This is a note to let you know that I've just added the patch titled
driver core: platform: change logic implementing platform_driver_probe
to the 5.10-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
driver-core-platform-change-logic-implementing-platf.patch
and it can be found in the queue-5.10 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@xxxxxxxxxxxxxxx> know about it.
commit 70fc1c8589b4de6e2f301a4307f3774ef5bce168
Author: Uwe Kleine-König <u.kleine-koenig@xxxxxxxxxxxxxx>
Date: Thu Nov 19 13:46:10 2020 +0100
driver core: platform: change logic implementing platform_driver_probe
[ Upstream commit 16085668eacdc56c46652d0f3bfef81ecace57de ]
Instead of overwriting the core driver's probe function handle probing
devices for drivers loaded by platform_driver_probe() in the platform
driver probe function.
The intended goal is to not have to change the probe function to
simplify converting the platform bus to use bus functions.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@xxxxxxxxxxxxxx>
Link: https://lore.kernel.org/r/20201119124611.2573057-2-u.kleine-koenig@xxxxxxxxxxxxxx
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
Stable-dep-of: 55c421b36448 ("mmc: davinci: Don't strip remove function when driver is builtin")
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
diff --git a/drivers/base/platform.c b/drivers/base/platform.c
index fa023cf80dc48..16426eb934632 100644
--- a/drivers/base/platform.c
+++ b/drivers/base/platform.c
@@ -743,12 +743,25 @@ struct platform_device *platform_device_register_full(
}
EXPORT_SYMBOL_GPL(platform_device_register_full);
+static int platform_probe_fail(struct platform_device *pdev);
+
static int platform_drv_probe(struct device *_dev)
{
struct platform_driver *drv = to_platform_driver(_dev->driver);
struct platform_device *dev = to_platform_device(_dev);
int ret;
+ /*
+ * A driver registered using platform_driver_probe() cannot be bound
+ * again later because the probe function usually lives in __init code
+ * and so is gone. For these drivers .probe is set to
+ * platform_probe_fail in __platform_driver_probe(). Don't even
+ * prepare clocks and PM domains for these to match the traditional
+ * behaviour.
+ */
+ if (unlikely(drv->probe == platform_probe_fail))
+ return -ENXIO;
+
ret = of_clk_set_defaults(_dev->of_node, false);
if (ret < 0)
return ret;
@@ -822,7 +835,7 @@ void platform_driver_unregister(struct platform_driver *drv)
}
EXPORT_SYMBOL_GPL(platform_driver_unregister);
-static int platform_drv_probe_fail(struct device *_dev)
+static int platform_probe_fail(struct platform_device *pdev)
{
return -ENXIO;
}
@@ -887,10 +900,9 @@ int __init_or_module __platform_driver_probe(struct platform_driver *drv,
* new devices fail.
*/
spin_lock(&drv->driver.bus->p->klist_drivers.k_lock);
- drv->probe = NULL;
+ drv->probe = platform_probe_fail;
if (code == 0 && list_empty(&drv->driver.p->klist_devices.k_list))
retval = -ENODEV;
- drv->driver.probe = platform_drv_probe_fail;
spin_unlock(&drv->driver.bus->p->klist_drivers.k_lock);
if (code != retval)
