On Tue, 29 Oct 2024, Arnd Bergmann wrote:

> From: Arnd Bergmann <arnd@arndb.de>
> 
> node_to_amd_nb() is defined to NULL in non-AMD configs:
> 
> drivers/platform/x86/amd/hsmp/plat.c: In function 'init_platform_device':
> drivers/platform/x86/amd/hsmp/plat.c:165:68: error: dereferencing 'void *' pointer [-Werror]
>   165 |                 sock->root                      = node_to_amd_nb(i)->root;
>       |                                                                    ^~
> drivers/platform/x86/amd/hsmp/plat.c:165:68: error: request for member 'root' in something not a structure or union
> 
> Change the definition to something that builds. This does introduce a
> NULL pointer dereference but the code is never called since the driver
> won't probe successfully.

I don't like this very wording because what the code very much does is 
NULL check on node_to_amd_nb() which leads to immediate failure of 
.probe(). (We don't call other deferences after a NULL check "NULL pointer 
dereference" either so none is introduced by this patch, IMO.)

With that fixed,

Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>

-- 
 i.

> Fixes: 7d3135d16356 ("platform/x86/amd/hsmp: Create separate ACPI, plat and common drivers")
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
>  arch/x86/include/asm/amd_nb.h | 5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/x86/include/asm/amd_nb.h b/arch/x86/include/asm/amd_nb.h
> index 6f3b6aef47ba..d0caac26533f 100644
> --- a/arch/x86/include/asm/amd_nb.h
> +++ b/arch/x86/include/asm/amd_nb.h
> @@ -116,7 +116,10 @@ static inline bool amd_gart_present(void)
>  
>  #define amd_nb_num(x)		0
>  #define amd_nb_has_feature(x)	false
> -#define node_to_amd_nb(x)	NULL
> +static inline struct amd_northbridge *node_to_amd_nb(int node)
> +{
> +	return NULL;
> +}
>  #define amd_gart_present(x)	false
>  
>  #endif
>