I noticed a warning while cross-compiling all arm defconfigs.
The mmp2_defconfig gave this warning:
drivers/video/backlight/max8925_bl.c: In function 'max8925_backlight_probe':
drivers/video/backlight/max8925_bl.c:177:3: warning: statement with no effect [-Wunused-value]
This appears to have been introduced by the above commit when !CONFIG_OF
Looking at this more closely, I am not sure how this was ever intended
to be handled or how the errors returned in the CONFIG_OF case were
intended to be handled as the return from max8925_backlight_dt_init is
always ignored.
I think this needs some more attention, but do not feel like I know
enough about it or have any means to test it to weigh in.
Thanks,
Robin
commit 47ec340cb8e232671e7c4a4689ff32c3bdf329da
Author: Qing Xu <[email protected]>
Date: Mon Feb 4 23:40:45 2013 +0800
mfd: max8925: Support dt for backlight
Add device tree support in max8925 backlight.
Signed-off-by: Qing Xu <[email protected]>
Signed-off-by: Haojian Zhuang <[email protected]>
Signed-off-by: Samuel Ortiz <[email protected]>
diff --git a/drivers/video/backlight/max8925_bl.c b/drivers/video/backlight/max8925_bl.c
index 2c9bce0..5ca11b0 100644
--- a/drivers/video/backlight/max8925_bl.c
+++ b/drivers/video/backlight/max8925_bl.c
@@ -101,6 +101,29 @@ static const struct backlight_ops max8925_backlight_ops = {
.get_brightness = max8925_backlight_get_brightness,
};
+#ifdef CONFIG_OF
+static int max8925_backlight_dt_init(struct platform_device *pdev,
+ struct max8925_backlight_pdata *pdata)
+{
+ struct device_node *nproot = pdev->dev.parent->of_node, *np;
+ int dual_string;
+
+ if (!nproot)
+ return -ENODEV;
+ np = of_find_node_by_name(nproot, "backlight");
+ if (!np) {
+ dev_err(&pdev->dev, "failed to find backlight node\n");
+ return -ENODEV;
+ }
+
+ of_property_read_u32(np, "maxim,max8925-dual-string", &dual_string);
+ pdata->dual_string = dual_string;
+ return 0;
+}
+#else
+#define max8925_backlight_dt_init(x, y) (-1)
+#endif
+
static int max8925_backlight_probe(struct platform_device *pdev)
{
struct max8925_chip *chip = dev_get_drvdata(pdev->dev.parent);
@@ -147,6 +170,13 @@ static int max8925_backlight_probe(struct platform_device *pdev)
platform_set_drvdata(pdev, bl);
value = 0;
+ if (pdev->dev.parent->of_node && !pdata) {
+ pdata = devm_kzalloc(&pdev->dev,
+ sizeof(struct max8925_backlight_pdata),
+ GFP_KERNEL);
+ max8925_backlight_dt_init(pdev, pdata);
+ }
+
if (pdata) {
if (pdata->lxw_scl)
value |= (1 << 7);
@@ -158,7 +188,6 @@ static int max8925_backlight_probe(struct platform_device *pdev)
ret = max8925_set_bits(chip->i2c, data->reg_mode_cntl, 0xfe, value);
if (ret < 0)
goto out_brt;
-
backlight_update_status(bl);
return 0;
out_brt:
Hello,
On Tue, May 07, 2013 at 04:17:34AM -0500, Robin Holt wrote:
> I noticed a warning while cross-compiling all arm defconfigs.
>
> The mmp2_defconfig gave this warning:
>
> drivers/video/backlight/max8925_bl.c: In function 'max8925_backlight_probe':
> drivers/video/backlight/max8925_bl.c:177:3: warning: statement with no effect [-Wunused-value]
>
> This appears to have been introduced by the above commit when !CONFIG_OF
>
> Looking at this more closely, I am not sure how this was ever intended
> to be handled or how the errors returned in the CONFIG_OF case were
> intended to be handled as the return from max8925_backlight_dt_init is
> always ignored.
>
> I think this needs some more attention, but do not feel like I know
> enough about it or have any means to test it to weigh in.
>
> Thanks,
> Robin
>
>
> commit 47ec340cb8e232671e7c4a4689ff32c3bdf329da
> Author: Qing Xu <[email protected]>
> Date: Mon Feb 4 23:40:45 2013 +0800
>
> mfd: max8925: Support dt for backlight
>
> Add device tree support in max8925 backlight.
