Here we change the DEVPATH for netdev's in the environment of the dev.d/
scripts to the name the device is renamed to. The original name doesn't
exist in the kernel after rename.
Here is a patch to change the netdev handling in the database and for
the dev.d/ calls. I applies on top of the udevd.patch, cause klibc has
no sysinfo().
o netdev's are also put into our database now. I want this for the
udevruler gui to get a list of all handled devices.
All devices in the db are stamped with the system uptime value at
the creation time. 'udevinfo -d' prints it.
o the DEVPATH value is the key for udevdb, but if we rename
a netdev, the name is replaced in the kernel, so we add
the changed name to the db to match with the remove event.
NOTE: The dev.d/ scripts still get the original name from the
hotplug call. Should we replace DEVPATH with the new name too?
o We now only add a device to the db, if we have successfully created
the main node or successfully renamed a netdev. This is the main part
of the patch, cause I needed to clean the retval passing trough all
the functions used for node creation.
o DEVNODE sounds a bit ugly for netdev's so I exported DEVNAME too.
Can we change the name?
o I've added a UDEV_NO_DEVD to possibly skip the script execution
and used it in udev-test.pl.
udevstart is the same horror now, if you have scripts with logging
statements in dev.d/ it takes minutes to finish, can we skip the
scripts here too?
o The get_device_type() function is changed to be more strict, cause
'udevinfo -a -p /block/' gets a class device for it and tries to
print the major/minor values.
o bugfix, the RESULT value has now a working newline removal and a test
for this case.
[PATCH] udevd race conditions and performance, assorted cleanups - take 2
here is a patch on top of your nice improvements.
I fixed the whitespace and it hopefully fixes the stupid timestamp bug in
udevd. Some stupid OS sets the hwclock to localtime and linux changes it
to UTC while starting. If any events are pending they may be delayed by
the users time distance from UTC :) So we use the uptime seconds now.
[PATCH] udevd race conditions and performance, assorted cleanups
This patch covers a number of areas:
1) sysfs.h is fixed up to use the common dbg() macro. This fixes the
case where DEBUG is defined but USE_LOG isn't.
2) udevstart.c is modified to include the proper headers, rather than
getting them indirectly which can break depending on Makefile flags
3) udevd.c gets some major changes:
a) I added a pipe from the signal handler. This fixes the race
conditions that I mentioned earlier. Basically, the point of the pipe
is to force the select() call to return immediately if a signal handler
fired before we actually started the select() call. This then lets us
run the appropriate code based on flags set in the signal handler proper.
b) I added a number of flags to coalesce calls to common routines. This
should make things slightly more efficient.
c) since most calls will tend to come in with a sequence number larger
than what has been received, I switched msg_queue_insert() to scan the
msg_list backwards to improve performance.
On Fri, Mar 26, 2004 at 06:36:32PM +0500, Ananth N Mavinakayanahalli wrote:
> On Fri, Mar 26, 2004 at 11:24:39AM +0100, Kay Sievers wrote:
> > On Fri, Mar 26, 2004 at 11:21:29AM +0500, Ananth N Mavinakayanahalli wrote:
> > > On Fri, Mar 26, 2004 at 04:47:35AM +0100, Kay Sievers wrote:
> > > > On Fri, Mar 26, 2004 at 01:26:46AM +0100, Carl-Daniel Hailfinger wrote:
> > > > > Greg KH wrote:
> > > > > > On Tue, Mar 23, 2004 at 01:51:01PM -0800, Daniel Stekloff wrote:
> > > >
> > > > No, it breaks the net device handling. I think we should change
> > > > libsysfs instead, not to return a class device for '/block', if
> > > > we want to fix it.
> > >
> > > /sys/block is considered a sysfs "class" and not a class_device. So,
> > > going by udevinfo's help, -p expects path to a class_device and _not_
> > > a class itself and hence option /sys/block with -p is not a valid query.
> > >
> > > Kay?
> >
> > Yes, it's invalid, but we shouldn't print major minor for a invalid
> > path. sysfs_open_class_device_path("/block") returns a device. If this is
> > the right behavior for libsysfs, I will change the get_device_type("/block")
> > not to return a 'b'-type.
>
> Libsysfs validates the path given to it for opening a class_device to be
> a valid directory; it does not however validate if the path is a valid
> class_device path. So, in the case of udevinfo, a 'b' type should not
> be returned if the path is just /sys/block or /sys/block/
On Thu, Mar 25, 2004 at 02:52:13AM +0100, Kay Sievers wrote:
> Please have look if it still works for you, I only did a very quick
> test.
