chiark / gitweb /
elogind.git
11 years agoavoid leaking netlink socket fd to external programs
Alan Jenkins [Mon, 6 Apr 2009 09:18:41 +0000 (10:18 +0100)]
avoid leaking netlink socket fd to external programs

The netlink socket is now used by udev event processes. We should take
care not to pass it to the programs they execute. This is the same way
the inotify fd was handled.

Signed-off-by: Alan Jenkins <alan-jenkins@tuffmail.co.uk>
11 years agoIMPORT: 2048 -> 4096 bytes buffer
Kay Sievers [Mon, 30 Mar 2009 18:09:08 +0000 (20:09 +0200)]
IMPORT: 2048 -> 4096 bytes buffer

11 years ago"UDEV_MONITOR_KERNEL/UDEV" -> "kernel/udev"
Kay Sievers [Sun, 29 Mar 2009 18:08:52 +0000 (20:08 +0200)]
"UDEV_MONITOR_KERNEL/UDEV" -> "kernel/udev"

11 years agosend monitor events back to netlink socket
Kay Sievers [Sun, 29 Mar 2009 02:24:39 +0000 (04:24 +0200)]
send monitor events back to netlink socket

Instead of of our own private monitor socket, we send the
processed event back to our netlink socket, to the multicast
group 2 -- so any number of users can listen to udev events,
just like they can listen to kernel emitted events on group 1.

11 years agousb_id: remove unused variable
Kay Sievers [Sat, 28 Mar 2009 19:57:01 +0000 (20:57 +0100)]
usb_id: remove unused variable

11 years agolibudev: GPL -> LGPL
Kay Sievers [Thu, 26 Mar 2009 18:29:36 +0000 (19:29 +0100)]
libudev: GPL -> LGPL

11 years agousb_id: add ID_VENDOR_ID, ID_MODEL_ID, ID_USB_INTERFACE_NUM, ID_USB_DRIVER
Kay Sievers [Wed, 25 Mar 2009 20:24:21 +0000 (21:24 +0100)]
usb_id: add ID_VENDOR_ID, ID_MODEL_ID, ID_USB_INTERFACE_NUM, ID_USB_DRIVER

$ extras/usb_id/usb_id -x /class/tty/ttyUSB1
ID_VENDOR=HUAWEI_Technology
ID_VENDOR_ENC=HUAWEI\x20Technology
ID_VENDOR_ID=12d1
ID_MODEL=HUAWEI_Mobile
ID_MODEL_ENC=HUAWEI\x20Mobile
ID_MODEL_ID=1003
ID_REVISION=0000
ID_SERIAL=HUAWEI_Technology_HUAWEI_Mobile
ID_TYPE=generic
ID_BUS=usb
ID_USB_INTERFACES=:ffffff:080650:
ID_USB_INTERFACE_NUM=01
ID_USB_DRIVER=option

11 years agorules: replace IDE driver with media match
Kay Sievers [Wed, 25 Mar 2009 12:40:55 +0000 (13:40 +0100)]
rules: replace IDE driver with media match

11 years agorules: serial - support ttyACM devices
Kay Sievers [Tue, 24 Mar 2009 00:55:47 +0000 (01:55 +0100)]
rules: serial - support ttyACM devices

11 years agorules: remove ram* from persisten disk links blacklist
Kay Sievers [Mon, 23 Mar 2009 20:09:02 +0000 (21:09 +0100)]
rules: remove ram* from persisten disk links blacklist

11 years agorules: exclude mtd* from persistent disk links
David Brownell [Mon, 23 Mar 2009 20:03:00 +0000 (21:03 +0100)]
rules: exclude mtd* from persistent disk links

11 years agorules: rename ide-floppy to ide-gd
Borislav Petkov [Sun, 22 Mar 2009 12:22:49 +0000 (13:22 +0100)]
rules: rename ide-floppy to ide-gd

The driver's name changed in the 2.6.28 timeframe.

