chiark / gitweb /
Kay Sievers [Sat, 6 Jun 2009 18:36:46 +0000 (20:36 +0200)]
path_id: add comments
Kay Sievers [Sat, 6 Jun 2009 18:30:56 +0000 (20:30 +0200)]
path_id: update SCSI handling
Kay Sievers [Sat, 6 Jun 2009 14:07:06 +0000 (16:07 +0200)]
path_id: implement in C using libudev
Kay Sievers [Sat, 6 Jun 2009 02:54:41 +0000 (04:54 +0200)]
cleanup $attr{} substitution
Kay Sievers [Sat, 6 Jun 2009 02:52:52 +0000 (04:52 +0200)]
allow format substitution in path of ATTR{<path>}=="<value>"
Kay Sievers [Fri, 5 Jun 2009 21:27:11 +0000 (23:27 +0200)]
udevd: use enum instead of char in struct declaration
Marcel Holtmann [Fri, 5 Jun 2009 13:09:05 +0000 (15:09 +0200)]
rules: make RFKILL control device world readable
Kay Sievers [Thu, 4 Jun 2009 01:21:33 +0000 (03:21 +0200)]
rules: do not call path_id for virtual devices
Kay Sievers [Thu, 4 Jun 2009 00:49:42 +0000 (02:49 +0200)]
udevd: close netlink socket in worker and set cloexec
Kay Sievers [Wed, 3 Jun 2009 23:44:04 +0000 (01:44 +0200)]
udevd: convert to event worker processes
Event processes now get re-used after they handled an event. This reduces
pressure on the CPU significantly because cloned event processes no longer
cause page faults in the main daemon. After the events have settled, the
no longer needed worker processes get killed.
Kay Sievers [Mon, 1 Jun 2009 19:10:29 +0000 (21:10 +0200)]
str[sp]cpyl: add __attribute__ ((sentinel))
Kay Sievers [Fri, 29 May 2009 16:54:03 +0000 (18:54 +0200)]
udevd: remove check for dev_t, DEVPATH_OLD takes care of that
Kay Sievers [Fri, 29 May 2009 12:05:39 +0000 (14:05 +0200)]
udevadm: settle - add --exit-if-exists=<file>
Kay Sievers [Fri, 29 May 2009 11:21:38 +0000 (13:21 +0200)]
cdrom_id: add error message if open() fails
Kay Sievers [Thu, 28 May 2009 17:21:35 +0000 (19:21 +0200)]
udevadm: control - add comment to man page about --reload-rules
Alan Jenkins [Thu, 28 May 2009 16:59:06 +0000 (17:59 +0100)]
udevd: queue-export - fix crash
The math in skip_to() was the wrong way round and allocated a
variable size array on the stack with a massively negative size.
Signed-off-by: Alan Jenkins <alan-jenkins@tuffmail.co.uk>
Alan Jenkins [Thu, 28 May 2009 10:50:10 +0000 (12:50 +0200)]
udev-selinux.c: remove libudev header
Signed-off-by: Alan Jenkins <alan-jenkins@tuffmail.co.uk>
Kay Sievers [Mon, 25 May 2009 22:50:45 +0000 (00:50 +0200)]
cdrom_id: suppress ID_CDROM_MEDIA_STATE=blank for plain non-writable CDROM media
Some broken fake cdrom drives return ID_CDROM_MEDIA_STATE=blank.
$ /lib/udev/cdrom_id -d /dev/sr1
main: probing: '/dev/sr1'
cd_inquiry: INQUIRY: [Nokia ][S60 ][1.0 ]
cd_profiles: GET CONFIGURATION: number of profiles 76
cd_profiles: current profile 0x08
cd_media_toc: READ TOC: len: 12
cd_media_toc: last track 1 starts at block 0
cd_media_info: disk type 00
ID_CDROM=1
ID_CDROM_MRW=1
ID_CDROM_MRW_W=1
ID_CDROM_MEDIA=1
ID_CDROM_MEDIA_CD=1
ID_CDROM_MEDIA_STATE=blank
Others work fine, but ID_CDROM_MEDIA_STATE is not needed for non-writable CDROM media:
$ /lib/udev/cdrom_id -d /dev/sr1
main: probing: '/dev/sr1'
cd_inquiry: INQUIRY: [SanDisk ][U3 Cruzer Micro ][8.02]
cd_profiles: GET CONFIGURATION: number of profiles 72
cd_profiles: current profile 0x08
cd_media_toc: READ TOC: len: 20
cd_media_toc: track=1 info=0x4(data) start_block=0
cd_media_toc: last track 1 starts at block 0
cd_media_info: disk type 00
ID_CDROM=1
ID_CDROM_MRW=1
ID_CDROM_MRW_W=1
ID_CDROM_MEDIA=1
ID_CDROM_MEDIA_CD=1
ID_CDROM_MEDIA_STATE=complete
ID_CDROM_MEDIA_SESSION_COUNT=1
ID_CDROM_MEDIA_TRACK_COUNT=1
ID_CDROM_MEDIA_TRACK_COUNT_DATA=1
Benjamin Gilbert [Fri, 22 May 2009 18:54:46 +0000 (20:54 +0200)]
test: check string substitutions in OWNER and GROUP
String substitutions in OWNER and GROUP keys were broken in udev 137-142.
