[PATCH 6/12] drivers/platform/x86: Correct redundant test

[PATCH 6/12] drivers/platform/x86: Correct redundant test

[Julia Lawall]

From: Julia Lawall <julia@diku.dk>

device and acpi_driver_data(device) were tested just a few lines above.

A simplified version of the semantic match that finds this problem is as
follows: (http://www.emn.fr/x-info/coccinelle/)

// <smpl>
@r exists@
local idexpression x;
expression E;
@@

if (x == NULL || ...) { ... when forall
   return ...; }
... when != \(x=E\|x--\|x++\|--x\|++x\|x-=E\|x+=E\|x|=E\|x&=E\|&x\)
(
*x == NULL
|
*x != NULL
)
// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>

---
 drivers/platform/x86/fujitsu-laptop.c          |    3 ---
 1 files changed, 0 insertions(+), 3 deletions(-)

diff --git a/drivers/platform/x86/fujitsu-laptop.c b/drivers/platform/x86/fujitsu-laptop.c
index 218b9a1..5306901 100644
--- a/drivers/platform/x86/fujitsu-laptop.c
+++ b/drivers/platform/x86/fujitsu-laptop.c
@@ -745,9 +745,6 @@ static int acpi_fujitsu_remove(struct acpi_device *device, int type)
 
 	fujitsu = acpi_driver_data(device);
 
-	if (!device || !acpi_driver_data(device))
-		return -EINVAL;
-
 	fujitsu->acpi_handle = NULL;
 
 	return 0;

Re: [PATCH 6/12] drivers/platform/x86: Correct redundant test

[Paulo Marques]

Julia Lawall wrote:
> [...]
> ---
>  drivers/platform/x86/fujitsu-laptop.c          |    3 ---
>  1 files changed, 0 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/platform/x86/fujitsu-laptop.c b/drivers/platform/x86/fujitsu-laptop.c
> index 218b9a1..5306901 100644
> --- a/drivers/platform/x86/fujitsu-laptop.c
> +++ b/drivers/platform/x86/fujitsu-laptop.c
> @@ -745,9 +745,6 @@ static int acpi_fujitsu_remove(struct acpi_device *device, int type)
>  
>  	fujitsu = acpi_driver_data(device);
>  
> -	if (!device || !acpi_driver_data(device))
> -		return -EINVAL;
> -

Shouldn't this still do a:

   if (!fujitsu)
     return -EINVAL;

to avoid dereferencing a NULL pointer below?

>  	fujitsu->acpi_handle = NULL;
>  
>  	return 0;

-- 
Paulo Marques - www.grupopie.com

"All I ask is a chance to prove that money can't make me happy."

Re: [PATCH 6/12] drivers/platform/x86: Correct redundant test

[Julia Lawall]

On Tue, 28 Jul 2009, Paulo Marques wrote:

> Julia Lawall wrote:
> > [...]
> > ---
> >  drivers/platform/x86/fujitsu-laptop.c          |    3 ---
> >  1 files changed, 0 insertions(+), 3 deletions(-)
> > 
> > diff --git a/drivers/platform/x86/fujitsu-laptop.c b/drivers/platform/x86/fujitsu-laptop.c
> > index 218b9a1..5306901 100644
> > --- a/drivers/platform/x86/fujitsu-laptop.c
> > +++ b/drivers/platform/x86/fujitsu-laptop.c
> > @@ -745,9 +745,6 @@ static int acpi_fujitsu_remove(struct acpi_device *device, int type)
> >  
> >  	fujitsu = acpi_driver_data(device);
> >  
> > -	if (!device || !acpi_driver_data(device))
> > -		return -EINVAL;
> > -
> 
> Shouldn't this still do a:
> 
>    if (!fujitsu)
>      return -EINVAL;

acpi_driver_data just accesses a field of its argument.  Is there a worry 
that from one call to the next it could have a different value?

Perhaps it would be better to first test !device, then initialize fujitsu, 
and then test the result of fujitsu?  The acpi_driver_data, which might 
someday do something more complicated, would only be called once.

julia

Re: [PATCH 6/12] drivers/platform/x86: Correct redundant test

[Jonathan Woithe]

Hi guys

> Julia Lawall wrote:
> > [...]
> > ---
> >  drivers/platform/x86/fujitsu-laptop.c          |    3 ---
> >  1 files changed, 0 insertions(+), 3 deletions(-)
> > 
> > diff --git a/drivers/platform/x86/fujitsu-laptop.c b/drivers/platform/x86/fujitsu-laptop.c
> > index 218b9a1..5306901 100644
> > --- a/drivers/platform/x86/fujitsu-laptop.c
> > +++ b/drivers/platform/x86/fujitsu-laptop.c
> > @@ -745,9 +745,6 @@ static int acpi_fujitsu_remove(struct acpi_device *device, int type)
> >  
> >  	fujitsu = acpi_driver_data(device);
> >  
> > -	if (!device || !acpi_driver_data(device))
> > -		return -EINVAL;
> > -
> 
> Shouldn't this still do a:
> 
>    if (!fujitsu)
>      return -EINVAL;
> 
> to avoid dereferencing a NULL pointer below?

