Assigning the device's dma_range_map from the iterator variable after
the loop means it always points to the empty terminator at the end of
the map, which is not what we want. Similarly, freeing the iterator on
error when it points to somwhere in the middle of the allocated array
won't work either. Fix this.

Fixes: bf2ee8d0 ("ACPI: scan: Support multiple DMA windows with different offsets")
Signed-off-by: NRobin Murphy <robin.murphy@arm.com>
Reviewed-by: NJianmin Lv <lvjianmin@loongson.cn>
Tested-by: NJeremy Linton <jeremy.linton@arm.com>
Tested-by: NYicong Yang <yangyicong@hisilicon.com>
Reviewed-by: NLorenzo Pieralisi <lpieralisi@kernel.org>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

In commit b83e2b30 ("ACPI: scan: Add function to fetch dependent
of ACPI device") we added a means of fetching the first device to
declare itself dependent on another ACPI device in the _DEP method.
One assumption in that patch was that there would only be a single
consuming device, but this has not held.

Replace that function with a new function that fetches the next consumer
of a supplier device. Where no "previous" consumer is passed in, it
behaves identically to the original function.
Reviewed-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NDaniel Scally <djrscally@gmail.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

In DT systems configurations, of_dma_get_range() returns struct
bus_dma_region DMA regions; they are used to set-up devices
DMA windows with different offset available for translation between DMA
address and CPU address.

In ACPI systems configuration, acpi_dma_get_range() does not return
DMA regions yet and that precludes setting up the dev->dma_range_map
pointer and therefore DMA regions with multiple offsets.

Update acpi_dma_get_range() to return struct bus_dma_region
DMA regions like of_dma_get_range() does.

After updating acpi_dma_get_range(), acpi_arch_dma_setup() is changed for
ARM64, where the original dma_addr and size are removed as these
arguments are now redundant, and pass 0 and U64_MAX for dma_base
and size of arch_setup_dma_ops; this is a simplification consistent
with what other ACPI architectures also pass to iommu_setup_dma_ops().
Reviewed-by: NRobin Murphy <robin.murphy@arm.com>
Signed-off-by: NJianmin Lv <lvjianmin@loongson.cn>
Reviewed-by: NLorenzo Pieralisi <lpieralisi@kernel.org>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

The parent field in struct acpi_device is, in fact, redundant,
because the dev.parent field in it effectively points to the same
object and it is used by the driver core.

Accordingly, the parent field can be dropped from struct acpi_device
and for this purpose define acpi_dev_parent() to retrieve a parent
struct acpi_device pointer from the dev.parent field in struct
acpi_device.  Next, update all of the users of the parent field
in struct acpi_device to use acpi_dev_parent() instead of it and
drop it.

While at it, drop the ACPI_IS_ROOT_DEVICE() macro that is only used
in one place in a confusing way.

No intentional functional impact.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: NMark Brown <broonie@kernel.org>
Acked-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Acked-by: NWei Liu <wei.liu@kernel.org>
Reviewed-by: NPunit Agrawal <punit.agrawal@bytedance.com>

Instead of having acpi_device_add() defined as a wrapper around
__acpi_device_add(), export acpi_tie_acpi_dev() so it can be called
directly by acpi_add_power_resource(), fold acpi_device_add() into the
latter and rename __acpi_device_add() to acpi_device_add().

No intentional functional impact.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NPunit Agrawal <punit.agrawal@bytedance.com>

The initialization of ACPI device objects is split between
acpi_init_device_object() and __acpi_device_add() that initializes
the dev field in struct acpi_device.  The "release" function pointer
is passed to __acpi_device_add() for this reason.

However, that split is artificial and all of the initialization can
be carried out by acpi_init_device_object(), so rearrange the code
to that end.  In particular, make acpi_init_device_object() take the
"release" pointer as an argument, along with the "type" which is
related to it, instead of __acpi_device_add().

No intentional functional impact.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NPunit Agrawal <punit.agrawal@bytedance.com>

The acpi_bus_get_parent() name doesn't really reflect the
purpose of the function so change it to a more accurate
acpi_find_parent_acpi_dev().

While at it, rearrange the code inside that function to make it
easier to read.

