 -0300)]
README: document that CONFIG_SECCOMP_FILTER is required for SECCOMP support

8 years agonspawn: detect SECCOMP availability, skip audit filter if unavailable
Felipe Sateler [Tue, 6 Sep 2016 23:25:22 +0000 (20:25 -0300)]
nspawn: detect SECCOMP availability, skip audit filter if unavailable

Fail hard if SECCOMP was detected but could not be installed

8 years agoseccomp: also detect if seccomp filtering is enabled
Felipe Sateler [Wed, 31 Aug 2016 13:00:35 +0000 (10:00 -0300)]
seccomp: also detect if seccomp filtering is enabled

In https://github.com/systemd/systemd/pull/4004 , a runtime detection
method for seccomp was added. However, it does not detect the case
where CONFIG_SECCOMP=y but CONFIG_SECCOMP_FILTER=n. This is possible
if the architecture does not support filtering yet.
Add a check for that case too.

While at it, change get_proc_field usage to use PR_GET_SECCOMP prctl,
as that should save a few system calls and (unnecessary) allocations.
Previously, reading of /proc/self/stat was done as recommended by
prctl(2) as safer. However, given that we need to do the prctl call
anyway, lets skip opening, reading and parsing the file.

Code for checking inspired by
https://outflux.net/teach-seccomp/autodetect.html

8 years agoMerge pull request #4084 from ssahani/netfix
Daniel Mack [Fri, 2 Sep 2016 20:08:18 +0000 (22:08 +0200)]
Merge pull request #4084 from ssahani/netfix

fix #4080

8 years agofix #4080
Susant Sahani [Fri, 2 Sep 2016 10:36:58 +0000 (16:06 +0530)]
fix #4080

ethtool_sset_info  adding some extra space to it.

also fix valgrind warning
```
Unloaded link configuration context.
==31690==
==31690== HEAP SUMMARY:
==31690==     in use at exit: 8,192 bytes in 2 blocks
==31690==   total heap usage: 431 allocs, 429 frees, 321,164 bytes allocated
==31690==
==31690== 4,096 bytes in 1 blocks are still reachable in loss record 1 of 2
==31690==    at 0x4C2BBAD: malloc (vg_replace_malloc.c:299)
==31690==    by 0x166B32: mempool_alloc_tile (mempool.c:62)
==31690==    by 0x166BBC: mempool_alloc0_tile (mempool.c:81)
==31690==    by 0x15B8FC: hashmap_base_new (hashmap.c:732)
==31690==    by 0x15B9F7: internal_hashmap_new (hashmap.c:766)
==31690==    by 0x151291: conf_files_list_strv_internal (conf-files.c:103)
==31690==    by 0x1514BA: conf_files_list_strv (conf-files.c:135)
==31690==    by 0x13A1CF: link_config_load (link-config.c:227)
==31690==    by 0x135B68: builtin_net_setup_link_init
(udev-builtin-net_setup_link.c:77)
==31690==    by 0x1306B3: udev_builtin_init (udev-builtin.c:57)
==31690==    by 0x11E984: adm_builtin (udevadm-test-builtin.c:72)
==31690==    by 0x117B4D: run_command (udevadm.c:75)
```
Fixes #4080

8 years agoshell-completion: add systemd-resolve --status (#4085)
Marcos Mello [Fri, 2 Sep 2016 10:18:11 +0000 (07:18 -0300)]
shell-completion: add systemd-resolve --status (#4085)

From be371fe.

8 years agoMerge pull request #4075 from Noeljunior/master
Peter Hutterer [Thu, 1 Sep 2016 09:07:49 +0000 (19:07 +1000)]
Merge pull request #4075 from Noeljunior/master

hwdb: add axis corrections for the Lenovo Yoga 500-14ISK touchpad

8 years agosystemctl: usable status command for special units (#4072)
Cireo [Thu, 1 Sep 2016 09:04:36 +0000 (02:04 -0700)]
systemctl: usable status command for special units (#4072)

Prior to this commit, users could be given an unusable command to run if
they attempted to stop or start special services.  For example:

    $ systemctl stop -- -.mount
    Failed to stop -.mount: Operation refused, unit -.mount may be \
        requested by dependency only.
    See system logs and 'systemctl status -.mount' for details.

