Attribute "kb_wake_angle" in /sys/<cros_ec hw path>/chromeos/cros_ec,
used to set at which angle the embedded controller must wake up the
host, is only set when there is at least 2 accelerometers in the
chromebook.
The detection of these sensors is done in cros_ec_sensorhub, driver that
can be probed after the cros_ec_sysfs driver that sets the attribute.
Therefore, we need to upgrade the cros_ec sysfs groups in the sensorhub
probe routine.
The first patch cleans up cros_ec_sysfs by using .dev_groups driver
field, the second patch fixes the problem.
Gwendal Grignou (2):
platform/chrome: Use dev_groups to set device sysfs attributes
platform/chrome: Update cros_ec sysfs attributes on sensors discovery
drivers/platform/chrome/cros_ec_sensorhub.c | 5 ++++-
drivers/platform/chrome/cros_ec_sysfs.c | 22 +++++++--------------
include/linux/platform_data/cros_ec_proto.h | 2 ++
3 files changed, 13 insertions(+), 16 deletions(-)
--
2.31.0.208.g409f899ff0-goog
When cros_ec_sysfs probe is called before cros_ec_sensorhub probe
routine, the |kb_wake_angle| attribute will not be displayed, even if
there are two accelerometers in the chromebook.
Call sysfs_update_groups() when accelerometers are enumerated if the
cros_ec sysfs attributes groups have already been created.
Fixes: d60ac88a62df ("mfd / platform / iio: cros_ec: Register sensor through sensorhub")
Cc: [email protected]
Signed-off-by: Gwendal Grignou <[email protected]>
---
Changes in v2:
Rebase after .dev_groups is used.
drivers/platform/chrome/cros_ec_sensorhub.c | 6 +++++-
drivers/platform/chrome/cros_ec_sysfs.c | 10 ++++++++++
include/linux/platform_data/cros_ec_proto.h | 2 ++
3 files changed, 16 insertions(+), 1 deletion(-)
diff --git a/drivers/platform/chrome/cros_ec_sensorhub.c b/drivers/platform/chrome/cros_ec_sensorhub.c
index 9c4af76a9956e..c69fec935fb50 100644
--- a/drivers/platform/chrome/cros_ec_sensorhub.c
+++ b/drivers/platform/chrome/cros_ec_sensorhub.c
@@ -106,8 +106,11 @@ static int cros_ec_sensorhub_register(struct device *dev,
sensor_type[sensorhub->resp->info.type]++;
}
- if (sensor_type[MOTIONSENSE_TYPE_ACCEL] >= 2)
+ if (sensor_type[MOTIONSENSE_TYPE_ACCEL] >= 2) {
ec->has_kb_wake_angle = true;
+ if (ec->groups && sysfs_update_groups(&ec->class_dev.kobj, ec->groups))
+ dev_warn(dev, "Unable to update cros-ec-sysfs");
+ }
if (cros_ec_check_features(ec,
EC_FEATURE_REFINED_TABLET_MODE_HYSTERESIS)) {
diff --git a/drivers/platform/chrome/cros_ec_sysfs.c b/drivers/platform/chrome/cros_ec_sysfs.c
index 114b9dbe981e7..b363f70270a38 100644
--- a/drivers/platform/chrome/cros_ec_sysfs.c
+++ b/drivers/platform/chrome/cros_ec_sysfs.c
@@ -340,11 +340,21 @@ static const struct attribute_group *cros_ec_attr_groups[] = {
NULL,
};
+static int cros_ec_sysfs_probe(struct platform_device *pd)
+{
+ struct cros_ec_dev *ec_dev = dev_get_drvdata(pd->dev.parent);
+
+ ec_dev->groups = cros_ec_attr_groups;
+
+ return 0;
+}
+
static struct platform_driver cros_ec_sysfs_driver = {
.driver = {
.name = DRV_NAME,
.dev_groups = cros_ec_attr_groups,
},
+ .probe = cros_ec_sysfs_probe,
};
module_platform_driver(cros_ec_sysfs_driver);
diff --git a/include/linux/platform_data/cros_ec_proto.h b/include/linux/platform_data/cros_ec_proto.h
index 02599687770c5..de940112f53ae 100644
--- a/include/linux/platform_data/cros_ec_proto.h
+++ b/include/linux/platform_data/cros_ec_proto.h
@@ -191,6 +191,7 @@ struct cros_ec_platform {
/**
* struct cros_ec_dev - ChromeOS EC device entry point.
* @class_dev: Device structure used in sysfs.
+ * @groups: sysfs attributes groups for this EC.
* @ec_dev: cros_ec_device structure to talk to the physical device.
* @dev: Pointer to the platform device.
* @debug_info: cros_ec_debugfs structure for debugging information.
@@ -200,6 +201,7 @@ struct cros_ec_platform {
*/
struct cros_ec_dev {
struct device class_dev;
+ const struct attribute_group **groups;
struct cros_ec_device *ec_dev;
struct device *dev;
struct cros_ec_debugfs *debug_info;
--
2.31.0.208.g409f899ff0-goog
Instead of manually call sysfs_create_group()/sysfs_remove_group(), set
.dev_groups driver data structure, and let the bus base code create
sysfs attributes.