No intentional functional impact.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NPunit Agrawal <punit.agrawal@bytedance.com>

Because acpi_bus_get_acpi_device() is completely analogous to
acpi_fetch_acpi_dev(), rename it to acpi_get_acpi_dev() and
add a kerneldoc comment to it.

Accordingly, rename acpi_bus_put_acpi_device() to acpi_put_acpi_dev()
and update all of the users of these two functions.

While at it, move the acpi_fetch_acpi_dev() header next to the
acpi_get_acpi_dev() header in the header file holding them.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: NGuenter Roeck <linux@roeck-us.net>
Reviewed-by: NPunit Agrawal <punit.agrawal@bytedance.com>

Add CLSA0101 id to the ignore_serial_bus_ids
so serial-multi-instantiate can correctly
instantiate the driver.
Signed-off-by: NLucas Tanure <tanureal@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220727095924.80884-4-tanureal@opensource.cirrus.comSigned-off-by: NTakashi Iwai <tiwai@suse.de>

Drop the children and node list heads that have no more users
from struct acpi_device and the code manipulating them from
__acpi_device_add() and acpi_device_del().
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>

Instead of walking the list of children of an ACPI device directly, use
acpi_dev_for_each_child() or acpi_dev_for_each_child_reverse() to carry
out an action for all of the given ACPI device's children.

This will help to eliminate the children list head from struct
acpi_device as it is redundant and it is used in questionable ways
in some places (in particular, locking is needed for walking the
list pointed to it safely, but it is often missing).
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>

Replace the last instance of acpi_bus_get_device(), added recently
by commit 87e59b36 ("spi: Support selection of the index of the
ACPI Spi Resource before alloc"), with acpi_fetch_acpi_dev() and
finally drop acpi_bus_get_device() that has no more users.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: NMark Brown <broonie@kernel.org>

Revert commit e38f9ff6 ("ACPI: scan: Do not add device IDs from _CID
if _HID is not valid"), because it has introduced regressions on
multiple systems, even though it only has effect on clearly invalid
firmware.
Reported-by: NPierre-Louis Bossart <notifications@github.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

The drivers/platform/surface/surface3_button.c code is alsmost a 1:1 copy
of the soc_button_array code.

The only big difference is that it binds to an i2c_client rather then to
a platform_device. The cause of this is the ACPI resources for the MSHW0028
device containing a bogus I2cSerialBusV2 resource which causes the kernel
to instantiate an i2c_client for it instead of a platform_device.

Add "MSHW0028" to the ignore_serial_bus_ids[] list in drivers/apci/scan.c,
so that a platform_device will be instantiated and add support for
the MSHW0028 HID to soc_button_array.

This fully replaces surface3_button, which will be removed in a separate
commit (since it binds to the now no longer created i2c_client it no
longer does anyyhing after this commit).

Note the MSHW0028 id is used by Microsoft to describe the tablet buttons on
both the Surface 3 and the Surface 3 Pro and the actual API/implementation
for the Surface 3 Pro is quite different. The changes in this commit should
not impact the separate surfacepro3_button driver:

1. Because of the bogus I2cSerialBusV2 resource problem that driver binds
   to the acpi_device itself, so instantiating a platform_device instead of
   an i2c_client does not matter.

2. The soc_button_array driver will not bind to the MSHW0028 device on
   the Surface 3 Pro, because it has no GPIO resources.
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Reviewed-by: Maximilian Luz <luzmaximilian@gmail.com
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>
Link: https://lore.kernel.org/r/20220224110241.9613-2-hdegoede@redhat.com

As recommended in include/linux/idr.h, use ida_alloc() instead of
ida_simple_get() for creating unique device object names and for
symmetry replace ida_simple_remove() with ida_free() (and fix up
the related overly long code line while at it).

Also drop the ACPI_MAX_DEVICE_INSTANCES limit that is not necessary
any more and may not be sufficient for future platforms.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>

The ACPI device with CSC3551 or CLSA0100 are sound cards
with multiple instances of CS35L41 connected by I2C or SPI
to the main CPU.

We add an ID to the ignore_serial_bus_ids list to enumerate
all I2C or SPI devices correctly.

