chiark / gitweb /
elogind.git
12 years agoconfigure: allow to enable/disable extras individually
Kay Sievers [Fri, 29 Apr 2011 12:27:11 +0000 (14:27 +0200)]
configure: allow to enable/disable extras individually

      - remove gobject introspection switch
      - disable hid2hci by default (moved to bluez)
      - disable action-modeswitch by default (will move to usb_modeswitch)
      - disable edd_id by default (problems with disk signatures)
      - disable legacy floppy by default (no more nasty device node hacks by default)

12 years agoudevd: improve error message in case exec() fails
Kay Sievers [Thu, 28 Apr 2011 12:58:06 +0000 (14:58 +0200)]
udevd: improve error message in case exec() fails

12 years agoset event timeout to 60 sec and settle timeout to 120
Kay Sievers [Wed, 27 Apr 2011 22:04:17 +0000 (00:04 +0200)]
set event timeout to 60 sec and settle timeout to 120

12 years agohid2hci: prepare move to bluez package
Kay Sievers [Tue, 26 Apr 2011 23:10:04 +0000 (01:10 +0200)]
hid2hci: prepare move to bluez package

12 years agoremove dead fstab_import files
Kay Sievers [Tue, 26 Apr 2011 21:04:44 +0000 (23:04 +0200)]
remove dead fstab_import files

13 years agolibudev: queue - _unref() should return the object
Kay Sievers [Sat, 23 Apr 2011 22:36:22 +0000 (00:36 +0200)]
libudev: queue - _unref() should return the object

13 years agolibudev: list - use bit flags for 'sort' and 'unique'
Kay Sievers [Sat, 23 Apr 2011 22:22:23 +0000 (00:22 +0200)]
libudev: list - use bit flags for 'sort' and 'unique'

13 years agosimplify rules file overwrite logic
Kay Sievers [Sat, 23 Apr 2011 21:33:11 +0000 (23:33 +0200)]
simplify rules file overwrite logic

The udev list already provides all we need, if we use the basename
as the key, and request sorting and uniqueness.

13 years agorelease 168
Kay Sievers [Fri, 22 Apr 2011 16:29:05 +0000 (18:29 +0200)]
release 168

13 years agoudevd, udev-event: sync waitpid() error handling
Kay Sievers [Thu, 21 Apr 2011 12:05:34 +0000 (14:05 +0200)]
udevd, udev-event: sync waitpid() error handling

13 years agoudevd: clarify worker exit status
Harald Hoyer [Thu, 21 Apr 2011 08:06:12 +0000 (10:06 +0200)]
udevd: clarify worker exit status

13 years agosystemd: Reqires= -> Wants=udev.socket
Kay Sievers [Wed, 20 Apr 2011 23:13:28 +0000 (01:13 +0200)]
systemd: Reqires= -> Wants=udev.socket

13 years agopath_id: rework SAS device handling
Michael Reed [Wed, 20 Apr 2011 18:25:16 +0000 (20:25 +0200)]
path_id: rework SAS device handling

Lun numbers < 256 are displayed as "lun-x".  Lun numbers >= 256 use
the lun number format previously used in fibre channel by-path
links, 0xabcd012300000000. This lun numbering scheme is implemented
for sas, fibre channel, iscsi.

I want it to be clearly understood that this patch changes the
/dev/disk/by-path names for fc devices.  Here is an example of the
change:
  Old: pci-0000:0c:00.1-fc-0x204700a0b81130aa:0x00fa000000000000
  New: pci-0000:0c:00.1-fc-0x204700a0b81130aa-lun-250

13 years agoudevd: log signal number when spawned processes fail
Kay Sievers [Wed, 20 Apr 2011 16:36:01 +0000 (18:36 +0200)]
udevd: log signal number when spawned processes fail

13 years agoudevd: rules files - accept empty or /dev/null links
Kay Sievers [Wed, 20 Apr 2011 15:58:58 +0000 (17:58 +0200)]
udevd: rules files - accept empty or /dev/null links