Explicitly test for this, since such breakage will not manifest in typical
rulesets.
Kay Sievers [Fri, 22 May 2009 14:44:53 +0000 (16:44 +0200)]
rules: add SUBSYSTEM match to scsi rules
On Fri, May 22, 2009 at 16:15, Alan Jenkins <alan-jenkins@tuffmail.co.uk> wrote:
> I've been looking at what is responsible for all the path lookup activity in
> coldplug. On my debian stable system, it looks like every device gets its
> parent looked up in sysfs. I think this is due to SUBSYSTEMS matches.
>
> I see the udev default rules are different, but it looks like they still
> test for SUBSYSTEMS on every single device. Should we add SUBSYSTEM="scsi_generic"
> to these three rules?
Kay Sievers [Thu, 21 May 2009 21:52:09 +0000 (23:52 +0200)]
update NEWS
Kay Sievers [Thu, 21 May 2009 21:49:53 +0000 (23:49 +0200)]
udevadm: settle - make --timeout=0 working
Alan Jenkins [Thu, 21 May 2009 20:22:37 +0000 (22:22 +0200)]
udevd: implement a more efficient queue file format
Directory lookups show up in profiling. The queue files are responsible
for a large proportion of file-related system calls in udev coldplug.
Instead of creating a file for each event, append their details to a
log file. The file is periodically rebuilt (garbage-collected) to
prevent it from growing indefinitely.
This single queue file replaces both the queue directory and the
uevent_seqnum file. On desktop systems the file tends not to grow
beyond one page. So it should also save a small amount of memory in
tmpfs.
Tests on a running EeePC indicate average savings of 5% *udevd* cpu time
as measured by oprofile. __link_path_walk is reduced from 1.5% to
1.3%. It is not completely clear where the rest of the gains come from.
In tests running ~400 events, the queue file is rebuilt about 5 times.
Signed-off-by: Alan Jenkins <alan-jenkins@tuffmail.co.uk>
Alan Jenkins [Thu, 21 May 2009 16:31:43 +0000 (18:31 +0200)]
udevd: remove tiny bit of dead code
Kay Sievers [Wed, 20 May 2009 23:44:45 +0000 (01:44 +0200)]
fix location of database files
Kay Sievers [Wed, 20 May 2009 23:20:18 +0000 (01:20 +0200)]
keep the ifdef'd udevd testing/profiling hack
Kay Sievers [Wed, 20 May 2009 21:45:32 +0000 (23:45 +0200)]
require key names in uppercase
Drop pretty expensive case-insensitive matching, and key names in
mixed or lowercase are not supported anyway.
Kay Sievers [Wed, 20 May 2009 21:12:22 +0000 (23:12 +0200)]
remove unused GL_FORMAT from rules parser
Kay Sievers [Wed, 20 May 2009 17:43:26 +0000 (19:43 +0200)]
update TODO
Kay Sievers [Wed, 20 May 2009 16:58:52 +0000 (18:58 +0200)]
use openat(), unlinkat(), fstatat()
Kay Sievers [Wed, 20 May 2009 16:10:26 +0000 (18:10 +0200)]
edd_id: use openat()
Kay Sievers [Wed, 20 May 2009 15:57:52 +0000 (17:57 +0200)]
use more efficient string copying
Alan Jenkins [Wed, 20 May 2009 14:12:22 +0000 (16:12 +0200)]
udevadm: settle - fix timeout
The timeout wasn't working when settle was run as root:
# udevadm control --stop-exec-queue
# udevadm trigger
# udevadm settle --timeout=1
... (hangs)
Signed-off-by: Alan Jenkins <alan-jenkins@tuffmail.co.uk>
Kay Sievers [Tue, 19 May 2009 10:56:17 +0000 (12:56 +0200)]
remove format char string truncation syntax
Scott James Remnant [Thu, 14 May 2009 11:42:44 +0000 (12:42 +0100)]
OWNER/GROUP: fix if logic
The introduction of the --resolve-names=early/never code introduced a
bug to the OWNER/GROUP lookup. Previously if the name had contained $,
lookup would have been performed later; after the patch, the key ended
up being ignored!
