chiark / gitweb /
elogind.git
18 years ago[PATCH] pattern match for label method
kay.sievers@vrfy.org [Fri, 5 Dec 2003 03:21:31 +0000 (19:21 -0800)]
[PATCH] pattern match for label method

  switch LABEL search to pattern match
  add a test for pattern match in LABEL
  remove useless rule from udev.rules

18 years ago[PATCH] a bug in linefeed removal
kay.sievers@vrfy.org [Fri, 5 Dec 2003 03:21:27 +0000 (19:21 -0800)]
[PATCH] a bug in linefeed removal

While I was adding pattern match to the LABEL method i hit a bug.
We modify a string returned from libsysfs, so with every iteration is is
truncated by one char:

Dec  4 02:27:16 pim udev[23307]: do_label: dev->bus='scsi' sysfs_device->bus='scsi'
Dec  4 02:27:16 pim udev[23307]: do_label: look for device attribute 'vendor'
Dec  4 02:27:16 pim udev[23307]: do_label: xxx 'IBM-ESXS '
Dec  4 02:27:16 pim udev[23307]: do_label: compare attribute 'vendor' value 'IBM-ESX' with '?IBM-ESXS'
Dec  4 02:27:16 pim udev[23307]: do_label: dev->bus='scsi' sysfs_device->bus='scsi'
Dec  4 02:27:16 pim udev[23307]: do_label: look for device attribute 'vendor'
Dec  4 02:27:16 pim udev[23307]: do_label: xxx 'IBM-ESX'
Dec  4 02:27:16 pim udev[23307]: do_label: compare attribute 'vendor' value 'IBM-ES' with 'IBM-ESXS?'
Dec  4 02:27:16 pim udev[23307]: do_label: dev->bus='scsi' sysfs_device->bus='scsi'
Dec  4 02:27:16 pim udev[23307]: do_label: look for device attribute 'vendor'
Dec  4 02:27:16 pim udev[23307]: do_label: xxx 'IBM-ES'
Dec  4 02:27:16 pim udev[23307]: do_label: compare attribute 'vendor' value 'IBM-E' with 'IBM-ES??'
Dec  4 02:27:16 pim udev[23307]: do_label: dev->bus='scsi' sysfs_device->bus='scsi'
Dec  4 02:27:16 pim udev[23307]: do_label: look for device attribute 'vendor'
Dec  4 02:27:16 pim udev[23307]: do_label: xxx 'IBM-E'
Dec  4 02:27:16 pim udev[23307]: do_label: compare attribute 'vendor' value 'IBM-' with 'IBM-ESXSS'

I changed the behavior to remove only the line feed.

03-bug-in-linefeed-removal.diff
  remove only the line feed from string not every last char

18 years ago[PATCH] 008_bk mark
greg@kroah.com [Fri, 5 Dec 2003 03:21:10 +0000 (19:21 -0800)]
[PATCH] 008_bk mark

18 years ago[PATCH] v008 release 008
greg@kroah.com [Fri, 5 Dec 2003 02:01:29 +0000 (18:01 -0800)]
[PATCH] v008 release

18 years ago[PATCH] update ChangeLog for v008
greg@kroah.com [Thu, 4 Dec 2003 05:57:12 +0000 (21:57 -0800)]
[PATCH] update ChangeLog for v008

18 years ago[PATCH] fix up some duplicated function compiler warnings in libsysfs
greg@kroah.com [Thu, 4 Dec 2003 05:51:37 +0000 (21:51 -0800)]
[PATCH] fix up some duplicated function compiler warnings in libsysfs

18 years ago[PATCH] fix some compiler warnings in the tdb code.
greg@kroah.com [Thu, 4 Dec 2003 05:45:20 +0000 (21:45 -0800)]
[PATCH] fix some compiler warnings in the tdb code.

18 years ago[PATCH] Added Kay's name to the man page.
greg@kroah.com [Thu, 4 Dec 2003 03:27:24 +0000 (19:27 -0800)]
[PATCH] Added Kay's name to the man page.

18 years ago[PATCH] update the wildcard documentation in the man page to show the new styles...
greg@kroah.com [Thu, 4 Dec 2003 03:26:02 +0000 (19:26 -0800)]
[PATCH] update the wildcard documentation in the man page to show the new styles supported.

18 years ago[PATCH] fix permission handling logic
greg@kroah.com [Thu, 4 Dec 2003 02:33:58 +0000 (18:33 -0800)]
[PATCH] fix permission handling logic

Now we can handle wildcards properly within the permission file.

18 years ago[PATCH] enable default_mode ability to actually build
greg@kroah.com [Thu, 4 Dec 2003 02:30:49 +0000 (18:30 -0800)]
[PATCH] enable default_mode ability to actually build

18 years ago[PATCH] add support for the default_mode variable, as it is documented...
greg@kroah.com [Thu, 4 Dec 2003 01:41:02 +0000 (17:41 -0800)]
[PATCH] add support for the default_mode variable, as it is documented...