    $ systemctl status -.mount
    systemctl: invalid option -- '.'

This adds a '--' to the example command in these situations.

8 years agoMerge pull request #4071 from poettering/some-fixes-3
Martin Pitt [Thu, 1 Sep 2016 06:30:12 +0000 (08:30 +0200)]
Merge pull request #4071 from poettering/some-fixes-3

three trivial fixes

8 years agohwdb: add axis range corrections for Lenovo L430 (#4074)
Peter Hutterer [Thu, 1 Sep 2016 06:29:13 +0000 (16:29 +1000)]
hwdb: add axis range corrections for Lenovo L430 (#4074)

https://bugs.freedesktop.org/show_bug.cgi?id=96220

8 years agohwdb: add axis corrections for the Lenovo Yoga 500-14ISK touchpad
Noeljunior [Thu, 1 Sep 2016 02:02:10 +0000 (03:02 +0100)]
hwdb: add axis corrections for the Lenovo Yoga 500-14ISK touchpad

8 years agohostnamed: prefer more precise DMI info over ACPI OSPM info
Lennart Poettering [Wed, 31 Aug 2016 18:23:19 +0000 (20:23 +0200)]
hostnamed: prefer more precise DMI info over ACPI OSPM info

(also: add comments about the used numbers)

Fixes: #3930

8 years agoresolved: include d.f.ip6.arpa in default NTA list
Lennart Poettering [Wed, 31 Aug 2016 18:12:11 +0000 (20:12 +0200)]
resolved: include d.f.ip6.arpa in default NTA list

Fixes: #4049

8 years agosd-bus: bump message queue size
Lennart Poettering [Wed, 31 Aug 2016 18:09:31 +0000 (20:09 +0200)]
sd-bus: bump message queue size

Let's bump it further, as this the current limit turns out to be problematic
IRL. Let's bump it to more than twice what we know of is needed.

Fixes: #4068

8 years agomachinectl: split OS field in two; print ip addresses (#4058)
Seraphime Kirkovski [Wed, 31 Aug 2016 18:06:57 +0000 (21:06 +0300)]
machinectl: split OS field in two; print ip addresses (#4058)

This splits the OS field in two : one for the distribution name
and one for the the version id.
Dashes are written for missing fields.
This also prints ip addresses of known machines. The `--max-addresses`
option specifies how much ip addresses we want to see. The default is 1.
When more than one address is written for a machine, a `,` follows it.
If there are more ips than `--max-addresses`, `...` follows the last
address.

8 years agonetworkd: add options to bridge (#4051)
Tobias Jungel [Wed, 31 Aug 2016 18:06:23 +0000 (20:06 +0200)]
networkd: add options to bridge (#4051)

This patch allows to configure AgeingTimeSec, Priority and DefaultPVID for
bridge interfaces.

8 years agoinstall: fix disable when /etc/systemd/system is a symlink (#4063)
Lennart Poettering [Wed, 31 Aug 2016 11:02:02 +0000 (13:02 +0200)]
install: fix disable when /etc/systemd/system is a symlink (#4063)

8 years agonetworkd: do not drop config for unmanaged interfaces (#4033)
Lennart Poettering [Wed, 31 Aug 2016 10:36:55 +0000 (12:36 +0200)]
networkd: do not drop config for unmanaged interfaces (#4033)

Flushing foreign configuration for unmanaged interfaces is outright
evil, especially when it's a regular occurence with Wi-Fi.

Fixes: 3104883ddc24 "networkd: remove route if carrier is lost"
Ref: #3831

8 years agohwdb: 60-keyboard: remove line causing a syntax error (#3999)
Lennart Poettering [Wed, 31 Aug 2016 10:29:51 +0000 (12:29 +0200)]
hwdb: 60-keyboard: remove line causing a syntax error (#3999)

8 years agocore: introduce MemorySwapMax= (#3659)
Lennart Poettering [Wed, 31 Aug 2016 10:28:54 +0000 (12:28 +0200)]
core: introduce MemorySwapMax= (#3659)

Similar to MemoryMax=, MemorySwapMax= limits swap usage. This controls
controls "memory.swap.max" attribute in unified cgroup.