Kay Sievers [Thu, 14 May 2009 10:57:19 +0000 (12:57 +0200)]
rule-generator: cd - skip by-path links if we create by-id links
Kay Sievers [Wed, 13 May 2009 22:32:25 +0000 (00:32 +0200)]
Revert "rules: v4l do not mix vbi and video nodes"
This reverts commit
6205f1186e4980544ea425d31770358d1b2579e4.
Kay Sievers [Wed, 13 May 2009 16:01:32 +0000 (18:01 +0200)]
fix possible endless loop for GOTO to non-existent LABEL
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=526365
Kay Sievers [Wed, 13 May 2009 15:55:08 +0000 (17:55 +0200)]
rules: v4l do not mix vbi and video nodes
Kay Sievers [Wed, 13 May 2009 01:13:08 +0000 (03:13 +0200)]
version bump
Kay Sievers [Wed, 13 May 2009 00:35:37 +0000 (02:35 +0200)]
release 142
Kay Sievers [Tue, 12 May 2009 11:28:01 +0000 (13:28 +0200)]
delete libvolume_id
Kay Sievers [Mon, 11 May 2009 02:12:23 +0000 (04:12 +0200)]
delete vol_id and require util-linux-ng's blkid
Kay Sievers [Mon, 11 May 2009 01:53:55 +0000 (03:53 +0200)]
update TODO
Kay Sievers [Thu, 7 May 2009 11:24:40 +0000 (13:24 +0200)]
add "static" back to the inline functions
<Keybuk> kay: udev git head ftbfs
<Keybuk> udev-watch.o: In function `udev_selinux_init':
<Keybuk> /../udev/udev.h:130: multiple definition of `udev_selinux_init'
Kay Sievers [Tue, 5 May 2009 10:43:42 +0000 (12:43 +0200)]
rules: fix ieee1394 rules
Kay Sievers [Tue, 5 May 2009 00:39:16 +0000 (02:39 +0200)]
remove asmlinkage
Kay Sievers [Mon, 4 May 2009 20:08:05 +0000 (22:08 +0200)]
remove all PHYSDEVPATH handling and warning about
Kay Sievers [Mon, 4 May 2009 12:10:05 +0000 (14:10 +0200)]
oops, removed ppp entry from rules got committed
Kay Sievers [Mon, 4 May 2009 02:52:31 +0000 (04:52 +0200)]
handle devtmpfs nodes
UDev follows the kernel given name, and re-uses the kernel created
device node. If the kernel and spcecified udev rules disagree, the
udev specified node node is created and the kernel-created on is
deleted.
Lubomir Rintel [Sun, 3 May 2009 13:37:30 +0000 (15:37 +0200)]
rule-generator: net - whitelist NICs that violate MAC local scheme
According to list of assigned ethernet codes [1] referred to by
IANA [2] certain global addresses do not follow the assignement
scheme and use numbers reserved for local use. Several such adapters
are quite widely used, generate rules for them.