11 years agoudevadm: monitor - clarify printed header
Kay Sievers [Wed, 18 Mar 2009 17:47:23 +0000 (18:47 +0100)]
udevadm: monitor - clarify printed header

11 years agoudevadm: test - handling trailing '/' in devpath
Kay Sievers [Wed, 18 Mar 2009 17:42:16 +0000 (18:42 +0100)]
udevadm: test - handling trailing '/' in devpath

On Wed, Mar 18, 2009 at 16:00, Matthias Schwarzott <zzam@gentoo.org> wrote:
found out how the error occurs:
It is a difference between
A. udevadm test /sys/class/mem/null/
and
B. udevadm test /sys/class/mem/null

Case A was the case that showed the error behaviour. It seems udevadm is
confused by the trailing slash. This behaviour seems to be there since ages.

11 years agoversion bump
Kay Sievers [Wed, 18 Mar 2009 16:54:00 +0000 (17:54 +0100)]
version bump

11 years agorules: fix extra quote in 50-udev-default.rules
Kay Sievers [Wed, 18 Mar 2009 01:32:12 +0000 (02:32 +0100)]
rules: fix extra quote in 50-udev-default.rules

Patch from Gerardo Exequiel Pozzi.

11 years agoudevadm: trigger - add "--action" to --help
Robby Workman [Tue, 17 Mar 2009 09:31:32 +0000 (10:31 +0100)]
udevadm: trigger - add "--action" to --help

11 years agovol_id: fix language in manpage
Adam Buchbinder [Mon, 16 Mar 2009 20:22:54 +0000 (16:22 -0400)]
vol_id: fix language in manpage

The documentation for the --offset option wasn't in complete
sentences.

Signed-off-by: Adam Buchbinder <adam.buchbinder@gmail.com>
11 years agocreate_floppy_devices: expand manpage
Adam Buchbinder [Mon, 16 Mar 2009 20:22:53 +0000 (16:22 -0400)]
create_floppy_devices: expand manpage

Include a table of what the CMOS types are, and note that nothing
will be created unless the -t option is specified. Also clean up
the formatting and bump the date.

Signed-off-by: Adam Buchbinder <adam.buchbinder@gmail.com>
11 years agocdrom_id: update manpage
Adam Buchbinder [Mon, 16 Mar 2009 20:22:52 +0000 (16:22 -0400)]
cdrom_id: update manpage

Include usage information on the -d/--debug flag. Also bump the date.

Signed-off-by: Adam Buchbinder <adam.buchbinder@gmail.com>
11 years agousb_id: add manpage
Adam Buchbinder [Mon, 16 Mar 2009 20:22:51 +0000 (16:22 -0400)]
usb_id: add manpage

Signed-off-by: Adam Buchbinder <adam.buchbinder@gmail.com>
11 years agorules: md-raid.rules fix
Michal Soltys [Mon, 16 Mar 2009 00:29:19 +0000 (01:29 +0100)]
rules: md-raid.rules fix

md/array_state in case of partition doesn't exist, so all uevents
for partitions didn't execute any SYMLINK rules

Signed-off-by: Michal Soltys <soltys@ziu.info>
11 years agorelease 140
Kay Sievers [Fri, 13 Mar 2009 12:17:31 +0000 (13:17 +0100)]
release 140

11 years agoudevadm: settle - synchronise with the udev daemon
Scott James Remnant [Tue, 10 Mar 2009 13:00:16 +0000 (13:00 +0000)]
udevadm: settle - synchronise with the udev daemon

There's still a slight race condition when using udevadm settle, if the
udev daemon has a pending inotify event but hasn't yet generated the
"change" uevent for it, the kernel and udev sequence numbers will match
and settle will exit.

Now udevadm settle will send a control message to udevd, which will
respond by sending SIGUSR1 back to the waiting udevadm settle once it
has completed the main loop iteration in which it received the control
message.

If there were no pending inotify events, this will simply wake up the
udev daemon and allow settle to continue.  If there are pending inotify
events, they are handled first in the main loop so when settle is
continued they will have been turned into uevents and the kernel
sequence number will have been incremented.

Since the inotify event is pending for udevd when the close() system
call returns (it's queued as part of the kernel handling for that system
call), and since the kernel sequence number is incremented by writing to
the uevent file (as udevd does), this solves the race.

When the settle continues, if there were pending inotify events that
udevd had not read, they are now pending uevents which settle can wait
for.

