[PATCH] can: softing: remove redundant NULL check

[PATCH] can: softing: remove redundant NULL check

[Daniil Dulov]

In this case dev cannot be NULL, so remove redundant check.

Found by Linux Verification Center (linuxtesting.org) with SVACE.

Fixes: 03fd3cf5a179 ("can: add driver for Softing card")
Signed-off-by: Daniil Dulov <d.dulov@aladdin.ru>
---
 drivers/net/can/softing/softing_fw.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/can/softing/softing_fw.c b/drivers/net/can/softing/softing_fw.c
index bad69a4abec1..5a3f9e4b0b62 100644
--- a/drivers/net/can/softing/softing_fw.c
+++ b/drivers/net/can/softing/softing_fw.c
@@ -436,7 +436,7 @@ int softing_startstop(struct net_device *dev, int up)
 		return ret;
 
 	bus_bitmask_start = 0;
-	if (dev && up)
+	if (up)
 		/* prepare to start this bus as well */
 		bus_bitmask_start |= (1 << priv->index);
 	/* bring netdevs down */
-- 
2.25.1

Re: [PATCH] can: softing: remove redundant NULL check

[Simon Horman]

On Sun, Feb 11, 2024 at 07:05:35AM -0800, Daniil Dulov wrote:
> In this case dev cannot be NULL, so remove redundant check.
> 
> Found by Linux Verification Center (linuxtesting.org) with SVACE.
> 
> Fixes: 03fd3cf5a179 ("can: add driver for Softing card")
> Signed-off-by: Daniil Dulov <d.dulov@aladdin.ru>

Hi Daniil,

I am not sure that dev cannot be NULL.
But I do see that the code assumes it is not, and would crash if it is.
So I think that, functionally, your statement is correct.

	priv = netdev_priv(dev);
	card = priv->card;

Reviewed-by: Simon Horman <horms@kernel.org>

Re: [PATCH] can: softing: remove redundant NULL check

[Oliver Hartkopp]

Hi Simon,

I have a general question on the "Fixes:" tag in this patch:

On 16.02.24 18:27, Simon Horman wrote:
> On Sun, Feb 11, 2024 at 07:05:35AM -0800, Daniil Dulov wrote:
>> In this case dev cannot be NULL, so remove redundant check.
>>
>> Found by Linux Verification Center (linuxtesting.org) with SVACE.
>>
>> Fixes: 03fd3cf5a179 ("can: add driver for Softing card")

IMHO this is simply an improvement which is done by all patches applied 
to the kernel but it does not really "fix" anything from a functional 
standpoint.

Shouldn't we either invent a new tag or better leave it out to not 
confuse the stable maintainers?

Best regards,
Oliver

>> Signed-off-by: Daniil Dulov <d.dulov@aladdin.ru>
> 
> Hi Daniil,
> 
> I am not sure that dev cannot be NULL.
> But I do see that the code assumes it is not, and would crash if it is.
> So I think that, functionally, your statement is correct.
> 
> 	priv = netdev_priv(dev);
> 	card = priv->card;
> 
> Reviewed-by: Simon Horman <horms@kernel.org>
> 

Re: [PATCH] can: softing: remove redundant NULL check

[Simon Horman]

On Fri, Feb 16, 2024 at 08:47:43PM +0100, Oliver Hartkopp wrote:
> Hi Simon,
> 
> I have a general question on the "Fixes:" tag in this patch:
> 
> On 16.02.24 18:27, Simon Horman wrote:
> > On Sun, Feb 11, 2024 at 07:05:35AM -0800, Daniil Dulov wrote:
> > > In this case dev cannot be NULL, so remove redundant check.
> > > 
> > > Found by Linux Verification Center (linuxtesting.org) with SVACE.
> > > 
> > > Fixes: 03fd3cf5a179 ("can: add driver for Softing card")
> 
> IMHO this is simply an improvement which is done by all patches applied to
> the kernel but it does not really "fix" anything from a functional
> standpoint.
> 
> Shouldn't we either invent a new tag or better leave it out to not confuse
> the stable maintainers?

Hi Oliver,

sorry for missing that in my review.

Yes, I agree that this is probably not a fix, for which my
rule of thumb is something that addresses a user-visible problem.
So I agree it should not have a fixes tag.

I would suggest that we can just change the text to something that
has no tag. Something like:

...

Introduced by 03fd3cf5a179 ("can: add driver for Softing card")

Signed-of-by: ...