[1] http://www.cavebear.com/archive/cavebear/Ethernet/vendor.html
[2] http://www.iana.org/assignments/ethernet-numbers
Kay Sievers [Sun, 3 May 2009 12:44:23 +0000 (14:44 +0200)]
rules: remove initramfs comment
Kay Sievers [Sat, 2 May 2009 03:23:02 +0000 (05:23 +0200)]
udevd: revert initial device node creation
Kay Sievers [Fri, 1 May 2009 15:52:21 +0000 (17:52 +0200)]
rules: add "event*" match
Kay Sievers [Fri, 1 May 2009 03:09:51 +0000 (05:09 +0200)]
build: use AC_MSG_RESULT
Kay Sievers [Wed, 29 Apr 2009 13:57:41 +0000 (15:57 +0200)]
update TODO
Kay Sievers [Tue, 28 Apr 2009 13:59:45 +0000 (15:59 +0200)]
accept DEVNAME from the kernel as a hint for the node name
Kay Sievers [Tue, 28 Apr 2009 09:24:25 +0000 (11:24 +0200)]
rules: add rfcomm* to group dialout
https://bugs.launchpad.net/bugs/368109
Kay Sievers [Mon, 27 Apr 2009 01:50:48 +0000 (03:50 +0200)]
rules: input - fix comments
Kay Sievers [Mon, 27 Apr 2009 01:23:41 +0000 (03:23 +0200)]
rules: input - add links for USB/platform non-kbd/mouse devices
Kay Sievers [Sun, 26 Apr 2009 18:53:30 +0000 (20:53 +0200)]
rule_generator: net - fix usb comment generation
On Sat, Apr 4, 2009 at 22:17, Omair Eshkenazi <stimpson@phys.huji.ac.il> wrote:
> I noticed that in (70-)persistent-net.rules, the comments for USB devices
> are missing the device/vendor id's. Example:
> # USB device 0x:0x (rt73usb)
Kay Sievers [Sun, 26 Apr 2009 11:59:39 +0000 (13:59 +0200)]
drop "extern" keyword from non-static function
Kay Sievers [Fri, 24 Apr 2009 00:27:44 +0000 (02:27 +0200)]
update NEWS
Andre Przywara [Thu, 23 Apr 2009 22:29:53 +0000 (00:29 +0200)]
rules: create /dev/cpu/<n>/cpuid world readable
I don't see any security implications, to be actually useful,
/dev/cpu/<n>/cpuid should be world readable. The cpuid instruction
can be called from userspace anyway, so there is nothing to hide.
The device does not support any write operation, so 0444 should
suffice.
Signed-off-by: Andre Przywara <andre.przywara@amd.com>
Kay Sievers [Thu, 23 Apr 2009 16:05:10 +0000 (18:05 +0200)]
libudev: monitor - export filter_update()
Kay Sievers [Thu, 23 Apr 2009 13:16:26 +0000 (15:16 +0200)]
libudev: queue - clarify comments
Kay Sievers [Thu, 23 Apr 2009 13:04:16 +0000 (15:04 +0200)]
cdrom_id: skip media tests if CDROM_DRIVE_STATUS != CDS_DISC_OK
Kay Sievers [Thu, 23 Apr 2009 11:23:45 +0000 (13:23 +0200)]
libudev: queue - fix get_seqnum_is_finished()
On Thu, Apr 23, 2009 at 11:39, Alan Jenkins <alan-jenkins@tuffmail.co.uk> wrote:
> I'm puzzled by this function:
>
> /* if we have not seen this seqnum, check if it is/was already queued */
> if (seqnum < udev_queue->last_seen_udev_seqnum) {
> udev_queue_get_udev_seqnum(udev_queue);
> if (seqnum < udev_queue->last_seen_udev_seqnum)
>
> Shouldn't the test be (seqnum > udev_queue->last_seen_udev_seqnum) ?
Kay Sievers [Thu, 23 Apr 2009 02:07:51 +0000 (04:07 +0200)]
libudev: monitor - add udev_monitor_filter_remove
Kay Sievers [Thu, 23 Apr 2009 02:07:15 +0000 (04:07 +0200)]
udevadm: monitor - add <subsytem>:<devtype> support
Kay Sievers [Thu, 23 Apr 2009 01:33:58 +0000 (03:33 +0200)]
libudev: monitor - do not filter messages with wrong magic
Kay Sievers [Thu, 23 Apr 2009 01:24:38 +0000 (03:24 +0200)]
libudev: monitor - switch to filter_add_match_subsystem_devtype()
Kay Sievers [Wed, 22 Apr 2009 16:38:16 +0000 (18:38 +0200)]
libudev: monitor - use simpler hash
John Wright [Wed, 22 Apr 2009 12:53:35 +0000 (14:53 +0200)]
edd_id: add cciss devices
Since cciss devices can be BIOS boot devices just as well as sd* and
hd*, the edd_id program should be run on them so that the later
/dev/disk/by-id/edd-* rules will work.