Signed-off-by: Scott James Remnant <scott@ubuntu.com>
11 years agoudevadm: settle - use timeout signal, instead of loop counter
Kay Sievers [Wed, 11 Mar 2009 14:06:14 +0000 (15:06 +0100)]
udevadm: settle - use timeout signal, instead of loop counter

11 years agolibudev: ctrl - return error after sending ctrl message
Kay Sievers [Wed, 11 Mar 2009 12:32:39 +0000 (13:32 +0100)]
libudev: ctrl - return error after sending ctrl message

11 years agoupdate TODO
Kay Sievers [Sat, 7 Mar 2009 01:35:18 +0000 (02:35 +0100)]
update TODO

11 years agoudevadm: add text for new options to command and man page
Kay Sievers [Sat, 7 Mar 2009 01:32:40 +0000 (02:32 +0100)]
udevadm: add text for new options to command and man page

11 years agolibvolume_id: bump age
Harald Hoyer [Fri, 6 Mar 2009 11:42:00 +0000 (12:42 +0100)]
libvolume_id: bump age

Because you added the symbol volume_id_set_uuid_sub() in commit
14e18278, you have to increase the library minor version.

11 years agoudevd: switch watch symlinks to devpath
Kay Sievers [Wed, 4 Mar 2009 23:35:58 +0000 (00:35 +0100)]
udevd: switch watch symlinks to devpath

Some broken tools get confused following links to /sys, switch
to link targets carrying the devpath instead of the syspath, like
the queue links.

11 years agoudevadm: settle - add --seq-start= --seq-end=
Kay Sievers [Wed, 4 Mar 2009 22:48:53 +0000 (23:48 +0100)]
udevadm: settle - add --seq-start= --seq-end=

It might be useful in some cases not to wait for "all" events.

  $ read START < /sys/kernel/uevent_seqnum
  $ (do something which may create uevents)
  $ udevadm settle --seq-start=$START

11 years agolibudev: queue - use lstat() to check existence of symlink
Kay Sievers [Wed, 4 Mar 2009 22:32:16 +0000 (23:32 +0100)]
libudev: queue - use lstat() to check existence of symlink

11 years agorules: Fedora update
Kay Sievers [Wed, 4 Mar 2009 11:44:08 +0000 (12:44 +0100)]
rules: Fedora update

11 years agoupdate TODO
Kay Sievers [Wed, 4 Mar 2009 11:39:52 +0000 (12:39 +0100)]
update TODO

11 years agovolume_id: ntfs - fix uuid setting
Kay Sievers [Tue, 3 Mar 2009 19:11:09 +0000 (20:11 +0100)]
volume_id: ntfs - fix uuid setting

In my scenario, the ntfs prober did *not* detect the presence of a
ntfs filesystem (i.e. vol_id --probe-all returned *only* ext3).
However, if you examine the source of the ntfs prober, it overwrites
the uuid field of the volume_id object long before it actually
decides there's a valid filesystem there - this resulted in vol_id
returning the rather bizarre combination of type=ext3, but a uuid
populated by the ntfs prober.

https://bugs.edge.launchpad.net/ubuntu/+source/udev/+bug/337015

11 years agoupdate TODO
Kay Sievers [Tue, 3 Mar 2009 17:26:11 +0000 (18:26 +0100)]
update TODO

11 years agofix compile error in debug mode
Michael Prokop [Tue, 3 Mar 2009 15:50:58 +0000 (16:50 +0100)]
fix compile error in debug mode

When building with './configure --enable-debug && make' it fails with:
  udev-rules.c: In function ‘dump_token’:
  udev-rules.c:366: error: ‘struct <anonymous>’ has no member named ‘i’

Signed-off-by: Michael Prokop <mika@grml.org>
11 years agoversion bump
Kay Sievers [Mon, 2 Mar 2009 23:55:42 +0000 (00:55 +0100)]
version bump

11 years agorelease 139
Kay Sievers [Thu, 26 Feb 2009 22:28:20 +0000 (23:28 +0100)]
release 139

11 years agoudevadm: info - export all devices with --export-db
Kay Sievers [Thu, 26 Feb 2009 21:59:40 +0000 (22:59 +0100)]
udevadm: info - export all devices with --export-db