The same IDs are also added into serial-multi-instantiate
so that the driver can correctly enumerate the ACPI.
Signed-off-by: NLucas Tanure <tanureal@opensource.cirrus.com>
Signed-off-by: NStefan Binding <sbinding@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220121172431.6876-10-sbinding@opensource.cirrus.comReviewed-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>

Rename I2C multi instantiate driver to serial-multi-instantiate for
upcoming addition of SPI support
Signed-off-by: NLucas Tanure <tanureal@opensource.cirrus.com>
Signed-off-by: NStefan Binding <sbinding@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220121172431.6876-6-sbinding@opensource.cirrus.comReviewed-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>

Rename the "out" label in acpi_scan_init() to "unlock", which is
a better match for its purpose, and fix up its alignment.

No functional impact.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>

It should be perfectly fine to use ACPI if the "fixed" power or sleep
buttons cannot be initialized.  Moreover, running acpi_bus_scan()
successfully on ACPI_ROOT_OBJECT generally causes many devices to
be enumerated and probed, possibly including the entire PCI bus, so
unregistering acpi_root if the registration of the "fixed" buttons
fails is rather unhelpful.

For this reason, do not fail acpi_scan_init() when
acpi_bus_scan_fixed() fails and turn the latter into a void function.

While at it, drop the outdated and misleading comment from
acpi_bus_scan_fixed().
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>

The only caller of acpi_scan_init(), acpi_init(), doesn't check its
return value, so turn it into a void function.

This avoids complaints from the Smatch static checker that the
function should return a negative error code when it fails, which
is not really a problem in this particular case.

No intentional functional impact.

Link: https://lore.kernel.org/linux-acpi/20220106082317.GA9123@kili/Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>

BCM4752 and LNV4752 ACPI nodes describe a Broadcom 4752 GPS module
attached to an UART of the system.

The GPS modules talk a custom protocol which only works with a closed-
source Android gpsd daemon which knows this protocol.

The ACPI nodes also describe GPIOs to turn the GPS on/off these are
handled by the net/rfkill/rfkill-gpio.c code. This handling predates the
addition of enumeration of ACPI instantiated serdevs to the kernel and
was broken by that addition, because the ACPI scan code now no longer
instantiates platform_device-s for these nodes.

Rename the i2c_multi_instantiate_ids HID list to ignore_serial_bus_ids
and add the BCM4752 and LNV4752 HIDs, so that rfkill-gpio gets
a platform_device to bind to again; and so that a tty cdev for gpsd
gets created for these.

Fixes: e361d1f8 ("ACPI / scan: Fix enumeration for special UART devices")
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Storing CRC-32 hash of the Physical Location of Device
object (_PLD) with devices that have it. The hash is stored
to a new struct acpi_device member "pld_crc".

The hash makes it easier to find devices that share a
location, as there is no need to evaluate the entire object
every time. Knowledge about devices that share a location
can be used in device drivers that need to know the
connections to other components inside a system. USB3 ports
will for example always share their location with a USB2
port.
Acked-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: NHeikki Krogerus <heikki.krogerus@linux.intel.com>
Link: https://lore.kernel.org/r/20211223081620.45479-3-heikki.krogerus@linux.intel.comSigned-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Introduce acpi_fetch_acpi_dev() as a more reasonable replacement for
acpi_bus_get_device() and modify the code in scan.c to use it instead
of the latter.

No expected functional impact.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>

The clk and regulator frameworks expect clk/regulator consumer-devices
to have info about the consumed clks/regulators described in the device's
fw_node.

To work around cases where this info is not present in the firmware tables,
which is often the case on x86/ACPI devices, both frameworks allow the
provider-driver to attach info about consumers to the clks/regulators
when registering these.

This causes problems with the probe ordering wrt drivers for consumers
of these clks/regulators. Since the lookups are only registered when the
provider-driver binds, trying to get these clks/regulators before then
results in a -ENOENT error for clks and a dummy regulator for regulators.

One case where we hit this issue is camera sensors such as e.g. the OV8865
sensor found on the Microsoft Surface Go. The sensor uses clks, regulators
and GPIOs provided by a TPS68470 PMIC which is described in an INT3472
ACPI device. There is special platform code handling this and setting
platform_data with the necessary consumer info on the MFD cells
instantiated for the PMIC under: drivers/platform/x86/intel/int3472.

For this to work properly the ov8865 driver must not bind to the I2C-client
for the OV8865 sensor until after the TPS68470 PMIC gpio, regulator and
clk MFD cells have all been fully setup.

