Message ID | 20210521171418.393871-8-hdegoede@redhat.com (mailing list archive) |
---|---|
State | Superseded |
Headers | show |
Series | iio: accel: bmc150: Add support for yoga's with dual accelerometers with an ACPI HID of DUAL250E | expand |
On Fri, 21 May 2021 19:14:17 +0200 Hans de Goede <hdegoede@redhat.com> wrote: > Some 360 degree hinges (yoga) style 2-in-1 devices use 2 bmc150 accels > to allow the OS to determine the angle between the display and the base > of the device, so that the OS can determine if the 2-in-1 is in laptop > or in tablet-mode. > > On Windows both accelerometers are read (polled) by a special service > and this service calls the DSM (Device Specific Method), which in turn > translates the angles to one of laptop/tablet/tent/stand mode and then > notifies the EC about the new mode and the EC then enables or disables > the builtin keyboard and touchpad based in the mode. > > When the 2-in-1 is powered-on or resumed folded in tablet mode the > EC senses this independent of the DSM by using a HALL effect sensor > which senses that the keyboard has been folded away behind the display. > > At power-on or resume the EC disables the keyboard based on this and > the only way to get the keyboard to work after this is to call the > DSM to re-enable it. > > Call the DSM on probe() and resume() to fix the keyboard not working > when powered-on / resumed in tablet-mode. > > This patch was developed and tested on a Lenovo Yoga 300-IBR. > > Signed-off-by: Hans de Goede <hdegoede@redhat.com> Putting aside my general grumpiness at this stuff having to pollute accelerometer drivers (and the broken DSM implementation on the hardware!) this is a fairly clean implementation so I guess we can survive it. Jonathan > --- > drivers/iio/accel/bmc150-accel-core.c | 3 + > drivers/iio/accel/bmc150-accel-i2c.c | 109 ++++++++++++++++++++++++++ > drivers/iio/accel/bmc150-accel.h | 3 + > 3 files changed, 115 insertions(+) > > diff --git a/drivers/iio/accel/bmc150-accel-core.c b/drivers/iio/accel/bmc150-accel-core.c > index 0291512648b2..932007895f18 100644 > --- a/drivers/iio/accel/bmc150-accel-core.c > +++ b/drivers/iio/accel/bmc150-accel-core.c > @@ -1803,6 +1803,9 @@ static int bmc150_accel_resume(struct device *dev) > bmc150_accel_fifo_set_mode(data); > mutex_unlock(&data->mutex); > > + if (data->resume_callback) > + data->resume_callback(dev); > + > return 0; > } > #endif > diff --git a/drivers/iio/accel/bmc150-accel-i2c.c b/drivers/iio/accel/bmc150-accel-i2c.c > index 1dd7b8a9a382..31256c32a33c 100644 > --- a/drivers/iio/accel/bmc150-accel-i2c.c > +++ b/drivers/iio/accel/bmc150-accel-i2c.c > @@ -28,6 +28,107 @@ static const struct acpi_device_id bmc150_acpi_dual_accel_ids[] = { > { }, > }; > > +/* > + * The DUAL250E ACPI device for 360° hinges type 2-in-1s with 1 accelerometer > + * in the display and 1 in the hinge has an ACPI-method (DSM) to tell the > + * ACPI code about the angle between the 2 halves. This will make the ACPI > + * code enable/disable the keyboard and touchpad. We need to call this to avoid > + * the keyboard being disabled when the 2-in-1 is turned-on or resumed while > + * fully folded into tablet mode (which gets detected with a HALL-sensor). > + * If we don't call this then the keyboard won't work even when the 2-in-1 is > + * changed to be used in laptop mode after the power-on / resume. > + * > + * This DSM takes 2 angles, selected by setting aux0 to 0 or 1, these presumably > + * define the angle between the gravity vector measured by the accelerometer in > + * the display (aux0=0) resp. the base (aux0=1) and some reference vector. > + * The 2 angles get subtracted from each other so the reference vector does > + * not matter and we can simply leave the second angle at 0. > + */ > + > +#define BMC150_DSM_GUID "7681541e-8827-4239-8d9d-36be7fe12542" > +#define DUAL250E_SET_ANGLE_FN_INDEX 3 > + > +struct dual250e_set_angle_args { > + u32 aux0; > + u32 ang0; > + u32 rawx; > + u32 rawy; > + u32 rawz; > +} __packed; > + > +static