The ->init() open codes the functionality of DMI matching code.
Moreover, all DMI quirks are using the same callback and driver_data.
With this in mind, refactor the DMI matching code.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20240423210706.3709568-1-andriy.shevchenko@linux.intel.com
Signed-off-by: Lee Jones <lee@kernel.org>

scnprintf() never returns negative value, drop the respective dead code.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20240223195113.880121-7-andriy.shevchenko@linux.intel.com
Signed-off-by: Lee Jones <lee@kernel.org>

Since platform_device_unregister() is NULL-aware, we don't need
to duplicate this check. Remove it and fold the rest of the code.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20240223195113.880121-6-andriy.shevchenko@linux.intel.com
Signed-off-by: Lee Jones <lee@kernel.org>

Use the `PLATFORM_DEVID_NONE` constant instead of hard-coding -1
when creating a platform device.

No functional changes are intended.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20240223195113.880121-5-andriy.shevchenko@linux.intel.com
Signed-off-by: Lee Jones <lee@kernel.org>

Use platform_device_register_full() instead of open coding this
function.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20240223195113.880121-4-andriy.shevchenko@linux.intel.com
Signed-off-by: Lee Jones <lee@kernel.org>

Instead of creating driver-specific device attributes with
sysfs_create_group() have device core do this by setting up dev_groups
pointer in the driver structure.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20240223195113.880121-3-andriy.shevchenko@linux.intel.com
Signed-off-by: Lee Jones <lee@kernel.org>

There is no need to include and use entire ACPI stack in the driver.
Replace respective pieces by agnostic code. No functional change
indented.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20240223195113.880121-2-andriy.shevchenko@linux.intel.com
Signed-off-by: Lee Jones <lee@kernel.org>

The current implementation to retrieve ACPI resources is faulty
and may cause issues that even can lead to non-booting systems.

When adding data from an ACPI device, the resources are already
assigned to the platform device. Therefore there is no need to
retrieve the resource list from ACPI and manually assign it to
the platform device. Also there shouldn't be any BIOS in the wild
anymore, that does not have resources added to the KEMPLD ACPI
data.

In particular this fixes an issue where the retrieval of the
resource list using /proc/ioports is disturbed and does not list
the assigned resource for the kempld device or even no resources
at all.
On some distributions this also leads to problems during system
initialization (e.g. with udev) and causes the system to not
boot at all.

I have reproduced the issue with the following kernel versions:
    5.10.209
    5.15.148
    6.1.25
    6.6.17
    6.7.5
    6.8-rc5

The patch applies to all of those versions and seems to resolve
the issue.

Signed-off-by: Michael Brunner <michael.brunner@kontron.com>
Link: https://lore.kernel.org/r/af8756be81c9062f9543d2e5d9373cf5e7877b1e.camel@kontron.com
Signed-off-by: Lee Jones <lee@kernel.org>

The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.

To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().

Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Link: https://lore.kernel.org/r/20231123165627.492259-8-u.kleine-koenig@pengutronix.de
Signed-off-by: Lee Jones <lee@kernel.org>

Follow the advice of the Documentation/filesystems/sysfs.rst and show()
should only use sysfs_emit() or sysfs_emit_at() when formatting the
value to be returned to user space.

Signed-off-by: ye xingchen <ye.xingchen@zte.com.cn>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/202212021528368292334@zte.com.cn