morgan/systemd
1
0
Fork 0
mirror of https://github.com/morgan9e/systemd synced 2026-04-14 00:14:32 +09:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #19081 from keszybz/three-comment-updates

Browse Source 
Three comment updates
This commit is contained in:
Luca Boccassi
2021-03-22 22:36:01 +00:00
committed by GitHub
parent 965984240e ccd593a61c
commit d4bb2b0b4e
3 changed files with 19 additions and 24 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
NEWS
View File

@@ -251,7 +251,7 @@ CHANGES WITH 248:
be restored for individual services with NoExecPaths=/dev (or by allow-
listing and excluding /dev from ExecPaths=).
* Permissions for /dev/vsock are now set to 0666, and /dev/vhost-vsock
* Permissions for /dev/vsock are now set to 0o666, and /dev/vhost-vsock
and /dev/vhost-net are owned by the kvm group.
* The hardware database has been extended with a list of fingerprint

25
man/systemd.device.xml
View File

@@ -26,11 +26,10 @@
<refsect1>
<title>Description</title>
<para>A unit configuration file whose name ends in
<literal>.device</literal> encodes information about a device unit
as exposed in the
sysfs/<citerefentry><refentrytitle>udev</refentrytitle><manvolnum>7</manvolnum></citerefentry>
device tree.</para>
<para>A unit configuration file whose name ends in <literal>.device</literal> encodes information about a
device unit as exposed in the
sysfs/<citerefentry><refentrytitle>udev</refentrytitle><manvolnum>7</manvolnum></citerefentry> device
tree. This may be used to define dependencies between devices and other units.</para>
<para>This unit type has no specific options. See
<citerefentry><refentrytitle>systemd.unit</refentrytitle><manvolnum>5</manvolnum></citerefentry>
@@ -40,14 +39,10 @@
sections. A separate [Device] section does not
exist, since no device-specific options may be configured.</para>
<para>systemd will dynamically create device units for all kernel
devices that are marked with the "systemd" udev tag (by default
all block and network devices, and a few others). This may be used
to define dependencies between devices and other units. To tag a
udev device, use <literal>TAG+="systemd"</literal> in the udev
rules file, see
<citerefentry><refentrytitle>udev</refentrytitle><manvolnum>7</manvolnum></citerefentry>
for details.</para>
<para>systemd will dynamically create device units for all kernel devices that are marked with the
<literal>systemd</literal> udev tag (by default all block and network devices, and a few others). Note
that <emphasis>if <filename>systemd-udev.service</filename> is not running, no device units will be
available (for example in a typical container)</emphasis>.</para>
<para>Device units are named after the <filename>/sys/</filename>
and <filename>/dev/</filename> paths they control. Example: the
@@ -57,6 +52,10 @@
name see
<citerefentry><refentrytitle>systemd.unit</refentrytitle><manvolnum>5</manvolnum></citerefentry>.</para>
<para>To tag a udev device, use <literal>TAG+="systemd"</literal> in the udev rules file, see
<citerefentry><refentrytitle>udev</refentrytitle><manvolnum>7</manvolnum></citerefentry> for details.
</para>
<para>Device units will be reloaded by systemd whenever the
corresponding device generates a <literal>changed</literal> event.
Other units can use <varname>ReloadPropagatedFrom=</varname> to react

16
src/basic/log.c
View File

@@ -252,18 +252,14 @@ int log_open(void) {
/* Do not call from library code. */
/* This function is often called in preparation for being able
* to log. Let's make sure we don't clobber errno, so that a call
* to a logging function immediately following a log_open() call
* can still easily reference an error that happened immediately
* before the log_open() call. */
/* This function is often called in preparation for logging. Let's make sure we don't clobber errno,
* so that a call to a logging function immediately following a log_open() call can still easily
* reference an error that happened immediately before the log_open() call. */
PROTECT_ERRNO;
/* If we don't use the console we close it here, to not get
* killed by SAK. If we don't use syslog we close it here so
* that we are not confused by somebody deleting the socket in
* the fs, and to make sure we don't use it if prohibit_ipc is
* set. If we don't use /dev/kmsg we still keep it open,
/* If we don't use the console, we close it here to not get killed by SAK. If we don't use syslog, we
* close it here too, so that we are not confused by somebody deleting the socket in the fs, and to
* make sure we don't use it if prohibit_ipc is set. If we don't use /dev/kmsg we still keep it open,
* because there is no reason to close it. */
if (log_target == LOG_TARGET_NULL) {