bool bmc150_acpi_set_angle_dsm(struct i2c_client *client, u32 aux0, u32 ang0) > +{ > + struct acpi_device *adev = ACPI_COMPANION(&client->dev); > + struct dual250e_set_angle_args args = { > + .aux0 = aux0, > + .ang0 = ang0, > + }; > + union acpi_object args_obj, *obj; > + guid_t guid; > + > + if (strcmp(acpi_device_hid(adev), "DUAL250E")) > + return false; > + > + guid_parse(BMC150_DSM_GUID, &guid); > + > + if (!acpi_check_dsm(adev->handle, &guid, 0, BIT(DUAL250E_SET_ANGLE_FN_INDEX))) > + return false; > + > + /* > + * Note this triggers the following warning: > + * "ACPI Warning: \_SB.PCI0.I2C2.ACC1._DSM: Argument #4 type mismatch - > + * Found [Buffer], ACPI requires [Package]" > + * This is unavoidable since the _DSM implementation expects a "naked" > + * buffer, so wrapping it in a package will _not_ work. ouch. > + */ > + args_obj.type = ACPI_TYPE_BUFFER; > + args_obj.buffer.length = sizeof(args); > + args_obj.buffer.pointer = (u8 *)&args; > + > + obj = acpi_evaluate_dsm(adev->handle, &guid, 0, DUAL250E_SET_ANGLE_FN_INDEX, &args_obj); > + if (!obj) { > + dev_err(&client->dev, "Failed to call DSM to enable keyboard and touchpad\n"); > + return false; > + } > + > + ACPI_FREE(obj); > + return true; > +} > + > +static bool bmc150_acpi_enable_keyboard(struct i2c_client *client) > +{ > + /* > + * The EC must see a change for it to re-enable the kbd, so first set the > + * angle to 270° (tent/stand mode) and then change it to 90° (laptop mode). > + */ > + if (!bmc150_acpi_set_angle_dsm(client, 0, 270)) > + return false; > + > + /* The EC needs some time to notice the angle being changed */ > + msleep(100); > + > + return bmc150_acpi_set_angle_dsm(client, 0, 90); > +} > + > +static void bmc150_acpi_resume_work(struct work_struct *work) > +{ > + struct bmc150_accel_data *data = > + container_of(work, struct bmc150_accel_data, resume_work.work); > + > + bmc150_acpi_enable_keyboard(data->second_device); > +} > + > +static void bmc150_acpi_resume_handler(struct device *dev) > +{ > + struct bmc150_accel_data *data = iio_priv(dev_get_drvdata(dev)); > + > + /* > + * Delay the bmc150_acpi_enable_keyboard() call till after the system > + * resume has completed, otherwise it will not work. > + */ > + schedule_delayed_work(&data->resume_work, msecs_to_jiffies(1000)); > +} > + > /* > * Some acpi_devices describe 2 accelerometers in a single ACPI device, try instantiating > * a second i2c_client for an I2cSerialBusV2 ACPI resource with index 1. > @@ -55,12 +156,20 @@ static void bmc150_acpi_dual_accel_probe(struct i2c_client *client) > board_info.irq = acpi_dev_gpio_irq_get_by(adev, NULL, 1); > > data->second_device = i2c_acpi_new_device(&client->dev, 1, &board_info); > + > + if (!IS_ERR(data->second_device) && bmc150_acpi_enable_keyboard(data->second_device)) { > + INIT_DELAYED_WORK(&data->resume_work, bmc150_acpi_resume_work); > + data->resume_callback = bmc150_acpi_resume_handler; > + } > } > > static void bmc150_acpi_dual_accel_remove(struct i2c_client *client) > { > struct bmc150_accel_data *data = iio_priv(i2c_get_clientdata(client)); > > + if (data->resume_callback) > + cancel_delayed_work_sync(&data->resume_work); > + > i2c_unregister_device(data->second_device); > } > #else > diff --git a/drivers/iio/accel/bmc150-accel.h b/drivers/iio/accel/bmc150-accel.h > index 5da6fd32bac5..d67d6ed6ae77 100644 > --- a/drivers/iio/accel/bmc150-accel.h > +++ b/drivers/iio/accel/bmc150-accel.h > @@ -6,6 +6,7 @@ > #include <linux/iio/iio.h> > #include <linux/mutex.h> > #include <linux/regulator/consumer.h> > +#include <linux/workqueue.h> > > struct regmap; > struct i2c_client; > @@ -62,6 +63,8 @@ struct bmc150_accel_data { > int64_t timestamp, old_timestamp; /* Only used in hw fifo mode. */ > const struct bmc150_accel_chip_info *chip_info; > struct i2c_client *second_device; > + void (*resume_callback)(struct device *dev); > + struct delayed_work resume_work; > struct iio_mount_matrix orientation; > }; >