8 years agolink : add support to configure LRO and GRO Offload features (#4064)
Lennart Poettering [Wed, 31 Aug 2016 10:13:08 +0000 (12:13 +0200)]
link : add support to configure LRO and GRO Offload features (#4064)

The patch supports to configure

GenericReceiveOffload
LargeReceiveOffload

8 years agoload-fragment: Resolve specifiers in OnCalendar and On*Sec (#4045)
Lennart Poettering [Wed, 31 Aug 2016 10:07:39 +0000 (12:07 +0200)]
load-fragment: Resolve specifiers in OnCalendar and On*Sec (#4045)

Resolves #3534

8 years agolink : add support to configure LRO and GRO Offload features
Susant Sahani [Wed, 31 Aug 2016 07:17:02 +0000 (12:47 +0530)]
link : add support to configure LRO and GRO Offload features

The patch supports to configure

GenericReceiveOffload
LargeReceiveOffload

8 years agolink : add support to configure Offload features (#4017)
Susant Sahani [Tue, 30 Aug 2016 14:52:04 +0000 (20:22 +0530)]
link : add support to configure Offload features (#4017)

This patch supports these features to be on or off

Generic Segmentation Offload
TCP Segmentation Offload
UDP Segmentation Offload

fixes #432

8 years agobootctl: properly skip duplicate entries (#4048)
Mantas Mikulėnas [Tue, 30 Aug 2016 14:50:19 +0000 (17:50 +0300)]
bootctl: properly skip duplicate entries (#4048)

'continue' is a fancy no-op here – it only skips through the inner loop,
not the outer one, so entries already in BootOrder get printed twice.

This partially reverts f939cff71577 "bootctl: various coding style
updates".

8 years agoMerge pull request #4053 from brulon/force-unmount
Lennart Poettering [Tue, 30 Aug 2016 14:44:51 +0000 (16:44 +0200)]
Merge pull request #4053 from brulon/force-unmount

add ForceUnmount= setting for mount units

8 years agoinstall: fix disable when /etc/systemd/system is a symlink
Lukas Nykryn [Tue, 30 Aug 2016 13:04:07 +0000 (15:04 +0200)]
install: fix disable when /etc/systemd/system is a symlink

8 years agoMerge pull request #4052 from yann-morin-1998/yem/o-tmpfile
Lennart Poettering [Tue, 30 Aug 2016 07:47:25 +0000 (09:47 +0200)]
Merge pull request #4052 from yann-morin-1998/yem/o-tmpfile

importd: fix build failure with missing O_TMPFILE (branch yem/o-tmpfile)

8 years agoMerge pull request #4054 from yuwata/man-fix
Martin Pitt [Tue, 30 Aug 2016 04:27:04 +0000 (06:27 +0200)]
Merge pull request #4054 from yuwata/man-fix

man page fixes

8 years agocore: introduce MemorySwapMax=
WaLyong Cho [Mon, 4 Jul 2016 07:03:54 +0000 (07:03 +0000)]
core: introduce MemorySwapMax=

Similar to MemoryMax=, MemorySwapMax= limits swap usage. This controls
controls "memory.swap.max" attribute in unified cgroup.

8 years agobasic/fileio: we always have O_TMPFILE now
Yann E. MORIN [Mon, 29 Aug 2016 10:34:50 +0000 (12:34 +0200)]
basic/fileio: we always have O_TMPFILE now

fileio makes use of O_TMPFILE when it is available.

We now always have O_TMPFILE, defined in missing.h if missing
from the toolchain headers.

Have fileio include missing.h and drop the guards around the
use of O_TMPFILE.

