Initializing rv to AE_OK is pointless because later function results are
assigned to them and only then the variable is used
Signed-off-by: Giedrius Statkevičius <[email protected]>
---
drivers/platform/x86/asus-laptop.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/platform/x86/asus-laptop.c b/drivers/platform/x86/asus-laptop.c
index f2b5d0a..d86d42e 100644
--- a/drivers/platform/x86/asus-laptop.c
+++ b/drivers/platform/x86/asus-laptop.c
@@ -771,7 +771,7 @@ static int asus_read_brightness(struct backlight_device *bd)
{
struct asus_laptop *asus = bl_get_data(bd);
unsigned long long value;
- acpi_status rv = AE_OK;
+ acpi_status rv;
rv = acpi_evaluate_integer(asus->handle, METHOD_BRIGHTNESS_GET,
NULL, &value);
@@ -865,7 +865,7 @@ static ssize_t infos_show(struct device *dev, struct device_attribute *attr,
int len = 0;
unsigned long long temp;
char buf[16]; /* enough for all info */
- acpi_status rv = AE_OK;
+ acpi_status rv;
/*
* We use the easy way, we don't care of off and count,
@@ -1265,7 +1265,7 @@ static DEVICE_ATTR_RO(ls_value);
static int asus_gps_status(struct asus_laptop *asus)
{
unsigned long long status;
- acpi_status rv = AE_OK;
+ acpi_status rv;
rv = acpi_evaluate_integer(asus->handle, METHOD_GPS_STATUS,
NULL, &status);
--
2.8.0
`out' was assigned value but it was never used so remove it
Signed-off-by: Giedrius Statkevičius <[email protected]>
---
drivers/platform/x86/asus-laptop.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/drivers/platform/x86/asus-laptop.c b/drivers/platform/x86/asus-laptop.c
index d86d42e..39ddcee 100644
--- a/drivers/platform/x86/asus-laptop.c
+++ b/drivers/platform/x86/asus-laptop.c
@@ -946,11 +946,8 @@ static ssize_t sysfs_acpi_set(struct asus_laptop *asus,
const char *method)
{
int rv, value;
- int out = 0;
rv = parse_arg(buf, count, &value);
- if (rv > 0)
- out = value ? 1 : 0;
if (write_acpi_int(asus->handle, method, value))
return -ENODEV;
--
2.8.0
On Thu, Apr 07, 2016 at 11:20:01PM +0300, Giedrius Statkevičius wrote:
> `out' was assigned value but it was never used so remove it
>
> Signed-off-by: Giedrius Statkevičius <[email protected]>
> ---
> drivers/platform/x86/asus-laptop.c | 3 ---
> 1 file changed, 3 deletions(-)
>
> diff --git a/drivers/platform/x86/asus-laptop.c b/drivers/platform/x86/asus-laptop.c
> index d86d42e..39ddcee 100644
> --- a/drivers/platform/x86/asus-laptop.c
> +++ b/drivers/platform/x86/asus-laptop.c
> @@ -946,11 +946,8 @@ static ssize_t sysfs_acpi_set(struct asus_laptop *asus,
> const char *method)
> {
> int rv, value;
> - int out = 0;
>
> rv = parse_arg(buf, count, &value);
> - if (rv > 0)
> - out = value ? 1 : 0;
>
> if (write_acpi_int(asus->handle, method, value))
out is indeed unused, however the rv > 0 condition is relevant as <=0 will pass
value uninitialized to write_acpi_int.
--
Darren Hart
Intel Open Source Technology Center
On Sat, Apr 09, 2016 at 08:21:21PM -0700, Darren Hart wrote:
> On Thu, Apr 07, 2016 at 11:20:01PM +0300, Giedrius Statkevičius wrote:
> > `out' was assigned value but it was never used so remove it
> >
> > Signed-off-by: Giedrius Statkevičius <[email protected]>
> > ---
> > drivers/platform/x86/asus-laptop.c | 3 ---
> > 1 file changed, 3 deletions(-)
> >
> > diff --git a/drivers/platform/x86/asus-laptop.c b/drivers/platform/x86/asus-laptop.c
> > index d86d42e..39ddcee 100644
> > --- a/drivers/platform/x86/asus-laptop.c
> > +++ b/drivers/platform/x86/asus-laptop.c
> > @@ -946,11 +946,8 @@ static ssize_t sysfs_acpi_set(struct asus_laptop *asus,
> > const char *method)
> > {
> > int rv, value;
> > - int out = 0;
> >
> > rv = parse_arg(buf, count, &value);
> > - if (rv > 0)
> > - out = value ? 1 : 0;
> >
> > if (write_acpi_int(asus->handle, method, value))
>
> out is indeed unused, however the rv > 0 condition is relevant as <=0 will pass
> value uninitialized to write_acpi_int.
