[PATCH] iio: staging: ad7298_ring: Fix maybe-uninitialized warning

[PATCH] iio: staging: ad7298_ring: Fix maybe-uninitialized warning

[michael.hennerich]

From: Michael Hennerich <michael.hennerich@analog.com>

drivers/staging/iio/adc/ad7298_ring.c:97:37: warning: 'time_ns' may
be used uninitialized in this function [-Wmaybe-uninitialized]

Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
---
 drivers/staging/iio/adc/ad7298_ring.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/staging/iio/adc/ad7298_ring.c b/drivers/staging/iio/adc/ad7298_ring.c
index fd1d855..506016f 100644
--- a/drivers/staging/iio/adc/ad7298_ring.c
+++ b/drivers/staging/iio/adc/ad7298_ring.c
@@ -76,7 +76,7 @@ static irqreturn_t ad7298_trigger_handler(int irq, void *p)
 	struct iio_dev *indio_dev = pf->indio_dev;
 	struct ad7298_state *st = iio_priv(indio_dev);
 	struct iio_buffer *ring = indio_dev->buffer;
-	s64 time_ns;
+	s64 time_ns = 0;
 	__u16 buf[16];
 	int b_sent, i;
 
-- 
1.7.0.4

Re: [PATCH] iio: staging: ad7298_ring: Fix maybe-uninitialized warning

[Peter Meerwald]

> drivers/staging/iio/adc/ad7298_ring.c:97:37: warning: 'time_ns' may
> be used uninitialized in this function [-Wmaybe-uninitialized]

> diff --git a/drivers/staging/iio/adc/ad7298_ring.c b/drivers/staging/iio/adc/ad7298_ring.c
> index fd1d855..506016f 100644
> --- a/drivers/staging/iio/adc/ad7298_ring.c
> +++ b/drivers/staging/iio/adc/ad7298_ring.c
> @@ -76,7 +76,7 @@ static irqreturn_t ad7298_trigger_handler(int irq, void *p)
>  	struct iio_dev *indio_dev = pf->indio_dev;
>  	struct ad7298_state *st = iio_priv(indio_dev);
>  	struct iio_buffer *ring = indio_dev->buffer;
> -	s64 time_ns;
> +	s64 time_ns = 0;
>  	__u16 buf[16];
>  	int b_sent, i;

I'd rather suggest 
s64 time_ns = iio_get_time_ns();

and simplify the if (indio_dev->scan_timestamp) block

p.

-- 

Peter Meerwald
+43-664-2444418 (mobile)

Re: [PATCH] iio: staging: ad7298_ring: Fix maybe-uninitialized warning

[Michael Hennerich]

On 07/16/2012 11:26 AM, Peter Meerwald wrote:
>> drivers/staging/iio/adc/ad7298_ring.c:97:37: warning: 'time_ns' may
>> be used uninitialized in this function [-Wmaybe-uninitialized]
>> diff --git a/drivers/staging/iio/adc/ad7298_ring.c b/drivers/staging/iio/adc/ad7298_ring.c
>> index fd1d855..506016f 100644
>> --- a/drivers/staging/iio/adc/ad7298_ring.c
>> +++ b/drivers/staging/iio/adc/ad7298_ring.c
>> @@ -76,7 +76,7 @@ static irqreturn_t ad7298_trigger_handler(int irq, void *p)
>>   	struct iio_dev *indio_dev = pf->indio_dev;
>>   	struct ad7298_state *st = iio_priv(indio_dev);
>>   	struct iio_buffer *ring = indio_dev->buffer;
>> -	s64 time_ns;
>> +	s64 time_ns = 0;
>>   	__u16 buf[16];
>>   	int b_sent, i;
> I'd rather suggest
> s64 time_ns = iio_get_time_ns();
>
> and simplify the if (indio_dev->scan_timestamp) block
>
> p.
Well - calling a function to initialize a variable which isn't being 
used anywhere doesn't make more sense.

IMHO a more adequate fix would be to remove the timestamp argument from 
store_to(),
since it isn't used anywhere.

-- 
Greetings,
Michael

--
Analog Devices GmbH      Wilhelm-Wagenfeld-Str. 6      80807 Muenchen
Sitz der Gesellschaft: Muenchen; Registergericht: Muenchen HRB 40368;
Geschaeftsfuehrer:Dr.Carsten Suckrow, Thomas Wessel, William A. Martin,
Margaret Seif

Re: [PATCH] iio: staging: ad7298_ring: Fix maybe-uninitialized warning

[Peter Meerwald]

> Well - calling a function to initialize a variable which isn't being used
> anywhere doesn't make more sense.
 
> IMHO a more adequate fix would be to remove the timestamp argument from
> store_to(),
> since it isn't used anywhere.

so
indio_dev->buffer->access->store_to(ring, (u8 *)buf, time_ns);
should become
iio_push_to_buffer(buffer, (u8 *)data->buffer, time_ns);
and you suggest to drop the last argument?

I wasn't aware that timestamp is not used; at least iio_store_to_kfifo() 
and iio_store_to_sw_ring() just ignore it

p.

-- 

Peter Meerwald
+43-664-2444418 (mobile)

Re: [PATCH] iio: staging: ad7298_ring: Fix maybe-uninitialized warning

[Jonathan Cameron]

On 7/16/2012 11:02 AM, Peter Meerwald wrote:
>
>> Well - calling a function to initialize a variable which isn't being used
>> anywhere doesn't make more sense.
>
>> IMHO a more adequate fix would be to remove the timestamp argument from
>> store_to(),
>> since it isn't used anywhere.
>
> so
> indio_dev->buffer->access->store_to(ring, (u8 *)buf, time_ns);
> should become
> iio_push_to_buffer(buffer, (u8 *)data->buffer, time_ns);
> and you suggest to drop the last argument?
>
> I wasn't aware that timestamp is not used; at least iio_store_to_kfifo()
> and iio_store_to_sw_ring() just ignore it
>
Yes. Been meaning to clean that up for a while, but as one of the
multiple buffer patches removes all those calls anyway I'll
kill it whilst doing that.