[ Upstream commit 5d010473cdeaabf6a2d3a9e2aed2186c1b73c213 ]

Calling fwnode_get_next_child_node() in ACPI implementation of the fwnode
property API is somewhat problematic as the latter is used in the
impelementation of the former. Instead of using
fwnode_get_next_child_node() in acpi_graph_get_next_endpoint(), call
acpi_get_next_subnode() directly instead.

Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com>
Link: https://patch.msgid.link/20251001104320.1272752-3-sakari.ailus@linux.intel.com
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

[ Upstream commit 593ee49222a0d751062fd9a5e4a963ade4ec028a ]

acpi_fwnode_graph_parse_endpoint() calls fwnode_get_parent() to obtain the
parent fwnode but returns without calling fwnode_handle_put() on it. This
potentially leads to a fwnode refcount leak and prevents the parent node
from being released properly.

Call fwnode_handle_put() on the parent fwnode before returning to prevent
the leak from occurring.

Fixes: 3b27d00e7b6d ("device property: Move fwnode graph ops to firmware specific locations")
Signed-off-by: Haotian Zhang <vulab@iscas.ac.cn>
Reviewed-by: Sakari Ailus <sakari.ailus@linux.intel.com>
[ rjw: Changelog edits ]
Link: https://patch.msgid.link/20251111075000.1828-1-vulab@iscas.ac.cn
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

A refactoring bug introduced an argument order mistake in the call to
acpi_fwnode_get_reference_args() from __acpi_node_get_property_reference().
This caused incorrect behavior when resolving ACPI property references.

Fix the issue by correcting the argument order.

Fixes: e121be784d35 ("ACPI: property: Refactor acpi_fwnode_get_reference_args() to support nargs_prop")
Reported-by: Thomas Richard <thomas.richard@bootlin.com>
Closes: https://lore.kernel.org/all/1241f2b6-9b4e-4623-8a83-77db8774ac32@bootlin.com/
Tested-by: Thomas Richard <thomas.richard@bootlin.com>
Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://patch.msgid.link/20251017100744.71871-1-sunilvl@ventanamicro.com
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

Pull more ACPI updates from Rafael Wysocki:
 "These fix a driver bug, clean up two pieces of code and improve the
  fwnode API consistency:

   - Add missing synchronization between interface updates in the ACPI
     battery driver (Rafael Wysocki)

   - Remove open coded check for cpu_feature_enabled() from
     acpi_processor_power_init_bm_check() (Mario Limonciello)

   - Remove redundant rcu_read_lock/unlock() under spinlock from
     ghes_notify_hed() in the ACPI APEI support code (pengdonglin)

   - Make the .get_next_child_node() callback in the ACPI fwnode backend
     skip ACPI devices that are not present for consistency with the
     analogous callback in the OF fwnode backend (Sakari Ailus)"

* tag 'acpi-6.18-rc1-2' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  ACPI: property: Return present device nodes only on fwnode interface
  ACPI: APEI: Remove redundant rcu_read_lock/unlock() under spinlock
  ACPI: battery: Add synchronization between interface updates
  x86/acpi/cstate: Remove open coded check for cpu_feature_enabled()

Pull more RISC-V updates from Paul Walmsley:

 - Support for the RISC-V-standardized RPMI interface.

   RPMI is a platform management communication mechanism between OSes
   running on application processors, and a remote platform management
   processor. Similar to ARM SCMI, TI SCI, etc. This includes irqchip,
   mailbox, and clk changes.

 - Support for the RISC-V-standardized MPXY SBI extension.

   MPXY is a RISC-V-specific standard implementing a shared memory
   mailbox between S-mode operating systems (e.g., Linux) and M-mode
   firmware (e.g., OpenSBI). It is part of this PR since one of its use
   cases is to enable M-mode firmware to act as a single RPMI client for
   all RPMI activity on a core (including S-mode RPMI activity).
   Includes a mailbox driver.

 - Some ACPI-related updates to enable the use of RPMI and MPXY.

 - The addition of Linux-wide memcpy_{from,to}_le32() static inline
   functions, for RPMI use.

