According to the specification[1] the 'capabilities' describe the physical
device as a whole and the 'device_caps' describe the current device node.
The existence of 'device_caps' is indicated by the V4L2_CAP_DEVICE_CAPS
capability flag.
Use the 'device_caps' if available to generate the correct
ID_V4L_CAPABILITIES for the current device node.
This is relevant for UVC devices with current kernels: Two /dev/videoX
devices exist for those. One for video and one for metadata. The
V4L2_CAP_VIDEO_CAPTURE flag is present in the 'capabilities' for both
device nodes but only in the 'device_caps' of the video device node.
Without this, the ID_V4L_CAPABILITIES of the metadata device node
incorrectly contains 'capture'.
[1] https://www.linuxtv.org/downloads/v4l-dvb-apis-new/uapi/v4l/vidioc-querycap.html
systemd-commit: 646876105fe1799355715839477958adf1d096fd
Author: Michael Olbrich <m.olbrich@pengutronix.de>
Date: Sun Feb 3 10:52:02 2019 +0100
Backport of afe87974dd.
Will address shown error
> Error, empty key or value ' XKB_FIXED_VARIANT' in '/usr/lib/udev/hwdb.d/60-keyboard.hwdb'
caued by recent changes in sys-apps/hwids-20201207 [Link 1].
Link 1: f30071bbeb (diff-7b9ba9c30888b1b5b1fa008f185e4efaff34eaff2c13f35c38b2ae0416cd891eR1840)
Signed-off-by: Thomas Deutschmann <whissi@gentoo.org>
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
The following device properly got assigned ID_PATH with eudev 3.2.1,
but does not get it any more with recent versions:
P: /devices/platform/rt5651-sound/sound/card0
E: DEVPATH=/devices/platform/rt5651-sound/sound/card0
E: SOUND_FORM_FACTOR=internal
E: SOUND_INITIALIZED=1
E: SUBSYSTEM=sound
E: USEC_INITIALIZED=507638688065
This results in pulseaudio giving a non-stable card name by using a
device number.
I tracked the problem to commit
47367bc4df, which changed (without
explaining why it does so) the rules for "internal" sound cards,
notably preventing `IMPORT{builtin}="path_id"` from being reached.
Signed-off-by: Yann Dirson <yann@blade-group.com>
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
Fixes `test: too many arguments` when building Linux-PAM using sbase.
This is due to a non-POSIX syntax test ... -a ... and test ... -o ....
> The XSI extensions specifying the -a and -o binary primaries and the
> '(' and ')' operators have been marked obsolescent.
See https://pubs.opengroup.org/onlinepubs/9699919799/utilities/test.html
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
This fixes#182 by ensuring @bindir@ is fully expanded at installation
time. See "Installation Directory Variables" in the GNU Autoconf
manual, in particular the note about AC_CONFIG_FILES.
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
This check fails for buildroot systems where /dev is not mounted
as a tmpfs filesystem. Dropping this check should be safe even
on regular systems.
This solves issue #172.
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
As explained in issue 6267 of systemd [1], Linux < 3.11 does not
provide definitions for BTN_DPAD_{UP,RIGHT}, which were introduced in
[2].
This patch fixes this issue.
[1] https://github.com/systemd/systemd/pull/6267
[2] 9ee2748712
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
If a device exposes more than 16 mouse buttons, we run into the BTN_JOYSTICK
range, also labelling it as joystick. And since 774ff9b this results in only
ID_INPUT_JOYSTICK but no ID_INPUT_MOUSE.
Fixes#8460
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
short options which were added to the udevd manpage, were not added to
the .xml file, and everytime manpages get regenerated the change gets
overwritten.
also update pregenerated version of udev manpage
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
Commit bb070c1 forgot to include the installation of 60-input-id.rules
and 70-joystick.rules via the Makefile. Without this, commits dba4728
and 5df0137 leave the system without any useable input devices.
Update to match systemd v235-1952-gba3182b91
Signed-off-by: Marcus Folkesson <marcus.folkesson@gmail.com>
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
Update to match systemd v235-1952-gba3182b91
Signed-off-by: Marcus Folkesson <marcus.folkesson@gmail.com>
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
Some headerfiles are missing "#pragma once" or corresponding
"#ifndef..." which may result in build error if multiple source files
include the same file.
Use "#pragma once" for all header files.
Signed-off-by: Marcus Folkesson <marcus.folkesson@gmail.com>
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
When using --debug, we want the output to our console.
Signed-off-by: Marcus Folkesson <marcus.folkesson@gmail.com>
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
Last time $UDEV_LOG was referred in code was in
dacd994fda.
Remove all traces for this legacy option.
Signed-off-by: Marcus Folkesson <marcus.folkesson@gmail.com>
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
Solve a bug that could result in really nasty behavior.
Signed-off-by: Marcus Folkesson <marcus.folkesson@gmail.com>
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
Resolve warning:
util.c:1730:23: warning: passing argument 1 of ‘__xpg_basename’ discards
‘const’ qualifier from pointer target type [-Wdiscarded-qualifiers]
Signed-off-by: Marcus Folkesson <marcus.folkesson@gmail.com>
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
For all those who prefer emails instead of web GUIs.
Signed-off-by: Marcus Folkesson <marcus.folkesson@gmail.com>
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