Hmm, yes it should.  Well spotted.  And I'm not certain how the duplicate
test on "device" got in there in the first place.  I suspect it came about
due to some structural changes made a few versions ago and I failed to
notice that the second check became redundant.

So, combining this with the above patch we should instead do

Signed-off-by: jwoithe@physics.adelaide.edu.au <Jonathan Woithe>

--- a/drivers/platform/x86/fujitsu-laptop.c	2009-06-12 19:51:45.333234000 +0930
+++ b/drivers/platform/x86/fujitsu-laptop.c	2009-07-29 10:14:30.610249941 +0930
@@ -745,7 +745,7 @@ static int acpi_fujitsu_remove(struct ac
 
 	fujitsu = acpi_driver_data(device);
 
-	if (!device || !acpi_driver_data(device))
+	if (!fujitsu)
 		return -EINVAL;
 
 	fujitsu->acpi_handle = NULL;

Regards
  jonathan

Re: [PATCH 6/12] drivers/platform/x86: Correct redundant test

[Julia Lawall]

On Wed, 29 Jul 2009, Jonathan Woithe wrote:

> Hi guys
> 
> > Julia Lawall wrote:
> > > [...]
> > > ---
> > >  drivers/platform/x86/fujitsu-laptop.c          |    3 ---
> > >  1 files changed, 0 insertions(+), 3 deletions(-)
> > > 
> > > diff --git a/drivers/platform/x86/fujitsu-laptop.c b/drivers/platform/x86/fujitsu-laptop.c
> > > index 218b9a1..5306901 100644
> > > --- a/drivers/platform/x86/fujitsu-laptop.c
> > > +++ b/drivers/platform/x86/fujitsu-laptop.c
> > > @@ -745,9 +745,6 @@ static int acpi_fujitsu_remove(struct acpi_device *device, int type)
> > >  
> > >  	fujitsu = acpi_driver_data(device);
> > >  
> > > -	if (!device || !acpi_driver_data(device))
> > > -		return -EINVAL;
> > > -
> > 
> > Shouldn't this still do a:
> > 
> >    if (!fujitsu)
> >      return -EINVAL;
> > 
> > to avoid dereferencing a NULL pointer below?
> 
> Hmm, yes it should.  Well spotted.  And I'm not certain how the duplicate
> test on "device" got in there in the first place.  I suspect it came about
> due to some structural changes made a few versions ago and I failed to
> notice that the second check became redundant.

If you are going to check fujitsu afterwards, then I think there is no 
need to test the result of acpi_driver_data before.

julia

 
> So, combining this with the above patch we should instead do
> 
> Signed-off-by: jwoithe@physics.adelaide.edu.au <Jonathan Woithe>
> 
> --- a/drivers/platform/x86/fujitsu-laptop.c	2009-06-12 19:51:45.333234000 +0930
> +++ b/drivers/platform/x86/fujitsu-laptop.c	2009-07-29 10:14:30.610249941 +0930
> @@ -745,7 +745,7 @@ static int acpi_fujitsu_remove(struct ac
>  
>  	fujitsu = acpi_driver_data(device);
>  
> -	if (!device || !acpi_driver_data(device))
> +	if (!fujitsu)
>  		return -EINVAL;
>  
>  	fujitsu->acpi_handle = NULL;
> 
> Regards
>   jonathan
> --
> To unsubscribe from this list: send the line "unsubscribe kernel-janitors" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

Re: [PATCH 6/12] drivers/platform/x86: Correct redundant test

[Jonathan Woithe]