>
> Signed-off-by: Qing Xu <[email protected]>
> Signed-off-by: Haojian Zhuang <[email protected]>
> Signed-off-by: Samuel Ortiz <[email protected]>
>
> diff --git a/drivers/video/backlight/max8925_bl.c b/drivers/video/backlight/max8925_bl.c
> index 2c9bce0..5ca11b0 100644
> --- a/drivers/video/backlight/max8925_bl.c
> +++ b/drivers/video/backlight/max8925_bl.c
> @@ -101,6 +101,29 @@ static const struct backlight_ops max8925_backlight_ops = {
> .get_brightness = max8925_backlight_get_brightness,
> };
>
> +#ifdef CONFIG_OF
> +static int max8925_backlight_dt_init(struct platform_device *pdev,
> + struct max8925_backlight_pdata *pdata)
> +{
> + struct device_node *nproot = pdev->dev.parent->of_node, *np;
> + int dual_string;
> +
> + if (!nproot)
> + return -ENODEV;
> + np = of_find_node_by_name(nproot, "backlight");
> + if (!np) {
> + dev_err(&pdev->dev, "failed to find backlight node\n");
> + return -ENODEV;
> + }
> +
> + of_property_read_u32(np, "maxim,max8925-dual-string", &dual_string);
> + pdata->dual_string = dual_string;
> + return 0;
> +}
> +#else
> +#define max8925_backlight_dt_init(x, y) (-1)
It's probably best to make this:
static inline max8925_backlight_dt_init(struct platform_device *pdev,
struct max8925_backlight_pdata *pdata)
{
return -ENODEV;
}
> +#endif
> +
> static int max8925_backlight_probe(struct platform_device *pdev)
> {
> struct max8925_chip *chip = dev_get_drvdata(pdev->dev.parent);
> @@ -147,6 +170,13 @@ static int max8925_backlight_probe(struct platform_device *pdev)
> platform_set_drvdata(pdev, bl);
>
> value = 0;
> + if (pdev->dev.parent->of_node && !pdata) {
> + pdata = devm_kzalloc(&pdev->dev,
> + sizeof(struct max8925_backlight_pdata),
> + GFP_KERNEL);
> + max8925_backlight_dt_init(pdev, pdata);
> + }
> +
> if (pdata) {
> if (pdata->lxw_scl)
> value |= (1 << 7);
> @@ -158,7 +188,6 @@ static int max8925_backlight_probe(struct platform_device *pdev)
> ret = max8925_set_bits(chip->i2c, data->reg_mode_cntl, 0xfe, value);
> if (ret < 0)
> goto out_brt;
> -
> backlight_update_status(bl);
> return 0;
> out_brt:
>
> _______________________________________________
> linux-arm-kernel mailing list
> [email protected]
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
>
--
Pengutronix e.K. | Uwe Kleine-K?nig |
Industrial Linux Solutions | http://www.pengutronix.de/ |
On Tue, May 7, 2013 at 5:24 PM, Uwe Kleine-K?nig
<[email protected]> wrote:
> Hello,
>
> On Tue, May 07, 2013 at 04:17:34AM -0500, Robin Holt wrote:
>> I noticed a warning while cross-compiling all arm defconfigs.
>>
>> The mmp2_defconfig gave this warning:
>>
>> drivers/video/backlight/max8925_bl.c: In function 'max8925_backlight_probe':
>> drivers/video/backlight/max8925_bl.c:177:3: warning: statement with no effect [-Wunused-value]
>>
>> This appears to have been introduced by the above commit when !CONFIG_OF
>>
>> Looking at this more closely, I am not sure how this was ever intended
>> to be handled or how the errors returned in the CONFIG_OF case were
>> intended to be handled as the return from max8925_backlight_dt_init is
>> always ignored.
>>
>> I think this needs some more attention, but do not feel like I know
>> enough about it or have any means to test it to weigh in.
>>
>> Thanks,
>> Robin
>>
>>
>> commit 47ec340cb8e232671e7c4a4689ff32c3bdf329da
>> Author: Qing Xu <[email protected]>
>> Date: Mon Feb 4 23:40:45 2013 +0800
>>
>> mfd: max8925: Support dt for backlight
>>
>> Add device tree support in max8925 backlight.
>>
>> Signed-off-by: Qing Xu <[email protected]>
>> Signed-off-by: Haojian Zhuang <[email protected]>
>> Signed-off-by: Samuel Ortiz <[email protected]>
>>
>> diff --git a/drivers/video/backlight/max8925_bl.c b/drivers/video/backlight/max8925_bl.c
>> index 2c9bce0..5ca11b0 100644
>> --- a/drivers/video/backlight/max8925_bl.c
>> +++ b/drivers/video/backlight/max8925_bl.c
>> @@ -101,6 +101,29 @@ static const struct backlight_ops max8925_backlight_ops = {
>> .get_brightness = max8925_backlight_get_brightness,
>> };
>>
>> +#ifdef CONFIG_OF
>> +static int max8925_backlight_dt_init(struct platform_device *pdev,
>> + struct max8925_backlight_pdata *pdata)
>> +{
>> + struct device_node *nproot = pdev->dev.parent->of_node, *np;
>> + int dual_string;
>> +
>> + if (!nproot)
>> + return -ENODEV;
>> + np = of_find_node_by_name(nproot, "backlight");
>> + if (!np) {
>> + dev_err(&pdev->dev, "failed to find backlight node\n");
>> + return -ENODEV;
>> + }
>> +
>> + of_property_read_u32(np, "maxim,max8925-dual-string", &dual_string);
>> + pdata->dual_string = dual_string;
>> + return 0;
>> +}
>> +#else
>> +#define max8925_backlight_dt_init(x, y) (-1)
> It's probably best to make this:
>
> static inline max8925_backlight_dt_init(struct platform_device *pdev,
> struct max8925_backlight_pdata *pdata)
> {
> return -ENODEV;
> }
>
I've submitted this patch to fix this issue for a long time.
