Re: [PATCH 3/4] printk: use a clever macro

[Alex Elder <elder@linaro.org>]

On 07/09/2014 12:58 PM, Geert Uytterhoeven wrote:
> Hi Alex,
> 
> On Wed, Jul 9, 2014 at 3:04 PM, Alex Elder <elder@linaro.org> wrote:
>> --- a/kernel/printk/printk.c
>> +++ b/kernel/printk/printk.c
>> @@ -453,11 +453,7 @@ static int log_store(int facility, int level,
>>         return msg->text_len;
>>  }
>>
>> -#ifdef CONFIG_SECURITY_DMESG_RESTRICT
>> -int dmesg_restrict = 1;
>> -#else
>> -int dmesg_restrict;
>> -#endif
>> +int dmesg_restrict = IS_ENABLED(CONFIG_SECURITY_DMESG_RESTRICT);
> 
> Doesn't this move dmesg_restrict from the bss to the data section
> in case CONFIG_SECURITY_DMESG_RESTRICT is not enabled, due
> to the explicit initialization to zero?

I honestly don't know.  Is that even a well-defined behavior?
Couldn't the compiler, noting an explicit 0 initialization,
put it into BSS anyway?

In any case, does this distinction matter?

					-Alex

> Gr{oetje,eeting}s,
> 
>                         Geert
> 
> --
> Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
> 
> In personal conversations with technical people, I call myself a hacker. But
> when I'm talking to journalists I just say "programmer" or something like that.
>                                 -- Linus Torvalds
>