chiark / gitweb /
elogind.git
15 years agorules: isdn - remove /dev/isdn/capi20 symlink
Kay Sievers [Mon, 22 Dec 2008 16:11:01 +0000 (17:11 +0100)]
rules: isdn - remove /dev/isdn/capi20 symlink

15 years agorules: put alsa in group "audio"
Kay Sievers [Mon, 22 Dec 2008 15:19:38 +0000 (16:19 +0100)]
rules: put alsa in group "audio"

15 years agorules: remove /dev/dsp /dev/audio
Kay Sievers [Mon, 22 Dec 2008 15:16:46 +0000 (16:16 +0100)]
rules: remove /dev/dsp /dev/audio

15 years agorules: second round merging with Ubuntu rules
Kay Sievers [Mon, 22 Dec 2008 15:15:37 +0000 (16:15 +0100)]
rules: second round merging with Ubuntu rules

15 years agorule_generator: fix enumeration for write_cd_rules
Harald Hoyer [Mon, 22 Dec 2008 11:26:06 +0000 (12:26 +0100)]
rule_generator: fix enumeration for write_cd_rules

quote +, as it would be taken as a part of the regexp otherwise
https://bugzilla.redhat.com/show_bug.cgi?id=477535

15 years agorules: do not put raw1394 in "video" group
Kay Sievers [Mon, 22 Dec 2008 13:58:11 +0000 (14:58 +0100)]
rules: do not put raw1394 in "video" group

A note on /dev/raw1394's security implications:
1. You cannot access local memory through raw1394, except
for ROMs and CSRs that are exposed to other nodes any way.

2. It is extremely hard to manipulate data on attached
SBP-2 devices (FireWire storage devices).

3. You can disturb operation of the FireWire bus, e.g.
creating a DoS situation for audio/video applications, for
SBP-2 devices, or eth1394 network interfaces.

4. If another PC is attached to the FireWire bus, it may be
possible to read or overwrite the entire RAM of that remote PC.
This depends on the PC's configuration. Most FireWire controllers
support this feature (yes, it's not a bug, or at least wasn't
intended to be one...) but not all OSs enable the feature.

Actually, a cheap setup to achieve #1 by #4 is to have two
FireWire controllers in the PC and connect them.

https://bugs.launchpad.net/ubuntu/+source/kino/+bug/6290/comments/21

15 years agorules: remove legacy symlinks
Kay Sievers [Mon, 22 Dec 2008 12:34:54 +0000 (13:34 +0100)]
rules: remove legacy symlinks

15 years agorules: rtc - create rtc compat link only for cmos type rtc
Kay Sievers [Sun, 21 Dec 2008 12:48:55 +0000 (13:48 +0100)]
rules: rtc - create rtc compat link only for cmos type rtc

15 years agorules: remove /dev/raw/raxctl symlink, it's a devfs leftover
Kay Sievers [Sun, 21 Dec 2008 12:42:08 +0000 (13:42 +0100)]
rules: remove /dev/raw/raxctl symlink, it's a devfs leftover

15 years agorules: more changes toward Ubuntu rules merge
Kay Sievers [Sun, 21 Dec 2008 12:17:37 +0000 (13:17 +0100)]
rules: more changes toward Ubuntu rules merge

dv1394*: no kernel name symlink
lp*: no par* symlink
hwrng: no kernel name symlink

15 years agorules: more changes toward Ubuntu rules merge
Kay Sievers [Sun, 21 Dec 2008 12:01:40 +0000 (13:01 +0100)]
rules: more changes toward Ubuntu rules merge

specialix_rioctl: no kernel name symlink
specialix_sxctl: no kernel name symlink
bus/usb: 0644 -> 0664
ppdev: lp
dri: 0666 -> 0660
js: no kernel name symlink

15 years ago"'/sbin/modprobe abnormal' exit" - also print program options
Kay Sievers [Fri, 19 Dec 2008 23:03:53 +0000 (00:03 +0100)]
"'/sbin/modprobe abnormal' exit" - also print program options

