[PATCH] iio: sca3000: Remove an erroneous 'get_device()'

[PATCH] iio: sca3000: Remove an erroneous 'get_device()'

[Christophe JAILLET]

This looks really unusual to have a 'get_device()' hidden in a 'dev_err()'
call.
Remove it.

While at it add a missing \n at the end of the message.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
---
This patch is purely speculative.
I've looked a bit arround and see no point for this get_device() but other
eyes are welcomed :)
---
 drivers/iio/accel/sca3000.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/iio/accel/sca3000.c b/drivers/iio/accel/sca3000.c
index 66d768d971e1..6e429072e44a 100644
--- a/drivers/iio/accel/sca3000.c
+++ b/drivers/iio/accel/sca3000.c
@@ -980,7 +980,7 @@ static int sca3000_read_data(struct sca3000_state *st,
 	st->tx[0] = SCA3000_READ_REG(reg_address_high);
 	ret = spi_sync_transfer(st->us, xfer, ARRAY_SIZE(xfer));
 	if (ret) {
-		dev_err(get_device(&st->us->dev), "problem reading register");
+		dev_err(&st->us->dev, "problem reading register\n");
 		return ret;
 	}
 
-- 
2.25.1

Re: [PATCH] iio: sca3000: Remove an erroneous 'get_device()'

[Andy Shevchenko]

On Wed, May 6, 2020 at 6:55 AM Christophe JAILLET
<christophe.jaillet@wanadoo.fr> wrote:
>
> This looks really unusual to have a 'get_device()' hidden in a 'dev_err()'
> call.
> Remove it.
>
> While at it add a missing \n at the end of the message.
>

It should have Fixes tag because it is a quite an issue (get_device()
breaks reference counting with all problems we may expect).

> Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
> ---
> This patch is purely speculative.
> I've looked a bit arround and see no point for this get_device() but other
> eyes are welcomed :)
> ---
>  drivers/iio/accel/sca3000.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/iio/accel/sca3000.c b/drivers/iio/accel/sca3000.c
> index 66d768d971e1..6e429072e44a 100644
> --- a/drivers/iio/accel/sca3000.c
> +++ b/drivers/iio/accel/sca3000.c
> @@ -980,7 +980,7 @@ static int sca3000_read_data(struct sca3000_state *st,
>         st->tx[0] = SCA3000_READ_REG(reg_address_high);
>         ret = spi_sync_transfer(st->us, xfer, ARRAY_SIZE(xfer));
>         if (ret) {
> -               dev_err(get_device(&st->us->dev), "problem reading register");
> +               dev_err(&st->us->dev, "problem reading register\n");
>                 return ret;
>         }
>
> --
> 2.25.1
>


-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH] iio: sca3000: Remove an erroneous 'get_device()'

[Marion & Christophe JAILLET]

Le 06/05/2020 à 12:38, Andy Shevchenko a écrit :
> On Wed, May 6, 2020 at 6:55 AM Christophe JAILLET
> <christophe.jaillet@wanadoo.fr> wrote:
>> This looks really unusual to have a 'get_device()' hidden in a 'dev_err()'
>> call.
>> Remove it.
>>
>> While at it add a missing \n at the end of the message.
>>
> It should have Fixes tag because it is a quite an issue (get_device()
> breaks reference counting with all problems we may expect).

Agreed and I usually do, but here, I've lost track when this driver has 
gone out of staging.

Based on:
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/log/drivers/iio/accel/sca3000.c
The issue was already there on 2016/10/23, but when I try to go one step 
further:

https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/drivers/staging/iio/accel/sca3000.c?id,cf61442ff142d2dde7c47471c2798a4d78b0ad
^^^^         ^^^^^^^
works but if I try to see the log for that:
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/log/drivers/staging/iio/accel/sca3000.c
^^^         ^^^^^^^
is empty.

Most of the time, when I do it like that it works just fine, but not on 
this file.

Any other way to navigate in history of moved file would be appreciated.

CJ

>
>> Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
>> ---
>> This patch is purely speculative.
>> I've looked a bit arround and see no point for this get_device() but other
>> eyes are welcomed :)
>> ---
>>   drivers/iio/accel/sca3000.c | 2 +-
>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/iio/accel/sca3000.c b/drivers/iio/accel/sca3000.c
>> index 66d768d971e1..6e429072e44a 100644
>> --- a/drivers/iio/accel/sca3000.c
>> +++ b/drivers/iio/accel/sca3000.c
>> @@ -980,7 +980,7 @@ static int sca3000_read_data(struct sca3000_state *st,
>>          st->tx[0] = SCA3000_READ_REG(reg_address_high);
>>          ret = spi_sync_transfer(st->us, xfer, ARRAY_SIZE(xfer));
>>          if (ret) {
>> -               dev_err(get_device(&st->us->dev), "problem reading register");
>> +               dev_err(&st->us->dev, "problem reading register\n");
>>                  return ret;
>>          }
>>
>> --
>> 2.25.1
>>
>

