13 Oct, 2007
3 commits
-
This makes it a bit more sane when trying to figure out how to clean up
the ktype mess.Based on a larger patch from Kay Sievers
Cc: Kay Sievers
Signed-off-by: Greg Kroah-Hartman -
Move uevent specific logic from the core into kobject_uevent.c, which
does no longer require to link the unused string array if hotplug
is not compiled in.Signed-off-by: Kay Sievers
Signed-off-by: Greg Kroah-Hartman -
This changes the uevent buffer functions to use a struct instead of a
long list of parameters. It does no longer require the caller to do the
proper buffer termination and size accounting, which is currently wrong
in some places. It fixes a known bug where parts of the uevent
environment are overwritten because of wrong index calculations.Many thanks to Mathieu Desnoyers for finding bugs and improving the
error handling.Signed-off-by: Kay Sievers
Cc: Mathieu Desnoyers
Cc: Cornelia Huck
Signed-off-by: Greg Kroah-Hartman
08 Oct, 2007
1 commit
-
We should only reparent to a class former class devices that
form the base of class hierarchy. Nested devices should still
grow from their real parents.Signed-off-by: Dmitry Torokhov
Tested-by: Andrey Borzenkov
Tested-by: Anssi Hannula
Signed-off-by: Linus Torvalds
27 Sep, 2007
1 commit
-
As Stephen Hemminger says, this is a "belt and suspenders" patch that
zeroes the envp array at allocation time, even though all the users
should NULL-terminate it anyway (and we've hopefully fixed everybody
that doesn't do that).And we'll apparently clean the whole envp thing up for 2.6.24 anyway.
But let's just be robust, and do both this *and* make sure that all
users are doing the right thing.Acked-by: Stephen Hemminger
Acked-by: Greg Kroah-Hartman
Signed-off-by: Linus Torvalds
20 Sep, 2007
1 commit
-
Nested class devices used to have 'device' symlink point to a real
(physical) device instead of a parent class device. When converting
subsystems to struct device we need to keep doing what class devices did if
CONFIG_SYSFS_DEPRECATED is Y, otherwise parts of udev break.Signed-off-by: Dmitry Torokhov
Cc: Kay Sievers
Acked-by: Greg KH
Tested-by: Anssi Hannula
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
31 Jul, 2007
2 commits
-
driver core: revert "device" link creation check
Commit 2ee97caf0a6602f749ddbfdb1449e383e1212707 introduced an extra
check on when to create the "device" symlink. Unfortunately, this
breaks input, so let's revert to the old behaviour.Signed-off-by: Cornelia Huck
Acked-by: Kay Sievers
Signed-off-by: Greg Kroah-Hartman -
This prevents the extern declaration in the driver core.
Cc: Kay Sievers
Cc: Andrew Morton
Signed-off-by: Greg Kroah-Hartman
19 Jul, 2007
2 commits
-
Check for return value of sysfs_create_link() in device_add() and
device_rename(). Add helper functions device_add_class_symlinks() and
device_remove_class_symlinks() to make the code easier to read.[akpm@linux-foundation.org: fix unused var warnings]
Signed-off-by: Cornelia Huck
Acked-by: Jeff Garzik
Signed-off-by: Andrew Morton
Signed-off-by: Greg Kroah-Hartman -
This allows the uevent file to handle any type of uevent action to be
triggered by userspace instead of just the "add" uevent.Signed-off-by: Kay Sievers
Signed-off-by: Greg Kroah-Hartman
12 Jul, 2007
3 commits
-
devt_attr and uevent_attr are either allocated dynamically with or
embedded in device and class_device as they needed their owner field
set to the module implementing the driver. Now that sysfs implements
immediate disconnect and owner field removed from struct attribute,
there is no reason to do this. Remove these attributes from
[class_]device and use static attribute structures instead.Signed-off-by: Tejun Heo
Signed-off-by: Greg Kroah-Hartman -
sysfs is now completely out of driver/module lifetime game. After
deletion, a sysfs node doesn't access anything outside sysfs proper,
so there's no reason to hold onto the attribute owners. Note that
often the wrong modules were accounted for as owners leading to
accessing removed modules.This patch kills now unnecessary attribute->owner. Note that with
this change, userland holding a sysfs node does not prevent the
backing module from being unloaded.For more info regarding lifetime rule cleanup, please read the
following message.http://article.gmane.org/gmane.linux.kernel/510293
(tweaked by Greg to not delete the field just yet, to make it easier to
merge things properly.)Signed-off-by: Tejun Heo
Cc: Cornelia Huck
Cc: Andrew Morton
Signed-off-by: Greg Kroah-Hartman -
This converts code of the form
if ((error = some_func()))
goto fixup;
to
error = some_func();
if (error)
goto fixup;Signed-off-by: Cornelia Huck
Signed-off-by: Andrew Morton
Signed-off-by: Greg Kroah-Hartman
09 Jun, 2007
1 commit
-
Class-devices created by "struct class_device" are going to be replaced
by "struct device". Keep the deprecated PHYSDEV* variables for the already
"deprecated" struct class_device" devices.Signed-off-by: Kay Sievers
Signed-off-by: Greg Kroah-Hartman
03 May, 2007
2 commits
-
We need to work on cleaning up the relationship between kobjects, ksets and
ktypes. The removal of 'struct subsystem' is the first step of this,
especially as it is not really needed at all.Thanks to Kay for fixing the bugs in this patch.
