power_supply_powers calls added in 35b4c01e29bdd9632dabf9784ed3486333f00427
have to be called after power device is created. This patch also fixes the
second call - it has to be "ac" instead of "battery"
Signed-off-by: Przemo Firszt <[email protected]>
Signed-off-by: Chris Bagwell <[email protected]>
---
drivers/hid/hid-wacom.c | 7 ++++---
1 files changed, 4 insertions(+), 3 deletions(-)
diff --git a/drivers/hid/hid-wacom.c b/drivers/hid/hid-wacom.c
index b47e58b..acab74c 100644
--- a/drivers/hid/hid-wacom.c
+++ b/drivers/hid/hid-wacom.c
@@ -531,7 +531,6 @@ static int wacom_probe(struct hid_device *hdev,
wdata->battery.type = POWER_SUPPLY_TYPE_BATTERY;
wdata->battery.use_for_apm = 0;
- power_supply_powers(&wdata->battery, &hdev->dev);
ret = power_supply_register(&hdev->dev, &wdata->battery);
if (ret) {
@@ -540,6 +539,8 @@ static int wacom_probe(struct hid_device *hdev,
goto err_battery;
}
+ power_supply_powers(&wdata->battery, &hdev->dev);
+
wdata->ac.properties = wacom_ac_props;
wdata->ac.num_properties = ARRAY_SIZE(wacom_ac_props);
wdata->ac.get_property = wacom_ac_get_property;
@@ -547,14 +548,14 @@ static int wacom_probe(struct hid_device *hdev,
wdata->ac.type = POWER_SUPPLY_TYPE_MAINS;
wdata->ac.use_for_apm = 0;
- power_supply_powers(&wdata->battery, &hdev->dev);
-
ret = power_supply_register(&hdev->dev, &wdata->ac);
if (ret) {
hid_warn(hdev,
"can't create ac battery attribute, err: %d\n", ret);
goto err_ac;
}
+
+ power_supply_powers(&wdata->ac, &hdev->dev);
#endif
return 0;
--
1.7.6.4
On Sun, 5 Feb 2012, Przemo Firszt wrote:
> power_supply_powers calls added in 35b4c01e29bdd9632dabf9784ed3486333f00427
> have to be called after power device is created. This patch also fixes the
> second call - it has to be "ac" instead of "battery"
>
> Signed-off-by: Przemo Firszt <[email protected]>
> Signed-off-by: Chris Bagwell <[email protected]>
[ adding Jeremy to CC ]
> ---
> drivers/hid/hid-wacom.c | 7 ++++---
> 1 files changed, 4 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/hid/hid-wacom.c b/drivers/hid/hid-wacom.c
> index b47e58b..acab74c 100644
> --- a/drivers/hid/hid-wacom.c
> +++ b/drivers/hid/hid-wacom.c
> @@ -531,7 +531,6 @@ static int wacom_probe(struct hid_device *hdev,
> wdata->battery.type = POWER_SUPPLY_TYPE_BATTERY;
> wdata->battery.use_for_apm = 0;
>
> - power_supply_powers(&wdata->battery, &hdev->dev);
>
> ret = power_supply_register(&hdev->dev, &wdata->battery);
> if (ret) {
> @@ -540,6 +539,8 @@ static int wacom_probe(struct hid_device *hdev,
> goto err_battery;
> }
>
> + power_supply_powers(&wdata->battery, &hdev->dev);
> +
> wdata->ac.properties = wacom_ac_props;
> wdata->ac.num_properties = ARRAY_SIZE(wacom_ac_props);
> wdata->ac.get_property = wacom_ac_get_property;
> @@ -547,14 +548,14 @@ static int wacom_probe(struct hid_device *hdev,
> wdata->ac.type = POWER_SUPPLY_TYPE_MAINS;
> wdata->ac.use_for_apm = 0;
>
> - power_supply_powers(&wdata->battery, &hdev->dev);
> -
> ret = power_supply_register(&hdev->dev, &wdata->ac);
> if (ret) {
> hid_warn(hdev,
> "can't create ac battery attribute, err: %d\n", ret);
> goto err_ac;
> }
> +
> + power_supply_powers(&wdata->ac, &hdev->dev);
> #endif
> return 0;
Hmm, seems valid. How did you notice? Have you seen crashes because of
wild pointers?
Thanks,
--
Jiri Kosina
SUSE Labs
> On Sun, 5 Feb 2012, Przemo Firszt wrote:
[..]