15 years agorules: first simple step merging with Ubuntu rules
Kay Sievers [Fri, 19 Dec 2008 15:49:56 +0000 (16:49 +0100)]
rules: first simple step merging with Ubuntu rules

15 years agorules: remove ide-scsi
Kay Sievers [Fri, 19 Dec 2008 03:37:09 +0000 (04:37 +0100)]
rules: remove ide-scsi

15 years agobuild: just use autoreconf -i
Kay Sievers [Fri, 19 Dec 2008 00:30:41 +0000 (01:30 +0100)]
build: just use autoreconf -i

Thanks to Scott for mentioning!

15 years agopath_id: rework SAS persistent names
Jeremy Higdon [Thu, 18 Dec 2008 12:42:28 +0000 (13:42 +0100)]
path_id: rework SAS persistent names

15 years agobuild: add -lsepol
Kay Sievers [Tue, 16 Dec 2008 15:17:53 +0000 (16:17 +0100)]
build: add -lsepol

Harald needs it for static binaries.

15 years agolibudev: include <sys/types.h>
Kay Sievers [Tue, 16 Dec 2008 13:48:33 +0000 (14:48 +0100)]
libudev: include <sys/types.h>

15 years agovolume_id: support sub-uuid's and plug in btrfs device uuid
Kay Sievers [Tue, 9 Dec 2008 17:14:23 +0000 (18:14 +0100)]
volume_id: support sub-uuid's and plug in btrfs device uuid

15 years agorules: add "do not edit header"
Kay Sievers [Tue, 9 Dec 2008 14:35:56 +0000 (15:35 +0100)]
rules: add "do not edit header"

15 years agovolume_id: btrfs - update format
Kay Sievers [Tue, 9 Dec 2008 14:33:23 +0000 (15:33 +0100)]
volume_id: btrfs - update format

15 years agorules: move aoe rules to default rules file
Kay Sievers [Tue, 9 Dec 2008 03:18:21 +0000 (04:18 +0100)]
rules: move aoe rules to default rules file

15 years agorules: add rules for AoE devices
Ryan Thomas [Tue, 9 Dec 2008 00:10:03 +0000 (19:10 -0500)]
rules: add rules for AoE devices

In the interest of standardizing udev rules, please consider the
following patch that adds udev rules for the ATA over Ethernet character
and block devices.  The aoe module has been a long-time member of the
kernel and needs inclusion in the standard udev rules.

15 years agoudevadm: settle - allow --timeout=0 and --quiet
Kay Sievers [Mon, 8 Dec 2008 15:48:54 +0000 (16:48 +0100)]
udevadm: settle - allow --timeout=0 and --quiet

15 years agomake: do not delete autotools generated file with distclean
Kay Sievers [Sat, 6 Dec 2008 03:03:08 +0000 (04:03 +0100)]
make: do not delete autotools generated file with distclean

                                        [...] running the command
`make maintainer-clean' should not delete `configure' even if
`configure' can be remade using a rule in the Makefile.  More
generally, `make maintainer-clean' should not delete anything that
needs to exist in order to run `configure' and then begin to build
the program.  This is the only exception; `maintainer-clean' should
delete everything else that can be rebuilt.

15 years agomake: install serial rules file
Kay Sievers [Thu, 4 Dec 2008 01:04:50 +0000 (02:04 +0100)]
make: install serial rules file

15 years agorules: add persistent /dev/serial/{by-id,by-path} rules
Kay Sievers [Thu, 4 Dec 2008 00:43:33 +0000 (01:43 +0100)]
rules: add persistent /dev/serial/{by-id,by-path} rules

