[PATCH] HID: wacom: Fix wacom_probe error handling

In the case of goto destroy_leds, we also need to call hid_hw_stop(hdev) and
remove sysfs attribute files we created for speed and oleds.

Signed-off-by: Axel Lin <redacted>
---
 drivers/hid/hid-wacom.c |   12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/hid/hid-wacom.c b/drivers/hid/hid-wacom.c
index a0fafac..fae7fd7 100644
--- a/drivers/hid/hid-wacom.c
+++ b/drivers/hid/hid-wacom.c

@@ -902,10 +902,18 @@ OLED_INIT(7);
 err_ac:
 	power_supply_unregister(&wdata->battery);
 err_battery:
-	device_remove_file(&hdev->dev, &dev_attr_speed);
-	hid_hw_stop(hdev);
 destroy_leds:
 	wacom_destroy_leds(hdev);
+	device_remove_file(&hdev->dev, &dev_attr_oled0_img);
+	device_remove_file(&hdev->dev, &dev_attr_oled1_img);
+	device_remove_file(&hdev->dev, &dev_attr_oled2_img);
+	device_remove_file(&hdev->dev, &dev_attr_oled3_img);
+	device_remove_file(&hdev->dev, &dev_attr_oled4_img);
+	device_remove_file(&hdev->dev, &dev_attr_oled5_img);
+	device_remove_file(&hdev->dev, &dev_attr_oled6_img);
+	device_remove_file(&hdev->dev, &dev_attr_oled7_img);
+	device_remove_file(&hdev->dev, &dev_attr_speed);
+	hid_hw_stop(hdev);
 err_free:
 	kfree(wdata);
 	return ret;

-- 
1.7.9.5

Dnia 17 Września 2012, 3:29 am, Pn, Axel Lin napisał(a):

Hi Alex,
The initial idea was that the OLEDs/LED are not critical for tablet
functioning and that's why there was no call of hid_stop_hw if the
OLED/LED initialisation faild.
What's is the general kernel/HID policy: is it OK to have a device working
with some non-essential features missing?

-- 
Regards,
Przemo Firszt

--
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

2012/9/17 Przemo Firszt [off-list ref]:

Dnia 17 Września 2012, 3:29 am, Pn, Axel Lin napisał(a):

quoted

In the case of goto destroy_leds, we also need to call hid_hw_stop(hdev)
and
remove sysfs attribute files we created for speed and oleds.

Signed-off-by: Axel Lin <redacted>
---
 drivers/hid/hid-wacom.c |   12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/hid/hid-wacom.c b/drivers/hid/hid-wacom.c
index a0fafac..fae7fd7 100644
--- a/drivers/hid/hid-wacom.c
+++ b/drivers/hid/hid-wacom.c

@@ -902,10 +902,18 @@ OLED_INIT(7);
 err_ac:
      power_supply_unregister(&wdata->battery);
 err_battery:
-     device_remove_file(&hdev->dev, &dev_attr_speed);
-     hid_hw_stop(hdev);
 destroy_leds:
      wacom_destroy_leds(hdev);
+     device_remove_file(&hdev->dev, &dev_attr_oled0_img);
+     device_remove_file(&hdev->dev, &dev_attr_oled1_img);
+     device_remove_file(&hdev->dev, &dev_attr_oled2_img);
+     device_remove_file(&hdev->dev, &dev_attr_oled3_img);
+     device_remove_file(&hdev->dev, &dev_attr_oled4_img);
+     device_remove_file(&hdev->dev, &dev_attr_oled5_img);
+     device_remove_file(&hdev->dev, &dev_attr_oled6_img);
+     device_remove_file(&hdev->dev, &dev_attr_oled7_img);
+     device_remove_file(&hdev->dev, &dev_attr_speed);
+     hid_hw_stop(hdev);
 err_free:
      kfree(wdata);
      return ret;
--

1.7.9.5

Hi Alex,

It's Axel.

The initial idea was that the OLEDs/LED are not critical for tablet
functioning and that's why there was no call of hid_stop_hw if the
OLED/LED initialisation faild.

Since it is not critical, current code ignore checking return value when
device_create_file for speed and oleds.

But current code does goto destroy_leds if wacom_initialize_leds fails.
Which means it returns error code in wacom_probe() and thus it should
clean up all the allocated resources in the error patch.

If you want to ignore wacom_initialize_leds() failure, you should not
call goto destroy_leds.

Comment?

Axel
--
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

On Mon, 2012-09-17 at 16:39 +0800, Axel Lin wrote:

2012/9/17 Przemo Firszt [off-list ref]:

quoted

Dnia 17 Września 2012, 3:29 am, Pn, Axel Lin napisał(a):

quoted

In the case of goto destroy_leds, we also need to call hid_hw_stop(hdev)
and
remove sysfs attribute files we created for speed and oleds.

Signed-off-by: Axel Lin <redacted>
---
 drivers/hid/hid-wacom.c |   12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/hid/hid-wacom.c b/drivers/hid/hid-wacom.c
index a0fafac..fae7fd7 100644
--- a/drivers/hid/hid-wacom.c
+++ b/drivers/hid/hid-wacom.c

@@ -902,10 +902,18 @@ OLED_INIT(7);
 err_ac:
      power_supply_unregister(&wdata->battery);
 err_battery:
-     device_remove_file(&hdev->dev, &dev_attr_speed);
-     hid_hw_stop(hdev);
 destroy_leds:
      wacom_destroy_leds(hdev);
+     device_remove_file(&hdev->dev, &dev_attr_oled0_img);
+     device_remove_file(&hdev->dev, &dev_attr_oled1_img);
+     device_remove_file(&hdev->dev, &dev_attr_oled2_img);
+     device_remove_file(&hdev->dev, &dev_attr_oled3_img);
+     device_remove_file(&hdev->dev, &dev_attr_oled4_img);
+     device_remove_file(&hdev->dev, &dev_attr_oled5_img);
+     device_remove_file(&hdev->dev, &dev_attr_oled6_img);
+     device_remove_file(&hdev->dev, &dev_attr_oled7_img);
+     device_remove_file(&hdev->dev, &dev_attr_speed);
+     hid_hw_stop(hdev);
 err_free:
      kfree(wdata);
      return ret;
--

1.7.9.5

Hi Alex,

It's Axel.

Hi Axel,
I'm really sorry for that mistake :-(

quoted

The initial idea was that the OLEDs/LED are not critical for tablet
functioning and that's why there was no call of hid_stop_hw if the
OLED/LED initialisation faild.

Since it is not critical, current code ignore checking return value when
device_create_file for speed and oleds.

But current code does goto destroy_leds if wacom_initialize_leds fails.
Which means it returns error code in wacom_probe() and thus it should
clean up all the allocated resources in the error patch.

If you want to ignore wacom_initialize_leds() failure, you should not
call goto destroy_leds.

Comment?

See my reply to the v2 patch. Thanks for catching that! 
--
Kind regards,
Przemo

--
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html