The OV8865 on the Microsoft Surface Go is just one example, all X86
devices using the Intel IPU3 camera block found on recent Intel SoCs
have similar issues where there is an INT3472 HID ACPI-device, which
describes the clks and regulators, and the driver for this INT3472 device
must be fully initialized before the sensor driver (any sensor driver)
binds for things to work properly.

On these devices the ACPI nodes describing the sensors all have a _DEP
dependency on the matching INT3472 ACPI device (there is one per sensor).

This allows solving the probe-ordering problem by delaying the enumeration
(instantiation of the I2C-client in the ov8865 example) of ACPI-devices
which have a _DEP dependency on an INT3472 device.

The new acpi_dev_ready_for_enumeration() helper used for this is also
exported because for devices, which have the enumeration_by_parent flag
set, the parent-driver will do its own scan of child ACPI devices and
it will try to enumerate those during its probe(). Code doing this such
as e.g. the i2c-core-acpi.c code must call this new helper to ensure
that it too delays the enumeration until all the _DEP dependencies are
met on devices which have the new honor_deps flag set.
Acked-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20211203102857.44539-2-hdegoede@redhat.com

Section 6.1.2 of ACPI 6.4 explicitly requires _HID to be present for
_CID to be defined, so don't add device IDs from _CID to the device
IDs list of a device if _HID is not valid.

Link: https://uefi.org/specs/ACPI/6.4/06_Device_Configuration/Device_Configuration.html#cid-compatible-idSigned-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>

Revert commit c10383e8 ("ACPI: scan: Release PM resources blocked
by unused objects"), because it causes boot issues to appear on some
platforms.
Reported-by: NKyle D. Pelton <kyle.d.pelton@intel.com>
Reported-by: NSaranya Gopal <saranya.gopal@intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Store a device's desired enumeration power state in struct
acpi_device_power during acpi_device object's initialisation.
Signed-off-by: NSakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

In read_domain_devices(), acpi_bus_get_device() is called to obtain
the ACPI device object attached to the given ACPI handle and
subsequently that object is passed to get_device() for reference
counting, but there is a window between the acpi_bus_get_device()
and get_device() calls in which the ACPI device object in question
may go away.

To address this issue, make read_domain_devices() use
acpi_bus_get_acpi_device() to reference count and return the given
ACPI device object in one go and export that function to modules.

While at it, also make read_domain_devices() and
remove_domain_devices() use acpi_dev_put() instead of calling
put_device() directly on the ACPI device objects returned by
acpi_bus_get_acpi_device().
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/11871063.O9o76ZdvQC@kreacherSigned-off-by: NGuenter Roeck <linux@roeck-us.net>

On some systems the ACPI namespace contains device objects that are
not used in certain configurations of the system.  If they start off
in the D0 power state configuration, they will stay in it until the
system reboots, because of the lack of any mechanism possibly causing
their configuration to change.  If that happens, they may prevent
some power resources from being turned off or generally they may
prevent the platform from getting into the deepest low-power states
thus causing some energy to be wasted.

Address this issue by changing the configuration of unused ACPI
device objects to the D3cold power state one after carrying out
the ACPI-based enumeration of devices.

BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=214091
Link: https://lore.kernel.org/linux-acpi/20211007205126.11769-1-mario.limonciello@amd.com/Reported-by: NMario Limonciello <mario.limonciello@amd.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Tested-by: NMario Limonciello <mario.limonciello@amd.com>

Code that use linux/nls.h was moved to device_sysfs.c by commit
c2efefb3 ("ACPI / scan: Move sysfs-related device code to a separate file")

Remove this include so that complier has easier times and it would be
easier to grep where nls code is used.
Signed-off-by: NKari Argillander <kari.argillander@gmail.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

The ACPI Virtual I/O Translation Table describes topology of
para-virtual platforms, similarly to vendor tables DMAR, IVRS and IORT.
For now it describes the relation between virtio-iommu and the endpoints
it manages.

Three steps are needed to configure DMA of endpoints:

(1) acpi_viot_init(): parse the VIOT table, find or create the fwnode
    associated to each vIOMMU device. This needs to happen after
    acpi_scan_init(), because it relies on the struct device and their
    fwnode to be available.

(2) When probing the vIOMMU device, the driver registers its IOMMU ops
    within the IOMMU subsystem. This step doesn't require any
    intervention from the VIOT driver.