 $ tree /dev/serial/
 /dev/serial/
 |-- by-id
 |   |-- usb-067b_2303-if00-port0 -> ../../ttyUSB0
 |   |-- usb-FTDI_FT232R_USB_UART_A7005uBP-if00-port0 -> ../../ttyUSB3
 |   |-- usb-HUAWEI_Technology_HUAWEI_Mobile-if00-port0 -> ../../ttyUSB1
 |   `-- usb-HUAWEI_Technology_HUAWEI_Mobile-if01-port0 -> ../../ttyUSB2
 `-- by-path
     |-- pci-0000:00:1d.0-usb-0:1:1.0-port0 -> ../../ttyUSB3
     |-- pci-0000:00:1d.7-usb-0:2.2.2:1.0-port0 -> ../../ttyUSB0
     |-- pci-0000:00:1d.7-usb-0:2.3:1.0-port0 -> ../../ttyUSB1
     `-- pci-0000:00:1d.7-usb-0:2.3:1.1-port0 -> ../../ttyUSB2

 $ tree /dev/serial/
 /dev/serial/
 |-- by-id
 |   |-- usb-Inside_Out_Networks_Edgeport_4_04-01-006467-if00-port0 -> ../../ttyUSB0
 |   |-- usb-Inside_Out_Networks_Edgeport_4_04-01-006467-if00-port1 -> ../../ttyUSB1
 |   |-- usb-Inside_Out_Networks_Edgeport_4_04-01-006467-if00-port2 -> ../../ttyUSB2
 |   |-- usb-Inside_Out_Networks_Edgeport_4_04-01-006467-if00-port3 -> ../../ttyUSB3
 |   |-- usb-Keyspan__a_division_of_InnoSys_Inc._USB_4-port_Serial_Adapter-if00-port0 -> ../../ttyUSB8
 |   |-- usb-Keyspan__a_division_of_InnoSys_Inc._USB_4-port_Serial_Adapter-if00-port1 -> ../../ttyUSB9
 |   |-- usb-Keyspan__a_division_of_InnoSys_Inc._USB_4-port_Serial_Adapter-if00-port2 -> ../../ttyUSB10
 |   |-- usb-Keyspan__a_division_of_InnoSys_Inc._USB_4-port_Serial_Adapter-if00-port3 -> ../../ttyUSB11
 |   `-- usb-Prolific_Technology_Inc._USB-Serial_Controller-if00-port0 -> ../../ttyUSB7
 `-- by-path
     |-- pci-0000:00:1d.2-usb-0:1.3:1.0-port0 -> ../../ttyUSB7
     |-- pci-0000:00:1d.7-usb-0:6.1.4.1:1.0-port0 -> ../../ttyUSB4
     |-- pci-0000:00:1d.7-usb-0:6.1.4.1:1.0-port1 -> ../../ttyUSB5
     |-- pci-0000:00:1d.7-usb-0:6.1.4.1:1.0-port2 -> ../../ttyUSB6
     |-- pci-0000:00:1d.7-usb-0:6.1.4.4:1.0-port0 -> ../../ttyUSB0
     |-- pci-0000:00:1d.7-usb-0:6.1.4.4:1.0-port1 -> ../../ttyUSB1
     |-- pci-0000:00:1d.7-usb-0:6.1.4.4:1.0-port2 -> ../../ttyUSB2
     |-- pci-0000:00:1d.7-usb-0:6.1.4.4:1.0-port3 -> ../../ttyUSB3
     |-- pci-0000:00:1d.7-usb-0:6.3:1.0-port0 -> ../../ttyUSB8
     |-- pci-0000:00:1d.7-usb-0:6.3:1.0-port1 -> ../../ttyUSB9
     |-- pci-0000:00:1d.7-usb-0:6.3:1.0-port2 -> ../../ttyUSB10
     `-- pci-0000:00:1d.7-usb-0:6.3:1.0-port3 -> ../../ttyUSB11

15 years agovolume_id: HPFS code clean up
Karel Zak [Wed, 3 Dec 2008 10:49:50 +0000 (11:49 +0100)]
volume_id: HPFS code clean up

* fix typo (hs vs. hbb)

 * set id->{version,usage,type} when all tests pass

 * be paranoid and don't use hs->version when "hs" source buffer was
   possibly modified by next volume_id_get_buffer() call.