Re: [PATCH] iio: sca3000: Remove an erroneous 'get_device()'

[Jonathan Cameron]

On Wed, 6 May 2020 19:31:38 +0200
Marion & Christophe JAILLET <christophe.jaillet@wanadoo.fr> wrote:

> Le 06/05/2020 à 12:38, Andy Shevchenko a écrit :
> > On Wed, May 6, 2020 at 6:55 AM Christophe JAILLET
> > <christophe.jaillet@wanadoo.fr> wrote:  
> >> This looks really unusual to have a 'get_device()' hidden in a 'dev_err()'
> >> call.
> >> Remove it.
> >>
> >> While at it add a missing \n at the end of the message.
> >>  
> > It should have Fixes tag because it is a quite an issue (get_device()
> > breaks reference counting with all problems we may expect).  
> 
> Agreed and I usually do, but here, I've lost track when this driver has 
> gone out of staging.
> 
> Based on:
> https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/log/drivers/iio/accel/sca3000.c
> The issue was already there on 2016/10/23, but when I try to go one step 
> further:
> 
> https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/drivers/staging/iio/accel/sca3000.c?id,cf61442ff142d2dde7c47471c2798a4d78b0ad
> ^^^^         ^^^^^^^
> works but if I try to see the log for that:
> https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/log/drivers/staging/iio/accel/sca3000.c
> ^^^         ^^^^^^^
> is empty.
> 
> Most of the time, when I do it like that it works just fine, but not on 
> this file.

Oddity of the web interface. Edit the url to get a log form a commit

https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/log/drivers/staging/iio/accel/sca3000.c?id,cf61442ff142d2dde7c47471c2798a4d78b0ad
                                                                    ^^^
Then for more fun you have to deal with another rename
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/drivers/staging/iio/accel/?idÎd5c03d360aeebaac6faa7dd8d6d0a77733ab16

Next bit of log...
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/log/drivers/staging/iio/accel/sca3000_ring.c?idÎd5c03d360aeebaac6faa7dd8d6d0a77733ab16 

It goes all the way..

https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/drivers/staging/iio/accel/sca3000_core.c?idW4fb258d63658e4564c32c1940068a3bad666a0

Fixes: 574fb258d636 ("Staging: IIO: VTI sca3000 series accelerometer driver (spi)")

Which was in the initial IIO patch set nearly 11 years ago in the merge window for 2.6.32.

What can I say - I was young and just as capable of writing dumb bugs as I am today :)

I'll get this queued up when I'm on the right computer...

Thanks,

Jonathan


> 
> Any other way to navigate in history of moved file would be appreciated.
> 
> CJ
> 
> >  
> >> Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
> >> ---
> >> This patch is purely speculative.
> >> I've looked a bit arround and see no point for this get_device() but other
> >> eyes are welcomed :)
> >> ---
> >>   drivers/iio/accel/sca3000.c | 2 +-
> >>   1 file changed, 1 insertion(+), 1 deletion(-)
> >>
> >> diff --git a/drivers/iio/accel/sca3000.c b/drivers/iio/accel/sca3000.c
> >> index 66d768d971e1..6e429072e44a 100644
> >> --- a/drivers/iio/accel/sca3000.c
> >> +++ b/drivers/iio/accel/sca3000.c
> >> @@ -980,7 +980,7 @@ static int sca3000_read_data(struct sca3000_state *st,
> >>          st->tx[0] = SCA3000_READ_REG(reg_address_high);
> >>          ret = spi_sync_transfer(st->us, xfer, ARRAY_SIZE(xfer));
> >>          if (ret) {
> >> -               dev_err(get_device(&st->us->dev), "problem reading register");
> >> +               dev_err(&st->us->dev, "problem reading register\n");
> >>                  return ret;
> >>          }
> >>
> >> --
> >> 2.25.1
> >>  
> >  

Re: [PATCH] iio: sca3000: Remove an erroneous 'get_device()'

[Jonathan Cameron]

On Fri, 8 May 2020 14:02:21 +0100
Jonathan Cameron <Jonathan.Cameron@Huawei.com> wrote:

