There's no need to save the old sigmask, if we are going to die. Let's
simplify this. Also, reset all the signal handlers, so that we don't
leave SIG_IGN set for some of them across reexec.
(cherry picked from commit 9bfcda9528)
Process 1 (aka init) needs to be started with an empty signal mask.
That includes the process 1 that's started after the initrd is finished.
When the initrd is using systemd (as it does with dracut based initrds)
then it is systemd that calls the real init. Normally this is systemd
again, except when the user uses for instance "init=/bin/bash" on the
kernel command line.
(cherry picked from commit 5a85ca1cb6)
Update for the current behavior of path_strv_resolve which now returns
paths relative to the given root, not the full absolute paths.
(cherry picked from commit 375eadd911)
This adds the same root argument to search_and_fopen that
conf_files_list already has. Tools that use those two functions as a
pair can now be easily modified to load configuration files from an
alternate root filesystem tree.
(cherry picked from commit 4cf7ea556a)
This restores the original root handling logic that was present prior to
112cfb18 when path expansion moved to path_strv_canonicalize_absolute.
That behavior partially went away in 12ed81d9.
Alternatively all users of conf_files_list* could be updated to
concatenate the paths themselves as unit_file_query_preset did but since
no user needs the un-concatenated form that is pointless duplication.
(cherry picked from commit cba2ef0272)
Conflicts:
src/shared/install.c
Otherwise the add_symlink() function tries to make directories for
each slash even for the slash after the @ symbol in the final link
name, failing for /dev/3270/tty1.
Based on a patch by Werner Fink <werner@suse.de>.
(cherry picked from commit a2ae516a25)
https://bugs.freedesktop.org/show_bug.cgi?id=77092
On Thu, Jun 05, 2014 at 08:37:20AM +0200, Lennart Poettering wrote:
> The patch is line-broken, please send an uncorrupted patch!
I am very sorry, I forgot that my client limits line width. I will use
mutt now on.
> clamp_brightness() clamps the brightness value to the range of the
> actual device. This is a recent addition that was added to deal with
> driver updates where the resolution is changed. I don't think this part
> should be dropped for LED devices. The clamp_brightness() call hence
> should be called unconditionally, however, internally it should use a
> different min_brightness value if something is an !backlight devices...
Thank you for explanation, this sounds very reasonable to me. Please,
see updated patch:
(cherry picked from commit 4cd2b2cf8c)
If too high a brightness value has been saved (e.g. due to kernel
mechanism changing from one kernel version to another, or booting the
userspace on another system), the brightness update fails and the
process exits.
Clamp saved brightness between the policy minimum introduced in
commit 7b909d7407
Author: Josh Triplett <josh@joshtriplett.org>
Date: Tue Mar 11 21:16:33 2014 -0700
backlight: Avoid restoring brightness to an unreadably dim level
and the absolute maximum.
https://bugs.freedesktop.org/show_bug.cgi?id=78200
(cherry picked from commit 0c9d8f1d4b)
On virtually any newer Asus mainboard, the eeepc-wmi driver is loaded.
It exposes a backlight device despite the lack of any physical backlight
devices. This fake backlight device has max_brightness set to 0. Since
the introduction of the clamp_brightness function, systemd-backlight
tries to write '1' to brightness and fails.
This patch changes systemd-backlight to exit gracefully when
max_brightness is 0 before performing any action. This affects
both the load and save actions.
(cherry picked from commit 3cadce7d33)
Some systems turn the backlight all the way off at the lowest levels.
Clamp saved brightness to at least 1 or 5% of max_brightness. This
avoids preserving an unreadably dim screen, which would otherwise force
the user to disable state restoration.
(cherry picked from commit 7b909d7407)
udev_device_get_sysattr_value returns NULL on failure, but doesn't
provide an error code; thus, when printing an error from it, don't print
an unrelated error code from a previous call.
(cherry picked from commit cddc35f75e)
Let's always make sure to look in all search directories for the full
unit names first, before looking for templates for them.
(cherry picked from commit e50bd77516)
In programs like eog and gimp the transparant background did not
look very good.
Similar fix from the one done in systemd-analyze (418e3750)
(cherry picked from commit a7997073ac)
This is based on parts of similar patches from Michael Marineau and
Lukas Nykrin, but simply uses strappenda3().
(cherry picked from commit cebed5005b)
If "systemctl switch-root" is called with a specific "INIT" or
/proc/cmdline contains "init=", then systemd would not serialize
itsself.
Let systemctl check, if the new init is in the standard systemd
installation path and if so, clear the INIT parameter,
to let systemd serialize itsself.
(cherry picked from commit f39d4a08e7)
For strappenda3, and it seems a good fix in general.
files_same() returns
1, if the files are the same
0, if the files have different inode/dev numbers
errno, for any stat error
(cherry picked from commit 9d9951a460)
this gives us a little bit more freedom to move things around later on,
as we don't hardcode the systemd paths in old RPMs that shall work with
new systemds.
(cherry picked from commit 8cf7c96517)
int unit_file_mask(...) in ./src/shared/install.c calls
get_config_path(...) which can in 4 error cases return without setting
"ret", and thus "prefix" can be uninitialized when unit_file_mask(...)
finishes (which it does directly after the error is returned from
get_config_path(...)).
(cherry picked from commit 223217749e)
static bool enable_name_policy(...) in ./src/udev/net/link-config.c
calls proc_cmdline(...) to get "line" initialized, but
proc_cmdline(...) does not guarantee that atleast when both
conditions (detect_container(NULL) > 0) and
read_full_file(...) returned < 0.
(cherry picked from commit f8a0bb5285)
static int killall(....) in ./src/core/killall.c tries to get "s"
initialized by calling get_process_comm(...) which calls
read_one_line_file(...) which if it fails will mean it is left
uninitialized.
It is then used in argument to strna(s) call where it is
dereferenced(!), in addition to nothing else initializing it before
the scope it is in finishes.
(cherry picked from commit 3e09eb5c83)
There might be implementations around where the network-online logic
might not talk to any network configuration service (and thus not have
to wait for it), hence let's explicitly order network-online.target
after network.target to avoid any ambiguities.
(cherry picked from commit 58e027023b)
Conflicts:
units/systemd-networkd-wait-online.service.in
The current vm detection lacks the distinction between Xen dom0 and Xen domU.
Both, dom0 and domU are running inside the hypervisor.
Therefore systemd-detect-virt and the ConditionVirtualization directive detect
dom0 as a virtual machine.
dom0 is not using virtual devices but is accessing the real hardware.
Therefore dom0 should be considered the virtualisation host and not a virtual
machine.
https://bugs.freedesktop.org/show_bug.cgi?id=77271
(cherry picked from commit 37287585b6)
For names like /var/lib/container/something, the message
becomes quite long. Better to split it.
Also reword the message not to suggest that ^]^]^] only works
in the beginning.
(cherry picked from commit 45f1386c9a)
Conflicts:
src/nspawn/nspawn.c
I am getting
"Error calling EVIOCSKEYCODE (scan code 0xc022d, key code 418): Invalid
argument", the error message does not tell on which specific device the
problem is, add that info.
(cherry picked from commit a52ec8ed88)
Both systemd-analyze and systemd-run only access org.freedesktop.systemd1
on the bus. This patch allows using systemd-run --user and systemd-analyze
--user even if the user session's bus is not properly integrated with the
systemd user unit.
https://bugs.freedesktop.org/show_bug.cgi?id=79252 and other reports...
(cherry picked from commit 1f89214e6e)
Zbigniew Jędrzejewski-Szmek