Signed-off-by: Karel Zak <kzak@redhat.com>
15 years agorules: fix isdn rules
Kay Sievers [Wed, 3 Dec 2008 00:32:00 +0000 (01:32 +0100)]
rules: fix isdn rules

On Tue, Dec 2, 2008 at 21:07, Matthias Schwarzott <zzam@gentoo.org> wrote:
> It seems that the rules related to capi devices are not correct:
>
> KERNEL=="capi",         NAME="capi20", SYMLINK+="isdn/capi20"
> KERNEL=="capi*",        NAME="capi/%n"
>
> Changing the second rule to match only on KERNEL=="capi[0-9]*" is reported to
> make it work.
> So I can only guess that the problem is the second rule overwriting the NAME
> set by the first one.

15 years agoupdate configure and NEWS
Kay Sievers [Tue, 2 Dec 2008 18:26:33 +0000 (19:26 +0100)]
update configure and NEWS

15 years agoscsi_id: do not fail if no serial is found like for optical drives
Kay Sievers [Tue, 2 Dec 2008 18:23:38 +0000 (19:23 +0100)]
scsi_id: do not fail if no serial is found like for optical drives

15 years agorules: ATA_COMPAT do not try to match on sr*, it will never have vendor ATA
Kay Sievers [Tue, 2 Dec 2008 16:33:11 +0000 (17:33 +0100)]
rules: ATA_COMPAT do not try to match on sr*, it will never have vendor ATA

15 years agorelease 135
Kay Sievers [Tue, 2 Dec 2008 01:35:48 +0000 (02:35 +0100)]
release 135

15 years agoupdate configure and NEWS
Kay Sievers [Tue, 2 Dec 2008 00:10:24 +0000 (01:10 +0100)]
update configure and NEWS

15 years agofix network interface name swapping
Kay Sievers [Tue, 2 Dec 2008 00:04:23 +0000 (01:04 +0100)]
fix network interface name swapping

15 years agorules: infiniband.rules -> 40-infiniband.rules
Kay Sievers [Fri, 28 Nov 2008 02:34:03 +0000 (03:34 +0100)]
rules: infiniband.rules -> 40-infiniband.rules

15 years agorules: add usb device nodes
Piter PUNK [Fri, 28 Nov 2008 01:01:51 +0000 (02:01 +0100)]
rules: add usb device nodes

15 years agorules: add infiniband rules
Kay Sievers [Fri, 28 Nov 2008 00:46:38 +0000 (01:46 +0100)]
rules: add infiniband rules

15 years agorules: fix cciss disk/by-id/ links
Kay Sievers [Fri, 28 Nov 2008 00:18:48 +0000 (01:18 +0100)]
rules: fix cciss disk/by-id/ links

15 years agorules: /dev/null -> X0R
Piter PUNK [Fri, 28 Nov 2008 00:18:15 +0000 (01:18 +0100)]
rules: /dev/null -> X0R

/dev/X0R        null            symbolic        Required by iBCS-2
Note: /dev/X0R is <letter X>-<digit 0>-<letter R>

15 years agousb_id: add "break" to currently unused case labels
Kay Sievers [Wed, 26 Nov 2008 12:38:49 +0000 (13:38 +0100)]
usb_id: add "break" to currently unused case labels

15 years agousb_id: fix switch statement for video type
Marcel Holtmann [Wed, 26 Nov 2008 12:31:37 +0000 (13:31 +0100)]
usb_id: fix switch statement for video type

Introducing the video type, creating a fall-through case where other
devices might now be declared as type video.