11 years agousb_id: fix comment
Kay Sievers [Thu, 26 Feb 2009 21:10:40 +0000 (22:10 +0100)]
usb_id: fix comment

11 years agousb_id: fix NULL string usage
Kay Sievers [Thu, 26 Feb 2009 19:57:48 +0000 (20:57 +0100)]
usb_id: fix NULL string usage

11 years agousb_id: return values if called directly for an usb_device
Kay Sievers [Thu, 26 Feb 2009 16:58:22 +0000 (17:58 +0100)]
usb_id: return values if called directly for an usb_device

11 years agousb_id: add ID_USB_INTERFACES=:0e0100:0e0200:010100:010200:
Kay Sievers [Thu, 26 Feb 2009 16:05:58 +0000 (17:05 +0100)]
usb_id: add ID_USB_INTERFACES=:0e0100:0e0200:010100:010200:

11 years agouse AC_USE_SYSTEM_EXTENSIONS instead of AC_GNU_SOURCE
Kay Sievers [Thu, 26 Feb 2009 13:17:04 +0000 (14:17 +0100)]
use AC_USE_SYSTEM_EXTENSIONS instead of AC_GNU_SOURCE

11 years agoadd m4/ subdir
Kay Sievers [Thu, 26 Feb 2009 13:11:43 +0000 (14:11 +0100)]
add m4/ subdir

11 years agoadd "nowatch" to disable a default installed watch with a later rule
Kay Sievers [Wed, 25 Feb 2009 10:18:15 +0000 (11:18 +0100)]
add "nowatch" to disable a default installed watch with a later rule

11 years agouse the event udev_device to disable the watch on "remove"
Kay Sievers [Tue, 24 Feb 2009 14:11:17 +0000 (15:11 +0100)]
use the event udev_device to disable the watch on "remove"

11 years agoremove static local variable
Kay Sievers [Tue, 24 Feb 2009 11:58:48 +0000 (12:58 +0100)]
remove static local variable

11 years agoversion bump
Kay Sievers [Mon, 23 Feb 2009 20:27:03 +0000 (21:27 +0100)]
version bump

11 years agoLook at more inotify events in the buffer than just the first.
Scott James Remnant [Mon, 23 Feb 2009 18:33:49 +0000 (18:33 +0000)]
Look at more inotify events in the buffer than just the first.

11 years agolookup the old watch handle; reload only if has a path
Scott James Remnant [Mon, 23 Feb 2009 17:59:01 +0000 (17:59 +0000)]
lookup the old watch handle; reload only if has a path

11 years agoOutput watch handle in udevadm info.
Scott James Remnant [Mon, 23 Feb 2009 17:48:18 +0000 (17:48 +0000)]
Output watch handle in udevadm info.

11 years agoPut a log message in a more sensible place.
Scott James Remnant [Mon, 23 Feb 2009 17:46:00 +0000 (17:46 +0000)]
Put a log message in a more sensible place.

11 years agoUse the udevdb to speed up watch clearing.
Scott James Remnant [Mon, 23 Feb 2009 17:43:53 +0000 (17:43 +0000)]
Use the udevdb to speed up watch clearing.

Also return a udev_device when looking up by handle as well, so
everything works the same way.

11 years agoStore watch handle in db.
Scott James Remnant [Mon, 23 Feb 2009 17:34:07 +0000 (17:34 +0000)]
Store watch handle in db.

11 years agoAllow watch handle to be stored in the udevdb.
Scott James Remnant [Mon, 23 Feb 2009 17:31:26 +0000 (17:31 +0000)]
Allow watch handle to be stored in the udevdb.

11 years agoCleanup a little.
Scott James Remnant [Mon, 23 Feb 2009 17:21:24 +0000 (17:21 +0000)]
Cleanup a little.

Move some of the checks into udev_watch_begin() and udev_watch_clear()
so we don't have to repeat them all the time.

11 years agoClear existing inotify watch before processing.
Scott James Remnant [Mon, 23 Feb 2009 17:19:05 +0000 (17:19 +0000)]
Clear existing inotify watch before processing.

Likewise when re-processing an event, we want to drop the existing
inotify watch first to avoid side-effects.