> Hmm, seems valid. How did you notice? Have you seen crashes because of
> wild pointers?
Hi Jiri,
Yes, the driver was unusable - 100% crashes during connection.
Chris came up with the solution, I did coding & testing.
The crash details are here: http://pastebin.com/ZVNZWaPs
regards,
Przemo
On Mon, 6 Feb 2012, [email protected] wrote:
> > Hmm, seems valid. How did you notice? Have you seen crashes because of
> > wild pointers?
>
> Hi Jiri,
> Yes, the driver was unusable - 100% crashes during connection.
Okay, I thought that'd be the case.
> Chris came up with the solution, I did coding & testing.
>
> The crash details are here: http://pastebin.com/ZVNZWaPs
Thank you. Will be pushing to Linus soon.
--
Jiri Kosina
SUSE Labs
On 02/06/2012 04:42 AM, Jiri Kosina wrote:
> On Mon, 6 Feb 2012, [email protected] wrote:
>
>>> Hmm, seems valid. How did you notice? Have you seen crashes because of
>>> wild pointers?
>> Hi Jiri,
>> Yes, the driver was unusable - 100% crashes during connection.
> Okay, I thought that'd be the case.
Very sorry about that. I don't have a device to test with, so I should
have reviewed the code extra carefully.
Does the same bug apply to the Wii changes, which were of the same form?
J
> On 02/06/2012 04:42 AM, Jiri Kosina wrote:
>> On Mon, 6 Feb 2012, [email protected] wrote:
>>
>>>> Hmm, seems valid. How did you notice? Have you seen crashes because of
>>>> wild pointers?
>>> Hi Jiri,
>>> Yes, the driver was unusable - 100% crashes during connection.
>> Okay, I thought that'd be the case.
>
> Very sorry about that. I don't have a device to test with, so I should
> have reviewed the code extra carefully.
>
> Does the same bug apply to the Wii changes, which were of the same form?
>
Hi Jeremy,
The wii code looks the same, so probably it's affected as well. Can you
make a patch?
power_supply_powers call in wiimote driver:
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=blob;f=drivers/hid/hid-wiimote-core.c#l1229
regards,
Przemo Firszt
On Mon, 6 Feb 2012, [email protected] wrote:
> >>>> Hmm, seems valid. How did you notice? Have you seen crashes because of
> >>>> wild pointers?
> >>> Hi Jiri,
> >>> Yes, the driver was unusable - 100% crashes during connection.
> >> Okay, I thought that'd be the case.
> >
> > Very sorry about that. I don't have a device to test with, so I should
> > have reviewed the code extra carefully.
> >
> > Does the same bug apply to the Wii changes, which were of the same form?
> >
> Hi Jeremy,
> The wii code looks the same, so probably it's affected as well. Can you
> make a patch?
>
> power_supply_powers call in wiimote driver:
> http://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=blob;f=drivers/hid/hid-wiimote-core.c#l1229
I have now queued the patch below for the same pile as well.
Thanks for spotting it.
From: Jiri Kosina <[email protected]>
Subject: [PATCH] HID: wiimote: fix invalid power_supply_powers call
Analogically to d7cb3dbd1 ("HID: wacom: Fix invalid power_supply_powers
calls"), fix also the same occurence in wiimote driver.
Reported-by: [email protected]
Signed-off-by: Jiri Kosina <[email protected]>
---
drivers/hid/hid-wiimote-core.c | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/hid/hid-wiimote-core.c b/drivers/hid/hid-wiimote-core.c
index fc253b4..cac3589 100644
--- a/drivers/hid/hid-wiimote-core.c
+++ b/drivers/hid/hid-wiimote-core.c
@@ -1226,14 +1226,14 @@ static int wiimote_hid_probe(struct hid_device *hdev,
wdata->battery.type = POWER_SUPPLY_TYPE_BATTERY;
wdata->battery.use_for_apm = 0;
- power_supply_powers(&wdata->battery, &hdev->dev);
-
ret = power_supply_register(&wdata->hdev->dev, &wdata->battery);
if (ret) {
hid_err(hdev, "Cannot register battery device\n");
goto err_battery;
}
+ power_supply_powers(&wdata->battery, &hdev->dev);
+
ret = wiimote_leds_create(wdata);
if (ret)
goto err_free;
--
Jiri Kosina
SUSE Labs