On Fri, May 21, 2021 at 11:22 PM Hans de Goede <hdegoede@redhat.com> wrote: > > Some 360 degree hinges (yoga) style 2-in-1 devices use 2 bmc150 accels > to allow the OS to determine the angle between the display and the base > of the device, so that the OS can determine if the 2-in-1 is in laptop > or in tablet-mode. > > On Windows both accelerometers are read (polled) by a special service > and this service calls the DSM (Device Specific Method), which in turn > translates the angles to one of laptop/tablet/tent/stand mode and then > notifies the EC about the new mode and the EC then enables or disables > the builtin keyboard and touchpad based in the mode. > > When the 2-in-1 is powered-on or resumed folded in tablet mode the > EC senses this independent of the DSM by using a HALL effect sensor > which senses that the keyboard has been folded away behind the display. > > At power-on or resume the EC disables the keyboard based on this and > the only way to get the keyboard to work after this is to call the > DSM to re-enable it. > > Call the DSM on probe() and resume() to fix the keyboard not working > when powered-on / resumed in tablet-mode. > > This patch was developed and tested on a Lenovo Yoga 300-IBR. ... > + if (strcmp(acpi_device_hid(adev), "DUAL250E")) Can we use like in the other case in this series the acpi_dev_hid_uid_match() ? Because it's actually what you are doing here and it may be better to see the same approach for this HID done in different places in the code to recognize what it is about. > + return false; ... > + /* > + * The EC must see a change for it to re-enable the kbd, so first set the > + * angle to 270° (tent/stand mode) and then change it to 90° (laptop mode). > + */ > + if (!bmc150_acpi_set_angle_dsm(client, 0, 270)) > + return false; > + /* The EC needs some time to notice the angle being changed */ > + msleep(100); I feel that you conducted a research and answer to the following will be no, but... Do we have any means of polling something (embedded controller / ASL / etc) to actually see the ACK for the action? > + return bmc150_acpi_set_angle_dsm(client, 0, 90); ... > + schedule_delayed_work(&data->resume_work, msecs_to_jiffies(1000)); Isn't it the same as 1 * HZ ?
Hi, On 5/22/21 8:21 PM, Andy Shevchenko wrote: > On Fri, May 21, 2021 at 11:22 PM Hans de Goede <hdegoede@redhat.com> wrote: >> >> Some 360 degree hinges (yoga) style 2-in-1 devices use 2 bmc150 accels >> to allow the OS to determine the angle between the display and the base >> of the device, so that the OS can determine if the 2-in-1 is in laptop >> or in tablet-mode. >> >> On Windows both accelerometers are read (polled) by a special service >> and this service calls the DSM (Device Specific Method), which in turn >> translates the angles to one of laptop/tablet/tent/stand mode and then >> notifies the EC about the new mode and the EC then enables or disables >> the builtin keyboard and touchpad based in the mode. >> >> When the 2-in-1 is powered-on or resumed folded in tablet mode the >> EC senses this independent of the DSM by using a HALL effect sensor >> which senses that the keyboard has been folded away behind the display. >> >> At power-on or resume the EC disables the keyboard based on this and >> the only way to get the keyboard to work after this is to call the >> DSM to re-enable it. >> >> Call the DSM on probe() and resume() to fix the keyboard not working >> when powered-on / resumed in tablet-mode. >> >> This patch was developed and tested on a Lenovo Yoga 300-IBR. > > ... > >> + if (strcmp(acpi_device_hid(adev), "DUAL250E")) > > Can we use like in the other case in this series the acpi_dev_hid_uid_match() ? I agree that that would be more consistent, I'll fix this for 2. > Because it's actually what you are doing here and it may be better to > see the same approach for this HID done in different places in the > code to recognize what it is about. > >> + return false; > > ... > >> + /* >> + * The EC must see a change for it to re-enable the kbd, so first set the >> + * angle