15 years agorelease 134
Kay Sievers [Wed, 26 Nov 2008 03:36:25 +0000 (04:36 +0100)]
release 134

15 years agorules: exclude "btibm" devices from vol_id calls
Kay Sievers [Tue, 25 Nov 2008 16:25:14 +0000 (17:25 +0100)]
rules: exclude "btibm" devices from vol_id calls

15 years agorules: Gentoo update
Matthias Schwarzott [Tue, 25 Nov 2008 16:13:07 +0000 (17:13 +0100)]
rules: Gentoo update

15 years agosilence PHYSDEV* warning for WAIT_FOR* rules
Kay Sievers [Mon, 24 Nov 2008 20:24:04 +0000 (21:24 +0100)]
silence PHYSDEV* warning for WAIT_FOR* rules

15 years agofix handling of swapping node name with symlink name
Kay Sievers [Mon, 24 Nov 2008 01:21:12 +0000 (02:21 +0100)]
fix handling of swapping node name with symlink name

15 years agofloppy: use ARRAY_SIZE()
Kay Sievers [Sun, 23 Nov 2008 16:29:37 +0000 (17:29 +0100)]
floppy: use ARRAY_SIZE()

15 years agofloppy: fix array bounds check and minor calculation
Peter Breitenlohner [Sun, 23 Nov 2008 16:22:32 +0000 (17:22 +0100)]
floppy: fix array bounds check and minor calculation

Fix off-by-4 bug in floppy minors for fd[4-7].
Avoid segmentation fault for bad CMOS type.
Print mode in human readable, i.e., octal form.

15 years agoupdate NEWS TODO
Kay Sievers [Fri, 21 Nov 2008 23:34:42 +0000 (00:34 +0100)]
update NEWS TODO

15 years agopath_id: fix fibre channel handling
Kay Sievers [Fri, 21 Nov 2008 19:48:24 +0000 (20:48 +0100)]
path_id: fix fibre channel handling

15 years agoman: fix typos
Peter Breitenlohner [Fri, 21 Nov 2008 09:50:12 +0000 (10:50 +0100)]
man: fix typos

15 years agovolume_id: clear probing result before probing and do not probe a second time, if...
Kay Sievers [Fri, 21 Nov 2008 06:26:44 +0000 (07:26 +0100)]
volume_id: clear probing result before probing and do not probe a second time, if not needed

On Thu, Nov 20, 2008 at 14:17, Karel Zak <kzak@redhat.com> wrote:
>  I see the patch (volume_id_probe_filesystem()) and a few things come
>  to mind:
>
>  - shouldn't be the relevant parts (label, uuid, version, ...) of
>    the "struct volume_id" zeroized when you found a signature and
>    before you call the next probing function?
>
>  - it seems as overkill to use two for()s and probe two times for all
>    filesystems. What about to store the first result and re-use it?
>
>  - .. or at least never use the second for() when the fist for() found
>    nothing ;-)

15 years agovolume_id: update btrfs
Kay Sievers [Fri, 21 Nov 2008 06:26:09 +0000 (07:26 +0100)]
volume_id: update btrfs

15 years agovol_id: if regular files are probed, use stat() for the size value
Kay Sievers [Fri, 21 Nov 2008 04:02:02 +0000 (05:02 +0100)]
vol_id: if regular files are probed, use stat() for the size value

15 years agovolume_id: remove unused usage types
Kay Sievers [Fri, 21 Nov 2008 03:22:04 +0000 (04:22 +0100)]
volume_id: remove unused usage types

15 years agoinclude errno.h in sysdeps.h
Gabor Z. Papp [Fri, 21 Nov 2008 03:19:11 +0000 (04:19 +0100)]
include errno.h in sysdeps.h

15 years agoscsi_id: make sure, we do not have slashes in values
Kay Sievers [Thu, 20 Nov 2008 06:57:17 +0000 (07:57 +0100)]
scsi_id: make sure, we do not have slashes in values

15 years agoata_id: make sure, we do not have slashes in values
Kay Sievers [Thu, 20 Nov 2008 06:52:33 +0000 (07:52 +0100)]
ata_id: make sure, we do not have slashes in values

15 years agousb_id: replace chars in returned strings
Kay Sievers [Thu, 20 Nov 2008 01:44:08 +0000 (02:44 +0100)]
usb_id: replace chars in returned strings

Slashes are not funny in device names:
  |-- input
  |   |-- by-id
  |   |   `-- usb-Logitech_USB-PS
  |   |       |-- 2_Optical_Mouse-event-mouse -> ../../event2
  |   |       `-- 2_Optical_Mouse-mouse -> ../../mouse1

