ACPI / hotplug: Fix panic on eject to ejected device
When an eject request is sent to an ejected ACPI device, the following
panic occurs:
ACPI: \_SB_.SCK3.CPU3: ACPI_NOTIFY_EJECT_REQUEST event
BUG: unable to handle kernel NULL pointer dereference at 0000000000000070
IP: [<ffffffff813a7cfe>] acpi_device_hotplug+0x10b/0x33b
:
Call Trace:
[<ffffffff813a24da>] acpi_hotplug_work_fn+0x1c/0x27
[<ffffffff8109cbe5>] process_one_work+0x175/0x430
[<ffffffff8109d7db>] worker_thread+0x11b/0x3a0
This is becase device->handler is NULL in acpi_device_hotplug().
This case was used to fail in acpi_hotplug_notify_cb() as the target
had no acpi_deivce. However, acpi_device now exists after ejection.
Added a check to verify if acpi_device->handler is valid for an
eject request in acpi_hotplug_notify_cb(). Note that handler passed
from an argument is still valid while acpi_device->handler is NULL.
Fixes: 202317a573b2 (ACPI / scan: Add acpi_device objects for all device nodes in the namespace)
Signed-off-by: Toshi Kani <toshi.kani@hp.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c
index 7384158..57b053f 100644
--- a/drivers/acpi/scan.c
+++ b/drivers/acpi/scan.c
@@ -484,7 +484,6 @@
static void acpi_hotplug_notify_cb(acpi_handle handle, u32 type, void *data)
{
u32 ost_code = ACPI_OST_SC_NON_SPECIFIC_FAILURE;
- struct acpi_scan_handler *handler = data;
struct acpi_device *adev;
acpi_status status;
@@ -500,7 +499,10 @@
break;
case ACPI_NOTIFY_EJECT_REQUEST:
acpi_handle_debug(handle, "ACPI_NOTIFY_EJECT_REQUEST event\n");
- if (!handler->hotplug.enabled) {
+ if (!adev->handler)
+ goto err_out;
+
+ if (!adev->handler->hotplug.enabled) {
acpi_handle_err(handle, "Eject disabled\n");
ost_code = ACPI_OST_SC_EJECT_NOT_SUPPORTED;
goto err_out;