firmware: meson-sm: unmap out_base shmem in error path

When SM driver was introduced in [1], the code flow did not require
to unmap out_base shmem in case of errors inside probe().
During [2], the additional error path appeared, which requires unmap.

Patch adds iounmap() missed.

Links:
[1] https://lore.kernel.org/linux-amlogic/1466339944-602-2-git-send-email-carlo@caione.org/
[2] https://lore.kernel.org/linux-amlogic/1532613556-5398-1-git-send-email-narmstrong@baylibre.com/

Fixes: 0789724f86a5 ("firmware: meson_sm: Add serial number sysfs entry")
Signed-off-by: Evgeny Bachinin <EABachinin@salutedevices.com>
Acked-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20231108125604.162383-3-EABachinin@salutedevices.com
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>

firmware: meson_sm: refactor serial sysfs entry via dev_groups attrs

Introduce just another way to register sysfs serial entry:
the less code, the better in the absence of extra error-paths

Signed-off-by: Evgeny Bachinin <EABachinin@salutedevices.com>
Acked-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20231108125604.162383-2-EABachinin@salutedevices.com
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>

firmware: meson: Use device_get_match_data()

Use preferred device_get_match_data() instead of of_match_device() to
get the driver match data. With this, adjust the includes to explicitly
include the correct headers.

Signed-off-by: Rob Herring <robh@kernel.org>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20231006224644.445295-1-robh@kernel.org
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>

drivers: meson: sm: correct meson_sm_* API retval handling

1. Following the ARM SMC32 calling convention, the return value
from secure monitor is a 32-bit signed integer. This patch changes
the type of the return value of the function meson_sm_call().

2. Now, when meson_sm_call() returns a 32-bit signed integer, we need
to ensure that this value is not negative. It is important to check
that the return value is not negative in both the meson_sm_call_read()
and meson_sm_call_write() functions.

3. Add a comment explaining why it is necessary to check if the SMC
return value is equal to 0 in the function meson_sm_call_read().
It is not obvious when reading this code.

Signed-off-by: Alexey Romanov <avromanov@salutedevices.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20230830140850.17130-1-avromanov@salutedevices.com
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>

firmware: meson_sm: fix to avoid potential NULL pointer dereference

of_match_device() may fail and returns a NULL pointer.

Fix this by checking the return value of of_match_device.

Fixes: 8cde3c2153e8 ("firmware: meson_sm: Rework driver as a proper platform driver")
Signed-off-by: Zhang Shurong <zhang_shurong@foxmail.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/tencent_AA08AAA6C4F34D53ADCE962E188A879B8206@qq.com
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>

firmware: meson_sm: populate platform devices from sm device tree data

In some meson boards, secure monitor device has children, for example,
power secure controller. By default, secure monitor isn't the bus in terms
of device tree subsystem, so the of_platform initialization code doesn't
populate its device tree data. As a result, secure monitor's children
aren't probed at all.

Run the 'of_platform_populate()' routine manually to resolve such issues.

Signed-off-by: Dmitry Rokosov <ddrokosov@sberdevices.ru>
Acked-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Link: https://lore.kernel.org/r/20230324145557.27797-1-ddrokosov@sberdevices.ru
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>

firmware: meson_sm: stop using 0 as NULL pointer

Use NULL for NULL pointer to fix the following sparse warning:
drivers/firmware/meson/meson_sm.c:85:24: sparse: warning: Using plain integer as NULL pointer

Signed-off-by: Miles Chen <miles.chen@mediatek.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20230110031242.4917-1-miles.chen@mediatek.com
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>

firmware: meson-sm: enable build as module

Enable secure module driver as module.
Default remains built-in.

Signed-off-by: Kevin Hilman <khilman@baylibre.com>

firmware: meson_sm: Add secure power domain support

The Amlogic Meson A1/C1 Secure Monitor implements calls to control power
domain.