13 years agoudevd: always use udevd[] log prefix
Kay Sievers [Wed, 20 Apr 2011 15:39:44 +0000 (17:39 +0200)]
udevd: always use udevd[] log prefix

13 years agoudevd: netif rename - use ifindex for temporary name
Kay Sievers [Wed, 20 Apr 2011 15:30:30 +0000 (17:30 +0200)]
udevd: netif rename - use ifindex for temporary name

13 years agoudevadm: settle - kill alarm()
Kay Sievers [Wed, 20 Apr 2011 00:18:22 +0000 (02:18 +0200)]
udevadm: settle - kill alarm()

13 years agotimeout handling without alarm()
Kay Sievers [Tue, 19 Apr 2011 23:53:03 +0000 (01:53 +0200)]
timeout handling without alarm()

13 years agokeymap: Another ID for Logitech Wave keyboard
Martin Pitt [Mon, 18 Apr 2011 11:34:42 +0000 (13:34 +0200)]
keymap: Another ID for Logitech Wave keyboard

http://bugs.debian.org/623153

13 years agorule-generator: net - ignore Hyper-V virtual interfaces
Leonid Antonenkov [Mon, 18 Apr 2011 03:11:53 +0000 (05:11 +0200)]
rule-generator: net - ignore Hyper-V virtual interfaces

13 years agoMerge branch 'docs/README' of git://github.com/mfwitten/udev
Kay Sievers [Mon, 18 Apr 2011 03:06:59 +0000 (05:06 +0200)]
Merge branch 'docs/README' of git://github.com/mfwitten/udev

13 years agoudevd: ppoll() -> epoll + signalfd
Kay Sievers [Mon, 18 Apr 2011 00:14:24 +0000 (02:14 +0200)]
udevd: ppoll() -> epoll + signalfd

13 years agolibudev: run_program() - select() -> epoll
Kay Sievers [Mon, 18 Apr 2011 00:13:40 +0000 (02:13 +0200)]
libudev: run_program() - select() -> epoll

13 years agouse 'else if' in epoll event array loop
Kay Sievers [Sun, 17 Apr 2011 20:10:18 +0000 (22:10 +0200)]
use 'else if' in epoll event array loop

13 years agoudevadm: test - use printf() instead of info() for non-debug output
Kay Sievers [Sun, 17 Apr 2011 20:09:08 +0000 (22:09 +0200)]
udevadm:  test - use printf() instead of info() for non-debug output

13 years agolibudev: test - use epoll
Kay Sievers [Fri, 15 Apr 2011 10:18:06 +0000 (12:18 +0200)]
libudev: test - use epoll

13 years agoudevadm: monitor - use epoll
Kay Sievers [Fri, 15 Apr 2011 09:58:17 +0000 (11:58 +0200)]
udevadm: monitor - use epoll

13 years agosystemd: socket -> sockets
Kay Sievers [Thu, 14 Apr 2011 23:43:27 +0000 (01:43 +0200)]
systemd: socket -> sockets

13 years agolibudev: ctrl, monitor - use SOCK_NONBLOCK
Kay Sievers [Thu, 14 Apr 2011 23:14:10 +0000 (01:14 +0200)]
libudev: ctrl, monitor - use SOCK_NONBLOCK

13 years agofix more warnings
Kay Sievers [Thu, 14 Apr 2011 22:19:45 +0000 (00:19 +0200)]
fix more warnings

13 years agoudevd: get netlink socket from systemd
Kay Sievers [Thu, 14 Apr 2011 21:46:44 +0000 (23:46 +0200)]
udevd: get netlink socket from systemd

13 years agorequire explicit "db_persist" to exclude device info from --db-cleanup
Kay Sievers [Thu, 14 Apr 2011 12:44:56 +0000 (14:44 +0200)]
require explicit "db_persist" to exclude device info from --db-cleanup