18 years ago[PATCH] show permissions and groups in the label_test
greg@kroah.com [Thu, 4 Dec 2003 01:37:29 +0000 (17:37 -0800)]
[PATCH] show permissions and groups in the label_test

18 years ago[PATCH] remove some items off of the TODO list, as they are now done.
greg@kroah.com [Wed, 3 Dec 2003 16:18:14 +0000 (08:18 -0800)]
[PATCH] remove some items off of the TODO list, as they are now done.

18 years ago[PATCH] fix up the tests to work without all of the environ variables.
greg@kroah.com [Wed, 3 Dec 2003 16:13:53 +0000 (08:13 -0800)]
[PATCH] fix up the tests to work without all of the environ variables.

18 years ago[PATCH] get rid of the majority of the debug environment variables.
greg@kroah.com [Wed, 3 Dec 2003 16:13:11 +0000 (08:13 -0800)]
[PATCH] get rid of the majority of the debug environment variables.

Now there are only 3 valid environment test variables.  The rest can be
specified with the config file.

18 years ago[PATCH] pattern matching for namedev
kay.sievers@vrfy.org [Wed, 3 Dec 2003 14:22:53 +0000 (06:22 -0800)]
[PATCH] pattern matching for namedev

As promised yesterday, here is a patch to implement a more advanced
pattern matching instead of the simple '*'.

We can remove the "tty"="tty" line from udev.rules now and
replace "tty*" by "tty[0-9]*" to catch only the vc's.

  implement pattern matching in namedev
  '*'  - to match zero or more chars
  '?'  - to match exactly one char
  '[]' - character classes with ranges '[0-9]'and negation [!A]

18 years ago[PATCH] Update the man page to show the new config file, it's format, and how to...
greg@kroah.com [Wed, 3 Dec 2003 09:09:48 +0000 (01:09 -0800)]
[PATCH] Update the man page to show the new config file, it's format, and how to use it.

wow, update the docs to keep in line with reality, what a concept...

18 years ago[PATCH] fix up the tests to support the rules file name change
greg@kroah.com [Wed, 3 Dec 2003 09:09:05 +0000 (01:09 -0800)]
[PATCH] fix up the tests to support the rules file name change

18 years ago[PATCH] add support for a main udev config file, udev.conf.
greg@kroah.com [Wed, 3 Dec 2003 09:08:46 +0000 (01:08 -0800)]
[PATCH] add support for a main udev config file, udev.conf.

the older udev.config file is now called udev.rules.
This allows us to better control configuration values, and move away from
the environment variables.

18 years ago[PATCH] turn debugging messages off by default.
greg@kroah.com [Wed, 3 Dec 2003 02:41:29 +0000 (18:41 -0800)]
[PATCH] turn debugging messages off by default.

it's a bit noisy for the masses...

18 years ago[PATCH] split out the namedev config parsing logic to namedev_parse.c
greg@kroah.com [Wed, 3 Dec 2003 02:38:30 +0000 (18:38 -0800)]
[PATCH] split out the namedev config parsing logic to namedev_parse.c

18 years ago[PATCH] rename namedev's get_attr() to be main namedev_name_device() as that's what...
greg@kroah.com [Wed, 3 Dec 2003 02:18:21 +0000 (18:18 -0800)]
[PATCH] rename namedev's get_attr() to be main namedev_name_device() as that's what it really is.

18 years ago[PATCH] add devfs like tty rules as an example in the default config file.
greg@kroah.com [Wed, 3 Dec 2003 01:55:00 +0000 (17:55 -0800)]
[PATCH] add devfs like tty rules as an example in the default config file.

18 years ago[PATCH] catch replace device by wildcard
kay.sievers@vrfy.org [Wed, 3 Dec 2003 01:52:26 +0000 (17:52 -0800)]
[PATCH] catch replace device by wildcard

  catch device name by wildcard to support a whole class of devices
  by just one config line like:
  REPLACE, KERNEL="tty*", NAME="vc/%n"

18 years ago[PATCH] operate on the rules in the order they are in the config file (within the...
greg@kroah.com [Wed, 3 Dec 2003 01:45:19 +0000 (17:45 -0800)]
[PATCH] operate on the rules in the order they are in the config file (within the rule type)

18 years ago[PATCH] Add -nodefaultlibs while compiling against klibc
mort@wildopensource.com [Tue, 2 Dec 2003 08:59:36 +0000 (00:59 -0800)]
[PATCH] Add -nodefaultlibs while compiling against klibc

This patch adds -nodefaultlibs to LDFLAGS when compiling udev against
klibc.  This fixes the warning that I was getting when using $(LD)=gcc
in the versions after Makefile.klibc disappeared.  The problem was that
it was still including a "-lc" in the call to the linker.

18 years ago[PATCH] another patch for path problem
dsteklof@us.ibm.com [Tue, 2 Dec 2003 08:48:01 +0000 (00:48 -0800)]
[PATCH] another patch for path problem

The quick patch I sent you yesterday fixes it in one location, but
there are other points in the library that calls sysfs_get_mnt_path. We
need to address all the areas in the library where paths are used. The
following patch is a band-aid until we can get a proper path management
in the library.