to 270° (tent/stand mode) and then change it to 90° (laptop mode). >> + */ >> + if (!bmc150_acpi_set_angle_dsm(client, 0, 270)) >> + return false; > >> + /* The EC needs some time to notice the angle being changed */ >> + msleep(100); > > I feel that you conducted a research and answer to the following will > be no, but... > > Do we have any means of polling something (embedded controller / ASL / > etc) to actually see the ACK for the action? Nope, there is nothing to poll and I tried shorter time-outs and that lead to the EC sometimes not seeing the change. > >> + return bmc150_acpi_set_angle_dsm(client, 0, 90); > > ... > >> + schedule_delayed_work(&data->resume_work, msecs_to_jiffies(1000)); > > Isn't it the same as 1 * HZ ? Yes, but this makes it clearer that the delay is 1 second, IMHO using n * HZ is harder to read / reason about then having the delay right there in msecs. This also means less churn if it needs to change to a different amount of msecs later. Regards, Hans
diff --git a/drivers/iio/accel/bmc150-accel-core.c b/drivers/iio/accel/bmc150-accel-core.c index 0291512648b2..932007895f18 100644 --- a/drivers/iio/accel/bmc150-accel-core.c +++ b/drivers/iio/accel/bmc150-accel-core.c @@ -1803,6 +1803,9 @@ static int bmc150_accel_resume(struct device *dev) bmc150_accel_fifo_set_mode(data); mutex_unlock(&data->mutex); + if (data->resume_callback) + data->resume_callback(dev); + return 0; } #endif diff --git a/drivers/iio/accel/bmc150-accel-i2c.c b/drivers/iio/accel/bmc150-accel-i2c.c index 1dd7b8a9a382..31256c32a33c 100644 --- a/drivers/iio/accel/bmc150-accel-i2c.c +++ b/drivers/iio/accel/bmc150-accel-i2c.c @@ -28,6 +28,107 @@ static const struct acpi_device_id bmc150_acpi_dual_accel_ids[] = { { }, }; +/* + * The DUAL250E ACPI device for 360° hinges type 2-in-1s with 1 accelerometer + * in the display and 1 in the hinge has an ACPI-method (DSM) to tell the + * ACPI code about the angle between the 2 halves. This will make the ACPI + * code enable/disable the keyboard and touchpad. We need to call this to avoid + * the keyboard being disabled when the 2-in-1 is turned-on or resumed while + * fully folded into tablet mode (which gets detected with a HALL-sensor). + * If we don't call this then the keyboard won't work even when the 2-in-1 is + * changed to be used in laptop mode after the power-on / resume. + * + * This DSM takes 2 angles, selected by setting aux0 to 0 or 1, these presumably + * define the angle between the gravity vector measured by the accelerometer in + * the display (aux0=0) resp. the base (aux0=1) and some reference vector. + * The 2 angles get subtracted from each other so the reference vector does + * not matter and we can simply leave the second angle at 0. + */ + +#define BMC150_DSM_GUID "7681541e-8827-4239-8d9d-36be7fe12542" +#define DUAL250E_SET_ANGLE_FN_INDEX 3 + +struct dual250e_set_angle_args { + u32 aux0; + u32 ang0; + u32 rawx; + u32 rawy; + u32 rawz; +} __packed; + +static bool bmc150_acpi_set_angle_dsm(struct i2c_client *client, u32 aux0, u32 ang0) +{ + struct acpi_device *adev = ACPI_COMPANION(&client->dev); + struct dual250e_set_angle_args args = { + .aux0 = aux0, + .ang0 = ang0, + }; + union acpi_object args_obj, *obj; + guid_t guid; + + if (strcmp(acpi_device_hid(adev), "DUAL250E")) + return false; + + guid_parse(BMC150_DSM_GUID, &guid); + + if (!acpi_check_dsm(adev->handle, &guid, 0, BIT(DUAL250E_SET_ANGLE_FN_INDEX))) + return false; + + /* + * Note this triggers the following warning: + * "ACPI Warning: \_SB.PCI0.I2C2.ACC1._DSM: Argument #4 type mismatch - + * Found [Buffer], ACPI requires [Package]" + * This is unavoidable since the _DSM implementation expects a "naked" + * buffer, so wrapping it in a package will _not_ work. + */ + args_obj.type = ACPI_TYPE_BUFFER; + args_obj.buffer.length = sizeof(args); + args_obj.buffer.pointer = (u8 *)&args; + + obj = acpi_evaluate_dsm(adev->handle, &guid, 0, DUAL250E_SET_ANGLE_FN_INDEX, &args_obj); + if (!obj) { + dev_err(&client->dev, "Failed to call