15 years agolibudev: accept NULL whitelist in util_replace_chars()
Kay Sievers [Thu, 20 Nov 2008 01:43:34 +0000 (02:43 +0100)]
libudev: accept NULL whitelist in util_replace_chars()

15 years agorules: v4l - add by-id/ links for USB devices
Kay Sievers [Thu, 20 Nov 2008 01:26:13 +0000 (02:26 +0100)]
rules: v4l - add by-id/ links for USB devices

/dev/v4l
|-- by-id
|   |-- usb-046d_09a4_C4B15020-video-index0 -> ../../video0
|   `-- usb-05a9_a511-video-index0 -> ../../video1
`-- by-path
    |-- pci-0000:00:1d.0-usb-0:1:1.0-video-index0 -> ../../video1
    `-- pci-0000:00:1d.7-usb-0:2:1.0-video-index0 -> ../../video0

15 years agorules: merge group "video" into default rules
Kay Sievers [Wed, 19 Nov 2008 20:47:07 +0000 (21:47 +0100)]
rules: merge group "video" into default rules

15 years agorules: add persistent rules for memory stick block devices
Harald Hoyer [Wed, 19 Nov 2008 10:24:03 +0000 (11:24 +0100)]
rules: add persistent rules for memory stick block devices

commit 5a9aed145ac0ffb3e29b1c8e0f19b34e277f9117
Author: Harald Hoyer <harald@redhat.com>
Date:   Wed Nov 19 11:22:30 2008 +0100

    added persistent rules for memory stick block devices

15 years agobump version
Kay Sievers [Tue, 18 Nov 2008 18:16:42 +0000 (19:16 +0100)]
bump version

15 years agovolume_id: update btrfs magic
Kay Sievers [Tue, 18 Nov 2008 18:14:37 +0000 (19:14 +0100)]
volume_id: update btrfs magic

15 years agoautogen.sh: fix -print-multi-os-directory usage
Kay Sievers [Tue, 18 Nov 2008 02:52:05 +0000 (03:52 +0100)]
autogen.sh: fix -print-multi-os-directory usage

15 years agorelease 133
Kay Sievers [Tue, 18 Nov 2008 01:50:58 +0000 (02:50 +0100)]
release 133

15 years agoupdate NEWS
Kay Sievers [Tue, 18 Nov 2008 01:49:08 +0000 (02:49 +0100)]
update NEWS

15 years agofix $links substitution for devices without any link
Kay Sievers [Mon, 17 Nov 2008 14:24:47 +0000 (15:24 +0100)]
fix $links substitution for devices without any link

Reported-by: Michel Hermier <michel.hermier@gmail.com>
15 years agoudevd: fix cleanup of /dev/.udev/uevent_seqnum
Kay Sievers [Mon, 17 Nov 2008 13:43:58 +0000 (14:43 +0100)]
udevd: fix cleanup of /dev/.udev/uevent_seqnum

15 years agolimit $attr(<symlink>) magic to well-known links only
Kay Sievers [Mon, 17 Nov 2008 01:01:37 +0000 (02:01 +0100)]
limit $attr(<symlink>) magic to well-known links only

15 years agoudevadm: fix option parsing breakage with klibc
Sergey Vlasov [Fri, 14 Nov 2008 21:34:43 +0000 (00:34 +0300)]
udevadm: fix option parsing breakage with klibc

The klibc implementation of getopt_long() behaves slightly different
from the glibc one - in particular, it treats the change of the option
string argument between invocations as start of parsing a different
command line, and resets its state.  However, the udevadm code
expected getopt_long() invocations in subcommands to continue parsing
the rest of command line after initial options has been parsed at the
top level; with klibc this broke, causing all udevadm subcommands to
stop recognizing their options.