Signed-off-by: Jianxin Pan <jianxin.pan@amlogic.com>
Signed-off-by: Kevin Hilman <khilman@baylibre.com>
Link: https://lore.kernel.org/r/1579087831-94965-2-git-send-email-jianxin.pan@amlogic.com

firmware: meson_sm: use %*ph to print small buffer

Use %*ph format to print small buffer as hex string.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Kevin Hilman <khilman@baylibre.com>

firmware: meson_sm: Rework driver as a proper platform driver

The secure monitor driver is currently a frankenstein driver which is
registered as a platform driver but its functionality goes through a
global struct accessed by the consumer drivers using exported helper
functions.

Try to tidy up the driver moving the firmware struct into the driver
data and make the consumer drivers referencing the secure-monitor using
a new property in the DT.

Currently only the nvmem driver is using this API so we can fix it in
the same commit.

Reviewed-by: Jerome Brunet <jbrunet@baylibre.com>
Signed-off-by: Carlo Caione <ccaione@baylibre.com>
Signed-off-by: Kevin Hilman <khilman@baylibre.com>

firmware: meson_sm: Mark chip struct as static const

No need to be a global struct.

Reviewed-by: Jerome Brunet <jbrunet@baylibre.com>
Signed-off-by: Carlo Caione <ccaione@baylibre.com>
Signed-off-by: Kevin Hilman <khilman@baylibre.com>

treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 446

Based on 1 normalized pattern(s):

this program is free software you can redistribute it and or modify
it under the terms of the gnu general public license version 2 as
published by the free software foundation you should have received a
copy of the gnu general public license along with this program if
not see http www gnu org licenses

extracted by the scancode license scanner the SPDX license identifier

GPL-2.0-only

has been chosen to replace the boilerplate/reference in 30 file(s).

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Allison Randal <allison@lohutok.net>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190531190115.962665879@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

firmware: meson_sm: Add serial number sysfs entry

The Amlogic Meson SoC Secure Monitor implements a call to retrieve an unique
SoC ID starting from the GX Family and all new families.

The serial number is simply exposed as a sysfs entry under the firmware
sysfs directory.

Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
Signed-off-by: Kevin Hilman <khilman@baylibre.com>

firmware: meson-sm: rework meson_sm_init to use module_platform_driver_probe

Commit 3aa0582fdb82 ("of: platform: populate /firmware/ node from
of_platform_default_populate_init()") takes care of populating
all the devices under the /firmware/ node in of_platform_default_populate_init()

This patch reworks meson_sm_init to use module_platform_driver_probe as
the platform device is populated.

Cc: Carlo Caione <carlo@caione.org>
Cc: Kevin Hilman <khilman@baylibre.com>
Cc: linux-amlogic@lists.infradead.org
Signed-off-by: Sudeep Holla <sudeep.holla@arm.com>
Signed-off-by: Kevin Hilman <khilman@baylibre.com>

firmware: meson-sm: Allow 0 as valid return value

Some special SMC calls (i.e. the function used to retrieve the serial
number of the Amlogic SoCs) returns 0 in the register 0 also when the
data was successfully read instead of using the register to hold the
number of bytes returned in the bounce buffer as expected.

With the current implementation of the driver this is seen as an error
and meson_sm_call_read() returns an error even though the data was
correctly read.

To deal with this when we have no information about the amount of read
data (that is 0 is returned by the SMC call) we return to the caller
the requested amount of data and 0 as return value.

Signed-off-by: Carlo Caione <carlo@endlessm.com>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Kevin Hilman <khilman@baylibre.com>

firmware: meson-sm: Check for buffer output size

After the data is read by the secure monitor driver it is being copied
in the output buffer checking only the size of the bounce buffer but not
the size of the output buffer.

Fix this in the secure monitor driver slightly changing the API. Fix
also the efuse driver that it is the only driver using this API to not
break bisectability.

Signed-off-by: Carlo Caione <carlo@endlessm.com>
Acked-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org> # for nvmem
Acked-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Kevin Hilman <khilman@baylibre.com>

firmware: Amlogic: Add secure monitor driver

Introduce a driver to provide calls into secure monitor mode.

In the Amlogic SoCs these calls are used for multiple reasons: access to
NVMEM, set USB boot, enable JTAG, etc...

Acked-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Carlo Caione <carlo@endlessm.com>
[khilman: add in SZ_4K cleanup]
Signed-off-by: Kevin Hilman <khilman@baylibre.com>