That's indeed a problem. Somehow I missed that :( Should I make a v2 to include
another patch that checks for rv < 0 or send it as an independent one?
On Sun, Apr 10, 2016 at 11:48:52AM +0300, Giedrius Statkevičius wrote:
> On Sat, Apr 09, 2016 at 08:21:21PM -0700, Darren Hart wrote:
> > On Thu, Apr 07, 2016 at 11:20:01PM +0300, Giedrius Statkevičius wrote:
> > > `out' was assigned value but it was never used so remove it
> > >
> > > Signed-off-by: Giedrius Statkevičius <[email protected]>
> > > ---
> > > drivers/platform/x86/asus-laptop.c | 3 ---
> > > 1 file changed, 3 deletions(-)
> > >
> > > diff --git a/drivers/platform/x86/asus-laptop.c b/drivers/platform/x86/asus-laptop.c
> > > index d86d42e..39ddcee 100644
> > > --- a/drivers/platform/x86/asus-laptop.c
> > > +++ b/drivers/platform/x86/asus-laptop.c
> > > @@ -946,11 +946,8 @@ static ssize_t sysfs_acpi_set(struct asus_laptop *asus,
> > > const char *method)
> > > {
> > > int rv, value;
> > > - int out = 0;
> > >
> > > rv = parse_arg(buf, count, &value);
> > > - if (rv > 0)
> > > - out = value ? 1 : 0;
> > >
> > > if (write_acpi_int(asus->handle, method, value))
> >
> > out is indeed unused, however the rv > 0 condition is relevant as <=0 will pass
> > value uninitialized to write_acpi_int.
> That's indeed a problem. Somehow I missed that :( Should I make a v2 to include
> another patch that checks for rv < 0 or send it as an independent one?
>
Please just rewrite this one to do the right thing as a v2.
--
Darren Hart
Intel Open Source Technology Center
On Sun, Apr 10, 2016 at 11:48:52AM +0300, Giedrius Statkevičius wrote:
> On Sat, Apr 09, 2016 at 08:21:21PM -0700, Darren Hart wrote:
> > On Thu, Apr 07, 2016 at 11:20:01PM +0300, Giedrius Statkevičius wrote:
> > > `out' was assigned value but it was never used so remove it
> > >
> > > Signed-off-by: Giedrius Statkevičius <[email protected]>
> > > ---
> > > drivers/platform/x86/asus-laptop.c | 3 ---
> > > 1 file changed, 3 deletions(-)
> > >
> > > diff --git a/drivers/platform/x86/asus-laptop.c b/drivers/platform/x86/asus-laptop.c
> > > index d86d42e..39ddcee 100644
> > > --- a/drivers/platform/x86/asus-laptop.c
> > > +++ b/drivers/platform/x86/asus-laptop.c
> > > @@ -946,11 +946,8 @@ static ssize_t sysfs_acpi_set(struct asus_laptop *asus,
> > > const char *method)
> > > {
> > > int rv, value;
> > > - int out = 0;
> > >
> > > rv = parse_arg(buf, count, &value);
> > > - if (rv > 0)
> > > - out = value ? 1 : 0;
> > >
> > > if (write_acpi_int(asus->handle, method, value))
> >
> > out is indeed unused, however the rv > 0 condition is relevant as <=0 will pass
> > value uninitialized to write_acpi_int.
> That's indeed a problem. Somehow I missed that :( Should I make a v2 to include
> another patch that checks for rv < 0 or send it as an independent one?
>
Sorry, my response was not clear. Please resend both patches as a v2, with the
first unchanged, and this one updated to do the right thing. This just makes
managing the patches as series simpler.
Thank you,
--
Darren Hart
Intel Open Source Technology Center