8 years agomissing.h: add missing definitions for __O_TMPFILE
Yann E. MORIN [Sun, 28 Aug 2016 15:26:42 +0000 (17:26 +0200)]
missing.h: add missing definitions for __O_TMPFILE

Currently, a missing __O_TMPFILE was only defined for i386 and x86_64,
leaving any other architectures with an "old" toolchain fail miserably
at build time:
    src/import/export-raw.c: In function 'reflink_snapshot':
    src/import/export-raw.c:271:26: error: 'O_TMPFILE' undeclared (first use in this function)
             new_fd = open(d, O_TMPFILE|O_CLOEXEC|O_NOCTTY|O_RDWR, 0600);
                              ^

__O_TMPFILE (and O_TMPFILE) are available since glibc 2.19. However, a
lot of existing toolchains are still using glibc-2.18, and some even
before that, and it is not really possible to update those toolchains.

Instead of defining it only for i386 and x86_64, define __O_TMPFILE
with the specific values for those archs where it is different from the
generic value. Use the values as found in the Linux kernel (v4.8-rc3,
current as of time of commit).

---
Note: tested on ARM (build+run), with glibc-2.18 and linux headers 3.12.
Untested on other archs, though (I have no board to test this).

Changes v1 -> v2:
  - add a comment specifying some are hexa, others are octal.

8 years agoimportd/export-raw: needs missing.h for O_TMPFILE
Yann E. MORIN [Sun, 28 Aug 2016 14:26:04 +0000 (16:26 +0200)]
importd/export-raw: needs missing.h for O_TMPFILE

O_TMPFILE may be missing from the system headers, so use our fallback
definition.

---
Changes v1 -> v2:
  - move include with local includes

8 years agoman: systemd-journal-remote: do not use ulink tags for example addresses
Yu Watanabe [Mon, 29 Aug 2016 06:59:18 +0000 (15:59 +0900)]
man: systemd-journal-remote: do not use ulink tags for example addresses

Applying ulink tags to example addresses adds meaningless references in NOTES section of the man page.

8 years agoman: systemd.mount: DefaultTimeoutStart -> DefaultTimeoutStartSec
Yu Watanabe [Mon, 29 Aug 2016 02:02:36 +0000 (11:02 +0900)]
man: systemd.mount: DefaultTimeoutStart -> DefaultTimeoutStartSec

8 years agoMerge pull request #4046 from pilophae/ideapad500isk
Peter Hutterer [Mon, 29 Aug 2016 01:37:01 +0000 (11:37 +1000)]
Merge pull request #4046 from pilophae/ideapad500isk

hwdb: add axis range corrections for Lenovo Ideapad 500S-13ISK

8 years agohwdb: 60-keyboard: remove line causing a syntax error
Mantas Mikulėnas [Fri, 19 Aug 2016 18:33:28 +0000 (21:33 +0300)]
hwdb: 60-keyboard: remove line causing a syntax error

8 years agomount: add SloppyOptions= to mount_dump()
Barron Rulon [Sat, 27 Aug 2016 14:42:05 +0000 (10:42 -0400)]
mount: add SloppyOptions= to mount_dump()

8 years agomount: add new ForceUnmount= setting for mount units, mapping to umount(8)'s "-f...
Barron Rulon [Sat, 27 Aug 2016 14:27:49 +0000 (10:27 -0400)]
mount: add new ForceUnmount= setting for mount units, mapping to umount(8)'s "-f" switch

8 years agohwdb: add axis range corrections for Lenovo Ideapad 500S-13ISK
Erik Karlsson [Wed, 24 Aug 2016 18:18:22 +0000 (20:18 +0200)]
hwdb: add axis range corrections for Lenovo Ideapad 500S-13ISK

8 years agoload-fragment: Resolve specifiers in OnCalendar and On*Sec
Douglas Christman [Fri, 26 Aug 2016 16:13:16 +0000 (12:13 -0400)]
load-fragment: Resolve specifiers in OnCalendar and On*Sec

Resolves #3534

8 years agomount: add new LazyUnmount= setting for mount units, mapping to umount(8)'s "-l"...
brulon [Fri, 26 Aug 2016 15:57:22 +0000 (11:57 -0400)]
mount: add new LazyUnmount= setting for mount units, mapping to umount(8)'s "-l" switch (#3827)

8 years agojournal: add/use flushed_flag_is_set() helper (#4041)
Vito Caputo [Fri, 26 Aug 2016 15:51:13 +0000 (08:51 -0700)]
journal: add/use flushed_flag_is_set() helper (#4041)

Minor cleanup suggested by Lennart.

