The accelerometer event relies on on the ACERWMID_EVENT_GUID notify.
So, this patch changes the codes to setup accelerometer input device
when detected ACERWMID_EVENT_GUID. It avoids that the accel input
device created on every acer machines.
In addition, patch adds a clearly parsing logic of accelerometer hid
to acer_wmi_get_handle_cb callback function. It is positive matching
the "SENR" name with "BST0001" device to avoid non-supported hardware.
Reported-by: Bjørn Mork <[email protected]>
Tested-by: Bjørn Mork <[email protected]>
Cc: Darren Hart <[email protected]>
Signed-off-by: Lee, Chun-Yi <[email protected]>
---
drivers/platform/x86/acer-wmi.c | 22 ++++++++++++++++++----
1 file changed, 18 insertions(+), 4 deletions(-)
diff --git a/drivers/platform/x86/acer-wmi.c b/drivers/platform/x86/acer-wmi.c
index 79d64ea..3bb3162 100644
--- a/drivers/platform/x86/acer-wmi.c
+++ b/drivers/platform/x86/acer-wmi.c
@@ -1808,11 +1808,24 @@ static int __init acer_wmi_enable_lm(void)
return status;
}
+#define ACER_WMID_ACCEL_HID "BST0001"
+
static acpi_status __init acer_wmi_get_handle_cb(acpi_handle ah, u32 level,
void *ctx, void **retval)
{
+ struct acpi_device *dev;
+
+ if (strcmp(ctx, "SENR")) {
+ if (acpi_bus_get_device(ah, &dev))
+ return AE_OK;
+ if (!strcmp(ACER_WMID_ACCEL_HID, acpi_device_hid(dev)))
+ return AE_OK;
+ } else
+ return AE_OK;
+
*(acpi_handle *)retval = ah;
- return AE_OK;
+
+ return AE_CTRL_TERMINATE;
}
static int __init acer_wmi_get_handle(const char *name, const char *prop,
@@ -1839,7 +1852,7 @@ static int __init acer_wmi_accel_setup(void)
{
int err;
- err = acer_wmi_get_handle("SENR", "BST0001", &gsensor_handle);
+ err = acer_wmi_get_handle("SENR", ACER_WMID_ACCEL_HID, &gsensor_handle);
if (err)
return err;
@@ -2177,10 +2190,11 @@ static int __init acer_wmi_init(void)
err = acer_wmi_input_setup();
if (err)
return err;
+ err = acer_wmi_accel_setup();
+ if (err)
+ return err;
}
- acer_wmi_accel_setup();
-
err = platform_driver_register(&acer_platform_driver);
if (err) {
pr_err("Unable to register platform driver\n");
--
2.1.4
On Tue, Nov 01, 2016 at 12:33:32PM +0800, Lee, Chun-Yi wrote:
> The accelerometer event relies on on the ACERWMID_EVENT_GUID notify.
> So, this patch changes the codes to setup accelerometer input device
> when detected ACERWMID_EVENT_GUID. It avoids that the accel input
> device created on every acer machines.
>
> In addition, patch adds a clearly parsing logic of accelerometer hid
> to acer_wmi_get_handle_cb callback function. It is positive matching
> the "SENR" name with "BST0001" device to avoid non-supported hardware.
>
> Reported-by: Bj?rn Mork <[email protected]>
> Tested-by: Bj?rn Mork <[email protected]>
> Cc: Darren Hart <[email protected]>
> Signed-off-by: Lee, Chun-Yi <[email protected]>
> ---
> drivers/platform/x86/acer-wmi.c | 22 ++++++++++++++++++----
> 1 file changed, 18 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/platform/x86/acer-wmi.c b/drivers/platform/x86/acer-wmi.c
> index 79d64ea..3bb3162 100644
> --- a/drivers/platform/x86/acer-wmi.c
> +++ b/drivers/platform/x86/acer-wmi.c
> @@ -1808,11 +1808,24 @@ static int __init acer_wmi_enable_lm(void)
> return status;
> }
>
> +#define ACER_WMID_ACCEL_HID "BST0001"
> +
> static acpi_status __init acer_wmi_get_handle_cb(acpi_handle ah, u32 level,
> void *ctx, void **retval)
> {
> + struct acpi_device *dev;
> +
> + if (strcmp(ctx, "SENR")) {
> + if (acpi_bus_get_device(ah, &dev))
> + return AE_OK;
> + if (!strcmp(ACER_WMID_ACCEL_HID, acpi_device_hid(dev)))
> + return AE_OK;
> + } else
> + return AE_OK;
> +
> *(acpi_handle *)retval = ah;
> - return AE_OK;
> +
> + return AE_CTRL_TERMINATE;
I don't understand this change. Previously, we assigned retval to ah and
returned AE_OK. Now, we either do not assign retval and still return AE_OK, or
we do assign it and then return AE_CTRL_TERMINATE....