18 years ago[PATCH] Cset exclude: dsteklof@us.ibm.com|ChangeSet|20031126173159|56255
greg@kroah.com [Tue, 2 Dec 2003 08:47:52 +0000 (00:47 -0800)]
[PATCH] Cset exclude: dsteklof@us.ibm.com|ChangeSet|20031126173159|56255

18 years ago[PATCH] ARCH detection for ppc
olh@suse.de [Tue, 2 Dec 2003 08:44:48 +0000 (00:44 -0800)]
[PATCH] ARCH detection for ppc

I'm not sure why ppc is converted to powerpc, it breaks at least $(ARCH)
in klibc.

gcc -dumpmachine
powerpc-suse-linux

18 years ago[PATCH] Build failure - missing linux/limits.h include?
arnd@arndb.de [Tue, 2 Dec 2003 08:26:46 +0000 (00:26 -0800)]
[PATCH] Build failure - missing linux/limits.h include?

On Tuesday 25 November 2003 00:12, Chris Larson wrote:
> udev fails to compile here unless I'm doing a KLIBC build.  The reason
> appears to be that the normal limits.h in the gcc inc dir doesn't pull
> in linux/limits.h, whereas the limits.h out in the klibc include dirs
> does.  I'd think it'd be best to add a #include <linux/limits.h> to
> udev.h directly, since it uses PATH_MAX.

No, don't include kernel headers directly if you can avoid it.
The problem you are referring to seems to be with old tool chains,
I have the same symptom with my s390 gcc-2.95/glibc-2.1.3.
Including <sys/param.h> instead of <limits.h> seems to fix it.

18 years ago[PATCH] udev-007/extras/multipath update
christophe.varoqui@free.fr [Tue, 2 Dec 2003 08:16:32 +0000 (00:16 -0800)]
[PATCH] udev-007/extras/multipath update

here is a clean-up patch :

* removes sg_err.[ch] deps
* makes sure the core code play nice with klibc
* port the sysfs calls to dlist helpers
* links against udev's sysfs (need libsysfs.a & dlist.a)
* finally define DM_TARGET as "multipath" as Joe posted the code today
  (not tested yet)
* push version forward (do you want it in sync with udev version?)

libdevmapper doesn't play well with klibc, so I wasn't able to produce a
static binary yet. Help needed here ... as I don't want to fall back to
merge libdevmapper code in the core.

It compiles here and doesn't segfault.

18 years ago[PATCH] fix udev parallel builds with klibc
patmans@us.ibm.com [Tue, 2 Dec 2003 07:50:09 +0000 (23:50 -0800)]
[PATCH] fix udev parallel builds with klibc

I can't build udev with make -j9. Here's a patch to fix it.

18 years ago[PATCH] add test for checking the BUS value.
greg@kroah.com [Thu, 27 Nov 2003 01:45:26 +0000 (17:45 -0800)]
[PATCH] add test for checking the BUS value.

18 years ago[PATCH] fix problem where we were not looking at the BUS value.
greg@kroah.com [Thu, 27 Nov 2003 01:45:05 +0000 (17:45 -0800)]
[PATCH] fix problem where we were not looking at the BUS value.

18 years ago[PATCH] quick fix for libsysfs bus
dsteklof@us.ibm.com [Thu, 27 Nov 2003 01:31:59 +0000 (17:31 -0800)]
[PATCH] quick fix for libsysfs bus

There seems to be a
mismatch in udev and in libsysfs as to what to expect if the mnt point
has a slash on the end or not. If I use the included patch, it breaks
something in udev. If I patch sysfs_get_mnt_path I break udev as well
because what you're expecting. I need to sit down and go through
the library and creaate a rule as to trailing slashes. Adding the env
brought this to light.

18 years ago[PATCH] libsysfs changes for sysfsutils 0.3.0
dsteklof@us.ibm.com [Tue, 25 Nov 2003 07:47:43 +0000 (23:47 -0800)]
[PATCH] libsysfs changes for sysfsutils 0.3.0

Here's the patch to up the library to the sysfsutils-0_3_0 level. The
following changes:

1) adds class name to sysfs_class_device structure
2) adds bus to sysfs_device
3) gets rid of code that made assumptions as to bus addresses being
unique across buses, which isn't the case.

I still owe you:

1) change getpagesize->sysconf. This is in the CVS tree and part of other
changes we're currently testing. Patch will follow.
2) you need a function to get a sysfs_class_device's parent. We hadn't
considered class devices to have parents, the one example of a multilevel
is the block class. We will add this function and send the patch to you.