13 years agosystemd: add OOMScoreAdjust=-1000
Kay Sievers [Thu, 14 Apr 2011 12:44:21 +0000 (14:44 +0200)]
systemd: add OOMScoreAdjust=-1000

13 years agoudevd: move OOM disable into --daemon option
Kay Sievers [Thu, 14 Apr 2011 12:43:22 +0000 (14:43 +0200)]
udevd: move OOM disable into --daemon option

13 years ago"db_persist=" -> "db_persist"
Kay Sievers [Thu, 14 Apr 2011 10:42:59 +0000 (12:42 +0200)]
"db_persist=" -> "db_persist"

13 years agoudevd: do not nice processes
Kay Sievers [Thu, 14 Apr 2011 10:25:19 +0000 (12:25 +0200)]
udevd: do not nice processes

13 years agoudevadm: info --cleanup-db
Kay Sievers [Wed, 13 Apr 2011 20:33:01 +0000 (22:33 +0200)]
udevadm: info --cleanup-db

Most of the udev database from initramfs should be deleted before
starting udev in the real root. udevadm: info --cleanup-db deletes
all database entries in /run/udev. Events that processed IMPORT{db},
or mark devices explicitely as persistent, will be excluded.

13 years agolibudev: bump revision
Kay Sievers [Wed, 13 Apr 2011 16:46:02 +0000 (18:46 +0200)]
libudev: bump revision

13 years agoudevadm: settle - watch queue file
Kay Sievers [Wed, 13 Apr 2011 16:44:28 +0000 (18:44 +0200)]
udevadm: settle - watch queue file

13 years agolibudev: ctrl - fix refcounting in connection handling
Kay Sievers [Wed, 13 Apr 2011 16:21:47 +0000 (18:21 +0200)]
libudev: ctrl - fix refcounting in connection handling

13 years agoudevd: log warning if /run is not writable
Kay Sievers [Wed, 13 Apr 2011 13:01:00 +0000 (15:01 +0200)]
udevd: log warning if /run is not writable

13 years agotrivial cleanups
Kay Sievers [Tue, 12 Apr 2011 23:47:56 +0000 (01:47 +0200)]
trivial cleanups

13 years agoudevadm: control - add --exit
Kay Sievers [Tue, 12 Apr 2011 23:17:09 +0000 (01:17 +0200)]
udevadm: control - add --exit

13 years agoDocs: README: Add a comma to offset the modifier
Michael Witten [Fri, 8 Apr 2011 16:20:18 +0000 (16:20 +0000)]
Docs: README: Add a comma to offset the modifier

Signed-off-by: Michael Witten <mfwitten@gmail.com>
13 years agoDocs: README: `set up' is the verb; `setup' is a noun
Michael Witten [Fri, 8 Apr 2011 16:19:46 +0000 (16:19 +0000)]
Docs: README: `set up' is the verb; `setup' is a noun

Signed-off-by: Michael Witten <mfwitten@gmail.com>
13 years agoDocs: README: Flip sentence structure to improve wording
Michael Witten [Fri, 8 Apr 2011 16:18:32 +0000 (16:18 +0000)]
Docs: README: Flip sentence structure to improve wording

Signed-off-by: Michael Witten <mfwitten@gmail.com>
13 years agoDocs: README: `does never apply' -> `never applies'
Michael Witten [Fri, 8 Apr 2011 16:15:39 +0000 (16:15 +0000)]
Docs: README: `does never apply' -> `never applies'

Signed-off-by: Michael Witten <mfwitten@gmail.com>
13 years agoDocs: README: Clarify configuration of existing devices
Michael Witten [Fri, 8 Apr 2011 16:14:48 +0000 (16:14 +0000)]
Docs: README: Clarify configuration of existing devices