DSM to enable keyboard and touchpad\n"); + return false; + } + + ACPI_FREE(obj); + return true; +} + +static bool bmc150_acpi_enable_keyboard(struct i2c_client *client) +{ + /* + * The EC must see a change for it to re-enable the kbd, so first set the + * angle to 270° (tent/stand mode) and then change it to 90° (laptop mode). + */ + if (!bmc150_acpi_set_angle_dsm(client, 0, 270)) + return false; + + /* The EC needs some time to notice the angle being changed */ + msleep(100); + + return bmc150_acpi_set_angle_dsm(client, 0, 90); +} + +static void bmc150_acpi_resume_work(struct work_struct *work) +{ + struct bmc150_accel_data *data = + container_of(work, struct bmc150_accel_data, resume_work.work); + + bmc150_acpi_enable_keyboard(data->second_device); +} + +static void bmc150_acpi_resume_handler(struct device *dev) +{ + struct bmc150_accel_data *data = iio_priv(dev_get_drvdata(dev)); + + /* + * Delay the bmc150_acpi_enable_keyboard() call till after the system + * resume has completed, otherwise it will not work. + */ + schedule_delayed_work(&data->resume_work, msecs_to_jiffies(1000)); +} + /* * Some acpi_devices describe 2 accelerometers in a single ACPI device, try instantiating * a second i2c_client for an I2cSerialBusV2 ACPI resource with index 1. @@ -55,12 +156,20 @@ static void bmc150_acpi_dual_accel_probe(struct i2c_client *client) board_info.irq = acpi_dev_gpio_irq_get_by(adev, NULL, 1); data->second_device = i2c_acpi_new_device(&client->dev, 1, &board_info); + + if (!IS_ERR(data->second_device) && bmc150_acpi_enable_keyboard(data->second_device)) { + INIT_DELAYED_WORK(&data->resume_work, bmc150_acpi_resume_work); + data->resume_callback = bmc150_acpi_resume_handler; + } } static void bmc150_acpi_dual_accel_remove(struct i2c_client *client) { struct bmc150_accel_data *data = iio_priv(i2c_get_clientdata(client)); + if (data->resume_callback) + cancel_delayed_work_sync(&data->resume_work); + i2c_unregister_device(data->second_device); } #else diff --git a/drivers/iio/accel/bmc150-accel.h b/drivers/iio/accel/bmc150-accel.h index 5da6fd32bac5..d67d6ed6ae77 100644 --- a/drivers/iio/accel/bmc150-accel.h +++ b/drivers/iio/accel/bmc150-accel.h @@ -6,6 +6,7 @@ #include <linux/iio/iio.h> #include <linux/mutex.h> #include <linux/regulator/consumer.h> +#include <linux/workqueue.h> struct regmap; struct i2c_client; @@ -62,6 +63,8 @@ struct bmc150_accel_data { int64_t timestamp, old_timestamp; /* Only used in hw fifo mode. */ const struct bmc150_accel_chip_info *chip_info; struct i2c_client *second_device; + void (*resume_callback)(struct device *dev); + struct delayed_work resume_work; struct iio_mount_matrix orientation; };
Some 360 degree hinges (yoga) style 2-in-1 devices use 2 bmc150 accels to allow the OS to determine the angle between the display and the base of the device, so that the OS can determine if the 2-in-1 is in laptop or in tablet-mode. On Windows both accelerometers are read (polled) by a special service and this service calls the DSM (Device Specific Method), which in turn translates the angles to one of laptop/tablet/tent/stand mode and then notifies the EC about the new mode and the EC then enables or disables the builtin keyboard and touchpad based in the mode. When the 2-in-1 is powered-on or resumed folded in tablet mode the EC senses this independent of the DSM by using a HALL effect sensor which senses that the keyboard has been folded away behind the display. At power-on or resume the EC disables the keyboard based on this and the only way to get the keyboard to work after this is to call the DSM to re-enable it. Call the DSM on probe() and resume() to fix the keyboard not working when powered-on / resumed in tablet-mode. This patch was developed and tested on a Lenovo Yoga 300-IBR. Signed-off-by: Hans de Goede <hdegoede@redhat.com> --- drivers/iio/accel/bmc150-accel-core.c | 3 + drivers/iio/accel/bmc150-accel-i2c.c | 109 ++++++++++++++++++++++++++ drivers/iio/accel/bmc150-accel.h | 3 + 3 files changed, 115 insertions(+)