> On Wed, 6 May 2020 19:31:38 +0200
> Marion & Christophe JAILLET <christophe.jaillet@wanadoo.fr> wrote:
> 
> > Le 06/05/2020 à 12:38, Andy Shevchenko a écrit :  
> > > On Wed, May 6, 2020 at 6:55 AM Christophe JAILLET
> > > <christophe.jaillet@wanadoo.fr> wrote:    
> > >> This looks really unusual to have a 'get_device()' hidden in a 'dev_err()'
> > >> call.
> > >> Remove it.
> > >>
> > >> While at it add a missing \n at the end of the message.
> > >>    
> > > It should have Fixes tag because it is a quite an issue (get_device()
> > > breaks reference counting with all problems we may expect).    
> > 
> > Agreed and I usually do, but here, I've lost track when this driver has 
> > gone out of staging.
> > 
> > Based on:
> > https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/log/drivers/iio/accel/sca3000.c
> > The issue was already there on 2016/10/23, but when I try to go one step 
> > further:
> > 
> > https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/drivers/staging/iio/accel/sca3000.c?id=2ccf61442ff142d2dde7c47471c2798a4d78b0ad
> > ^^^^         ^^^^^^^
> > works but if I try to see the log for that:
> > https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/log/drivers/staging/iio/accel/sca3000.c
> > ^^^         ^^^^^^^
> > is empty.
> > 
> > Most of the time, when I do it like that it works just fine, but not on 
> > this file.  
> 
> Oddity of the web interface. Edit the url to get a log form a commit
> 
> https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/log/drivers/staging/iio/accel/sca3000.c?id=2ccf61442ff142d2dde7c47471c2798a4d78b0ad
>                                                                     ^^^
> Then for more fun you have to deal with another rename
> https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/drivers/staging/iio/accel/?id=ced5c03d360aeebaac6faa7dd8d6d0a77733ab16
> 
> Next bit of log...
> https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/log/drivers/staging/iio/accel/sca3000_ring.c?id=ced5c03d360aeebaac6faa7dd8d6d0a77733ab16 
> 
> It goes all the way..
> 
> https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/drivers/staging/iio/accel/sca3000_core.c?id=574fb258d63658e4564c32c1940068a3bad666a0
> 
> Fixes: 574fb258d636 ("Staging: IIO: VTI sca3000 series accelerometer driver (spi)")
> 
> Which was in the initial IIO patch set nearly 11 years ago in the merge window for 2.6.32.
> 
> What can I say - I was young and just as capable of writing dumb bugs as I am today :)
> 
> I'll get this queued up when I'm on the right computer...

Applied to the fixes-togreg branch of iio.git.

Thanks,

Jonathan

> 
> Thanks,
> 
> Jonathan
> 
> 
> > 
> > Any other way to navigate in history of moved file would be appreciated.
> > 
> > CJ
> >   
> > >    
> > >> Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
> > >> ---
> > >> This patch is purely speculative.
> > >> I've looked a bit arround and see no point for this get_device() but other
> > >> eyes are welcomed :)
> > >> ---
> > >>   drivers/iio/accel/sca3000.c | 2 +-
> > >>   1 file changed, 1 insertion(+), 1 deletion(-)
> > >>
> > >> diff --git a/drivers/iio/accel/sca3000.c b/drivers/iio/accel/sca3000.c
> > >> index 66d768d971e1..6e429072e44a 100644
> > >> --- a/drivers/iio/accel/sca3000.c
> > >> +++ b/drivers/iio/accel/sca3000.c
> > >> @@ -980,7 +980,7 @@ static int sca3000_read_data(struct sca3000_state *st,
> > >>          st->tx[0] = SCA3000_READ_REG(reg_address_high);
> > >>          ret = spi_sync_transfer(st->us, xfer, ARRAY_SIZE(xfer));
> > >>          if (ret) {
> > >> -               dev_err(get_device(&st->us->dev), "problem reading register");
> > >> +               dev_err(&st->us->dev, "problem reading register\n");
> > >>                  return ret;
> > >>          }
> > >>
> > >> --
> > >> 2.25.1
> > >>    
> > >    
> 
> 

Re: [PATCH] iio: sca3000: Remove an erroneous get_device() call in sca3000_read_data()

[Markus Elfring]

> This looks really unusual to have a 'get_device()' hidden in a 'dev_err()' call.

Is there a need to prevent similar function calls by the means of
advanced source code analysis?

Regards,
Markus

Re: [PATCH] iio: sca3000: Remove an erroneous get_device() call in sca3000_read_data()

[Jonathan Cameron]

On Wed, 6 May 2020 17:30:06 +0200
Markus Elfring <Markus.Elfring@web.de> wrote:

> > This looks really unusual to have a 'get_device()' hidden in a 'dev_err()' call.  
> 
> Is there a need to prevent similar function calls by the means of
> advanced source code analysis?

It's a seriously stupid bug so I'd have thought unlikely to
occur often enough to be worth it.

Jonathan

> 
> Regards,
> Markus