Signed-off-by: Michael Witten <mfwitten@gmail.com>
13 years agoDocs: README: Clean up `udev extras' requirements
Michael Witten [Fri, 8 Apr 2011 16:04:38 +0000 (16:04 +0000)]
Docs: README: Clean up `udev extras' requirements

Signed-off-by: Michael Witten <mfwitten@gmail.com>
13 years agoDocs: README: Remove commas and use subjective mood
Michael Witten [Fri, 8 Apr 2011 16:01:18 +0000 (16:01 +0000)]
Docs: README: Remove commas and use subjective mood

Signed-off-by: Michael Witten <mfwitten@gmail.com>
13 years agoDocs: README: Add missing `and'
Michael Witten [Fri, 8 Apr 2011 16:00:01 +0000 (16:00 +0000)]
Docs: README: Add missing `and'

Signed-off-by: Michael Witten <mfwitten@gmail.com>
13 years agoDocs: README: Use present tense
Michael Witten [Fri, 8 Apr 2011 15:59:01 +0000 (15:59 +0000)]
Docs: README: Use present tense

Signed-off-by: Michael Witten <mfwitten@gmail.com>
13 years agoDocs: README: Clean up a sentence
Michael Witten [Fri, 8 Apr 2011 15:56:16 +0000 (15:56 +0000)]
Docs: README: Clean up a sentence

Signed-off-by: Michael Witten <mfwitten@gmail.com>
13 years agoDocs: README: `,' -> `;'
Michael Witten [Fri, 8 Apr 2011 15:55:18 +0000 (15:55 +0000)]
Docs: README: `,' -> `;'

Signed-off-by: Michael Witten <mfwitten@gmail.com>
13 years agoDocs: README: `to replace' -> `replacing'
Michael Witten [Fri, 8 Apr 2011 15:54:04 +0000 (15:54 +0000)]
Docs: README: `to replace' -> `replacing'

Signed-off-by: Michael Witten <mfwitten@gmail.com>
13 years agoselinux: firmware - do not label files in runtime dir
Kay Sievers [Fri, 8 Apr 2011 01:08:16 +0000 (03:08 +0200)]
selinux: firmware - do not label files in runtime dir

13 years agoselinux: do not label files in runtime dir
Kay Sievers [Thu, 7 Apr 2011 23:03:49 +0000 (01:03 +0200)]
selinux: do not label files in runtime dir

Do not label any files in the udev runtime directory, but only nodes,
links and directories below /dev.

In case the runtime directory falls back to /dev/.udev, label this
directory once at udevd startup, but never anything below it.

13 years agokeymap: Support Dell Latitude XT2 tablet-mode navigation keys
Seth Forshee [Tue, 5 Apr 2011 16:43:00 +0000 (11:43 -0500)]
keymap: Support Dell Latitude XT2 tablet-mode navigation keys

The XT2 has a rocker (up/down/enter) and back button on the side
in tablet mode, none of which work currently. Add entries for
these keys.

There is some overlap here with scan codes used in other Dell
models, so these buttons are put in a new file specific to this
model.

Signed-off-by: Seth Forshee <seth.forshee@canonical.com>
Signed-off-by: Martin Pitt <martin.pitt@ubuntu.com>
13 years agoAvoid spinning up CD on pressing eject button
Martin Pitt [Mon, 4 Apr 2011 09:47:54 +0000 (11:47 +0200)]
Avoid spinning up CD on pressing eject button

When the CD tray door is locked and the hardware eject button is pressed, newer
kernels (2.6.38+) will send out a change event with a DISK_EJECT_REQUEST==1
property. Do not run cdrom_id and blkid in this case, as the media state and
contents does not change, and this only needlessly spins up the drive again
right before ejection.

13 years agoRevert "Do not build extras with --disable-extras"
Martin Pitt [Thu, 31 Mar 2011 17:23:19 +0000 (19:23 +0200)]
Revert "Do not build extras with --disable-extras"

This reverts commit 905818f5961446ac32d1b2d165185fffddc4915a.