Instead of relying on the glibc behavior, reset the getopt_long()
state properly before invoking the subcommand handler: move argv to
point to the subcommand name, decrease argc appropriately, and set
optind = 0.  This also fixes a minor bug visible with glibc - without
setting optind = 0 all getopt_long() calls in subcommand handlers were
behaving as if "+" was specified as the first character of the option
string (which disables option reordering), because that state was set
by the first getopt_long() call at the top level, and was not reset
when parsing subcommand options.

Signed-off-by: Sergey Vlasov <vsu@altlinux.ru>
15 years agolibudev: device - handle disk "device" link for partitions in deprecated sysfs layout
Kay Sievers [Fri, 14 Nov 2008 17:26:49 +0000 (18:26 +0100)]
libudev: device - handle disk "device" link for partitions in deprecated sysfs layout

Thanks to Mikhail Kolesnik <mike@openbunker.org> for finding this.

15 years agovol_id: clarify error message
Kay Sievers [Thu, 13 Nov 2008 23:57:34 +0000 (00:57 +0100)]
vol_id: clarify error message

15 years agovolume_id: ntfs - mark as no other fs must match
Kay Sievers [Thu, 13 Nov 2008 20:44:28 +0000 (21:44 +0100)]
volume_id: ntfs - mark as no other fs must match

15 years agofix spelling in comment
Kay Sievers [Thu, 13 Nov 2008 19:46:25 +0000 (20:46 +0100)]
fix spelling in comment

15 years agovolume_id: fat - accept empty FAT32 fsinfo signature
Kay Sievers [Thu, 13 Nov 2008 18:40:24 +0000 (19:40 +0100)]
volume_id: fat - accept empty FAT32 fsinfo signature

15 years agovolume_id: always check for all filesystem types and skip conflicting results
Kay Sievers [Thu, 13 Nov 2008 18:34:41 +0000 (19:34 +0100)]
volume_id: always check for all filesystem types and skip conflicting results

We probe for all known filesystems to find conflicting signatures. If
we find multiple matching signatures and one of the detected filesystem
types claims that it can not co-exist with any other filesystem type,
we do not return a probing result.

We can not afford to mount a volume with the wrong filesystem code and
possibly corrupt it. Linux ssytems have the problem of dozens of possible
filesystem types, and volumes with left-over signatures from former
filesystem types. Invalid signature need to be removed from the volume
to make the filesystem detection successful.

We do not want to read that many bytes from probed floppies, skip volumes
smaller than a usual floppy disk.

15 years agovolume_id: bump revision
Kay Sievers [Thu, 13 Nov 2008 17:22:39 +0000 (18:22 +0100)]
volume_id: bump revision

15 years agoremove len == 0 check, the index root is always '\0'
Kay Sievers [Thu, 13 Nov 2008 17:21:28 +0000 (18:21 +0100)]
remove len == 0 check, the index root is always '\0'

15 years agovolume_id: fat - move check for msdos signature (0x55 0xaa)
Karel Zak [Thu, 13 Nov 2008 16:32:14 +0000 (17:32 +0100)]
volume_id: fat - move check for msdos signature (0x55 0xaa)

Make the msdos signature (0x55 0xaa) at 510 and 511 optional when
the standard FAT magic string is present.

15 years agoconfigure: add linux-hotplug mail address
Kay Sievers [Thu, 13 Nov 2008 07:16:55 +0000 (08:16 +0100)]
configure: add linux-hotplug mail address

15 years agoupdate NEWS
Kay Sievers [Thu, 13 Nov 2008 07:03:49 +0000 (08:03 +0100)]
update NEWS

15 years agorules: remove pnp shell script, acpi loads these modules properly
Kay Sievers [Thu, 13 Nov 2008 06:47:16 +0000 (07:47 +0100)]
rules: remove pnp shell script, acpi loads these modules properly

15 years agorules: remove DEVTYPE disk/partition
Kay Sievers [Thu, 13 Nov 2008 06:45:58 +0000 (07:45 +0100)]
rules: remove DEVTYPE disk/partition

