Skip to contentSkip to conversation

ACPI: scan: Fix _STA getting called on devices with unmet dependencies

PATCH Applied Hans de Goede 2 participants 3 messages Mar 29, 2021 – Mar 30, 2021 ACPI

v2 of 3 Applied Latest is v3 →

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 ACPI: scan: Fix _STA getting called on devices with unmet dependencies

Commit 71da201f38df ("ACPI: scan: Defer enumeration of devices with _DEP lists") dropped the following 2 lines from acpi_init_device_object():

	/* Assume there are unmet deps until acpi_device_dep_initialize() runs */
	device->dep_unmet = 1;

Leaving the initial value of dep_unmet at the 0 from the kzalloc(). This causes the acpi_bus_get_status() call in acpi_add_single_object() to actually call _STA, even though there maybe unmet deps, leading to errors like these:

[    0.123579] ACPI Error: No handler for Region [ECRM] (00000000ba9edc4c)
               [GenericSerialBus] (20170831/evregion-166)
[    0.123601] ACPI Error: Region GenericSerialBus (ID=9) has no handler
               (20170831/exfldio-299)
[    0.123618] ACPI Error: Method parse/execution failed
               \_SB.I2C1.BAT1._STA, AE_NOT_EXIST (20170831/psparse-550)

Fix this by re-adding the dep_unmet = 1 initialization to acpi_init_device_object() and modifying acpi_bus_check_add() to make sure that dep_unmet always gets setup there, overriding the initial 1 value.

This re-fixes the issue initially fixed by
commit 63347db0affa ("ACPI / scan: Use acpi_bus_get_status() to initialize ACPI_TYPE_DEVICE devs"), which introduced the removed "device->dep_unmet = 1;" statement.

This issue was noticed; and the fix tested on a Dell Venue 10 Pro 5055.

Fixes 71da201f38df ("ACPI: scan: Defer enumeration of devices with _DEP lists")
---
 drivers/acpi/scan.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)
drivers/acpi/scan.c +9 -1
index 1584c9e463bd..aca4edc8fe6b 100644
@@ -1647,6 +1647,8 @@ void acpi_init_device_object(struct acpi_device *device, acpi_handle handle,
 	device_initialize(&device->dev);
 	dev_set_uevent_suppress(&device->dev, true);
 	acpi_init_coherency(device);
+	/* Assume there are unmet deps to start with. */
+	device->dep_unmet = 1;
 }
 
 void acpi_device_add_finalize(struct acpi_device *device)
@@ -1957,7 +1959,13 @@ static acpi_status acpi_bus_check_add(acpi_handle handle, bool check_dep,
 		return AE_CTRL_DEPTH;
 
 	acpi_scan_init_hotplug(device);
-	if (!check_dep)
+	/*
+	 * If check_dep is true at this point, the device has no dependencies,
+	 * or the creation of the device object would have been postponed above.
+	 */
+	if (check_dep)
+		device->dep_unmet = 0;
+	else
 		acpi_scan_dep_init(device);
 
 out:
Signature · 1 line
2.30.2
maintainer Applied
#1 lore raw
Quoted 28 lines · Hans de Goede

Commit 71da201f38df ("ACPI: scan: Defer enumeration of devices with _DEP lists") dropped the following 2 lines from acpi_init_device_object():

        /* Assume there are unmet deps until acpi_device_dep_initialize() runs */
        device->dep_unmet = 1;

Leaving the initial value of dep_unmet at the 0 from the kzalloc(). This causes the acpi_bus_get_status() call in acpi_add_single_object() to actually call _STA, even though there maybe unmet deps, leading to errors like these:

[    0.123579] ACPI Error: No handler for Region [ECRM] (00000000ba9edc4c)
               [GenericSerialBus] (20170831/evregion-166)
[    0.123601] ACPI Error: Region GenericSerialBus (ID=9) has no handler
               (20170831/exfldio-299)
[    0.123618] ACPI Error: Method parse/execution failed
               \_SB.I2C1.BAT1._STA, AE_NOT_EXIST (20170831/psparse-550)

Fix this by re-adding the dep_unmet = 1 initialization to acpi_init_device_object() and modifying acpi_bus_check_add() to make sure that dep_unmet always gets setup there, overriding the initial 1 value.

