Re: [PATCH] kunit/fortify: Remove __kmalloc_node() test

[Vlastimil Babka]

On 5/31/24 8:57 PM, Kees Cook wrote:
> __kmalloc_node() is considered an "internal" function to the Slab, so
> drop it from explicit testing.

So is __kmalloc() and so I have the removal of both as part of the cleanup here:

https://lore.kernel.org/all/20240527090127.21979-2-vbabka@xxxxxxx/

which reminds me I should put it to -next at this point. Review still welcome :)

> Signed-off-by: Kees Cook <kees@xxxxxxxxxx>
> ---
> Cc: Vlastimil Babka <vbabka@xxxxxxx>
> Cc: linux-mm@xxxxxxxxx
> Cc: linux-hardening@xxxxxxxxxxxxxxx
> ---
>  lib/fortify_kunit.c | 3 ---
>  1 file changed, 3 deletions(-)
> 
> diff --git a/lib/fortify_kunit.c b/lib/fortify_kunit.c
> index 39da5b3bc649..f9cc467334ce 100644
> --- a/lib/fortify_kunit.c
> +++ b/lib/fortify_kunit.c
> @@ -235,9 +235,6 @@ static void fortify_test_alloc_size_##allocator##_dynamic(struct kunit *test) \
>  		kmalloc_array_node(alloc_size, 1, gfp, NUMA_NO_NODE),	\
>  		kfree(p));						\
>  	checker(expected_size, __kmalloc(alloc_size, gfp),		\
> -		kfree(p));						\
> -	checker(expected_size,						\
> -		__kmalloc_node(alloc_size, gfp, NUMA_NO_NODE),		\
>  		kfree(p));						\
>  									\
>  	orig = kmalloc(alloc_size, gfp);				\