11 years agoDon't add inotify watch until RUN rules processed.
Scott James Remnant [Mon, 23 Feb 2009 17:17:20 +0000 (17:17 +0000)]
Don't add inotify watch until RUN rules processed.

The programs we RUN can open device nodes, and thus we can end up
chasing our own tail.

11 years agorelease 138
Kay Sievers [Wed, 18 Feb 2009 23:27:24 +0000 (00:27 +0100)]
release 138

11 years ago*_id: add model/vendor enc strings
David Zeuthen [Tue, 17 Feb 2009 19:15:17 +0000 (14:15 -0500)]
*_id: add model/vendor enc strings

So ID_MODEL and ID_VENDOR are pretty useful keys. However since we fix
them up (removing leading/trailing whitespace, converts spaces to
underscores) for use in device naming etc. we also force these fixups on
the desktop shell. And this looks pretty ugly.

The attached patch introduces the ID_MODEL_ENC and ID_VENDOR_ENC keys
that contains the encoded version of the raw strings obtained. It's
pretty similar in spirit to ID_FS_LABEL and its cousin ID_FS_LABEL_ENC.

With this patch a desktop shell can fix up these strings as it sees fit.
Note that some fixup is still needed though, for example

# /lib/udev/ata_id --export /dev/sda |grep ID_MODEL
ID_MODEL=INTEL_SSDSA2MH080G1GC
ID_MODEL_ENC=INTEL\x20SSDSA2MH080G1GC\x20\x20\x20\x20\x20\x20\x20\x20
\x20\x20\x20\x20\x20\x20\x20\x20\x20\x20\x20

Note the trailing and leading whitespace. Anyway with the attached patch
the desktop shell should be able to display "INTEL SSDSA2MH080G1GC"
rather than "INTEL_SSDSA2MH080G1GC" to the user.

11 years agorules: include loop block devices in persistent links
Kay Sievers [Mon, 16 Feb 2009 21:22:57 +0000 (22:22 +0100)]
rules: include loop block devices in persistent links

11 years agorules: add "watch" option to dm and md rules
Kay Sievers [Fri, 13 Feb 2009 15:02:36 +0000 (16:02 +0100)]
rules: add "watch" option to dm and md rules

11 years agouse global "reload_config" flag
Kay Sievers [Wed, 11 Feb 2009 20:56:35 +0000 (21:56 +0100)]
use global "reload_config" flag

11 years agorules: move OPTIONS to separate rule
Kay Sievers [Wed, 11 Feb 2009 20:54:43 +0000 (21:54 +0100)]
rules: move OPTIONS to separate rule

A failing IMPORT+ match would prevent the OPTIONS+= action
from being applied.

11 years agorules: update persistent storage rules to use inotify watches
Scott James Remnant [Wed, 11 Feb 2009 17:44:03 +0000 (17:44 +0000)]
rules: update persistent storage rules to use inotify watches

11 years agoudevd: optionally watch device nodes with inotify
Scott James Remnant [Wed, 11 Feb 2009 17:38:56 +0000 (17:38 +0000)]
udevd: optionally watch device nodes with inotify

This allows you to re-process the rules if the content of the device
has been changed, most useful for block subsystem to cause vol_id to
be run again.

11 years agofix NAME= and OPTION+="string_escape=..." logic
Kay Sievers [Sun, 8 Feb 2009 15:02:15 +0000 (16:02 +0100)]
fix NAME= and OPTION+="string_escape=..." logic

Scott discovered this silly bug.

11 years agoautogen.sh: add more warnings
Kay Sievers [Thu, 5 Feb 2009 18:01:27 +0000 (19:01 +0100)]
autogen.sh: add more warnings

11 years agorules: fix md "change"/"remove" handling
Kay Sievers [Thu, 5 Feb 2009 13:03:17 +0000 (14:03 +0100)]
rules: fix md "change"/"remove" handling

On Thu, Feb 5, 2009 at 08:43, Harald Hoyer <harald@redhat.com> wrote:
> Radek Vykydal <rvykydal@redhat.com> encountered a problem with md devices.
> If the raid is about to be removed a "change" and "remove" event is sent.