18 years ago[PATCH] add test and documentation for new %D devfs format modifier
greg@kroah.com [Tue, 25 Nov 2003 07:45:38 +0000 (23:45 -0800)]
[PATCH] add test and documentation for new %D devfs format modifier

18 years ago[PATCH] Add format modifier for devfs like naming
arnd@arndb.de [Tue, 25 Nov 2003 07:41:40 +0000 (23:41 -0800)]
[PATCH] Add format modifier for devfs like naming

On Monday 24 November 2003 01:29, Greg KH wrote:
> I think with the ability to capture the output of the CALLOUT rule,
> combined with the ability to put format modifiers in the CALLOUT program
> string, we now have everything in place to emulate the existing devfs
> naming scheme.  Anyone want to verify this or not?

I would prefer to have the ability of creating partition nodes in devfs
style built-in to udev. Devfs used to call the whole disk e.g.
"/dev/dasd/0123/disk" and the partitions "/dev/dasd/0123/part[1-3]".
This can obviously be done with a CALLOUT rule, but its common enough
to make it a format modifier. AFAIK, this scheme has been used for
ide, scsi and dasd disks, which is about 99% of all disks ever connected
to Linux.

18 years ago[PATCH] more config file parsing robustness
azarah@nosferatu.za.org [Tue, 25 Nov 2003 07:25:06 +0000 (23:25 -0800)]
[PATCH] more config file parsing robustness

udev kept on segfaulting when it was in use, and not having the time
(and building it with DEBUG=true showing nothing), I have not tracked it
until tonight.  Seems like I made a type-o, and forgotten the ':'
between one line's group and permission parameters.  Attached patch
should stop the segfault, and warn at that at least.

18 years ago[PATCH] changed the default location of the database to /udev/.udev.tdb to be LSB...
greg@kroah.com [Tue, 25 Nov 2003 07:14:44 +0000 (23:14 -0800)]
[PATCH] changed the default location of the database to /udev/.udev.tdb to be LSB compliant

Finally the Debian people can get off my back...

18 years ago[PATCH] udev.8 tweak numeric id text
kay.sievers@vrfy.org [Tue, 25 Nov 2003 06:27:25 +0000 (22:27 -0800)]
[PATCH] udev.8 tweak numeric id text

04-udev.8-tweak-numeric-id-text.diff
  o change "return value" to "returned string"
  o add textual owner/group example for udev.permissions
  o mention klibc compile as special case for udev.permissions
  o remove duplicated numeric owner clause

18 years ago[PATCH] udev-test.pl add subdir test
kay.sievers@vrfy.org [Tue, 25 Nov 2003 06:27:23 +0000 (22:27 -0800)]
[PATCH] udev-test.pl add subdir test

03-udev-test.pl-add-subdir-test.diff
  o duplicate existing test and change it to explicitely test the subdir handling

18 years ago[PATCH] namedev.c strcat tweak
kay.sievers@vrfy.org [Tue, 25 Nov 2003 06:27:20 +0000 (22:27 -0800)]
[PATCH] namedev.c strcat tweak

02-namedev.c-strcat-tweak.diff
  o cat the substitution to the already known end of the string instead of searching it another time

18 years ago[PATCH] overall whitespace + debug text conditioning
kay.sievers@vrfy.org [Tue, 25 Nov 2003 06:27:17 +0000 (22:27 -0800)]
[PATCH] overall whitespace + debug text conditioning

01-overall-whitespace+debug-text-conditioning.diff
  o cleanup whitespace
  o clarify a few comments
  o enclose all printed debug string values in ''

18 years ago[PATCH] klibc makefile fixes
arnd@arndb.de [Tue, 25 Nov 2003 06:10:23 +0000 (22:10 -0800)]
[PATCH] klibc makefile fixes

18 years ago[PATCH] get rid of functions in klibc_fixups that are now in klibc
greg@kroah.com [Tue, 25 Nov 2003 05:59:03 +0000 (21:59 -0800)]
[PATCH] get rid of functions in klibc_fixups that are now in klibc

18 years ago[PATCH] sync up with the 0.84 version of klibc
greg@kroah.com [Tue, 25 Nov 2003 05:57:33 +0000 (21:57 -0800)]
[PATCH] sync up with the 0.84 version of klibc

18 years ago[PATCH] fix udev init.d script to handle all class devices in sysfs.
greg@kroah.com [Mon, 24 Nov 2003 14:19:44 +0000 (06:19 -0800)]
[PATCH] fix udev init.d script to handle all class devices in sysfs.

18 years ago[PATCH] fix the test.block and test.tty scripts due to their moveing. Also add a...
greg@kroah.com [Mon, 24 Nov 2003 14:13:05 +0000 (06:13 -0800)]
[PATCH] fix the test.block and test.tty scripts due to their moveing. Also add a test.all script.

18 years ago[PATCH] udev-test.pl - tweaks
kay.sievers@vrfy.org [Mon, 24 Nov 2003 13:44:01 +0000 (05:44 -0800)]
[PATCH] udev-test.pl - tweaks

I found two missing characters and optimized $PWD.

18 years ago[PATCH] 007_bk version change to Makefile.
greg@kroah.com [Mon, 24 Nov 2003 13:38:01 +0000 (05:38 -0800)]
[PATCH] 007_bk version change to Makefile.

