chiark / gitweb /
Kay Sievers [Mon, 23 Mar 2009 20:09:02 +0000 (21:09 +0100)]
rules: remove ram* from persisten disk links blacklist
David Brownell [Mon, 23 Mar 2009 20:03:00 +0000 (21:03 +0100)]
rules: exclude mtd* from persistent disk links
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.
Kay Sievers [Wed, 18 Mar 2009 17:47:23 +0000 (18:47 +0100)]
udevadm: monitor - clarify printed header
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.
Kay Sievers [Wed, 18 Mar 2009 16:54:00 +0000 (17:54 +0100)]
version bump
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.
Robby Workman [Tue, 17 Mar 2009 09:31:32 +0000 (10:31 +0100)]
udevadm: trigger - add "--action" to --help
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>
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>
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>
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>
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>
Kay Sievers [Fri, 13 Mar 2009 12:17:31 +0000 (13:17 +0100)]
release 140
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>
Kay Sievers [Wed, 11 Mar 2009 14:06:14 +0000 (15:06 +0100)]
udevadm: settle - use timeout signal, instead of loop counter
Kay Sievers [Wed, 11 Mar 2009 12:32:39 +0000 (13:32 +0100)]
libudev: ctrl - return error after sending ctrl message
Kay Sievers [Sat, 7 Mar 2009 01:35:18 +0000 (02:35 +0100)]
update TODO
Kay Sievers [Sat, 7 Mar 2009 01:32:40 +0000 (02:32 +0100)]
udevadm: add text for new options to command and man page
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.
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.
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
Kay Sievers [Wed, 4 Mar 2009 22:32:16 +0000 (23:32 +0100)]
libudev: queue - use lstat() to check existence of symlink
Kay Sievers [Wed, 4 Mar 2009 11:44:08 +0000 (12:44 +0100)]
rules: Fedora update
Kay Sievers [Wed, 4 Mar 2009 11:39:52 +0000 (12:39 +0100)]
update TODO
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
Kay Sievers [Tue, 3 Mar 2009 17:26:11 +0000 (18:26 +0100)]
update TODO
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>
Kay Sievers [Mon, 2 Mar 2009 23:55:42 +0000 (00:55 +0100)]
version bump
Kay Sievers [Thu, 26 Feb 2009 22:28:20 +0000 (23:28 +0100)]
release 139
Kay Sievers [Thu, 26 Feb 2009 21:59:40 +0000 (22:59 +0100)]
udevadm: info - export all devices with --export-db
Kay Sievers [Thu, 26 Feb 2009 21:10:40 +0000 (22:10 +0100)]
usb_id: fix comment
Kay Sievers [Thu, 26 Feb 2009 19:57:48 +0000 (20:57 +0100)]
usb_id: fix NULL string usage
Kay Sievers [Thu, 26 Feb 2009 16:58:22 +0000 (17:58 +0100)]
usb_id: return values if called directly for an usb_device
Kay Sievers [Thu, 26 Feb 2009 16:05:58 +0000 (17:05 +0100)]
usb_id: add ID_USB_INTERFACES=:0e0100:0e0200:010100:010200:
Kay Sievers [Thu, 26 Feb 2009 13:17:04 +0000 (14:17 +0100)]
use AC_USE_SYSTEM_EXTENSIONS instead of AC_GNU_SOURCE
Kay Sievers [Thu, 26 Feb 2009 13:11:43 +0000 (14:11 +0100)]
add m4/ subdir
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
Kay Sievers [Tue, 24 Feb 2009 14:11:17 +0000 (15:11 +0100)]
use the event udev_device to disable the watch on "remove"
Kay Sievers [Tue, 24 Feb 2009 11:58:48 +0000 (12:58 +0100)]
remove static local variable
Kay Sievers [Mon, 23 Feb 2009 20:27:03 +0000 (21:27 +0100)]
version bump
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.
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
Scott James Remnant [Mon, 23 Feb 2009 17:48:18 +0000 (17:48 +0000)]
Output watch handle in udevadm info.
Scott James Remnant [Mon, 23 Feb 2009 17:46:00 +0000 (17:46 +0000)]
Put a log message in a more sensible place.
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.
Scott James Remnant [Mon, 23 Feb 2009 17:34:07 +0000 (17:34 +0000)]
Store watch handle in db.
Scott James Remnant [Mon, 23 Feb 2009 17:31:26 +0000 (17:31 +0000)]
Allow watch handle to be stored in the udevdb.
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.
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.
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.
Kay Sievers [Wed, 18 Feb 2009 23:27:24 +0000 (00:27 +0100)]
release 138
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.