> 
> Best regards,
> Oliver
> 
> > > Signed-off-by: Daniil Dulov <d.dulov@aladdin.ru>
> > 
> > Hi Daniil,
> > 
> > I am not sure that dev cannot be NULL.
> > But I do see that the code assumes it is not, and would crash if it is.
> > So I think that, functionally, your statement is correct.
> > 
> > 	priv = netdev_priv(dev);
> > 	card = priv->card;
> > 
> > Reviewed-by: Simon Horman <horms@kernel.org>
> > 
> 

Re: [PATCH] can: softing: remove redundant NULL check

[Oliver Hartkopp]

Hi Simon,

On 2024-02-19 18:00, Simon Horman wrote:
> On Fri, Feb 16, 2024 at 08:47:43PM +0100, Oliver Hartkopp wrote:
>> Hi Simon,
>>
>> I have a general question on the "Fixes:" tag in this patch:
>>
>> On 16.02.24 18:27, Simon Horman wrote:
>>> On Sun, Feb 11, 2024 at 07:05:35AM -0800, Daniil Dulov wrote:
>>>> In this case dev cannot be NULL, so remove redundant check.
>>>>
>>>> Found by Linux Verification Center (linuxtesting.org) with SVACE.
>>>>
>>>> Fixes: 03fd3cf5a179 ("can: add driver for Softing card")
>>
>> IMHO this is simply an improvement which is done by all patches applied to
>> the kernel but it does not really "fix" anything from a functional
>> standpoint.
>>
>> Shouldn't we either invent a new tag or better leave it out to not confuse
>> the stable maintainers?
> 
> Hi Oliver,
> 
> sorry for missing that in my review.
> 
> Yes, I agree that this is probably not a fix, for which my
> rule of thumb is something that addresses a user-visible problem.
> So I agree it should not have a fixes tag.
> 
> I would suggest that we can just change the text to something that
> has no tag. Something like:
> 
> ...
> 
> Introduced by 03fd3cf5a179 ("can: add driver for Softing card")
> 

Yes, but the "Introduced-by:" tag would be an optional tag for people 
that like blaming others, right?

IMHO we should think about completely removing the "Fixes:" tag, when it 
has no user-visible effect that might be a candidate for stable kernels. 
It is common improvement work. And it has been so for years.

Best regards,
Oliver

Re: [PATCH] can: softing: remove redundant NULL check

[Simon Horman]

On Mon, Feb 19, 2024 at 09:37:46PM +0100, Oliver Hartkopp wrote:
> Hi Simon,
> 
> On 2024-02-19 18:00, Simon Horman wrote:
> > On Fri, Feb 16, 2024 at 08:47:43PM +0100, Oliver Hartkopp wrote:
> > > Hi Simon,
> > > 
> > > I have a general question on the "Fixes:" tag in this patch:
> > > 
> > > On 16.02.24 18:27, Simon Horman wrote:
> > > > On Sun, Feb 11, 2024 at 07:05:35AM -0800, Daniil Dulov wrote:
> > > > > In this case dev cannot be NULL, so remove redundant check.
> > > > > 
> > > > > Found by Linux Verification Center (linuxtesting.org) with SVACE.
> > > > > 
> > > > > Fixes: 03fd3cf5a179 ("can: add driver for Softing card")
> > > 
> > > IMHO this is simply an improvement which is done by all patches applied to
> > > the kernel but it does not really "fix" anything from a functional
> > > standpoint.
> > > 
> > > Shouldn't we either invent a new tag or better leave it out to not confuse
> > > the stable maintainers?
> > 
> > Hi Oliver,
> > 
> > sorry for missing that in my review.
> > 
> > Yes, I agree that this is probably not a fix, for which my
> > rule of thumb is something that addresses a user-visible problem.
> > So I agree it should not have a fixes tag.
> > 
> > I would suggest that we can just change the text to something that
> > has no tag. Something like:
> > 
> > ...
> > 
> > Introduced by 03fd3cf5a179 ("can: add driver for Softing card")
> > 
> 
> Yes, but the "Introduced-by:" tag would be an optional tag for people that
> like blaming others, right?

Yes, That does seem useful to me.

> IMHO we should think about completely removing the "Fixes:" tag, when it has
> no user-visible effect that might be a candidate for stable kernels. It is
> common improvement work. And it has been so for years.

Likewise, that does sound like a good idea to me.