8 years agotests: mount proc before `exec valgrind` (#4044)
Evgeny Vereshchagin [Fri, 26 Aug 2016 15:49:56 +0000 (18:49 +0300)]
tests: mount proc before `exec valgrind` (#4044)

Fixes:
$ sudo make run QEMU_BIN=/usr/bin/qemu-system-x86_64  KERNEL_APPEND=systemd.unit=multi-user.target SKIP_INITRD=yes
TEST RUN: Basic systemd setup
+ /usr/bin/qemu-system-x86_64 -smp 1 -net none -m 512M -nographic -kernel /boot/vmlinuz-4.6.7-200.fc23.x86_64 -drive format=raw,cache=unsafe,file=/var/tmp/systemd-test.izx99J/rootdisk.img -append 'root=/dev/sda1 raid=noautodetect loglevel=2 init=/usr/lib/systemd/systemd-under-valgrind rw console=ttyS0 selinux=0 systemd.unified_cgroup_hierarchy=no systemd.unit=multi-user.target '
valgrind: warning (non-fatal): readlink("/proc/self/exe") failed.
valgrind: continuing, however --trace-children=yes will not work.
--1:0: aspacem Valgrind: FATAL: can't open /proc/self/maps
--1:0: aspacem Exiting now.
[    3.152367] Kernel panic - not syncing: Attempted to kill init! exitcode=0x00000100
[    3.152367]
[    3.152367] CPU: 0 PID: 1 Comm: memcheck-amd64- Not tainted 4.6.7-200.fc23.x86_64 #1
[    3.152367] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.8.2-20150714_191134- 04/01/2014
[    3.152367]  0000000000000086 0000000047ca1666 ffff88001ea43e00 ffffffff813d954e
[    3.152367]  ffffffff81a205a0 ffff88001ea43e98 ffff88001ea43e88 ffffffff811b5557
[    3.152367]  ffffffff00000010 ffff88001ea43e98 ffff88001ea43e30 0000000047ca1666
[    3.152367] Call Trace:
[    3.152367]  [<ffffffff813d954e>] dump_stack+0x63/0x85
[    3.152367]  [<ffffffff811b5557>] panic+0xde/0x220
[    3.152367]  [<ffffffff810ab9c3>] do_exit+0xb43/0xb50
[    3.152367]  [<ffffffff810aba57>] do_group_exit+0x47/0xb0
[    3.152367]  [<ffffffff810abad4>] SyS_exit_group+0x14/0x20
[    3.152367]  [<ffffffff817dae72>] entry_SYSCALL_64_fastpath+0x1a/0xa4
[    3.152367] Kernel Offset: disabled
[    3.152367] ---[ end Kernel panic - not syncing: Attempted to kill init! exitcode=0x00000100
[    3.152367]
QEMU: Terminated

8 years agoMerge pull request #3984 from poettering/refcnt
Evgeny Vereshchagin [Fri, 26 Aug 2016 13:17:05 +0000 (16:17 +0300)]
Merge pull request #3984 from poettering/refcnt

permit bus clients to pin units to avoid automatic GC

8 years agonspawn: split down SYSTEMD_NSPAWN_SHARE_SYSTEM (#4023)
Luca Bruno [Thu, 25 Aug 2016 22:08:26 +0000 (22:08 +0000)]
nspawn: split down SYSTEMD_NSPAWN_SHARE_SYSTEM (#4023)

This commit follows further on the deprecation path for --share-system,
by splitting and gating each share-able namespace behind its own
environment flag.