 - An ACPI Kconfig change to enable boot logos on any ACPI-using
   architecture (including RISC-V)

 - A RISC-V defconfig change to add GPIO keyboard and event device
   support, for front panel shutdown or reboot buttons

* tag 'riscv-for-linus-6.18-mw2' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux: (26 commits)
  clk: COMMON_CLK_RPMI should depend on RISCV
  ACPI: support BGRT table on RISC-V
  MAINTAINERS: Add entry for RISC-V RPMI and MPXY drivers
  RISC-V: Enable GPIO keyboard and event device in RV64 defconfig
  irqchip/riscv-rpmi-sysmsi: Add ACPI support
  mailbox/riscv-sbi-mpxy: Add ACPI support
  irqchip/irq-riscv-imsic-early: Export imsic_acpi_get_fwnode()
  ACPI: RISC-V: Add RPMI System MSI to GSI mapping
  ACPI: RISC-V: Add support to update gsi range
  ACPI: RISC-V: Create interrupt controller list in sorted order
  ACPI: scan: Update honor list for RPMI System MSI
  ACPI: Add support for nargs_prop in acpi_fwnode_get_reference_args()
  ACPI: property: Refactor acpi_fwnode_get_reference_args() to support nargs_prop
  irqchip: Add driver for the RPMI system MSI service group
  dt-bindings: Add RPMI system MSI interrupt controller bindings
  dt-bindings: Add RPMI system MSI message proxy bindings
  clk: Add clock driver for the RISC-V RPMI clock service group
  dt-bindings: clock: Add RPMI clock service controller bindings
  dt-bindings: clock: Add RPMI clock service message proxy bindings
  mailbox: Add RISC-V SBI message proxy (MPXY) based mailbox driver
  ...

fwnode_graph_get_next_subnode() may return fwnode backed by ACPI
device nodes and there has been no check these devices are present
in the system, unlike there has been on fwnode OF backend.

In order to provide consistent behaviour towards callers,
add a check for device presence by introducing
a new function acpi_get_next_present_subnode(), used as the
get_next_child_node() fwnode operation that also checks device
node presence.

Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com>
Link: https://patch.msgid.link/20251001102636.1272722-2-sakari.ailus@linux.intel.com
[ rjw: Kerneldoc comment and changelog edits ]
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

Currently, ACPI does not support the use of a nargs_prop (e.g.,
associated with a reference in fwnode_property_get_reference_args().
Instead, ACPI expects the number of arguments (nargs) to be explicitly
passed or known.

This behavior diverges from Open Firmware (OF), which allows the use of
a #*-cells property in the referenced node to determine the number of
arguments. Since fwnode_property_get_reference_args() is a common
interface used across both OF and ACPI firmware paradigms, it is
desirable to have a unified calling convention that works seamlessly for
both.

Add the support for ACPI to parse a nargs_prop from the referenced
fwnode, aligning its behavior with the OF backend. This allows drivers
and subsystems using fwnode_property_get_reference_args() to work in a
firmware-agnostic way without having to hardcode or special-case
argument counts for ACPI.

Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Acked-by: Jassi Brar <jassisinghbrar@gmail.com>
Link: https://lore.kernel.org/r/20250818040920.272664-16-apatel@ventanamicro.com
Signed-off-by: Paul Walmsley <pjw@kernel.org>

Currently, acpi_fwnode_get_reference_args() delegates to the internal
function __acpi_node_get_property_reference() to retrieve property
references. However, this function does not handle the nargs_prop (cells
property) parameter, and instead expects the number of arguments (nargs)
to be known or hardcoded.

As a result, when fwnode_property_get_reference_args() is used with a
valid nargs_prop, the ACPI backend ignores it, whereas the Device Tree
(DT) backend uses the #*-cells property from the reference node to
determine the number of arguments dynamically.

To support the nargs_prop in ACPI, refactor the code as follows:

- Move the implementation from __acpi_node_get_property_reference()
  into acpi_fwnode_get_reference_args().

- Update __acpi_node_get_property_reference() to call the (now updated)
  acpi_fwnode_get_reference_args() passing NULL as nargs_prop to keep
  the behavior of __acpi_node_get_property_reference() intact.

Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Suggested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Acked-by: Jassi Brar <jassisinghbrar@gmail.com>
Link: https://lore.kernel.org/r/20250818040920.272664-15-apatel@ventanamicro.com
Signed-off-by: Paul Walmsley <pjw@kernel.org>

Make acpi_nondev_subnode_extract() follow the usual code flow pattern
in which failure is handled at the point where it is detected.

No intentional functional impact.

Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Tested-by: Sakari Ailus <sakari.ailus@linux.intel.com>

In certain circumstances, the ACPI handle of a data-only node may be
NULL, in which case it does not make sense to attempt to attach that
node to an ACPI namespace object, so update the code to avoid attempts
to do so.

This prevents confusing and unuseful error messages from being printed.

Also document the fact that the ACPI handle of a data-only node may be
NULL and when that happens in a code comment.  In addition, make
acpi_add_nondev_subnodes() print a diagnostic message for each data-only
node with an unknown ACPI namespace scope.

Fixes: 1d52f10917a7 ("ACPI: property: Tie data nodes to acpi handles")
Cc: 6.0+ <stable@vger.kernel.org> # 6.0+
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Tested-by: Sakari Ailus <sakari.ailus@linux.intel.com>