11 years agotest: add test for empty and non-existent ATTR
Kay Sievers [Thu, 5 Feb 2009 11:40:15 +0000 (12:40 +0100)]
test: add test for empty and non-existent ATTR

11 years agorules: dm - add escape for uuid links with whitespace
Kay Sievers [Fri, 30 Jan 2009 12:18:59 +0000 (13:18 +0100)]
rules: dm - add escape for uuid links with whitespace

11 years agofix NAME="" logic
Kay Sievers [Fri, 30 Jan 2009 02:53:09 +0000 (03:53 +0100)]
fix NAME="" logic

Reported-By: Soh Kam Yung <sohkamyung@gmail.com>
11 years agorules: Gentoo update
Matthias Schwarzott [Tue, 27 Jan 2009 10:56:45 +0000 (11:56 +0100)]
rules: Gentoo update

11 years agoversion bump
Kay Sievers [Sun, 25 Jan 2009 20:25:41 +0000 (21:25 +0100)]
version bump

11 years agoman: fix grammar
Kay Sievers [Sun, 25 Jan 2009 20:21:43 +0000 (21:21 +0100)]
man: fix grammar

Reported-by: Robert Peterson <rpeterso@redhat.com>
11 years agodoc: writing udev rules - refer to 'udevadm info' instead of 'udevinfo'
Miklos Vajna [Fri, 23 Jan 2009 02:55:24 +0000 (03:55 +0100)]
doc: writing udev rules - refer to 'udevadm info' instead of 'udevinfo'

Signed-off-by: Miklos Vajna <vmiklos@frugalware.org>
11 years agovol_id: add missing id->type to swap0
Karel Zak [Fri, 23 Jan 2009 23:37:23 +0000 (00:37 +0100)]
vol_id: add missing id->type to swap0

Signed-off-by: Karel Zak <kzak@redhat.com>
11 years agovol_id: fix ddf version string
Karel Zak [Fri, 23 Jan 2009 23:37:22 +0000 (00:37 +0100)]
vol_id: fix ddf version string

The version string in DDF suberblock does not include the null terminator
and the snprintf() function writes at most "size" bytes *including* the
null byte.

old version:
ID_FS_VERSION=02.00.0
new version:
ID_FS_VERSION=02.00.00

Signed-off-by: Karel Zak <kzak@redhat.com>
11 years agorelease 137
Kay Sievers [Fri, 23 Jan 2009 02:05:05 +0000 (03:05 +0100)]
release 137

11 years agoudevadm: settle - fix typo
Kay Sievers [Fri, 23 Jan 2009 01:19:12 +0000 (02:19 +0100)]
udevadm: settle - fix typo

11 years agousb_id: handle ATAPI devices like SCSI devices
Kay Sievers [Thu, 22 Jan 2009 18:23:33 +0000 (19:23 +0100)]
usb_id: handle ATAPI devices like SCSI devices

Some weird mass storage devices identify the disks as ATAPI/cdrom
devices, and all disks in such enclusure get the same ids which
overwrite each other, in such cases, get the typ of device from the
scsi device, which has more knowledge what kind of device it really
is.

11 years agoudevd: add back SA_RESTART
Kay Sievers [Wed, 21 Jan 2009 17:55:59 +0000 (18:55 +0100)]
udevd: add back SA_RESTART

<Keybuk> poll, select, ppoll, pselect, epoll, etc. are *explicitly* NOT restarted in case of signal
<kay> ok

11 years agosplit up long line
Kay Sievers [Wed, 21 Jan 2009 14:11:38 +0000 (15:11 +0100)]
split up long line

11 years agoMerge commit '9032f119f07ad3b5116b3d4858816d851d4127de'
Kay Sievers [Wed, 21 Jan 2009 13:43:22 +0000 (14:43 +0100)]
Merge commit '9032f119f07ad3b5116b3d4858816d851d4127de'

11 years agoMerge commit '5f03ed8a56d308af72db8a48ab66ed68667af2c6'
Kay Sievers [Wed, 21 Jan 2009 13:42:30 +0000 (14:42 +0100)]
Merge commit '5f03ed8a56d308af72db8a48ab66ed68667af2c6'

Conflicts:
udev/udevd.c

11 years agoincrease netif renaming timeout from 30 to 90 seconds
Kay Sievers [Wed, 21 Jan 2009 10:17:58 +0000 (11:17 +0100)]
increase netif renaming timeout from 30 to 90 seconds

11 years agoreap children faster
Olaf Kirch [Wed, 21 Jan 2009 10:17:11 +0000 (11:17 +0100)]
reap children faster

udevd's event_queue_manager loop is pretty sensitive to the
delays of exiting child processes. I found that it helps boot
times if we try to reap children as quickly as possible.

This patch changes event_queue_manager to call sigchilds_waiting
if it finds a signal has been received.

Signed-off-by: Olaf Kirch <okir@suse.de>
11 years agoopen-code pollfd setup
Kay Sievers [Wed, 21 Jan 2009 10:09:27 +0000 (11:09 +0100)]
open-code pollfd setup

11 years agoExpose delayed name resolution
Scott James Remnant [Tue, 20 Jan 2009 21:36:09 +0000 (21:36 +0000)]
Expose delayed name resolution

udevd had the ability to delay name resolution until the event, but this
was never exposed.  Add a --resolve-names=late option to do so.

11 years agoAllow user and group lookup to be disabled.
Scott James Remnant [Tue, 20 Jan 2009 21:33:02 +0000 (21:33 +0000)]
Allow user and group lookup to be disabled.

In certain cut-down situations such as an installer or inside the
initramfs, we simply don't have any kind of name service.  While we
could use rules without OWNER or GROUP, it's better to have the same
rules as a full system and have udevd ignore those parts of the rules.

Adds a --resolve-names=never switch to udevd that has this effect.

11 years agoudevd: use ppoll instead of signal pipes
Olaf Kirch [Tue, 20 Jan 2009 11:49:20 +0000 (12:49 +0100)]
udevd: use ppoll instead of signal pipes

udevd uses a rather old-fashioned way of handling signals
while waiting for input through select (ie by using an unnamed
pipe, to which the signal handler writes one byte for every signal
received). This is rather awkward and may potentially even block
if we receive more signals than the kernel's pipe buffer.

This patch replaces all of that with ppoll, which was designed
for this purpose.

It also removes the SA_RESTART flag from all installed signal
handlers, because otherwise the ppoll call would just be restarted
after handling eg a SIGCHLD.

Signed-off-by: Olaf Kirch <okir@suse.de>
11 years agoremove stray initializer
Alan Jenkins [Fri, 16 Jan 2009 20:35:34 +0000 (20:35 +0000)]
remove stray initializer

Picked up by "gcc -Wextra".

udevadm.c:83: warning: initialized field overwritten
udevadm.c:83: warning: (near initialization for ‘cmds[2].help’)

This is just a cleanup.  It doesn't change the code generated by gcc.

11 years agoman: typo fixes
Alan Jenkins [Fri, 16 Jan 2009 20:26:16 +0000 (20:26 +0000)]
man: typo fixes

11 years agocleanup old names before creating the new names
Kay Sievers [Fri, 16 Jan 2009 20:15:53 +0000 (21:15 +0100)]
cleanup old names before creating the new names

Scott found that the node /dev/pktcdvd can not be replaced by
/dev/pktcdvd/control by changing the rules, and re-trigger the
event. We used to create the new names before we cleaned up the
old ones, which can not work if we need to create subdir with
the same name.

11 years agoremove name from index if the node name has changed
Kay Sievers [Fri, 16 Jan 2009 20:14:13 +0000 (21:14 +0100)]
remove name from index if the node name has changed

Scott discovered that a changed rule, which renames a device node
leaves the old node name around as a symlink. So drop the reference
to the old name.

11 years agoupdate NEWS
Kay Sievers [Fri, 16 Jan 2009 19:54:11 +0000 (20:54 +0100)]
update NEWS

11 years agoudevadm: test - remove --force option
Kay Sievers [Thu, 15 Jan 2009 19:20:16 +0000 (20:20 +0100)]
udevadm: test - remove --force option

Since a while we change the database with a "test" run, but do not update
the node and symlinks. We need to "force" all the time, to keep things
in sync.

11 years agovolume_id: support via raid version 2
Sven Jost [Thu, 15 Jan 2009 18:09:12 +0000 (19:09 +0100)]
volume_id: support via raid version 2