18 years ago[PATCH] v007 release 007
greg@kroah.com [Mon, 24 Nov 2003 08:03:24 +0000 (00:03 -0800)]
[PATCH] v007 release

18 years ago[PATCH] v007 Changelog
greg@kroah.com [Mon, 24 Nov 2003 08:03:10 +0000 (00:03 -0800)]
[PATCH] v007 Changelog

18 years ago[PATCH] update default config file with a CALLOUT rule, and more documentation.
greg@kroah.com [Mon, 24 Nov 2003 07:39:59 +0000 (23:39 -0800)]
[PATCH] update default config file with a CALLOUT rule, and more documentation.

18 years ago[PATCH] updated the man page with the latest format specifier changes.
greg@kroah.com [Mon, 24 Nov 2003 07:39:39 +0000 (23:39 -0800)]
[PATCH] updated the man page with the latest format specifier changes.

18 years ago[PATCH] added ability to put format specifiers in the CALLOUT program string.
greg@kroah.com [Mon, 24 Nov 2003 06:25:13 +0000 (22:25 -0800)]
[PATCH] added ability to put format specifiers in the CALLOUT program string.

18 years ago[PATCH] tweak udev-test.pl to report '0' errors if that's what happened.
greg@kroah.com [Mon, 24 Nov 2003 05:17:34 +0000 (21:17 -0800)]
[PATCH] tweak udev-test.pl to report '0' errors if that's what happened.

18 years ago[PATCH] - format char for CALLOUT output
kay.sievers@vrfy.org [Mon, 24 Nov 2003 05:14:33 +0000 (21:14 -0800)]
[PATCH] - format char for CALLOUT output

here is a patch for inserting the callout output into NAME=.
ID= supports the usual wildcard to compare with the output.

I've moved all wildcard matching to a function cause this was the third occurrence.
Also attached is the last whitespace cleanup and debug text corrections.
The callout patch depends on the whitespace patch.

CALLOUT, BUS="usb", PROGRAM="/bin/echo -n return", ID="ret*", NAME="webcam-%c-"

results in:

Nov 21 17:33:51 pim udev[20399]: get_major_minor: found major = 81, minor = 0
Nov 21 17:33:51 pim udev[20399]: exec_callout: callout to '/bin/echo -n return'
Nov 21 17:33:51 pim udev[20399]: exec_callout: callout returned 'return'
Nov 21 17:33:51 pim udev[20399]: get_attr: substitute callout output 'return'
Nov 21 17:33:51 pim udev[20399]: udev_add_device: name = webcam-return-
Nov 21 17:33:51 pim udev[20399]: create_node: mknod(/udev/webcam-return-, 020660, 81, 0)

18 years ago[PATCH] more namedev whitespace cleanups
kay.sievers@vrfy.org [Mon, 24 Nov 2003 05:14:10 +0000 (21:14 -0800)]
[PATCH] more namedev whitespace cleanups

attached is the last whitespace cleanup and debug text corrections.

18 years ago[PATCH] only build klibc_fixups.c if we are actually using klibc.
greg@kroah.com [Mon, 24 Nov 2003 04:56:52 +0000 (20:56 -0800)]
[PATCH] only build klibc_fixups.c if we are actually using klibc.

18 years ago[PATCH] add support for string group and string user names in udev.permissions.
greg@kroah.com [Mon, 24 Nov 2003 04:56:18 +0000 (20:56 -0800)]
[PATCH] add support for string group and string user names in udev.permissions.

This only works if you are using glibc.

Patch based on patch from Marco d'Itri <md@Linux.IT>

18 years ago[PATCH] add getgrnam and getpwnam to klibc_fixups files.
greg@kroah.com [Mon, 24 Nov 2003 04:44:41 +0000 (20:44 -0800)]
[PATCH] add getgrnam and getpwnam to klibc_fixups files.

Hopefully the klibc_fixups code will not be needed eventually.

18 years ago[PATCH] fix segfault in parsing bad udev.permissions file
md@Linux.IT [Mon, 24 Nov 2003 02:35:03 +0000 (18:35 -0800)]
[PATCH] fix segfault in parsing bad udev.permissions file

18 years ago[PATCH] remove Makefile.klibc
greg@kroah.com [Sun, 23 Nov 2003 13:59:07 +0000 (05:59 -0800)]
[PATCH] remove Makefile.klibc

18 years ago[PATCH] static klibc udev does not link against crt0.o
olh@suse.de [Sun, 23 Nov 2003 13:58:21 +0000 (05:58 -0800)]
[PATCH] static klibc udev does not link against crt0.o

 On Wed, Nov 19, Greg KH wrote:

> > I did 'make KLIBC=true' in the current bk tree.
>
> try 'make -f Makefile.klibc' in the current tree.  For some reason I
> couldn't figure out how to have Makefile work for both KLIBC=true and
> KLIBC=false.  But I didn't try too hard :)

I dont understand that.

please do rm -f Makefile.klibc; apply this patch and tell me what fails.
works for me.