Here is a unified version, with all the functions moved to udev_lib.c.
We have a generic function now, to call a given fnct(char *) for every
file ending with a specific suffix, sorted in lexical order. We use it
to execute the dev.d/ files and read our rules.d/ files. The binary
should be a bit smaller now.
I've also changed it, to not do the dev.d/ exec for net devices.
Hmm, Arndt Bergmann sent a patch like this one a few weeks ago and
I want to bring the question back, if we want to handle net device
naming with udev.
With this patch it is actually possible to specify something like this
in udev.rules:
[root@pim udev.kay]# ./udevinfo -a -p /sys/class/net/private
looking at class device '/sys/class/net/private':
SYSFS{addr_len}="6"
SYSFS{address}="00:0d:60:77:30:91"
SYSFS{broadcast}="ff:ff:ff:ff:ff:ff"
SYSFS{features}="0x3a9"
SYSFS{flags}="0x1003"
SYSFS{ifindex}="2"
SYSFS{iflink}="2"
SYSFS{mtu}="1500"
SYSFS{tx_queue_len}="1000"
SYSFS{type}="1"
follow the class device's "device"
looking at the device chain at '/sys/devices/pci0000:00/0000:00:1e.0/0000:02:01.0':
BUS="pci"
ID="0000:02:01.0"
SYSFS{class}="0x020000"
SYSFS{detach_state}="0"
SYSFS{device}="0x101e"
SYSFS{irq}="11"
SYSFS{subsystem_device}="0x0549"
SYSFS{subsystem_vendor}="0x1014"
SYSFS{vendor}="0x8086"
The matching device will be renamed to the given name. The device name
will not be put into the udev database, cause the kernel renames the
device and the sysfs name disappears.
I like it, cause it plugs in nicely. We have all the naming features
and sysfs queries and walks inside of udev. The sysfs timing races
are already solved and the management tools are working for net devices
too. nameif can only match the MAC address now. udev can match any sysfs
value of the device tree the net device is connected to.
But right, net devices do not have device nodes :)
I think this is what you want for udevinfo. Patched against the latest BK
tree. I tested it and it seemed to work.
One other question, shouldn't udevinfo.c:print_all_attributes() check to
make sure attr->method is SYSFS_METHOD_SHOW along with checking to see
if attr->value != NULL or doesn't that matter?
[PATCH] correct apply_format() for symlink only rules
Patch from Andrey, which restores the ability to use RESULT values in a
"symlink only" rule. We need to call apply_format() directly after
the matching rule, otherwise the RESULT value may be lost.
Is there any reason to parse the rules for a remove event?
Without it, our test script needs only 2.1 seconds instead of 2.5,
so we have 19 percent more time for testing now :)
olh@suse.de [Wed, 24 Mar 2004 02:52:52 +0000 (18:52 -0800)]
[PATCH] uninitialized variable for mknod and friend
mknod gets an uninitialized variable, which leads to interesting file
modes. the bug is in namedev, devices with no match must not use the
uninitialized stuff were dev points to.
On Mon, Mar 15, 2004 at 09:28:17PM +0100, Kay Sievers wrote:
> Here is a first simple and pretty stupid try to make a simple tool for
> composing of a udev rule.
>
> It reads the udevdb to get all currently handled devices and presents a
> list, where you can choose the device to compose the rule for.
>
> The composed rule is just printed out in a window, nothing else by now.
>
> Do we want something like this?
> Nevermind, I always wanted to know, how this newt thing works :)
Here is the next step, I still can't sleep and there are to many patches
pending to make something useful :)
Cause nobody wanted to play with me, I've made a screenshot.
The device list is sorted in alphabetical order now and if there are only
a few recently discovered devices, they are placed on top of the list.
For those who want to have a look:
http://vrfy.org/projects/udev/udevruler.png
The patch applies on top of today's mmap() patch. The db format is
changed to have the file and line number of the applied rule. So it
should be easy to edit the matching rule with this beast. It compiles
with "make all udevruler".
Hi, Greg. Appended is scsi-devfs.sh, a script for udev to implement
devfs-style names for SCSI hard discs, CD-ROM's and generic devices.
This has been tested with both hard discs and CD-ROM's. The SCSI
generic support should be OK for when there is sysfs/udev support for
SCSI generic devices. SCSI tapes are not yet implemented because I
don't have one to test with.
In addition, this script supports physical names, based on PCI bus
location, both longhand (/udev/bus/pci/...) and shorthand
(/udev/sd/pci/*).
