[PATCH -next] powerpc/spufs: Fix build warning when CONFIG_PROC_FS=n

[PATCH -next] powerpc/spufs: Fix build warning when CONFIG_PROC_FS=n

[YueHaibing]

arch/powerpc/platforms/cell/spufs/sched.c:1055:12: warning: ‘show_spu_loadavg’ defined but not used [-Wunused-function]
 static int show_spu_loadavg(struct seq_file *s, void *private)
            ^~~~~~~~~~~~~~~~

Mark this as __maybe_unused to fix this.

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
---
 arch/powerpc/platforms/cell/spufs/sched.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/powerpc/platforms/cell/spufs/sched.c b/arch/powerpc/platforms/cell/spufs/sched.c
index d058f6233e66..c46835625133 100644
--- a/arch/powerpc/platforms/cell/spufs/sched.c
+++ b/arch/powerpc/platforms/cell/spufs/sched.c
@@ -1052,7 +1052,7 @@ void spuctx_switch_state(struct spu_context *ctx,
 	}
 }
 
-static int show_spu_loadavg(struct seq_file *s, void *private)
+static int __maybe_unused show_spu_loadavg(struct seq_file *s, void *private)
 {
 	int a, b, c;
 
-- 
2.17.1

Re: [PATCH -next] powerpc/spufs: Fix build warning when CONFIG_PROC_FS=n

[Christophe Leroy]

Le 05/03/2022 à 13:31, YueHaibing a écrit :
> arch/powerpc/platforms/cell/spufs/sched.c:1055:12: warning: ‘show_spu_loadavg’ defined but not used [-Wunused-function]
>   static int show_spu_loadavg(struct seq_file *s, void *private)
>              ^~~~~~~~~~~~~~~~
> 
> Mark this as __maybe_unused to fix this.

Marking it as __maybe_unused doesn't fix it. It just pushes the dust 
under the carpet.

proc_create_single macro should be fix to avoid that warning.

> 
> Signed-off-by: YueHaibing <yuehaibing@huawei.com>
> ---
>   arch/powerpc/platforms/cell/spufs/sched.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/arch/powerpc/platforms/cell/spufs/sched.c b/arch/powerpc/platforms/cell/spufs/sched.c
> index d058f6233e66..c46835625133 100644
> --- a/arch/powerpc/platforms/cell/spufs/sched.c
> +++ b/arch/powerpc/platforms/cell/spufs/sched.c
> @@ -1052,7 +1052,7 @@ void spuctx_switch_state(struct spu_context *ctx,
>   	}
>   }
>   
> -static int show_spu_loadavg(struct seq_file *s, void *private)
> +static int __maybe_unused show_spu_loadavg(struct seq_file *s, void *private)
>   {
>   	int a, b, c;
>   

Re: [PATCH -next] powerpc/spufs: Fix build warning when CONFIG_PROC_FS=n

[Arnd Bergmann]

On Sun, Mar 6, 2022 at 9:04 PM Christophe Leroy
<christophe.leroy@csgroup.eu> wrote:
> Le 05/03/2022 à 13:31, YueHaibing a écrit :
> > arch/powerpc/platforms/cell/spufs/sched.c:1055:12: warning: ‘show_spu_loadavg’ defined but not used [-Wunused-function]
> >   static int show_spu_loadavg(struct seq_file *s, void *private)
> >              ^~~~~~~~~~~~~~~~
> >
> > Mark this as __maybe_unused to fix this.
>
> Marking it as __maybe_unused doesn't fix it. It just pushes the dust
> under the carpet.
>
> proc_create_single macro should be fix to avoid that warning.

We discussed that when proc_create_single() was introduced, but ended up
not doing it that way because there were already a lot of files using an #ifdef
around the function definitions. To change it back, one would have to audit
every user of proc_create_single() and remove the #ifdefs.

        Arnd

Re: [PATCH -next] powerpc/spufs: Fix build warning when CONFIG_PROC_FS=n

[Christophe Leroy]

Le 07/03/2022 à 14:10, Arnd Bergmann a écrit :
> On Sun, Mar 6, 2022 at 9:04 PM Christophe Leroy
> <christophe.leroy@csgroup.eu> wrote:
>> Le 05/03/2022 à 13:31, YueHaibing a écrit :
>>> arch/powerpc/platforms/cell/spufs/sched.c:1055:12: warning: ‘show_spu_loadavg’ defined but not used [-Wunused-function]
>>>    static int show_spu_loadavg(struct seq_file *s, void *private)
>>>               ^~~~~~~~~~~~~~~~
>>>
>>> Mark this as __maybe_unused to fix this.
>>
>> Marking it as __maybe_unused doesn't fix it. It just pushes the dust
>> under the carpet.
>>
>> proc_create_single macro should be fix to avoid that warning.
> 
> We discussed that when proc_create_single() was introduced, but ended up
> not doing it that way because there were already a lot of files using an #ifdef
> around the function definitions. To change it back, one would have to audit
> every user of proc_create_single() and remove the #ifdefs.
> 

Fair enough.

In that case, I'd prefer to go for a #ifdef as well for 
show_spu_loadavg() instead of going for a __maybe_unused flag.

Christophe

Re: [PATCH -next] powerpc/spufs: Fix build warning when CONFIG_PROC_FS=n

[YueHaibing]

On 2022/3/7 21:30, Christophe Leroy wrote:
> 
> 
> Le 07/03/2022 à 14:10, Arnd Bergmann a écrit :
>> On Sun, Mar 6, 2022 at 9:04 PM Christophe Leroy
>> <christophe.leroy@csgroup.eu> wrote:
>>> Le 05/03/2022 à 13:31, YueHaibing a écrit :
>>>> arch/powerpc/platforms/cell/spufs/sched.c:1055:12: warning: ‘show_spu_loadavg’ defined but not used [-Wunused-function]
>>>>    static int show_spu_loadavg(struct seq_file *s, void *private)
>>>>               ^~~~~~~~~~~~~~~~
>>>>
>>>> Mark this as __maybe_unused to fix this.
>>>
>>> Marking it as __maybe_unused doesn't fix it. It just pushes the dust
>>> under the carpet.
>>>
>>> proc_create_single macro should be fix to avoid that warning.
>>
>> We discussed that when proc_create_single() was introduced, but ended up
>> not doing it that way because there were already a lot of files using an #ifdef
>> around the function definitions. To change it back, one would have to audit
>> every user of proc_create_single() and remove the #ifdefs.
>>
> 
> Fair enough.
> 
> In that case, I'd prefer to go for a #ifdef as well for 
> show_spu_loadavg() instead of going for a __maybe_unused flag.

Ok, will do that in v2.
> 
> Christophe
>