Re: [PATCH] MIPS: Add prototypes for plat_post_relocation() and relocate_kernel()

[Thomas Bogendoerfer <tsbogend@alpha.franken.de>]

On Wed, Apr 03, 2024 at 02:26:33PM -0700, Nathan Chancellor wrote:
> When building malta_defconfig with CONFIG_RELOCATABLE=y, there are two
> warnings due to missing prototypes for functions only used when that
> configuration is enabled:
> 
>   arch/mips/kernel/relocate.c:42:12: warning: no previous prototype for 'plat_post_relocation' [-Wmissing-prototypes]
>      42 | int __weak plat_post_relocation(long offset)
>         |            ^~~~~~~~~~~~~~~~~~~~
>   arch/mips/kernel/relocate.c:324:14: warning: no previous prototype for 'relocate_kernel' [-Wmissing-prototypes]
>     324 | void *__init relocate_kernel(void)
>         |              ^~~~~~~~~~~~~~~
> 
> While relocate_kernel() is only called from assembly, it makes sense to
> keep the prototypes together in C to fix the warnings. Add them to
> silence the warnings.
> 
> Signed-off-by: Nathan Chancellor <nathan@kernel.org>
> ---
>  arch/mips/include/asm/setup.h | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/arch/mips/include/asm/setup.h b/arch/mips/include/asm/setup.h
> index 4dce41138bad..d8077136372c 100644
> --- a/arch/mips/include/asm/setup.h
> +++ b/arch/mips/include/asm/setup.h
> @@ -2,6 +2,7 @@
>  #ifndef _MIPS_SETUP_H
>  #define _MIPS_SETUP_H
>  
> +#include <linux/init.h>
>  #include <linux/types.h>
>  #include <uapi/asm/setup.h>
>  
> @@ -29,4 +30,9 @@ extern void per_cpu_trap_init(bool);
>  extern void cpu_cache_init(void);
>  extern void tlb_init(void);
>  
> +#ifdef CONFIG_RELOCATABLE
> +extern void * __init relocate_kernel(void);
> +extern int plat_post_relocation(long);
> +#endif
> +
>  #endif /* __SETUP_H */
> 
> ---
> base-commit: 39cd87c4eb2b893354f3b850f916353f2658ae6f
> change-id: 20240403-mips-kaslr-missing-prototypes-b15cbf88513f
> 
> Best regards,
> -- 
> Nathan Chancellor <nathan@kernel.org>

applied to mips-next.

Thomas.

-- 
Crap can work. Given enough thrust pigs will fly, but it's not necessarily a
good idea.                                                [ RFC1925, 2.3 ]