Fixes: 6fd7f2bbd442 ("mfd / platform: cros_ec: Move device sysfs attributes to its own driver")
Cc: [email protected]
Signed-off-by: Gwendal Grignou <[email protected]>
---
New in v2.
drivers/platform/chrome/cros_ec_sysfs.c | 28 +++++--------------------
1 file changed, 5 insertions(+), 23 deletions(-)
diff --git a/drivers/platform/chrome/cros_ec_sysfs.c b/drivers/platform/chrome/cros_ec_sysfs.c
index f07eabcf9494c..114b9dbe981e7 100644
--- a/drivers/platform/chrome/cros_ec_sysfs.c
+++ b/drivers/platform/chrome/cros_ec_sysfs.c
@@ -335,34 +335,16 @@ static const struct attribute_group cros_ec_attr_group = {
.is_visible = cros_ec_ctrl_visible,
};
-static int cros_ec_sysfs_probe(struct platform_device *pd)
-{
- struct cros_ec_dev *ec_dev = dev_get_drvdata(pd->dev.parent);
- struct device *dev = &pd->dev;
- int ret;
-
- ret = sysfs_create_group(&ec_dev->class_dev.kobj, &cros_ec_attr_group);
- if (ret < 0)
- dev_err(dev, "failed to create attributes. err=%d\n", ret);
-
- return ret;
-}
-
-static int cros_ec_sysfs_remove(struct platform_device *pd)
-{
- struct cros_ec_dev *ec_dev = dev_get_drvdata(pd->dev.parent);
-
- sysfs_remove_group(&ec_dev->class_dev.kobj, &cros_ec_attr_group);
-
- return 0;
-}
+static const struct attribute_group *cros_ec_attr_groups[] = {
+ &cros_ec_attr_group,
+ NULL,
+};
static struct platform_driver cros_ec_sysfs_driver = {
.driver = {
.name = DRV_NAME,
+ .dev_groups = cros_ec_attr_groups,
},
- .probe = cros_ec_sysfs_probe,
- .remove = cros_ec_sysfs_remove,
};
module_platform_driver(cros_ec_sysfs_driver);
--
2.31.0.208.g409f899ff0-goog
[+dtor]
Is this change acceptable? I was worried it could break when
asynchronous probing is used [https://lwn.net/Articles/629895/], but
since all probes are deferred in a single thread, it is safe.
Gwendal.
Gwendal.
On Mon, Apr 5, 2021 at 1:29 PM Gwendal Grignou <[email protected]> wrote:
>
> Attribute "kb_wake_angle" in /sys/<cros_ec hw path>/chromeos/cros_ec,
> used to set at which angle the embedded controller must wake up the
> host, is only set when there is at least 2 accelerometers in the
> chromebook.
>
> The detection of these sensors is done in cros_ec_sensorhub, driver that
> can be probed after the cros_ec_sysfs driver that sets the attribute.
> Therefore, we need to upgrade the cros_ec sysfs groups in the sensorhub
> probe routine.
>
> The first patch cleans up cros_ec_sysfs by using .dev_groups driver
> field, the second patch fixes the problem.
>
> Gwendal Grignou (2):
> platform/chrome: Use dev_groups to set device sysfs attributes
> platform/chrome: Update cros_ec sysfs attributes on sensors discovery
>
> drivers/platform/chrome/cros_ec_sensorhub.c | 5 ++++-
> drivers/platform/chrome/cros_ec_sysfs.c | 22 +++++++--------------
> include/linux/platform_data/cros_ec_proto.h | 2 ++
> 3 files changed, 13 insertions(+), 16 deletions(-)
>
> --
> 2.31.0.208.g409f899ff0-goog
>
Hi Gwendal,
On Thu, May 27, 2021 at 2:01 PM Gwendal Grignou <[email protected]> wrote:
>
> [+dtor]
> Is this change acceptable? I was worried it could break when
> asynchronous probing is used [https://lwn.net/Articles/629895/], but
> since all probes are deferred in a single thread, it is safe.
I think this is a bit awkward that we need to poke a separate sub-driver.
Have you considered having cros_ec_sensorhub.c create its own
attribute group (it does not have to have a name and it looks like one
can register as many unnamed groups as they want) and have wake angle
show and store methods directly in cros_ec_sensorhub.c?
Thanks,
Dmitry
The v2 patches I submitted were wrong, as cros_ec sysfs driver is
modifying the attributes of the class driver, not its own. I repost a
patch that takes that into account.
Gwendal.
On Sun, May 30, 2021 at 10:36 PM Dmitry Torokhov <[email protected]> wrote:
>
> Hi Gwendal,
>
> On Thu, May 27, 2021 at 2:01 PM Gwendal Grignou <[email protected]> wrote:
> >
> > [+dtor]
> > Is this change acceptable? I was worried it could break when
> > asynchronous probing is used [https://lwn.net/Articles/629895/], but
> > since all probes are deferred in a single thread, it is safe.
>
> I think this is a bit awkward that we need to poke a separate sub-driver.
>
> Have you considered having cros_ec_sensorhub.c create its own
> attribute group (it does not have to have a name and it looks like one
> can register as many unnamed groups as they want) and have wake angle
> show and store methods directly in cros_ec_sensorhub.c?
>
> Thanks,
> Dmitry