Signed-off-by: Greg Kroah-Hartman
-
Declaring an array of PAGE_SIZE does bad things for people running with
4k stacks...Thanks to Tilman Schmidt for tracking this down.
Cc: Tilman Schmidt
Cc: Kay Sievers
Signed-off-by: Greg Kroah-Hartman
28 Apr, 2007
10 commits
-
This patch (as896b) fixes an oversight in the design of
device_schedule_callback(). It is necessary to acquire a reference to the
module owning the callback routine, to prevent the module from being
unloaded before the callback can run.Signed-off-by: Alan Stern
Cc: Satyam Sharma
Cc: Neil Brown
Cc: Cornelia Huck
Signed-off-by: Andrew Morton
Signed-off-by: Greg Kroah-Hartman -
In the future we will allow the uevent type to be written to the uevent
file to trigger the different types of uevents. But for now, as we only
support the ADD event, warn if userspace tries to write anything else to
this file.Signed-off-by: Kay Sievers
Signed-off-by: Greg Kroah-Hartman -
This allows sysfs to show the environment variables that are available
if the uevent happens. This lets userspace not have to cache all of
this information as the kernel already knows it.Signed-off-by: Kay Sievers
Signed-off-by: Greg Kroah-Hartman -
Suppress uevents for devices if uevent_suppress is set via
dev_uevent_filter(). This makes the driver core suppress all device
uevents, not just the add event in device_add().Signed-off-by: Cornelia Huck
Cc: Kay Sievers
Signed-off-by: Greg Kroah-Hartman -
Don't fail bus_attach_device() if the device cannot be bound.
If dev->driver has been specified, reset it to NULL if device_bind_driver()
failed and add the device as an unbound device. As a result,
bus_attach_device() now cannot fail, and we can remove some checking from
device_add().Also remove an unneeded check in bus_rescan_devices_helper().
Signed-off-by: Cornelia Huck
Signed-off-by: Andrew Morton
Signed-off-by: Greg Kroah-Hartman -
If "name" of a device_type is specified, the uevent will
contain the device_type name in the DEVTYPE variable.
This helps userspace to distingiush between different types
of devices, belonging to the same subsystem.Signed-off-by: Kay Sievers
Signed-off-by: Greg Kroah-Hartman -
Driver core: use attribute groups in struct device_type
Attribute groups are more flexible than attribute lists
(an attribute list can be represented by anonymous group)
so switch struct device_type to use them.Also rework attribute creation for devices so that they all
cleaned up properly in case of errors.Signed-off-by: Dmitry Torokhov
Cc: Kay Sievers
Signed-off-by: Greg Kroah-Hartman -
As the new-style class devices (as opposed to old-style struct
class_device) are becoming more widely used, I noticed that the
dev_printk-based functions are not working properly with these.
New-style class devices have no driver nor bus, almost by definition,
and as a result dev_driver_string(), which is used as the first
parameter of dev_printk, resolves to an empty string. This causes
entries like the following to show in my logs:i2c-2: adapter [SMBus stub driver] registered
Notice the unaesthetical leading whitespace. In order to fix this
problem, I suggest that we extend dev_driver_string to deal with
new-style class devices:Signed-off-by: Jean Delvare
Signed-off-by: Greg Kroah-Hartman -
- uses a kset in "struct class" to keep track of all directories
belonging to this class
- merges with the /sys/devices/virtual logic.
