Re: [PATCH 2/3] platform/x86: wmi: Do not instantiate older WMI drivers multiple times

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Am 28.02.24 um 14:23 schrieb Hans de Goede:

Hi Armin,

On 2/27/24 23:47, Armin Wolf wrote:
Am 27.02.24 um 21:30 schrieb Pali Rohár:

On Monday 26 February 2024 20:35:56 Armin Wolf wrote:
Many older WMI drivers cannot be instantiated multiple times for
two reasons:

- they are using the legacy GUID-based WMI API
- they are singletons (with global state)

Prevent such WMI drivers from binding to WMI devices with a duplicated
GUID, as this would mean that the WMI driver will be instantiated at
least two times (one for the original GUID and one for the duplicated
GUID).
WMI drivers which can be instantiated multiple times can signal this
by setting a flag inside struct wmi_driver.
What do you think about opposite direction? Adding ".singleton = true"
into every driver which is not compatible with duplicated initialization
and having the default value that drivers are not singletons.

But if the direction it to not accept new "legacy" drivers and start get
rid of all "legacy" drivers by rewriting them, then it does not matter
if "no_singleton" or "is_singleton" is used...
Hi,

i want to make sure that i wont forget any legacy WMI drivers. This way, the
older legacy WMI drivers automatically initialize no_singleton with false.

Also i intent to not accept new legacy WMI drivers.
Somewhat offtopic question, how do you plan to handle the case where
there are 2 WMI GUIDs for what really is a single "thing",
specifically one main WMI GUID for a vendor specific interface
for say the embedded-controller and a separate GUID for events ?

IIRC we have several such cases. I thought we even have a case
where the main WMI GUID already is bound to using wmi_bus wmi_driver,
while the event guid is listened to by using wmi_install_notify_handler()
but I cannot find the code doing this, so I might be mistaken on this.

Regards,

Hans

I am aware of those WMI device constellations, for the classic event-data constellation
i think there should be two WMI drivers (one for the event and one for the data) which
are connected with each other through a standard notifier.

I plan on writing a WMI driver developing guide which will address this by giving examples
on what to do in such cases.

The problem is that many WMI interfaces consisting of multiple GUIDs are inherently unstable
since the vendor assumes that they will just be available after the Windows userspace has loaded,
so the WMI driver has no means of discovering which GUIDs are available at probe time (other GUIDs
can be discovered later by the WMI driver core, so the WMI driver cannot see them sooner).

The solution for this depends on the specific WMI device constellation.

Armin Wolf.


Tested on a ASUS Prime B650-Plus.

Signed-off-by: Armin Wolf <W_Armin@xxxxxx>
---
   drivers/hwmon/dell-smm-hwmon.c                 |  1 +
   drivers/platform/x86/dell/dell-wmi-ddv.c       |  1 +
   drivers/platform/x86/intel/wmi/sbl-fw-update.c |  1 +
   drivers/platform/x86/intel/wmi/thunderbolt.c   |  1 +
   drivers/platform/x86/wmi-bmof.c                |  1 +
   drivers/platform/x86/wmi.c                     | 12 ++++++++++++
   include/linux/wmi.h                            |  2 ++
   7 files changed, 19 insertions(+)

diff --git a/drivers/hwmon/dell-smm-hwmon.c b/drivers/hwmon/dell-smm-hwmon.c
index 6d8c0f328b7b..168d669c4eca 100644
--- a/drivers/hwmon/dell-smm-hwmon.c
+++ b/drivers/hwmon/dell-smm-hwmon.c
@@ -1587,6 +1587,7 @@ static struct wmi_driver dell_smm_wmi_driver = {
       },
       .id_table = dell_smm_wmi_id_table,
       .probe = dell_smm_wmi_probe,
+    .no_singleton = true,
   };

   /*
diff --git a/drivers/platform/x86/dell/dell-wmi-ddv.c b/drivers/platform/x86/dell/dell-wmi-ddv.c
index db1e9240dd02..0b2299f7a2de 100644
--- a/drivers/platform/x86/dell/dell-wmi-ddv.c
+++ b/drivers/platform/x86/dell/dell-wmi-ddv.c
@@ -882,6 +882,7 @@ static struct wmi_driver dell_wmi_ddv_driver = {
       },
       .id_table = dell_wmi_ddv_id_table,
       .probe = dell_wmi_ddv_probe,
+    .no_singleton = true,
   };
   module_wmi_driver(dell_wmi_ddv_driver);