Kay Sievers [Mon, 16 Feb 2009 21:22:57 +0000 (22:22 +0100)]
rules: include loop block devices in persistent links
Kay Sievers [Fri, 13 Feb 2009 15:02:36 +0000 (16:02 +0100)]
rules: add "watch" option to dm and md rules
Kay Sievers [Wed, 11 Feb 2009 20:56:35 +0000 (21:56 +0100)]
use global "reload_config" flag
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.
Scott James Remnant [Wed, 11 Feb 2009 17:44:03 +0000 (17:44 +0000)]
rules: update persistent storage rules to use inotify watches
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.
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.
Kay Sievers [Thu, 5 Feb 2009 18:01:27 +0000 (19:01 +0100)]
autogen.sh: add more warnings
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.
Kay Sievers [Thu, 5 Feb 2009 11:40:15 +0000 (12:40 +0100)]
test: add test for empty and non-existent ATTR
Kay Sievers [Fri, 30 Jan 2009 12:18:59 +0000 (13:18 +0100)]
rules: dm - add escape for uuid links with whitespace
Kay Sievers [Fri, 30 Jan 2009 02:53:09 +0000 (03:53 +0100)]
fix NAME="" logic
Reported-By: Soh Kam Yung <sohkamyung@gmail.com>
Matthias Schwarzott [Tue, 27 Jan 2009 10:56:45 +0000 (11:56 +0100)]
rules: Gentoo update
Kay Sievers [Sun, 25 Jan 2009 20:25:41 +0000 (21:25 +0100)]
version bump
Kay Sievers [Sun, 25 Jan 2009 20:21:43 +0000 (21:21 +0100)]
man: fix grammar
Reported-by: Robert Peterson <rpeterso@redhat.com>
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>
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>
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>
Kay Sievers [Fri, 23 Jan 2009 02:05:05 +0000 (03:05 +0100)]
release 137
Kay Sievers [Fri, 23 Jan 2009 01:19:12 +0000 (02:19 +0100)]
udevadm: settle - fix typo
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.
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
Kay Sievers [Wed, 21 Jan 2009 14:11:38 +0000 (15:11 +0100)]
split up long line
Kay Sievers [Wed, 21 Jan 2009 13:43:22 +0000 (14:43 +0100)]
Kay Sievers [Wed, 21 Jan 2009 13:42:30 +0000 (14:42 +0100)]
Kay Sievers [Wed, 21 Jan 2009 10:17:58 +0000 (11:17 +0100)]
increase netif renaming timeout from 30 to 90 seconds
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>
Kay Sievers [Wed, 21 Jan 2009 10:09:27 +0000 (11:09 +0100)]
open-code pollfd setup
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.
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.
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>
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.
Alan Jenkins [Fri, 16 Jan 2009 20:26:16 +0000 (20:26 +0000)]
man: typo fixes
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.
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.
Kay Sievers [Fri, 16 Jan 2009 19:54:11 +0000 (20:54 +0100)]
update NEWS
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.
Sven Jost [Thu, 15 Jan 2009 18:09:12 +0000 (19:09 +0100)]
volume_id: support via raid version 2
Kay Sievers [Thu, 15 Jan 2009 16:52:17 +0000 (17:52 +0100)]
do not complain about a missing /etc/udev/rules.d/
Scott mentioned that initramfs does not neccessarily have that
directory, so do not log an error.
Kay Sievers [Thu, 15 Jan 2009 16:06:14 +0000 (17:06 +0100)]
rules: add drm devices to group "video"
https://bugs.launchpad.net/bugs/317430
Kay Sievers [Wed, 14 Jan 2009 20:54:31 +0000 (21:54 +0100)]
libudev: use 4096 bytes buffer for attribute reading
Kay Sievers [Sat, 10 Jan 2009 16:44:09 +0000 (17:44 +0100)]
rules: fix typo in ide cd rule
Thanks to Scott, who found that.
Lennart Poettering [Sat, 10 Jan 2009 00:37:03 +0000 (01:37 +0100)]
fix naming for tape nst devices in /dev/tape/by-path/
$env{ID_PATH} includes the "-nst" suffix anyway, so we shouldn't append
it a second time as part of the rule creating the device file symlink.
Signed-off-by: Lennart Poettering <lennart@poettering.net>
Kay Sievers [Sat, 10 Jan 2009 00:38:54 +0000 (01:38 +0100)]
version bump
Kay Sievers [Tue, 6 Jan 2009 21:01:50 +0000 (22:01 +0100)]
release 136
Adam Buchbinder [Tue, 6 Jan 2009 18:43:01 +0000 (19:43 +0100)]
extras: fix mis-spelling of "environment"
Kay Sievers [Tue, 6 Jan 2009 15:52:40 +0000 (16:52 +0100)]
libudev: test - add udev_device_get_property_value()