(3) viot_iommu_configure(): before binding the endpoint to a driver,
    find the associated IOMMU ops. Register them, along with the
    endpoint ID, into the device's iommu_fwspec.

If step (3) happens before step (2), it is deferred until the IOMMU is
initialized, then retried.
Tested-by: NEric Auger <eric.auger@redhat.com>
Reviewed-by: NEric Auger <eric.auger@redhat.com>
Signed-off-by: NJean-Philippe Brucker <jean-philippe@linaro.org>
Acked-by: NRafael J. Wysocki <rafael@kernel.org>
Link: https://lore.kernel.org/r/20210618152059.1194210-4-jean-philippe@linaro.orgSigned-off-by: NJoerg Roedel <jroedel@suse.de>

Extract the code that sets up the IOMMU infrastructure from IORT, since
it can be reused by VIOT. Move it one level up into a new
acpi_iommu_configure_id() function, which calls the IORT parsing
function which in turn calls the acpi_iommu_fwspec_init() helper.
Signed-off-by: NJean-Philippe Brucker <jean-philippe@linaro.org>
Reviewed-by: NEric Auger <eric.auger@redhat.com>
Reviewed-by: NRobin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/20210618152059.1194210-3-jean-philippe@linaro.orgSigned-off-by: NJoerg Roedel <jroedel@suse.de>

Extract generic DMA setup code out of IORT, so it can be reused by VIOT.
Keep it in drivers/acpi/arm64 for now, since it could break x86
platforms that haven't run this code so far, if they have invalid
tables.
Reviewed-by: NEric Auger <eric.auger@redhat.com>
Signed-off-by: NJean-Philippe Brucker <jean-philippe@linaro.org>
Reviewed-by: NRobin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/20210618152059.1194210-2-jean-philippe@linaro.orgSigned-off-by: NJoerg Roedel <jroedel@suse.de>

Notice that the table field of struct acpi_table_events_work is never
read and its event field is always equal to ACPI_TABLE_EVENT_LOAD, so
both of them are redundant.

Accordingly, drop struct acpi_table_events_work and use struct
work_struct directly instead of it, simplify acpi_scan_table_handler()
and rename it to acpi_scan_table_notify().

Moreover, make acpi_bus_table_handler() check the event code against
ACPI_TABLE_EVENT_LOAD before calling acpi_scan_table_notify(), so it
is not necessary to do that check in the latter.

No intentional functional impact.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

If acpi_add_single_object() runs concurrently with respect to
acpi_scan_clear_dep() which deletes a dependencies list entry where
the device being added is the consumer, the device's dep_unmet
counter may not be updated to reflect that change.

Namely, if the dependencies list entry is deleted right after
calling acpi_scan_dep_init() and before calling acpi_device_add(),
acpi_scan_clear_dep() will not find the device object corresponding
to the consumer device ACPI handle and it will not update its
dep_unmet counter to reflect the deletion of the list entry.
Consequently, the dep_unmet counter of the device will never
become zero going forward which may prevent it from being
completely enumerated.

To address this problem, modify acpi_add_single_object() to run
acpi_tie_acpi_dev(), to attach the ACPI device object created by it
to the corresponding ACPI namespace node, under acpi_dep_list_lock
along with acpi_scan_dep_init() whenever the latter is called.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>

Move the invocation of acpi_attach_data() in acpi_device_add()
into a separate function.

No intentional functional impact.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>

In general, acpi_bus_attach() can only be run safely under
acpi_scan_lock, but that lock cannot be acquired under
acpi_dep_list_lock, so make acpi_scan_clear_dep() schedule deferred
execution of acpi_bus_attach() under acpi_scan_lock instead of
calling it directly.

This also fixes a possible race between acpi_scan_clear_dep() and
device removal that might cause a device object that went away to
be accessed, because acpi_scan_clear_dep() is changed to acquire
a reference on the consumer device object.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>

Because acpi_walk_dep_device_list() is only called by the code in the
file in which it is defined, make it static, drop the export of it
and drop its header from acpi.h.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>

Make acpi_dev_get_first_consumer_dev_cb() a bit more straightforward
and rewrite the comment in it.

No functional impact.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>

Since acpi_bus_put_acpi_device() is a synonym for acpi_dev_put(),
define it as static inline in analogy with the latter.

No functional impact.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>