We do want to build some extras, just not the ones with extra dependencies.
Thanks to Juergen Daubert for pointing this out.

13 years agoudevd: add 'N:' to optstring in getopt_long
Thomas Egerer [Thu, 31 Mar 2011 15:25:31 +0000 (17:25 +0200)]
udevd: add 'N:' to optstring in getopt_long

13 years agoAdd rule for Acer Aspire One ZG8 to use acer-aspire_5720 keymap
Lee, Chun-Yi [Thu, 31 Mar 2011 10:01:05 +0000 (18:01 +0800)]
Add rule for Acer Aspire One ZG8 to use acer-aspire_5720 keymap

Acer Aspire One ZG8's bluetooth HW key emit 0xD9 scancode, it must map
to KEY_BLUETOOTH like Acer Aspire 5720. So, add rule for Acer Aspire One
ZG8 use acer-aspire_5720 keymap.

Tested on Acer Aspire One ZG8 netbook.

Signed-off-by: Lee, Chun-Yi <jlee@novell.com>
Signed-off-by: Martin Pitt <martin.pitt@ubuntu.com>
13 years agosystemd: let settle depend on trigger, do not block basic with trigger
Kay Sievers [Thu, 31 Mar 2011 11:15:47 +0000 (13:15 +0200)]
systemd: let settle depend on trigger, do not block basic with trigger

13 years agoRun ata_id on non-removable USB devices
David Zeuthen [Wed, 30 Mar 2011 12:11:03 +0000 (08:11 -0400)]
Run ata_id on non-removable USB devices

If a USB device is marked as removable, it is often a SATA/PATA disk
in an enclosure (note that flash card readers and usb storage sticks
are always marked as removable). In this case, try running ata_id
(which sends ATA commands wrapped in SCSI ATA PASS THROUGH commands)
to get information about the disk. If this fails, just fall back to
running usb_id since it could be the device isn't an ATA device at
all or the device doesn't have a SAT layer.

This extra information is nice to have as it indicates if it is
suitable to send e.g. SMART commands to the disk, whether the disk
supports power management and so on. Additionally, the VPD and serial
number returned by ata_id is usually more accurate as it stems from
the disk itself instead of the enclosure.

Note that udisks has been doing this for a while

 KERNEL=="sd*[!0-9]", ATTR{removable}=="0", ENV{ID_BUS}=="usb", ENV{DEVTYPE}=="disk", IMPORT{program}="udisks-probe-ata-smart $tempnode"

so this change shouldn't be too disruptive (udisks-probe-ata-smart
also sends ATA commands via the ATA PASS THROUGH command).

Signed-off-by: David Zeuthen <davidz@redhat.com>
13 years agoversion bump
Kay Sievers [Wed, 30 Mar 2011 14:03:56 +0000 (16:03 +0200)]
version bump

13 years agorelease 167
Kay Sievers [Wed, 30 Mar 2011 13:58:14 +0000 (15:58 +0200)]
release 167

13 years agosystemd: udev.socket - disable implicit dependencies
Kay Sievers [Tue, 29 Mar 2011 15:59:29 +0000 (17:59 +0200)]
systemd: udev.socket - disable implicit dependencies

  Set hostname to <eeepc>.
  Found ordering cycle on sysinit.target/start
  Walked on cycle path to systemd-tmpfiles-setup.service/start
  Walked on cycle path to local-fs.target/start
  Walked on cycle path to fedora-storage-init.service/start
  Walked on cycle path to fedora-wait-storage.service/start
  Walked on cycle path to udev-settle.service/start
  Walked on cycle path to udev.service/start
  Walked on cycle path to udev.socket/start
  Walked on cycle path to sysinit.target/start
  Breaking ordering cycle by deleting job systemd-tmpfiles-setup.service/start

13 years agosystemd: do not enable udev-settle.service by default
Kay Sievers [Mon, 28 Mar 2011 23:14:53 +0000 (01:14 +0200)]
systemd: do not enable udev-settle.service by default

This barrier service is usually not enabled by default. If enabled,
it acts as a barrier for basic.target -- so all later services will
wait for udev completely finishing its coldplug run.

