Re: [PATCH] net/core/sysctl_net_core.c unused variable

Re: [PATCH] net/core/sysctl_net_core.c unused variable

[Nick Warne]

On 06/09/15 16:28, Joe Perches wrote:
> On Sun, 2015-09-06 at 16:16 +0100, Nick Warne wrote:
>> On 06/09/15 15:52, Joe Perches wrote:
>> > On Sun, 2015-09-06 at 15:13 +0100, Nick Warne wrote:
>> >> gcc version 4.8.2 (GCC) warns that 'static int one = 1;' is declared but
>> >> not used in file net/core/sysctl_net_core.c.
>> >
>> > Only when CONFIG_NET isn't set.
>>
>> CONFIG_NET=y
>>
>> Peculiar indeed.
>>
>> >> Reading the file, that is
>> >> the case.  Attached is a patch to remove it.
>> >
>> > $ git grep -w -n one net/core/sysctl_net_core.c
>> > net/core/sysctl_net_core.c:26:static int one = 1;
>> > net/core/sysctl_net_core.c:332:         .extra2         = &one
>> >
>> >> Signed-off-by:  Nick Warne <nick@linicks.net>
>> >
>> > Please use grep to augment reading.
>>
>> grep -w -n one net/core/sysctl_net_core.c
>> 26:static int one = 1;
>>
>> ?
>>
>> I just don't have the &one.
>>
>> I am confused now.
>
> What source tree are you using?

Latest longterm 3.18.21

> What changes in what branch exist?

I am not using git (if that is what you mean by 'branches') - just 
tarballs from kernel.org
>
> btw: please use scripts/get_maintainer.pl to better
> determine who should be cc'd on your patches.
>
> you left out netdev.

Sorry, my bad, I need to learn/read more.

Thanks for your help/advice :)

Nick
-- 
Gosh that takes me back... or is it forward?  That's the trouble with
time travel, you never can tell."
		-- Doctor Who "Androids of Tara"

Re: [PATCH] net/core/sysctl_net_core.c unused variable

[Joe Perches]

On Sun, 2015-09-06 at 16:36 +0100, Nick Warne wrote:
> On 06/09/15 16:28, Joe Perches wrote:
> > On Sun, 2015-09-06 at 16:16 +0100, Nick Warne wrote:
> >> On 06/09/15 15:52, Joe Perches wrote:
> >> > On Sun, 2015-09-06 at 15:13 +0100, Nick Warne wrote:
> >> >> gcc version 4.8.2 (GCC) warns that 'static int one = 1;' is declared but
> >> >> not used in file net/core/sysctl_net_core.c.
> >> >
> >> > Only when CONFIG_NET isn't set.
> >>
> >> CONFIG_NET=y
> >>
> >> Peculiar indeed.
> >>
> >> >> Reading the file, that is
> >> >> the case.  Attached is a patch to remove it.
> >> >
> >> > $ git grep -w -n one net/core/sysctl_net_core.c
> >> > net/core/sysctl_net_core.c:26:static int one = 1;
> >> > net/core/sysctl_net_core.c:332:         .extra2         = &one
> >> >
> >> >> Signed-off-by:  Nick Warne <nick@linicks.net>
> >> >
> >> > Please use grep to augment reading.
> >>
> >> grep -w -n one net/core/sysctl_net_core.c
> >> 26:static int one = 1;
> >>
> >> ?
> >>
> >> I just don't have the &one.
> >>
> >> I am confused now.
> >
> > What source tree are you using?
> 
> Latest longterm 3.18.21

OK, it's important to mention that otherwise the
assumption would be a git tree like net or net-next.

> > What changes in what branch exist?
> 
> I am not using git (if that is what you mean by 'branches') - just 
> tarballs from kernel.org

(OK, using git and linux-stable)

$ git grep -w -n one v3.18.21 -- net/core/sysctl_net_core.c
v3.18.21:net/core/sysctl_net_core.c:26:static int one = 1;

And the responsible commit:

commit c48cf4f27d4555a455c3fef71137bd0fc44d1656
("net: sysctl_net_core: check SNDBUF and RCVBUF for min length")

Re: [PATCH] net/core/sysctl_net_core.c unused variable

[Nick Warne]

On 06/09/15 16:51, Joe Perches wrote:
> On Sun, 2015-09-06 at 16:36 +0100, Nick Warne wrote:
>> On 06/09/15 16:28, Joe Perches wrote:
>> > On Sun, 2015-09-06 at 16:16 +0100, Nick Warne wrote:
>> >> On 06/09/15 15:52, Joe Perches wrote:
>> >> > On Sun, 2015-09-06 at 15:13 +0100, Nick Warne wrote:
>> >> >> gcc version 4.8.2 (GCC) warns that 'static int one = 1;' is declared but
>> >> >> not used in file net/core/sysctl_net_core.c.
>> >> >
>> >> > Only when CONFIG_NET isn't set.
>> >>
>> >> CONFIG_NET=y
>> >>
>> >> Peculiar indeed.
>> >>
>> >> >> Reading the file, that is
>> >> >> the case.  Attached is a patch to remove it.
>> >> >
>> >> > $ git grep -w -n one net/core/sysctl_net_core.c
>> >> > net/core/sysctl_net_core.c:26:static int one = 1;
>> >> > net/core/sysctl_net_core.c:332:         .extra2         = &one
>> >> >
>> >> >> Signed-off-by:  Nick Warne <nick@linicks.net>
>> >> >
>> >> > Please use grep to augment reading.
>> >>
>> >> grep -w -n one net/core/sysctl_net_core.c
>> >> 26:static int one = 1;
>> >>
>> >> ?
>> >>
>> >> I just don't have the &one.
>> >>
>> >> I am confused now.
>> >
>> > What source tree are you using?
>>
>> Latest longterm 3.18.21
>
> OK, it's important to mention that otherwise the
> assumption would be a git tree like net or net-next.
>
>> > What changes in what branch exist?
>>
>> I am not using git (if that is what you mean by 'branches') - just
>> tarballs from kernel.org
>
> (OK, using git and linux-stable)
>
> $ git grep -w -n one v3.18.21 -- net/core/sysctl_net_core.c
> v3.18.21:net/core/sysctl_net_core.c:26:static int one = 1;
>
> And the responsible commit:
>
> commit c48cf4f27d4555a455c3fef71137bd0fc44d1656
> ("net: sysctl_net_core: check SNDBUF and RCVBUF for min length")

Ah, OK. GCC was right - just the variable declaration was overlooked to 
be removed too.

http://git.kernel.org/cgit/linux/kernel/git/stable/linux-stable.git/commit/?id=c48cf4f27d4555a455c3fef71137bd0fc44d1656

My patch was right then (but wrong) :)

Thanks Joe,

Nick

-- 
Gosh that takes me back... or is it forward?  That's the trouble with
time travel, you never can tell."
		-- Doctor Who "Androids of Tara"