15 years agoudevadm: info - unify -V and --version
Kay Sievers [Thu, 13 Nov 2008 06:31:04 +0000 (07:31 +0100)]
udevadm: info - unify -V and --version

15 years agoset errno = ENOSYS in inotify stub
Kay Sievers [Thu, 13 Nov 2008 00:52:33 +0000 (01:52 +0100)]
set errno = ENOSYS in inotify stub

15 years agoscsi_id: we don't use DEVPATH env var anymore, update man page
Alan Jenkins [Wed, 12 Nov 2008 16:23:17 +0000 (16:23 +0000)]
scsi_id: we don't use DEVPATH env var anymore, update man page

Signed-off-by: Alan Jenkins <alan-jenkins@tuffmail.co.uk>
15 years agomerge trie nodes, childs and root into a single array
Kay Sievers [Wed, 12 Nov 2008 23:52:12 +0000 (00:52 +0100)]
merge trie nodes, childs and root into a single array

15 years agoreserve child slot 0
Kay Sievers [Wed, 12 Nov 2008 14:01:20 +0000 (15:01 +0100)]
reserve child slot 0

15 years agostring index - split nodes and childs to allow and unlimited number of childs
Kay Sievers [Wed, 12 Nov 2008 04:50:05 +0000 (05:50 +0100)]
string index - split nodes and childs to allow and unlimited number of childs

15 years agosilence "comparison between signed and unsigned"
Kay Sievers [Tue, 11 Nov 2008 22:47:01 +0000 (23:47 +0100)]
silence "comparison between signed and unsigned"

15 years agoudevd: de-duplicate strings in rules
Alan Jenkins [Tue, 11 Nov 2008 20:20:11 +0000 (20:20 +0000)]
udevd: de-duplicate strings in rules

On my Ubuntu installation this removes 15k of duplicate strings,
using a temporary index of about 25k.

Signed-off-by: Alan Jenkins <alan-jenkins@tuffmail.co.uk>
15 years agorelease 132
Kay Sievers [Fri, 7 Nov 2008 15:08:58 +0000 (16:08 +0100)]
release 132

15 years agoconvert debug string arrays to functions
Kay Sievers [Fri, 7 Nov 2008 14:59:58 +0000 (15:59 +0100)]
convert debug string arrays to functions

On Fri, Nov 7, 2008 at 13:07, Matthias Schwarzott <zzam@gentoo.org> wrote:
> I managed to let udev-131 segfault at startup.
>
> I configured it like this:
> CFLAGS="-Wall -ggdb" ./configure --prefix=/usr --sysconfdir=/etc --exec-prefix=
>
> Running it in gdb shows it segfaults at udev-rules.c:831
>
> (gdb) run
> Starting program: /tmp/udev-131/udev/udevd
>
> Program received signal SIGSEGV, Segmentation fault.
> 0x0804ea06 in get_key (udev=0x9175008, line=0xafcdc8f0, key=0xafcdc5d8,
> op=0xafcdc5d0, value=0xafcdc5d4)
>    at udev-rules.c:831
> 831             dbg(udev, "%s '%s'-'%s'\n", operation_str[*op], *key, *value);

If compiled without optimization, the dbg() macro dereferences variables
which are not available. Convert the string array to a function, which just
returns NULL if compiled without DEBUG.

15 years agofix size_t compiler warning on 32 bit platforms
Kay Sievers [Fri, 7 Nov 2008 01:58:54 +0000 (02:58 +0100)]
fix size_t compiler warning on 32 bit platforms

15 years agorelease 131
Kay Sievers [Fri, 7 Nov 2008 01:36:20 +0000 (02:36 +0100)]
release 131

15 years agoudevadm: settle - exit if udevd exits
Kay Sievers [Thu, 6 Nov 2008 18:41:58 +0000 (19:41 +0100)]
udevadm: settle - exit if udevd exits

15 years agocleanup /dev/.udev/queue on startup and exit
Kay Sievers [Thu, 6 Nov 2008 16:43:12 +0000 (17:43 +0100)]
cleanup /dev/.udev/queue on startup and exit