It might be enabled just unconditionally, or pulled-in on-demand by
broken or non-hotplug-aware services that assume a fully populated
/dev at startup.

13 years agolibudev: make valgrind happy
Kay Sievers [Mon, 28 Mar 2011 12:34:05 +0000 (14:34 +0200)]
libudev: make valgrind happy

Warning: noted but unhandled ioctl 0x641e with no size/direction hints
  Syscall param socketcall.setsockopt(optval) points to uninitialised byte(s)
  at 0x6A30BAA: setsockopt (syscall-template.S:82)
  by 0x4E36E1C: udev_monitor_filter_update (libudev-monitor.c:347)
  by 0x4E36ED8: udev_monitor_enable_receiving (libudev-monitor.c:373)
  by 0x8A9E68E: I830ScreenInit (intel_driver.c:838)
  by 0x42EE9E: AddScreen (dispatch.c:3890)
  by 0x471561: InitOutput (xf86Init.c:738)
  by 0x422C18: main (main.c:205)
  Location 0x7ff0000c2 is 2 bytes inside local var "filter"
  declared at libudev-monitor.c:256, in frame #1 of thread 1
  Uninitialised value was created by a stack allocation
  at 0x4E30860: ??? (in /lib64/libudev.so.0.10.0)

Thanks to Zdenek Kabelac.

13 years agolibudev: docs - add udev_get_run_path()
Kay Sievers [Fri, 25 Mar 2011 16:58:30 +0000 (17:58 +0100)]
libudev: docs - add udev_get_run_path()

13 years agolibudev: export udev_get_run_path()
Kay Sievers [Fri, 25 Mar 2011 14:43:01 +0000 (15:43 +0100)]
libudev: export udev_get_run_path()

13 years agokeymap: continue reading keymap after invalid scancodes
Seth Forshee [Thu, 24 Mar 2011 15:33:54 +0000 (16:33 +0100)]
keymap: continue reading keymap after invalid scancodes

Reading of keymaps is aborted when EINVAL is returned from EVIOCGETKEYCODE.
Scan codes are not always continuous ranges of values starting at 0, so this
can result in not getting the full keymap for a device. Instead, continue
processing if EINVAL is returned.

13 years agoman: remove trigger --type=failed handling
Kay Sievers [Wed, 23 Mar 2011 22:56:15 +0000 (23:56 +0100)]
man: remove trigger --type=failed handling

13 years agosystemd: use sockets.target not socket.target
Kay Sievers [Wed, 23 Mar 2011 19:35:19 +0000 (20:35 +0100)]
systemd: use sockets.target not socket.target

13 years agosystemd: bind udev control socket in systemd and split udev.service
Kay Sievers [Wed, 23 Mar 2011 15:40:23 +0000 (16:40 +0100)]
systemd: bind udev control socket in systemd and split udev.service

We should bind the udev socket from systemd, so we are sure
that the abstract namespace socket is always bound by a root
process and there is never a window during an update where
an untrusted process can steal our socket.

Also split the udev.service file, so that the daemon can be
updated/restarted without triggering any coldplug events.