[PATCH] replace fgets() with mmap() and introduce udev_lib.[hc]
Here we replace the various fgets() with a mmap() call for the config
file reading, due to the reported performance problems with klibc.
Thanks to Patrick's testing, it makes a very small, close to nothing
speed gain for libc users, but a 6 times speed increase for klibc users
with a 1000 line config file.
I've created a udev_lib.[hc] for this and also moved all the generic
stuff from udev.h in there and uninlined the functions.
Here are the missing pieces for udevtest. A simple man page is added,
the blacklist is removed, cause it can't work without having a subsystem.
The Makefile removes all manpages now with a uninstall and installs
udevtest in /usr/bin/.
Any old version from /sbin/ should be deleted by hand.
The only expected argument is the sysfs devpath, here I changed it to be
more tolerant to the input. The path may now be specified with or
without a leading slash and optionally with the /sys moutpoint prepended.
I hope this will end the confusion about the use of this program :)
On Thu, Mar 11, 2004 at 02:36:23PM +0100, Kay Sievers wrote:
> On Thu, 2004-03-11 at 15:02, Ananth N Mavinakayanahalli wrote:
> > On Thu, Mar 11, 2004 at 02:04:36PM +0100, Kay Sievers wrote:
> > > On Thu, Mar 11, 2004 at 11:53:50AM +0500, Ananth N Mavinakayanahalli wrote:
> > >
> > > > +#define safestrcpy(to, from) strncpy(to, from, sizeof(to)-1)
> > > > +#define safestrcat(to, from) strncat(to, from, sizeof(to) - strlen(to)-1)
> > >
> > > These strings are not terminated with '\0' if from is longer than
> > > the sizeof to.
> >
> > Did not do it on purpose as the "to" elements are either calloc'd or memset to
> > '0' explicitly in the library. Thats the reason I mentioned "scaled down" :)
>
> Ahh, sounds good.
>
> > > > +#define safestrncpy(to, from, maxsize) \
> > > > +do { \
> > > > + to[maxsize-1] = '\0'; \
> > > > + strncpy(to, from, maxsize-1); \
> > > > +} while (0)
> > > > +
> > > > +#define safestrncat(to, from, maxsize) \
> > > > +do { \
> > > > + to[maxsize-1] = '\0'; \
> > > > + strncat(to, from, maxsize - strlen(to)-1); \
> > > > +} while (0)
> > >
> > > We all expect a similar behavior like strncat/strncpy according to the
> > > names, but these macros are limiting by the target size and do not limit
> > > the count of chars copied.
> > > This is confusing I think and suggest using a different name like
> > > 'safestrcopymax()' or something.
> >
> > Good point.. will make the change
>
> Nice. I've had these *n* names too and I forgot about the logic and only
> 10 days later I introduced a ugly bug cause I can't limit the count of
> copied chars :)
Inlined is the patch for this... applies on the earlier _BIG_ patch.
Please find attached a _BIG_ patch to update udev's libsysfs. Patch applies
on udev-021 and contains:
1. Updates to get udev's libsysfs to the latest (to be released) level.
2. Changes for C++ compatibility (use "char" and not "unsigned char"
unless absolutely necessary).
3. More importantly, take care of buffer overflows. Libsysfs now uses a
scaled down version of Kay's "safe" macros.
Here I change the callout fork logic.
The current cersion is unable to read a pipe which is not flushed at once,
Now we read until it's closed.
The maximum argument count is calculated by the strlen now. We have 100
chars for our result buffer so we can't have more than 50 parameters.
So it's much more clear what will happen now and not some magic boundary
where we use shell behind it.
Parameter can be combined to one by using apostrophes.
this on works now:
BUS="scsi", PROGRAM="/bin/sh -c 'echo foo3 foo4 foo5 foo6 foo7 foo8 foo9 | sed s/foo9/bar9/'", KERNEL="sda3", NAME="%c{7}"
[PATCH] allow to specify node permissions in the rule
This allows to set the permissions along with the rule.
This is not a general replacement for the permissions config, but it
may be easier sometimes for the user to specify the permissions along
with the rule, cause the permissions config file wants the final node
name to match, which seems sometimes a bit difficult to guess, if
format % chars are used in the NAME field.
Any value not given in the rule is still be read from the permissions
file or set to the default. This one will also work:
Hey, I wrote the strn*() macros just 10 days ago and yesterday this trap
caught me with the %c{x} bug.
The names are misleading cause we all expect that the from field is limited by
the size argument, but we actually limit the overall size of the destination
string to prevent a overflow.
Here we rename all strn*() macros to str*max(). That should be
more self-explanatory.