I would have expected to see something more like:
if (we should setup the accelerometer)
reval = ah
return AE_OK
return AE_CTRL_TERMINATE
If I'm misunderstanding this, can you please try to explain - a comment in the
function would be useful.
Thanks,
--
Darren Hart
Intel Open Source Technology Center
Hi Darren,
Thanks for your review!
On Tue, Nov 01, 2016 at 10:28:51AM -0700, Darren Hart wrote:
> On Tue, Nov 01, 2016 at 12:33:32PM +0800, Lee, Chun-Yi wrote:
> > The accelerometer event relies on on the ACERWMID_EVENT_GUID notify.
> > So, this patch changes the codes to setup accelerometer input device
> > when detected ACERWMID_EVENT_GUID. It avoids that the accel input
> > device created on every acer machines.
> >
> > In addition, patch adds a clearly parsing logic of accelerometer hid
> > to acer_wmi_get_handle_cb callback function. It is positive matching
> > the "SENR" name with "BST0001" device to avoid non-supported hardware.
> >
> > Reported-by: Bj?rn Mork <[email protected]>
> > Tested-by: Bj?rn Mork <[email protected]>
> > Cc: Darren Hart <[email protected]>
> > Signed-off-by: Lee, Chun-Yi <[email protected]>
> > ---
> > drivers/platform/x86/acer-wmi.c | 22 ++++++++++++++++++----
> > 1 file changed, 18 insertions(+), 4 deletions(-)
> >
> > diff --git a/drivers/platform/x86/acer-wmi.c b/drivers/platform/x86/acer-wmi.c
> > index 79d64ea..3bb3162 100644
> > --- a/drivers/platform/x86/acer-wmi.c
> > +++ b/drivers/platform/x86/acer-wmi.c
> > @@ -1808,11 +1808,24 @@ static int __init acer_wmi_enable_lm(void)
> > return status;
> > }
> >
> > +#define ACER_WMID_ACCEL_HID "BST0001"
> > +
> > static acpi_status __init acer_wmi_get_handle_cb(acpi_handle ah, u32 level,
> > void *ctx, void **retval)
> > {
> > + struct acpi_device *dev;
> > +
> > + if (strcmp(ctx, "SENR")) {
Here must be "if (!strcmp(ctx, "SENR"))", it's my mistake.
I will send v2 patch.
> > + if (acpi_bus_get_device(ah, &dev))
> > + return AE_OK;
> > + if (!strcmp(ACER_WMID_ACCEL_HID, acpi_device_hid(dev)))
> > + return AE_OK;
> > + } else
> > + return AE_OK;
> > +
> > *(acpi_handle *)retval = ah;
> > - return AE_OK;
> > +
> > + return AE_CTRL_TERMINATE;
>
> I don't understand this change. Previously, we assigned retval to ah and
> returned AE_OK. Now, we either do not assign retval and still return AE_OK, or
> we do assign it and then return AE_CTRL_TERMINATE....
>
> I would have expected to see something more like:
>
> if (we should setup the accelerometer)
> reval = ah
> return AE_OK
>
> return AE_CTRL_TERMINATE
>
> If I'm misunderstanding this, can you please try to explain - a comment in the
> function would be useful.
>
> Thanks,
>
Base on the function name "acer_wmi_get_handle_cb", it will be a generic
callback function for any caller who uses acer_wmi_get_handle(). And
acer_wmi_get_handle() uses acpi_get_devices(), the acpi_get_devices() allows
that HID is NULL. So that's better to check the context name and HID in callback.
Like this (after my mistake got fixed):
static acpi_status __init acer_wmi_get_handle_cb(acpi_handle ah, u32 level,
void *ctx, void **retval)
{
+ struct acpi_device *dev;
+
/* if context is "SENR", then make sure that HID matches with BST0001 */
+ if (!strcmp(ctx, "SENR")) {
+ if (acpi_bus_get_device(ah, &dev))
+ return AE_OK;
/* if HID doesn't match then skip this device to next */
+ if (!strcmp(ACER_WMID_ACCEL_HID, acpi_device_hid(dev)))
+ return AE_OK;
/* if this device do not match any context, then just skip it to next */
+ } else
+ return AE_OK;
+
/* if match with context and HID, then return handler */
*(acpi_handle *)retval = ah;
- return AE_OK;
+
/* return AE_CTRL_TERMINATE to exit acpi_ns_walk_namespace() because we found handler */
+ return AE_CTRL_TERMINATE;
}
Thanks a lot!
Joey Lee