Hi Julia

> > > Julia Lawall wrote:
> > > > [...]
> > > > ---
> > > >  drivers/platform/x86/fujitsu-laptop.c          |    3 ---
> > > >  1 files changed, 0 insertions(+), 3 deletions(-)
> > > > 
> > > > diff --git a/drivers/platform/x86/fujitsu-laptop.c b/drivers/platform/x86/fujitsu-laptop.c
> > > > index 218b9a1..5306901 100644
> > > > --- a/drivers/platform/x86/fujitsu-laptop.c
> > > > +++ b/drivers/platform/x86/fujitsu-laptop.c
> > > > @@ -745,9 +745,6 @@ static int acpi_fujitsu_remove(struct acpi_device *device, int type)
> > > >  
> > > >  	fujitsu = acpi_driver_data(device);
> > > >  
> > > > -	if (!device || !acpi_driver_data(device))
> > > > -		return -EINVAL;
> > > > -
> > > 
> > > Shouldn't this still do a:
> > > 
> > >    if (!fujitsu)
> > >      return -EINVAL;
> > > 
> > > to avoid dereferencing a NULL pointer below?
> > 
> > Hmm, yes it should.  Well spotted.  And I'm not certain how the duplicate
> > test on "device" got in there in the first place.  I suspect it came about
> > due to some structural changes made a few versions ago and I failed to
> > notice that the second check became redundant.
> 
> If you are going to check fujitsu afterwards, then I think there is no 
> need to test the result of acpi_driver_data before.

Yes, of course.  I'll wake up soon, promise!

So we're left with this.

Signed-off-by: jwoithe@physics.adelaide.edu.au <Jonathan Woithe>

--- a/drivers/platform/x86/fujitsu-laptop.c	2009-06-12 19:51:45.333234000 +0930
+++ b/drivers/platform/x86/fujitsu-laptop.c	2009-07-29 12:10:11.504901871 +0930
@@ -740,12 +740,12 @@ static int acpi_fujitsu_remove(struct ac
 {
        struct fujitsu_t *fujitsu = NULL;
 
-	if (!device || !acpi_driver_data(device))
+	if (!device)
 		return -EINVAL;
 
 	fujitsu = acpi_driver_data(device);
 
-	if (!device || !acpi_driver_data(device))
+	if (!fujitsu)
 		return -EINVAL;
 
	fujitsu->acpi_handle = NULL;


Regards
  jonathan

Re: [PATCH 6/12] drivers/platform/x86: Correct redundant test

[Julia Lawall]

On Wed, 29 Jul 2009, Jonathan Woithe wrote:

> Hi Julia
> 
> > > > Julia Lawall wrote:
> > > > > [...]
> > > > > ---
> > > > >  drivers/platform/x86/fujitsu-laptop.c          |    3 ---
> > > > >  1 files changed, 0 insertions(+), 3 deletions(-)
> > > > > 
> > > > > diff --git a/drivers/platform/x86/fujitsu-laptop.c b/drivers/platform/x86/fujitsu-laptop.c
> > > > > index 218b9a1..5306901 100644
> > > > > --- a/drivers/platform/x86/fujitsu-laptop.c
> > > > > +++ b/drivers/platform/x86/fujitsu-laptop.c
> > > > > @@ -745,9 +745,6 @@ static int acpi_fujitsu_remove(struct acpi_device *device, int type)
> > > > >  
> > > > >  	fujitsu = acpi_driver_data(device);
> > > > >  
> > > > > -	if (!device || !acpi_driver_data(device))
> > > > > -		return -EINVAL;
> > > > > -
> > > > 
> > > > Shouldn't this still do a:
> > > > 
> > > >    if (!fujitsu)
> > > >      return -EINVAL;
> > > > 
> > > > to avoid dereferencing a NULL pointer below?
> > > 
> > > Hmm, yes it should.  Well spotted.  And I'm not certain how the duplicate
> > > test on "device" got in there in the first place.  I suspect it came about
> > > due to some structural changes made a few versions ago and I failed to
> > > notice that the second check became redundant.
> > 
> > If you are going to check fujitsu afterwards, then I think there is no 
> > need to test the result of acpi_driver_data before.
> 
> Yes, of course.  I'll wake up soon, promise!
> 
> So we're left with this.

Looks fine now, thanks.
julia


> Signed-off-by: jwoithe@physics.adelaide.edu.au <Jonathan Woithe>
> 
> --- a/drivers/platform/x86/fujitsu-laptop.c	2009-06-12 19:51:45.333234000 +0930
> +++ b/drivers/platform/x86/fujitsu-laptop.c	2009-07-29 12:10:11.504901871 +0930
> @@ -740,12 +740,12 @@ static int acpi_fujitsu_remove(struct ac
>  {
>         struct fujitsu_t *fujitsu = NULL;
>  
> -	if (!device || !acpi_driver_data(device))
> +	if (!device)
>  		return -EINVAL;
>  
>  	fujitsu = acpi_driver_data(device);
>  
> -	if (!device || !acpi_driver_data(device))
> +	if (!fujitsu)
>  		return -EINVAL;
>  
> 	fujitsu->acpi_handle = NULL;
> 
> 
> Regards
>   jonathan
> --
> To unsubscribe from this list: send the line "unsubscribe kernel-janitors" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>