diff --git a/drivers/platform/x86/intel/wmi/sbl-fw-update.c b/drivers/platform/x86/intel/wmi/sbl-fw-update.c
index 040153ad67c1..75c82c08117f 100644
--- a/drivers/platform/x86/intel/wmi/sbl-fw-update.c
+++ b/drivers/platform/x86/intel/wmi/sbl-fw-update.c
@@ -131,6 +131,7 @@ static struct wmi_driver intel_wmi_sbl_fw_update_driver = {
       .probe = intel_wmi_sbl_fw_update_probe,
       .remove = intel_wmi_sbl_fw_update_remove,
       .id_table = intel_wmi_sbl_id_table,
+    .no_singleton = true,
   };
   module_wmi_driver(intel_wmi_sbl_fw_update_driver);

diff --git a/drivers/platform/x86/intel/wmi/thunderbolt.c b/drivers/platform/x86/intel/wmi/thunderbolt.c
index e2ad3f46f356..08df560a2c7a 100644
--- a/drivers/platform/x86/intel/wmi/thunderbolt.c
+++ b/drivers/platform/x86/intel/wmi/thunderbolt.c
@@ -63,6 +63,7 @@ static struct wmi_driver intel_wmi_thunderbolt_driver = {
           .dev_groups = tbt_groups,
       },
       .id_table = intel_wmi_thunderbolt_id_table,
+    .no_singleton = true,
   };

   module_wmi_driver(intel_wmi_thunderbolt_driver);
diff --git a/drivers/platform/x86/wmi-bmof.c b/drivers/platform/x86/wmi-bmof.c
index 644d2fd889c0..df6f0ae6e6c7 100644
--- a/drivers/platform/x86/wmi-bmof.c
+++ b/drivers/platform/x86/wmi-bmof.c
@@ -94,6 +94,7 @@ static struct wmi_driver wmi_bmof_driver = {
       .probe = wmi_bmof_probe,
       .remove = wmi_bmof_remove,
       .id_table = wmi_bmof_id_table,
+    .no_singleton = true,
   };

   module_wmi_driver(wmi_bmof_driver);
diff --git a/drivers/platform/x86/wmi.c b/drivers/platform/x86/wmi.c
index 29dfe52eb802..349deced87e8 100644
--- a/drivers/platform/x86/wmi.c
+++ b/drivers/platform/x86/wmi.c
@@ -883,6 +883,18 @@ static int wmi_dev_probe(struct device *dev)
       struct wmi_driver *wdriver = drv_to_wdrv(dev->driver);
       int ret = 0;

+    /* Some older WMI drivers will break if instantiated multiple times,
+     * so they are blocked from probing WMI devices with a duplicated GUID.
+     *
+     * New WMI drivers should support being instantiated multiple times.
+     */
+    if (test_bit(WMI_GUID_DUPLICATED, &wblock->flags) && !wdriver->no_singleton) {
+        dev_warn(dev, "Legacy driver %s cannot be instantiated multiple times\n",
+             dev->driver->name);
+
+        return -ENODEV;
+    }
+
       if (wdriver->notify) {
           if (test_bit(WMI_NO_EVENT_DATA, &wblock->flags) && !wdriver->no_notify_data)
               return -ENODEV;
diff --git a/include/linux/wmi.h b/include/linux/wmi.h
index 781958310bfb..63cca3b58d6d 100644
--- a/include/linux/wmi.h
+++ b/include/linux/wmi.h
@@ -49,6 +49,7 @@ u8 wmidev_instance_count(struct wmi_device *wdev);
    * @driver: Driver model structure
    * @id_table: List of WMI GUIDs supported by this driver
    * @no_notify_data: Driver supports WMI events which provide no event data
+ * @no_singleton: Driver can be instantiated multiple times
    * @probe: Callback for device binding
    * @remove: Callback for device unbinding
    * @notify: Callback for receiving WMI events
@@ -59,6 +60,7 @@ struct wmi_driver {
       struct device_driver driver;
       const struct wmi_device_id *id_table;
       bool no_notify_data;
+    bool no_singleton;

       int (*probe)(struct wmi_device *wdev, const void *context);
       void (*remove)(struct wmi_device *wdev);
--
2.39.2







[Index of Archives]     [LM Sensors]     [Linux Sound]     [ALSA Users]     [ALSA Devel]     [Linux Audio Users]     [Linux Media]     [Kernel]     [Gimp]     [Yosemite News]     [Linux Media]

  Powered by Linux