18 years ago[PATCH] add udev-test perl script from Kay Sievers <kay.sievers@vrfy.org> which blows...
greg@kroah.com [Sun, 23 Nov 2003 13:47:28 +0000 (05:47 -0800)]
[PATCH] add udev-test perl script from Kay Sievers <kay.sievers@vrfy.org> which blows away my puny shell scripts.

18 years ago[PATCH] added debian's version of udev.permissions.
greg@kroah.com [Sun, 23 Nov 2003 02:18:53 +0000 (18:18 -0800)]
[PATCH] added debian's version of udev.permissions.

Hm, should probably move all of the config file stuff to a new directory soon...

18 years ago[PATCH] support arguments in callout exec
kay.sievers@vrfy.org [Fri, 21 Nov 2003 14:48:01 +0000 (06:48 -0800)]
[PATCH] support arguments in callout exec

here is argument support for CALLOUT exec:

CALLOUT, PROGRAM="/bin/echo -n xxx", BUS="usb", ID="xxx", NAME="webcam%n"

results in:

Nov 20 02:35:20 pim udev[30422]: get_major_minor: found major = 81, minor = 0
Nov 20 02:35:20 pim udev[30422]: exec_callout: callout to /bin/echo -n xxx
Nov 20 02:35:20 pim udev[30422]: exec_callout: callout returned 'xxx'
Nov 20 02:35:20 pim udev[30422]: get_attr: kernel number appended: 0

The feature is really nice, but the maximum argument count is hard coded to 8.

18 years ago[PATCH] namedev.c - change order of fields in CALLOUT
kay.sievers@vrfy.org [Fri, 21 Nov 2003 14:47:29 +0000 (06:47 -0800)]
[PATCH] namedev.c - change order of fields in CALLOUT

I want to bring the CALLOUT field ordering in line with the other
methods, cause the current parsing relies on the ordering it's good
to have it like the others. The BUS= is now the first expected field.

Also made the last two remaining field names to uppercase and the man page
callout example is updated.

18 years ago[PATCH] namedev.c whitespace + debug text cleanup
kay.sievers@vrfy.org [Fri, 21 Nov 2003 02:37:01 +0000 (18:37 -0800)]
[PATCH] namedev.c whitespace + debug text cleanup

here is mainly a whitespace cleanup for namedev.c. I changed the
dbg_parse() output a bit for better readability:

current:

Nov 19 19:00:59 pim udev[25582]: do_number: NUMBER path='/sys/devices/pci0000:00/0000:00:1d.1/usb2/2-1/2-1.1'
Nov 19 19:00:59 pim udev[25582]: do_number: NUMBER temp='/2-1.1' id='00:07.1'
Nov 19 19:00:59 pim udev[25582]: do_number: NUMBER temp='/2-1' id='00:07.1'
Nov 19 19:00:59 pim udev[25582]: do_number: NUMBER path='/sys/devices/pci0000:00/0000:00:1d.1/usb2/2-1/2-1.1'
Nov 19 19:00:59 pim udev[25582]: do_number: NUMBER temp='/2-1.1' id='00:0b.0'
Nov 19 19:00:59 pim udev[25582]: do_number: NUMBER temp='/2-1' id='00:0b.0'
Nov 19 19:00:59 pim udev[25582]: do_number: NUMBER path='/sys/devices/pci0000:00/0000:00:1d.1/usb2/2-1/2-1.1'
Nov 19 19:00:59 pim udev[25582]: do_number: NUMBER temp='/2-1.1' id='2-1.1'
Nov 19 19:00:59 pim udev[25582]: do_number: device id '2-1.1' becomes 'webcam%n' - owner='', group ='', mode=0

becomes:

Nov 19 19:23:40 pim udev[26091]: do_number: search '00:07.1' in '/2-1.1', path='/sys/devices/pci0000:00/0000:00:1d.1/usb2/2-1/2-1.1'
Nov 19 19:23:40 pim udev[26091]: do_number: search '00:07.1' in '/2-1', path='/sys/devices/pci0000:00/0000:00:1d.1/usb2/2-1'
Nov 19 19:23:40 pim udev[26091]: do_number: search '00:0b.0' in '/2-1.1', path='/sys/devices/pci0000:00/0000:00:1d.1/usb2/2-1/2-1.1'
Nov 19 19:23:40 pim udev[26091]: do_number: search '00:0b.0' in '/2-1', path='/sys/devices/pci0000:00/0000:00:1d.1/usb2/2-1'
Nov 19 19:23:40 pim udev[26091]: do_number: search '2-1.1' in '/2-1.1', path='/sys/devices/pci0000:00/0000:00:1d.1/usb2/2-1/2-1.1'
Nov 19 19:23:40 pim udev[26091]: do_number: found id '2-1.1', 'video0' becomes 'webcam%n' - owner='', group ='', mode=0

18 years ago[PATCH] man page with udev.permissions wildcard
kay.sievers@vrfy.org [Fri, 21 Nov 2003 02:36:55 +0000 (18:36 -0800)]
[PATCH] man page with udev.permissions wildcard

here is the permissions wildcard text for the man page.

18 years ago[PATCH] change to 006_bk version
greg@kroah.com [Fri, 21 Nov 2003 02:35:01 +0000 (18:35 -0800)]
[PATCH] change to 006_bk version

18 years ago[PATCH] v006 006
greg@kroah.com [Wed, 19 Nov 2003 17:35:52 +0000 (09:35 -0800)]
[PATCH] v006

18 years ago[PATCH] v006 changelog
greg@kroah.com [Wed, 19 Nov 2003 17:34:53 +0000 (09:34 -0800)]
[PATCH] v006 changelog

18 years ago[PATCH] add another line to udev.permissions in the proper format.
greg@kroah.com [Wed, 19 Nov 2003 17:25:19 +0000 (09:25 -0800)]
[PATCH] add another line to udev.permissions in the proper format.

18 years ago[PATCH] tweak replace_test
greg@kroah.com [Wed, 19 Nov 2003 17:24:56 +0000 (09:24 -0800)]
[PATCH] tweak replace_test

18 years ago[PATCH] fix permissions to work properly now.
greg@kroah.com [Wed, 19 Nov 2003 17:23:24 +0000 (09:23 -0800)]
[PATCH] fix permissions to work properly now.

Forgot about umask in mknod...

18 years ago[PATCH] add real udev.permissions file to test directory.
greg@kroah.com [Wed, 19 Nov 2003 14:38:45 +0000 (06:38 -0800)]
[PATCH] add real udev.permissions file to test directory.

18 years ago[PATCH] fix namedev.c to build with older version of gcc.
greg@kroah.com [Wed, 19 Nov 2003 14:38:18 +0000 (06:38 -0800)]
[PATCH] fix namedev.c to build with older version of gcc.

18 years ago[PATCH] apply permissions.conf support for wildcard and default name
kay.sievers@vrfy.org [Wed, 19 Nov 2003 14:19:06 +0000 (06:19 -0800)]
[PATCH] apply permissions.conf support for wildcard and default name

Permissions given in udev.permissions are not applied if no METHOD from
udev.config is found. I've added do_kernelname() to scan for known
permissions if we only use the default method.

Simple support for wildcards is also added:

#name:user:group:mode
hdb*:2702:2702:0660

results in:

drwxr-xr-x    2 root     root          240 Nov 19 03:45 .
drwxr-xr-x   23 root     root          528 Nov 17 03:36 ..
brw-r--r--    1 root     root       3,   0 Nov 19 03:45 hda
brw-r--r--    1 root     root       3,   1 Nov 19 03:45 hda1
brw-r--r--    1 root     root       3,   2 Nov 19 03:45 hda2
brw-r--r--    1 root     root       3,   4 Nov 19 03:45 hda4
brw-r-----    1 kay      kay        3,  64 Nov 19 03:45 hdb
brw-r-----    1 kay      kay        3,  65 Nov 19 03:45 hdb1
brw-r--r--    1 root     root      22,   0 Nov 19 03:45 hdc
crw-r--r--    1 root     root      81,   0 Nov 19 03:34 webcam0

18 years ago[PATCH] man page with included placeholder list
kay.sievers@vrfy.org [Wed, 19 Nov 2003 14:18:54 +0000 (06:18 -0800)]
[PATCH] man page with included placeholder list

  mention the printf-like support for string substitution in the NAME field

18 years ago[PATCH] add dumb test for all of the different modifiers.
greg@kroah.com [Wed, 19 Nov 2003 10:49:29 +0000 (02:49 -0800)]
[PATCH] add dumb test for all of the different modifiers.

18 years ago[PATCH] update the TODO list with more items that people can easily do.
greg@kroah.com [Wed, 19 Nov 2003 10:48:57 +0000 (02:48 -0800)]
[PATCH] update the TODO list with more items that people can easily do.

18 years ago[PATCH] faster test scripts
chris_friesen@sympatico.ca [Wed, 19 Nov 2003 10:39:39 +0000 (02:39 -0800)]
[PATCH] faster test scripts

I've attached a patch against 005 for both the block and tty scripts.  I
didn't bother running udev backgrounded, since as you say, the new code
runs a lot faster.

18 years ago[PATCH] move the test.block and test.tty scripts to the test/ directory
greg@kroah.com [Wed, 19 Nov 2003 08:40:39 +0000 (00:40 -0800)]
[PATCH] move the test.block and test.tty scripts to the test/ directory

18 years ago[PATCH] add remove actions to the test scripts
greg@kroah.com [Wed, 19 Nov 2003 08:39:43 +0000 (00:39 -0800)]
[PATCH] add remove actions to the test scripts

18 years ago[PATCH] turn DEBUG_PARSER off by default.
greg@kroah.com [Wed, 19 Nov 2003 08:27:38 +0000 (00:27 -0800)]
[PATCH] turn DEBUG_PARSER off by default.

18 years ago[PATCH] add some documentation for the %b modifier to the default config file.
greg@kroah.com [Wed, 19 Nov 2003 06:48:46 +0000 (22:48 -0800)]
[PATCH] add some documentation for the %b modifier to the default config file.

18 years ago[PATCH] fix make install rule for when the udev symlink is already there.
greg@kroah.com [Wed, 19 Nov 2003 05:45:46 +0000 (21:45 -0800)]
[PATCH] fix make install rule for when the udev symlink is already there.

18 years ago[PATCH] more robust config file parsing in namedev.c
arnd@arndb.de [Wed, 19 Nov 2003 05:39:30 +0000 (21:39 -0800)]
[PATCH] more robust config file parsing in namedev.c

After getting a number of different crashes for udev reading broken
udev.config files, I decided to try to make the parser a little
more robust.

The behaviour is changed to stop reading the configuration file
and logging the broken entry instead of silently ignoring it (is
that good? It's easy to just print and continue).
All strcpy()'s to a fixed length string are now implicitly limited
to the bounds of the target string.

I kept the -ENODEV return code for now, not sure if there should be
different ones.

18 years ago[PATCH] add bus id modifier
arnd@arndb.de [Wed, 19 Nov 2003 05:36:10 +0000 (21:36 -0800)]
[PATCH] add bus id modifier

On Tuesday 18 November 2003 02:14, Greg KH wrote:
> On Mon, Nov 17, 2003 at 06:33:32PM +0100, Arnd Bergmann wrote:
> > That would at least be part of the solution I'm looking for. How about
> > extra format characters for bus_id and for the result of a callout
> > program?
>
> Sure, I can see the use for that.  Want to send a patch?  :)

> Take a look at the current bk tree (which has moved to
> bk://linuxusb.bkbits.net/udev/ )  I've made finding that device a lot
> easier now, and it works for all rule types.

Great, just what I was missing. I didn't see the any link to the bk
repository. Here's the patch for the bus_id. I'll need to think about
the handling of callout results a bit more.

18 years ago[PATCH] change release target in makefile.
greg@kroah.com [Wed, 19 Nov 2003 05:35:44 +0000 (21:35 -0800)]
[PATCH] change release target in makefile.

18 years ago[PATCH] change debug level on printf values for now.
greg@kroah.com [Tue, 18 Nov 2003 09:22:02 +0000 (01:22 -0800)]
[PATCH] change debug level on printf values for now.

18 years ago[PATCH] updated demo config file.
greg@kroah.com [Tue, 18 Nov 2003 09:15:11 +0000 (01:15 -0800)]
[PATCH] updated demo config file.

18 years ago[PATCH] add some documentation of the modifiers to the default config file.
greg@kroah.com [Tue, 18 Nov 2003 09:12:31 +0000 (01:12 -0800)]
[PATCH] add some documentation of the modifiers to the default config file.

18 years ago[PATCH] add demo config file.
greg@kroah.com [Tue, 18 Nov 2003 09:09:58 +0000 (01:09 -0800)]
[PATCH] add demo config file.

18 years ago[PATCH] add printf option to label test to verify it works.
greg@kroah.com [Tue, 18 Nov 2003 09:07:49 +0000 (01:07 -0800)]
[PATCH] add printf option to label test to verify it works.

18 years ago[PATCH] fix up printf-like functionality due to previous changes.
greg@kroah.com [Tue, 18 Nov 2003 09:07:21 +0000 (01:07 -0800)]
[PATCH] fix up printf-like functionality due to previous changes.

18 years ago[PATCH] implement printf-like placeholder support for NAME
kay.sievers@vrfy.org [Tue, 18 Nov 2003 08:59:27 +0000 (00:59 -0800)]
[PATCH] implement printf-like placeholder support for NAME

> Problem is, if you use the LABEL rule to match a device, like a SCSI
> vendor, then all of the partitions, as well as the main block device,
> will end up with the same name.  That's why I added the "add the number"
> hack to the LABEL rule.
>
> So yes, your patch is correct in that we shouldn't always be adding the
> number to any match for LABEL (like for char devices), but if we do
> that, then we break partitions.  Your '%' patch fixes this, but I'd just
> like to extend it a bit.  Let me see what I can come up with...

Oh, I see. Do you mean something like this:

LABEL, BUS="usb", model="Creative Labs WebCam 3", NAME="webcam%n-%M:%m-test"

results in: "webcam0-81:0-test"

Nov 15 16:51:53 pim udev[16193]: get_class_dev: looking at /sys/class/video4linux/video0
Nov 15 16:51:53 pim udev[16193]: get_class_dev: class_dev->name = video0
Nov 15 16:51:53 pim udev[16193]: get_major_minor: dev = 81:0
Nov 15 16:51:53 pim udev[16193]: get_major_minor: found major = 81, minor = 0
Nov 15 16:51:53 pim udev[16193]: udev_add_device: name = webcam0-81:0-test
Nov 15 16:51:53 pim udev[16193]: create_node: mknod(/udev/webcam0-81:0-test, 020666, 81, 0)

  implement printf-like placeholder support for NAME
  %n-kernel number, %M-major number, %m-minor number