Kay Sievers [Wed, 22 Apr 2009 03:00:15 +0000 (05:00 +0200)]
udevadm monitor - add --subsystem-match=
Kay Sievers [Wed, 22 Apr 2009 02:32:25 +0000 (04:32 +0200)]
update TODO
Kay Sievers [Wed, 22 Apr 2009 02:11:18 +0000 (04:11 +0200)]
udevadm: monitor - print error if we can not bind to socket
Kay Sievers [Wed, 22 Apr 2009 01:50:11 +0000 (03:50 +0200)]
libudev: monitor - add client socket filter for subsystem value
Messages send back by the udev daemon to the netlink socket are
multiplexed by the kernel and delivered to multiple clients. The
clients can upload a socket filter to let the kernel drop messages
not belonging to a certain subsystem. This prevent needless wakeups
and message processing for users who are only interested in a
subset of available events.
Recent kernels allow untrusted users to listen to the netlink
messages.
The messages send by the udev daemon are versioned, to prevent any
custom software reading them without libudev. The message wire format
may change with any udev version update.
Kay Sievers [Tue, 21 Apr 2009 01:27:14 +0000 (03:27 +0200)]
cdrom_id: add and use ID_CDROM_MEDIA to decide if we run vol_id
Some broken mobile phones offer a faked cdrom drive with a media
without any tracks.
Kay Sievers [Sun, 19 Apr 2009 01:25:04 +0000 (03:25 +0200)]
udevd: at startup create /dev/null, /dev/console, /dev/kmsg
Kay Sievers [Sat, 18 Apr 2009 11:24:20 +0000 (13:24 +0200)]
update NEWS
Kay Sievers [Sat, 18 Apr 2009 11:23:55 +0000 (13:23 +0200)]
TODO: add packet filter
Kay Sievers [Sat, 18 Apr 2009 11:23:10 +0000 (13:23 +0200)]
test-libudev: update monitor source
Kay Sievers [Thu, 16 Apr 2009 22:29:56 +0000 (00:29 +0200)]
cdrom_id: add Xen cdrom support
Ian Campbell [Thu, 16 Apr 2009 20:46:18 +0000 (22:46 +0200)]
path_id: support identification of Xen virtual block devices
Kay Sievers [Thu, 16 Apr 2009 20:36:26 +0000 (22:36 +0200)]
update NEWS
Kay Sievers [Thu, 16 Apr 2009 20:35:59 +0000 (22:35 +0200)]
rules: remove DVB shell script
Kay Sievers [Thu, 16 Apr 2009 20:34:26 +0000 (22:34 +0200)]
udevd: log error for too old kernels or CONFIG_SYSFS_DEPRECATED
Kay Sievers [Wed, 15 Apr 2009 23:49:16 +0000 (01:49 +0200)]
libudev: monitor - clarify socket handling documentation
Kay Sievers [Wed, 15 Apr 2009 19:47:04 +0000 (21:47 +0200)]
libudev: path_encode - always return 0 if encoded string does not fit into size
Kay Sievers [Tue, 14 Apr 2009 19:02:39 +0000 (21:02 +0200)]
version bump
Kay Sievers [Wed, 8 Apr 2009 23:35:53 +0000 (16:35 -0700)]
release 141
Scott James Remnant [Wed, 8 Apr 2009 23:16:14 +0000 (16:16 -0700)]
libudev: monitor - ignore messages from unusual sources
For added protection, ignore any unicast message received on the
netlink socket or any multicast message on the kernel group not
received from the kernel.
Signed-off-by: Scott James Remnant <scott@ubuntu.com>
Kay Sievers [Wed, 8 Apr 2009 14:46:54 +0000 (07:46 -0700)]
libudev: monitor - unify socket message handling
Kay Sievers [Wed, 8 Apr 2009 13:06:01 +0000 (06:06 -0700)]
path_encode: fix max length calculation
Sebastian Krahmer wrote:
> it should reserve 4 times not 3 times len :)
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>
Kay Sievers [Mon, 30 Mar 2009 18:09:08 +0000 (20:09 +0200)]
IMPORT: 2048 -> 4096 bytes buffer
Kay Sievers [Sun, 29 Mar 2009 18:08:52 +0000 (20:08 +0200)]
"UDEV_MONITOR_KERNEL/UDEV" -> "kernel/udev"
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.
Kay Sievers [Sat, 28 Mar 2009 19:57:01 +0000 (20:57 +0100)]
usb_id: remove unused variable