8 years agounits: remove udev control socket when systemd stops the socket unit (#4039)
Michal Sekletar [Thu, 25 Aug 2016 22:07:58 +0000 (00:07 +0200)]
units: remove udev control socket when systemd stops the socket unit (#4039)

Mere presence of the socket in the filesystem makes
udev_queue_get_udev_is_active() return that udev is running. Note that,
udev on exit doesn't unlink control socket nor does systemd. Thus socket
stays around even when both daemon and socket are stopped. This causes
problems for cryptsetup because when it detects running udev it launches
synchronous operations that *really* require udev. This in turn may
cause blocking and subsequent timeout in systemd-cryptsetup on reboot
while machine is in a state that udev and its control socket units are
stopped, e.g. emergency mode.

Fixes #2477

8 years agojournal: implicitly flush to var on recovery (#4028)
Vito Caputo [Thu, 25 Aug 2016 15:37:57 +0000 (08:37 -0700)]
journal: implicitly flush to var on recovery (#4028)

When the system journal becomes re-opened post-flush with the runtime
journal open, it implies we've recovered from something like an ENOSPC
situation where the system journal rotate had failed, leaving the system
journal closed, causing the runtime journal to be opened post-flush.

For the duration of the unavailable system journal, we log to the
runtime journal.  But when the system journal gets opened (space made
available, for example), we need to close the runtime journal before new
journal writes will go to the system journal.  Calling
server_flush_to_var() after opening the system journal with a runtime
journal present, post-flush, achieves this while preserving the runtime
journal's contents in the system journal.

The combination of the present flushed flag file and the runtime journal
being open is a state where we should be logging to the system journal,
so it's appropriate to resume doing so once we've successfully opened
the system journal.

8 years agonetworkd: do not drop config for unmanaged interfaces
Mantas Mikulėnas [Wed, 24 Aug 2016 15:26:48 +0000 (18:26 +0300)]
networkd: do not drop config for unmanaged interfaces

Flushing foreign configuration for unmanaged interfaces is outright
evil, especially when it's a regular occurence with Wi-Fi.

Fixes: 3104883ddc24 "networkd: remove route if carrier is lost"
Ref: #3831

8 years agoudev: inform systemd how many workers we can potentially spawn (#4036)
Michal Sekletar [Thu, 25 Aug 2016 06:18:42 +0000 (08:18 +0200)]
udev: inform systemd how many workers we can potentially spawn (#4036)

8 years agodocs: add --with-support-url= to distribution porting guide. (#4035)
Stefan [Wed, 24 Aug 2016 18:49:53 +0000 (20:49 +0200)]
docs: add --with-support-url= to distribution porting guide. (#4035)

8 years agorun: respect quiet when starting timer (#4029)
Nick Owens [Wed, 24 Aug 2016 10:12:18 +0000 (03:12 -0700)]
run: respect quiet when starting timer (#4029)

8 years agohwdb: add axis ranges for the MacBook 4,1 (#4030)
Peter Hutterer [Wed, 24 Aug 2016 10:11:48 +0000 (20:11 +1000)]
hwdb: add axis ranges for the MacBook 4,1 (#4030)

https://bugzilla.redhat.com/show_bug.cgi?id=1357990

8 years agoUpdated Hungarian translation (#4026)
Balázs Úr [Tue, 23 Aug 2016 18:08:35 +0000 (20:08 +0200)]
Updated Hungarian translation (#4026)

8 years agocore,network: Use const qualifiers for block-local variables in macro functions ...
Felipe Sateler [Tue, 23 Aug 2016 09:29:30 +0000 (06:29 -0300)]
core,network: Use const qualifiers for block-local variables in macro functions (#4019)

Prevents discard-qualifiers warnings when the passed variable was const

8 years agocore: do not fail at step SECCOMP if there is no kernel support (#4004)
Felipe Sateler [Mon, 22 Aug 2016 19:40:58 +0000 (16:40 -0300)]
core: do not fail at step SECCOMP if there is no kernel support (#4004)

Fixes #3882

8 years agobus-util: turn on exit-on-disconnect for all command line tools
Lennart Poettering [Mon, 22 Aug 2016 15:21:27 +0000 (17:21 +0200)]
bus-util: turn on exit-on-disconnect for all command line tools

bus_connect_transport() is exclusively used from our command line tools, hence
let's set exit-on-disconnect for all of them, making behaviour a bit nicer in
case dbus-daemon goes down.