Samuel,
Should I send it again?
Regards
Haojian
On Tue, May 07, 2013 at 05:35:28PM +0800, Haojian Zhuang wrote:
> On Tue, May 7, 2013 at 5:24 PM, Uwe Kleine-K?nig
> <[email protected]> wrote:
> > Hello,
> >
> > On Tue, May 07, 2013 at 04:17:34AM -0500, Robin Holt wrote:
> >> I noticed a warning while cross-compiling all arm defconfigs.
> >>
> >> The mmp2_defconfig gave this warning:
> >>
> >> drivers/video/backlight/max8925_bl.c: In function 'max8925_backlight_probe':
> >> drivers/video/backlight/max8925_bl.c:177:3: warning: statement with no effect [-Wunused-value]
> >>
> >> This appears to have been introduced by the above commit when !CONFIG_OF
> >>
> >> Looking at this more closely, I am not sure how this was ever intended
> >> to be handled or how the errors returned in the CONFIG_OF case were
> >> intended to be handled as the return from max8925_backlight_dt_init is
> >> always ignored.
> >>
> >> I think this needs some more attention, but do not feel like I know
> >> enough about it or have any means to test it to weigh in.
> >>
> >> Thanks,
> >> Robin
> >>
> >>
> >> commit 47ec340cb8e232671e7c4a4689ff32c3bdf329da
> >> Author: Qing Xu <[email protected]>
> >> Date: Mon Feb 4 23:40:45 2013 +0800
> >>
> >> mfd: max8925: Support dt for backlight
> >>
> >> Add device tree support in max8925 backlight.
> >>
> >> Signed-off-by: Qing Xu <[email protected]>
> >> Signed-off-by: Haojian Zhuang <[email protected]>
> >> Signed-off-by: Samuel Ortiz <[email protected]>
> >>
> >> diff --git a/drivers/video/backlight/max8925_bl.c b/drivers/video/backlight/max8925_bl.c
> >> index 2c9bce0..5ca11b0 100644
> >> --- a/drivers/video/backlight/max8925_bl.c
> >> +++ b/drivers/video/backlight/max8925_bl.c
> >> @@ -101,6 +101,29 @@ static const struct backlight_ops max8925_backlight_ops = {
> >> .get_brightness = max8925_backlight_get_brightness,
> >> };
> >>
> >> +#ifdef CONFIG_OF
> >> +static int max8925_backlight_dt_init(struct platform_device *pdev,
> >> + struct max8925_backlight_pdata *pdata)
> >> +{
> >> + struct device_node *nproot = pdev->dev.parent->of_node, *np;
> >> + int dual_string;
> >> +
> >> + if (!nproot)
> >> + return -ENODEV;
> >> + np = of_find_node_by_name(nproot, "backlight");
> >> + if (!np) {
> >> + dev_err(&pdev->dev, "failed to find backlight node\n");
> >> + return -ENODEV;
> >> + }
> >> +
> >> + of_property_read_u32(np, "maxim,max8925-dual-string", &dual_string);
> >> + pdata->dual_string = dual_string;
> >> + return 0;
> >> +}
> >> +#else
> >> +#define max8925_backlight_dt_init(x, y) (-1)
> > It's probably best to make this:
> >
> > static inline max8925_backlight_dt_init(struct platform_device *pdev,
> > struct max8925_backlight_pdata *pdata)
> > {
> > return -ENODEV;
> > }
> >
>
> I've submitted this patch to fix this issue for a long time.
>
> Samuel,
>
> Should I send it again?
It fixes nothing. The return value is not used. There is more to this
bug report than the -1. You need to handle that error case. Otherwise,
you could just change it into a void return.
Robin
On Tue, May 07, 2013 at 05:35:28PM +0800, Haojian Zhuang wrote:
> > It's probably best to make this:
> >
> > static inline max8925_backlight_dt_init(struct platform_device *pdev,
> > struct max8925_backlight_pdata *pdata)
> > {
> > return -ENODEV;
> > }
> >
>
> I've submitted this patch to fix this issue for a long time.
>
> Samuel,
>
> Should I send it again?
To the backlight maintainers, yes. I carried 47ec340c as it was part of an MFD
patchset, but now that it's upstream I shouldn't take any patches for it.
Cheers,
Samuel.
--
Intel Open Source Technology Centre
http://oss.intel.com/