Re: [kernel-hardening] [PATCH 1/6] binfmt_elf: document load_bias a little bit

From: Kees Cook
Date: Fri Jun 02 2017 - 15:27:52 EST


On Fri, Jun 2, 2017 at 8:20 AM, <riel@xxxxxxxxxx> wrote:
> From: Rik van Riel <riel@xxxxxxxxxx>
>
> After me and another unnamed developer got confused by the subtraction
> of vaddr in this branch of the code, followed by adding vaddr back in
> a little bit later, for the third time, maybe it is time to document
> this quirky bit of code.
>
> Signed-off-by: Rik van Riel <riel@xxxxxxxxxx>

Acked-by: Kees Cook <keescook@xxxxxxxxxxxx>

Any and all improvements to this code and its documentation are welcome. :)

-Kees

> ---
> fs/binfmt_elf.c | 10 ++++++++--
> 1 file changed, 8 insertions(+), 2 deletions(-)
>
> diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
> index 5075fd5c62c8..8c3f4dbc7603 100644
> --- a/fs/binfmt_elf.c
> +++ b/fs/binfmt_elf.c
> @@ -930,10 +930,16 @@ static int load_elf_binary(struct linux_binprm *bprm)
> if (loc->elf_ex.e_type == ET_EXEC || load_addr_set) {
> elf_flags |= MAP_FIXED;
> } else if (loc->elf_ex.e_type == ET_DYN) {
> - /* Try and get dynamic programs out of the way of the
> + /*
> + * Try and get dynamic programs out of the way of the
> * default mmap base, as well as whatever program they
> * might try to exec. This is because the brk will
> - * follow the loader, and is not movable. */
> + * follow the loader, and is not movable.
> + *
> + * The load_bias is the difference between the address
> + * in the elf header and the address where the binary
> + * is mmapped.
> + */
> load_bias = ELF_ET_DYN_BASE - vaddr;
> if (current->flags & PF_RANDOMIZE)
> load_bias += arch_mmap_rnd();
> --
> 2.9.3
>



--
Kees Cook
Pixel Security