netdev - Re: [PATCH net-next 1/8] inet: preserve const qualifier in inet

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAHk-=wiPUfe8aji5KojAhDKjWhJJU2F9kfzyL660=jRkY+Uzyg@mail.gmail.com>
Date:   Wed, 15 Mar 2023 16:21:17 -0700
From:   Linus Torvalds <torvalds@...ux-foundation.org>
To:     Eric Dumazet <edumazet@...gle.com>
Cc:     Jakub Kicinski <kuba@...nel.org>,
        "David S . Miller" <davem@...emloft.net>,
        Paolo Abeni <pabeni@...hat.com>,
        David Ahern <dsahern@...nel.org>, netdev@...r.kernel.org,
        eric.dumazet@...il.com,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Subject: Re: [PATCH net-next 1/8] inet: preserve const qualifier in inet_sk()

On Wed, Mar 15, 2023 at 3:38 PM Eric Dumazet <edumazet@...gle.com> wrote:
>
> Maybe something like this?

Please no.

> +#define promote_to_type(ptr, oldtype, newtype)                 \
> +       _Generic(ptr,                                           \
> +                const oldtype *: ((const newtype *)(ptr)),     \
> +                oldtype *: ((newtype *)(ptr))                  \
> +       )

That's just a very ugly way to just do a cast. It's wrong.

> +#define inet_sk(sk) promote_to_type(sk, struct sock, struct inet_sock)

This is horrid.

Why isn't this just doing

   #define inet_sk(ptr) container_of(ptr, struct inet_sock, sk)

which is different from a plain cast in that it actually checks that
"yes, struct inet_sock has a member called 'sk' that has the right
type, so now we can convert from that sk to the containing structure".

That's very different from just randomly casting a pointer to another
pointer, like the current inet_sk() does, and like that disgusting
promote_to_type() macro does.

We really strive for proper type safety in the kernel. And that very
much means *not* doing random casts.

At least that "inet_sk(sk)" version using generics didn't take random
pointer types. But I really don't see why you don't just use
"container_of()", which is actually type-safe, and would allow "struct
inet_sock" to contain the "struct sock" somewhere else than in the
first field.

Hmm? Am I missing something that is happening in linux-next?

                Linus