13 years agorules: run ata_id only on SPC-3 or later optical drives
Kay Sievers [Wed, 23 Mar 2011 11:11:38 +0000 (12:11 +0100)]
rules: run ata_id only on SPC-3 or later optical drives

13 years agouse /run/udev/ if possible and fall back to /dev/.udev/
Kay Sievers [Wed, 23 Mar 2011 01:31:09 +0000 (02:31 +0100)]
use /run/udev/ if possible and fall back to /dev/.udev/

13 years agoudevd: initialize fds, for proper close() on exit
Kay Sievers [Wed, 23 Mar 2011 00:45:21 +0000 (01:45 +0100)]
udevd: initialize fds, for proper close() on exit

13 years agoudevd: use facility == LOG_DAEMON when writing to /dev/kmsg
Kay Sievers [Wed, 23 Mar 2011 00:41:25 +0000 (01:41 +0100)]
udevd: use facility == LOG_DAEMON when writing to /dev/kmsg

Syslog wants to distinguish the sorce of messages. We should
indicate that this is a userspace message (LOG_DAEMON) and not
a kernel message (LOG_KERNEL).

13 years agoupdate sd-daemon.[ch]
Kay Sievers [Wed, 23 Mar 2011 00:34:29 +0000 (01:34 +0100)]
update sd-daemon.[ch]

13 years agofstab_import: remove from configure
Kay Sievers [Sat, 19 Mar 2011 17:28:57 +0000 (18:28 +0100)]
fstab_import: remove from configure

13 years agosystemd: remove deprecated udev-retry.service
Kay Sievers [Sat, 19 Mar 2011 17:13:01 +0000 (18:13 +0100)]
systemd: remove deprecated udev-retry.service

13 years agofstab_import: disable build
Kay Sievers [Sat, 19 Mar 2011 17:11:12 +0000 (18:11 +0100)]
fstab_import: disable build

13 years agoinput_id: silent gcc warnings
Kay Sievers [Sat, 19 Mar 2011 16:04:56 +0000 (17:04 +0100)]
input_id: silent gcc warnings

13 years agoinput_id: Rewrite debug logging to use standard udev info()
Martin Pitt [Fri, 18 Mar 2011 15:18:17 +0000 (16:18 +0100)]
input_id: Rewrite debug logging to use standard udev info()

13 years agoinput_id: Cover key devices which only have KEY_* > 255
Martin Pitt [Fri, 18 Mar 2011 13:45:42 +0000 (14:45 +0100)]
input_id: Cover key devices which only have KEY_* > 255

We previously only tested the "key" capabilities for keys between 0 and 255 to
determine ID_INPUT_KEY. If there are none, also check for keys in the upper
block (KEY_OK/0x160 to KEY_TOUCHPAD_OFF/0x214).

13 years agoinput_id: Avoid memory overflow with too long capability masks
Martin Pitt [Fri, 18 Mar 2011 12:56:32 +0000 (13:56 +0100)]
input_id: Avoid memory overflow with too long capability masks

Joey Lee <jlee@novell.com> reported a problem on an MSI laptop which reports a
too long capabilities/key:

E: EV==3
E: KEY==180000 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

This is longer than KEY_MAX and thus caused a memory overflow. Guard against
this now and just ignore the excess blocks.

13 years agoinput_id: Add some debugging output
Martin Pitt [Fri, 18 Mar 2011 12:42:02 +0000 (13:42 +0100)]
input_id: Add some debugging output

If $DEBUG is set, output the capability bitmaps to stderr.

13 years agoinput_id: Consistently use tabs for indentation
Martin Pitt [Fri, 18 Mar 2011 12:32:55 +0000 (13:32 +0100)]
input_id: Consistently use tabs for indentation

13 years agoudev-acl/70-acl.rules: tag ID_REMOTE_CONTROL with acl
Harald Hoyer [Thu, 17 Mar 2011 19:36:20 +0000 (20:36 +0100)]
udev-acl/70-acl.rules: tag ID_REMOTE_CONTROL with acl

Used by "concordance" to access Logitech Harmony programmable remote
controls.

http://www.phildev.net/concordance/

13 years agoNEWS: clarify /dev/.run/ requirements
Kay Sievers [Wed, 16 Mar 2011 17:58:03 +0000 (18:58 +0100)]
NEWS: clarify /dev/.run/ requirements

13 years agocdrom_id: Don't ignore profiles when there is no media available
David Zeuthen [Tue, 15 Mar 2011 16:05:00 +0000 (12:05 -0400)]
cdrom_id: Don't ignore profiles when there is no media available

Just because the GET CONFIGURATION MMC-2 command returns the current
profile as 0 does not mean that we can ignore the profiles describing
the capabilities of the drive - it only means that there currently is
no recognized media in the drive. Therefore, do process the returned
profiles even when cur_profile is 0.

