| Age | Commit message (Collapse) | Author |
|---|
|
This makes the udev operation completely lockless by storing a
file for every node in /dev/.udevdb/* This solved the problem
with deadlocking concurrent udev processes waiting for each other
to release the file lock under heavy load.
|
|
|
|
Hi,
Here is a few updates for the udev.rules.gentoo from udev package.
I will summarise it briefly:
1) The last change you did to legacy tty's is wrong.
I say this because:
1a) The original devfs rules had both master and slave in /dev/pty:
nosferatu linux # grep devfs_name drivers/char/pty.c
pty_driver->devfs_name = "pty/m";
pty_slave_driver->devfs_name = "pty/s";
nosferatu linux #
1b) If you refer to '2.6.8.1-mm1 Tty problems?', you will see that
the /dev/tty/ directory our rules create, replaces this symlink:
nosferatu portage # ls -l /dev/tty
crw-rw-rw- 1 root tty 5, 0 Nov 14 17:06 /dev/tty
nosferatu portage #
which is used to determine the controlling tty.
2) Somebody added the /dev/cpu/microcode rule, but it was not run
as there was an older rule before that placing it in /dev/misc (which
is wrong). Just remove the first broken rule
3) Some form/tab cleanups. Reorder rules alphabetically according to
device class to make searching/editing easier.
|
|
|
|
|
|
|
|
|
|
|
|
I thought I had fixed this a long time ago...
|
|
should fix a lot of gentoo bug reports
|
|
Should prevent false error reports from happening.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
physical device
|
|
Will help distros that have a test phase of their build.
|
|
|
|
In the grand tradition of releasing free software projects on my birthday
for the past few years.
And yes, I skipped version 041, call it grumpyness in my old age...
|
|
|
|
|
|
|
|
|
|
Change to the same timeout loop we use in the rest of the code. Change
some comments and names to be more descriptive.
I'm mostly finished with the overall cleanup. I will post a new patch
for the udevd-nofork experiment, which will be much smaller now.
|
|
|
|
|
|
|
|
This patch exposes the wait_for_sysfs functions to all possible users,
so we need to maintain only one list of exceptions. The last list is
hereby removed from udev.c.
|
|
|
|
|
|
Here we get the logging for udev_volume_id working if it is compiled
with DEBUG=true. Also fixed is a name clash with a global variable.
|
|
Hi,
I got these when creating LVM2 snapshots:
,----
| Oct 17 04:23:09 server wait_for_sysfs[5120]: error: wait_for_sysfs needs an update to handle the device '/block/dm-6' properly, please report to <linux-hotplug-devel@lists.sourceforge.net>
| Oct 17 04:23:09 server wait_for_sysfs[5129]: error: wait_for_sysfs needs an update to handle the device '/block/dm-6' properly, please report to <linux-hotplug-devel@lists.sourceforge.net>
| Oct 17 04:23:09 server wait_for_sysfs[5131]: error: wait_for_sysfs needs an update to handle the device '/block/dm-7' properly, please report to <linux-hotplug-devel@lists.sourceforge.net>
| [...]
`----
|
|
|
|
I've been told that open() should not pass O_NONBLOCK.
|
|
We once implemented the devfsd feature to set the owner of a device node
to the "local" user. This was before we had the dev.d/ scripts. We discussed
a similar issue with D-BUS recently and this should be better handled depending
on the distributions way to do such a thing.
I'm for removing this here as this can be easily covered by a dev.d/
script.
Here is the patch if nobody objects :)
|
|
|
|
Here we supress the dev.d/ execution if we didn't change a network
interface's name with a rule. This should solve the issue of two
running dhclients for the same interface, cause the
/etc/dev.d/net/hotplug.dev script that fakes the hotplug event runs
with every udevstart for every interface and fakes a second identical
hotplug event on bootup.
With this patch netif interfaces are no longer stored in the udevdb.
It is not needed, cause we don't have permissions or symlinks :) and
all information is available in sysfs.
This patch also moves the dev_d execution calls out of the
udev_add/udev_remove. As with the former api-cleanup-patch we have
all processed data in one udev struct and can place the execution
calls where needed.
|
|
Here is the first patch to cleanup the internal processing of the
various stages of an udev event. It should not change any behavior,
but if your system depends on udev, please always test it before reboot :)
We pass only one generic structure around between add, remove,
namedev, db and dev_d handling and make all relevant data available
to all internal stages. All udev structures are renamed to "udev".
We replace the fake parameter by a flag in the udev structure.
We open the class device in the main binaries and not in udev_add, to
make it possible to use libsysfs for udevstart directory crawling.
The last sleep parameters are removed.
|
|
We also emit the same things the -mm tree does now, so to make that easier
to test for.
|
|
No need to wait for dummy net devices.
|
|
Seems that we never closed the opened syslog.
Here is a patch to do this in all our binaries.
|
|
On Sat, Oct 16, 2004 at 10:32:47AM -0500, Jason Dravet wrote:
> FYI
>
> wait_for_sysfs[2986]: either wait_for_sysfs (udev 038) needs an update to
> handle the device '/class/ppp/ppp' properly (5) or the sysfs-support of
> your device's driver needs to be fixed, please report to
> <linux-hotplug-devel@lists.sourceforge.net>
>
> I am running udev 38-2 and ppp-2.4.2-6.3
Yeah, the ppp-class is missing, we only covered the net ppp-devices.
|
|
On Sat, Oct 16, 2004 at 02:48:42PM -0400, Jeff wrote:
> Kay Sievers wrote:
> >On Sat, Oct 16, 2004 at 04:40:51AM +0200, Kay Sievers wrote:
> >
> >Here is a better version. We return successful in the case, that the
> >advertised device does not appear within the timeout or goes completely
> >away while waiting for the attributes. All magic numbers should be
> >readable as a string now.
> >
> >This hopefully fixes the last vc errors.
> >
> K, first patch produced a couple of device unavailables & a slew of no
> device symlinks. Second patch-no messages at all :)
>
> Thank god, if I have to reboot again I'm going to puke.
Nice to hear that we covered all the vc errors now. I think we should
apply the same logic to the /devices device wait loops, as they may go
with a remove event too.
Here is the 3rd version of the patch :)
|
|
|
|
Update udev's scsi_id to version 0.7, this fixes a bad bug in scsi_id on
64 bit systems.
|