This re-fixes the issue initially fixed by
commit 63347db0affa ("ACPI / scan: Use acpi_bus_get_status() to initialize ACPI_TYPE_DEVICE devs"), which introduced the removed "device->dep_unmet = 1;" statement.

Quoted 4 lines

This issue was noticed; and the fix tested on a Dell Venue 10 Pro 5055.

Fixes 71da201f38df ("ACPI: scan: Defer enumeration of devices with _DEP lists")

Applied as 5.12-rc material, thanks!

#2 lore raw
In reply to Hans de Goede

Hi Rafael,

Quoted 27 lines · Hans de Goede

Commit 71da201f38df ("ACPI: scan: Defer enumeration of devices with _DEP lists") dropped the following 2 lines from acpi_init_device_object():

	/* Assume there are unmet deps until acpi_device_dep_initialize() runs */
	device->dep_unmet = 1;

Leaving the initial value of dep_unmet at the 0 from the kzalloc(). This causes the acpi_bus_get_status() call in acpi_add_single_object() to actually call _STA, even though there maybe unmet deps, leading to errors like these:

[    0.123579] ACPI Error: No handler for Region [ECRM] (00000000ba9edc4c)
               [GenericSerialBus] (20170831/evregion-166)
[    0.123601] ACPI Error: Region GenericSerialBus (ID=9) has no handler
               (20170831/exfldio-299)
[    0.123618] ACPI Error: Method parse/execution failed
               \_SB.I2C1.BAT1._STA, AE_NOT_EXIST (20170831/psparse-550)

Fix this by re-adding the dep_unmet = 1 initialization to acpi_init_device_object() and modifying acpi_bus_check_add() to make sure that dep_unmet always gets setup there, overriding the initial 1 value.

This re-fixes the issue initially fixed by
commit 63347db0affa ("ACPI / scan: Use acpi_bus_get_status() to initialize ACPI_TYPE_DEVICE devs"), which introduced the removed "device->dep_unmet = 1;" statement.

Quoted 4 lines

This issue was noticed; and the fix tested on a Dell Venue 10 Pro 5055.

Fixes 71da201f38df ("ACPI: scan: Defer enumeration of devices with _DEP lists")

Self-nack, sorry, I know you already merged this, but I hope you can still replace it, as this is broken. It causes dep_unmet to never reach 0 for devices with deps because it is missing this:

drivers/acpi/scan.c +2 -0
@@ -1910,6 +1910,8 @@ static void acpi_scan_dep_init(struct acpi_device *adev)
 {
 	struct acpi_dep_data *dep;
 
+	adev->dep_unmet = 0;
+
 	mutex_lock(&acpi_dep_list_lock);
 
 	list_for_each_entry(dep, &acpi_dep_list, node) {

So now acpi_scan_dep_init() starts with 1 dep and then adds the actual number of deps on top.

I'll send a v3 fixing this as soon as I've tested it.

(I should have noticed the battery reporting was gone with v2, but I only checked that the errors were gone from dmesg...)

Let me know if you want a follow-up patch fixing this instead.

Regards,

Hans

Quoted 33 lines
---
 drivers/acpi/scan.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)
drivers/acpi/scan.c +9 -1
index 1584c9e463bd..aca4edc8fe6b 100644
@@ -1647,6 +1647,8 @@ void acpi_init_device_object(struct acpi_device *device, acpi_handle handle,
 	device_initialize(&device->dev);
 	dev_set_uevent_suppress(&device->dev, true);
 	acpi_init_coherency(device);
+	/* Assume there are unmet deps to start with. */
+	device->dep_unmet = 1;
 }
 
 void acpi_device_add_finalize(struct acpi_device *device)
@@ -1957,7 +1959,13 @@ static acpi_status acpi_bus_check_add(acpi_handle handle, bool check_dep,
 		return AE_CTRL_DEPTH;
 
 	acpi_scan_init_hotplug(device);
-	if (!check_dep)
+	/*
+	 * If check_dep is true at this point, the device has no dependencies,
+	 * or the creation of the device object would have been postponed above.
+	 */
+	if (check_dep)
+		device->dep_unmet = 0;
+	else
 		acpi_scan_dep_init(device);
 
 out: