[PATCH] init: verify that function is initcall_t at compile-time

[PATCH] init: verify that function is initcall_t at compile-time

[Marco Elver]

In the spirit of making it hard to misuse an interface, add a
compile-time assertion in the CONFIG_HAVE_ARCH_PREL32_RELOCATIONS case
to verify the initcall function matches initcall_t, because the inline
asm bypasses any type-checking the compiler would otherwise do. This
will help developers catch incorrect API use in all configurations.

A recent example of this is:
https://lkml.kernel.org/r/20210514140015.2944744-1-arnd@kernel.org

Signed-off-by: Marco Elver <elver@google.com>
Reviewed-by: Miguel Ojeda <ojeda@kernel.org>
Reviewed-by: Nathan Chancellor <nathan@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org>
---
 include/linux/init.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/include/linux/init.h b/include/linux/init.h
index 045ad1650ed1..d82b4b2e1d25 100644
--- a/include/linux/init.h
+++ b/include/linux/init.h
@@ -242,7 +242,8 @@ extern bool initcall_debug;
 	asm(".section	\"" __sec "\", \"a\"		\n"	\
 	    __stringify(__name) ":			\n"	\
 	    ".long	" __stringify(__stub) " - .	\n"	\
-	    ".previous					\n");
+	    ".previous					\n");	\
+	static_assert(__same_type(initcall_t, &fn));
 #else
 #define ____define_initcall(fn, __unused, __name, __sec)	\
 	static initcall_t __name __used 			\
-- 
2.31.1.818.g46aad6cb9e-goog

Re: [PATCH] init: verify that function is initcall_t at compile-time

[Sami Tolvanen]

Hi Marco,

On Fri, May 21, 2021 at 12:26 AM Marco Elver <elver@google.com> wrote:
>
> In the spirit of making it hard to misuse an interface, add a
> compile-time assertion in the CONFIG_HAVE_ARCH_PREL32_RELOCATIONS case
> to verify the initcall function matches initcall_t, because the inline
> asm bypasses any type-checking the compiler would otherwise do. This
> will help developers catch incorrect API use in all configurations.
>
> A recent example of this is:
> https://lkml.kernel.org/r/20210514140015.2944744-1-arnd@kernel.org
>
> Signed-off-by: Marco Elver <elver@google.com>
> Reviewed-by: Miguel Ojeda <ojeda@kernel.org>
> Reviewed-by: Nathan Chancellor <nathan@kernel.org>
> Tested-by: Nathan Chancellor <nathan@kernel.org>
> ---
>  include/linux/init.h | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/include/linux/init.h b/include/linux/init.h
> index 045ad1650ed1..d82b4b2e1d25 100644
> --- a/include/linux/init.h
> +++ b/include/linux/init.h
> @@ -242,7 +242,8 @@ extern bool initcall_debug;
>         asm(".section   \"" __sec "\", \"a\"            \n"     \
>             __stringify(__name) ":                      \n"     \
>             ".long      " __stringify(__stub) " - .     \n"     \
> -           ".previous                                  \n");
> +           ".previous                                  \n");   \
> +       static_assert(__same_type(initcall_t, &fn));
>  #else
>  #define ____define_initcall(fn, __unused, __name, __sec)       \
>         static initcall_t __name __used                         \

This looks like a nice improvement, thank you for sending the patch!

Reviewed-by: Sami Tolvanen <samitolvanen@google.com>

Sami

Re: [PATCH] init: verify that function is initcall_t at compile-time

[Paul E. McKenney]

On Fri, May 21, 2021 at 09:26:10AM +0200, Marco Elver wrote:
> In the spirit of making it hard to misuse an interface, add a
> compile-time assertion in the CONFIG_HAVE_ARCH_PREL32_RELOCATIONS case
> to verify the initcall function matches initcall_t, because the inline
> asm bypasses any type-checking the compiler would otherwise do. This
> will help developers catch incorrect API use in all configurations.
> 
> A recent example of this is:
> https://lkml.kernel.org/r/20210514140015.2944744-1-arnd@kernel.org
> 
> Signed-off-by: Marco Elver <elver@google.com>
> Reviewed-by: Miguel Ojeda <ojeda@kernel.org>
> Reviewed-by: Nathan Chancellor <nathan@kernel.org>
> Tested-by: Nathan Chancellor <nathan@kernel.org>

Tested-by: Paul E. McKenney <paulmck@kernel.org>

> ---
>  include/linux/init.h | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/include/linux/init.h b/include/linux/init.h
> index 045ad1650ed1..d82b4b2e1d25 100644
> --- a/include/linux/init.h
> +++ b/include/linux/init.h
> @@ -242,7 +242,8 @@ extern bool initcall_debug;
>  	asm(".section	\"" __sec "\", \"a\"		\n"	\
>  	    __stringify(__name) ":			\n"	\
>  	    ".long	" __stringify(__stub) " - .	\n"	\
> -	    ".previous					\n");
> +	    ".previous					\n");	\
> +	static_assert(__same_type(initcall_t, &fn));
>  #else
>  #define ____define_initcall(fn, __unused, __name, __sec)	\
>  	static initcall_t __name __used 			\
> -- 
> 2.31.1.818.g46aad6cb9e-goog
> 

Re: [PATCH] init: verify that function is initcall_t at compile-time

[Kees Cook]

On Fri, 21 May 2021 09:26:10 +0200, Marco Elver wrote:
> In the spirit of making it hard to misuse an interface, add a
> compile-time assertion in the CONFIG_HAVE_ARCH_PREL32_RELOCATIONS case
> to verify the initcall function matches initcall_t, because the inline
> asm bypasses any type-checking the compiler would otherwise do. This
> will help developers catch incorrect API use in all configurations.
> 
> A recent example of this is:
> https://lkml.kernel.org/r/20210514140015.2944744-1-arnd@kernel.org

Since this touches on CFI, I've taken this patch.

Applied to for-next/clang/features, thanks!

[1/1] init: verify that function is initcall_t at compile-time
      https://git.kernel.org/kees/c/72a12a91a634

-- 
Kees Cook