This fixes a bug where only

 ID_CDROM=1
 ID_CDROM_CD_R=1
 ID_CDROM_CD_RW=1
 ID_CDROM_DVD=1
 ID_CDROM_DVD_R=1
 ID_CDROM_DVD_RAM=1

was returned when there is no media in the drive instead of

 ID_CDROM=1
 ID_CDROM_CD=1
 ID_CDROM_CD_R=1
 ID_CDROM_CD_RW=1
 ID_CDROM_DVD=1
 ID_CDROM_DVD_R=1
 ID_CDROM_DVD_RW=1
 ID_CDROM_DVD_RAM=1
 ID_CDROM_DVD_PLUS_R=1
 ID_CDROM_DVD_PLUS_RW=1
 ID_CDROM_DVD_PLUS_R_DL=1
 ID_CDROM_BD=1
 ID_CDROM_BD_R=1
 ID_CDROM_BD_RE=1
 ID_CDROM_HDDVD=1

as is returned now.

Signed-off-by: David Zeuthen <davidz@redhat.com>
13 years agomove /dev/.udev/ to /dev/.run/udev/ and convert old udev database at udevd startup
Kay Sievers [Tue, 15 Mar 2011 16:26:08 +0000 (17:26 +0100)]
move /dev/.udev/ to /dev/.run/udev/ and convert old udev database at udevd startup

13 years agoRemap MSI Laptop touchpad on/off key to F22 and F23
Lee, Chun-Yi [Thu, 10 Mar 2011 02:17:12 +0000 (10:17 +0800)]
Remap MSI Laptop touchpad on/off key to F22 and F23

MSI Laptop driver will query the real touchpad state then emit
KEY_TOUCHPAD_ON/OFF key.

Currently, X has defined F22 for touchpad on and F23 for touchpad
off. This patch aligns MSI Laptop driver's key with F22 and F23.

Tested on MSI U160 netbook using msi-laptop driver.

Signed-off-by: Martin Pitt <martin.pitt@ubuntu.com>
13 years agoudevadm: info - make attribute array static and const
Thomas Egerer [Tue, 8 Mar 2011 08:57:17 +0000 (09:57 +0100)]
udevadm: info - make attribute array static and const

13 years agoudev-acl: move sg rule to optical drive rule
Kay Sievers [Mon, 7 Mar 2011 15:10:50 +0000 (16:10 +0100)]
udev-acl: move sg rule to optical drive rule

13 years agoudev-acl: add /dev/sgX nodes for CD-ROM
Andrey Borzenkov [Sun, 6 Mar 2011 09:01:22 +0000 (12:01 +0300)]
udev-acl: add /dev/sgX nodes for CD-ROM

Current wine is using /dev/sgX to access CD-ROM devices. Since
distributions switched to using ACL instead of group membership
to control device access, wine is not able to access them.

Add ACL to device nodes that already get GROUP="cdrom".

Ref: https://qa.mandriva.com/show_bug.cgi?id=62114

Signed-off-by: Andrey Borzenkov <arvidjaar@mail.ru>
13 years agolibudev: bump minor version
Kay Sievers [Sat, 5 Mar 2011 05:27:31 +0000 (06:27 +0100)]
libudev: bump minor version

13 years agolibudev: resolve ifindex in udev_device_new_from_id_filename()
Kay Sievers [Sat, 5 Mar 2011 05:26:39 +0000 (06:26 +0100)]
libudev: resolve ifindex in udev_device_new_from_id_filename()