- removes the namespace-dir if the last member of that class
leaves the directory.There may be locking or refcounting fixes left, I stopped when it seemed
to work with network and sound modules. :)From: Kay Sievers
Signed-off-by: Greg Kroah-Hartman -
- At the moment we jump here device was't added to
dev->class->devices list yet.Signed-off-by: Monakhov Dmitriy
Signed-off-by: Greg Kroah-Hartman
05 Apr, 2007
1 commit
-
Revert all this. It can cause device-mapper to receive a different major from
earlier kernels and it turns out that the Amanda backup program (via GNU tar,
apparently) checks major numbers on files when performing incremental backups.Which is a bit broken of Amanda (or tar), but this feature isn't important
enough to justify the churn.Cc: Ingo Molnar
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
16 Mar, 2007
1 commit
-
This patch (as868) adds a helper routine for device drivers that need
to set up a callback to perform some action in a different process's
context. This is intended for use by attribute methods that want to
unregister themselves or their parent device. Attribute method calls
are mutually exclusive with unregistration, so such actions cannot be
taken directly.Two attribute methods are converted to use the new helper routine: one
for SCSI device deletion and one for System/390 ccwgroup devices.Signed-off-by: Alan Stern
Cc: Hugh Dickins
Cc: Cornelia Huck
Cc: Oliver Neukum
Signed-off-by: Linus Torvalds
10 Mar, 2007
2 commits
-
In wireless we'd like to allow renaming of the phy devices we surface in
sysfs. The base wireless code, however, can be built modular and thus we
need device_rename exported.Signed-off-by: Johannes Berg
Signed-off-by: Greg Kroah-Hartman -
This moves the device symlink back to sysfs even if
CONFIG_SYSFS_DEPRECATED is enabled as too many userspace programs (well,
HAL), still rely on this link to be present.I will rework the ability for sysfs to change layouts like this in the
future, but for now, this patch should fix people's network connections.Cc: Kay Sievers
Cc: Matt Mackall
Signed-off-by: Greg Kroah-Hartman
09 Mar, 2007
1 commit
-
Some platform devices are driven without driver attached, so managed
resources can be acquired without driver attached. Make sure such
resources are released by calling devres_release_all() in
device_del().Signed-off-by: Tejun Heo
Signed-off-by: Jeff Garzik
24 Feb, 2007
1 commit
-
When a device fails to register the class symlinks where not cleaned up.
This left a symlink in the /sys/class/"device"/ directory that pointed
to no where. This caused the sysfs_follow_link Oops I reported earlier.
This patch cleanups up the symlink. Please apply. Thank you.Signed-Off: James Simmons
Signed-off-by: Greg Kroah-Hartman
21 Feb, 2007
1 commit
-
Several people have reported failures in dynamic major device number handling
due to the recent changes in there to avoid handing out the local/experimental
majors.Rolf reports that this is due to a gcc-4.1.0 bug.
The patch refactors that code a lot in an attempt to provoke the compiler into
behaving.Cc: Rolf Eike Beer
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
10 Feb, 2007
1 commit
-
Implement device resource management, in short, devres. A device
driver can allocate arbirary size of devres data which is associated
with a release function. On driver detach, release function is
invoked on the devres data, then, devres data is freed.devreses are typed by associated release functions. Some devreses are
better represented by single instance of the type while others need
multiple instances sharing the same release function. Both usages are
supported.devreses can be grouped using devres group such that a device driver
can easily release acquired resources halfway through initialization
or selectively release resources (e.g. resources for port 1 out of 4
ports).This patch adds devres core including documentation and the following
managed interfaces.* alloc/free : devm_kzalloc(), devm_kzfree()
* IO region : devm_request_region(), devm_release_region()
* IRQ : devm_request_irq(), devm_free_irq()
* DMA : dmam_alloc_coherent(), dmam_free_coherent(),
dmam_declare_coherent_memory(), dmam_pool_create(),
dmam_pool_destroy()
* PCI : pcim_enable_device(), pcim_pin_device(), pci_is_managed()
* iomap : devm_ioport_map(), devm_ioport_unmap(), devm_ioremap(),
devm_ioremap_nocache(), devm_iounmap(), pcim_iomap_table(),
pcim_iomap(), pcim_iounmap()Signed-off-by: Tejun Heo
Signed-off-by: Jeff Garzik
08 Feb, 2007
6 commits
-
For the block subsystem, we want to delay all uevents until the
disk has been scanned and allpartitons are already created before
the first event is sent out.Signed-off-by: Kay Sievers
Signed-off-by: Greg Kroah-Hartman -
This allows us to add type specific attributes, uevent vars and
release funtions.A subsystem can carry different types of devices like the "block"
subsys has disks and partitions. Both types create a different set
of attributes, but belong to the same subsystem.This corresponds to the low level objects:
kobject -> device (object/device data)
kobj_type -> device_type (type of object/device we are embedded in)
kset -> class/bus (list of objects/devices of a subsystem)Signed-off-by: Kay Sievers
Signed-off-by: Greg Kroah-Hartman -
Devices converted from class_device to device should have
the same uevent keys as the original class_device had. We
search up the parents until we find the first bus device and
add the (already deprecated) PHYDEV* values.Signed-off-by: Kay Sievers
Signed-off-by: Greg Kroah-Hartman -
Make make_class_name() return NULL on error and fixup callers in the
driver core.Signed-off-by: Cornelia Huck
Signed-off-by: Greg Kroah-Hartman -
If we allow NULL as the new parent in device_move(), we need to make sure
that the device is placed into the same place as it would if it was
newly registered:- Consider the device virtual tree. In order to be able to reuse code,
setup_parent() has been tweaked a bit.
- kobject_move() can fall back to the kset's kobject.
- sysfs_move_dir() uses the sysfs root dir as fallback.Signed-off-by: Cornelia Huck
Cc: Marcel Holtmann
Signed-off-by: Greg Kroah-Hartman -
device_is_registered() will always be false for a device with no bus. Remove
this check and trust the caller to know what they're doing.Signed-off-by: Cornelia Huck
Cc: Marcel Holtmann
Signed-off-by: Greg Kroah-Hartman
