Re: [PATCH 1/2 v2] Don't overflow in __libdw_in_section

[Mark Wielaard <mark@klomp.org>]

Hi Ulf,

(Meta, I have some trouble applying this with git am, it thinks the
patch is malformed. But I can apply by hand of course.)

On Fri, 2017-12-08 at 16:05 +0100, Ulf Hermann wrote:
> 
> +2017-05-09  Ulf Hermann  <ulf.hermann@qt.io>
> +
> +	* libdwP.h: Fix check for the upper border of the range in 
> __libdw_in_section.
> +
>   2017-11-03  Mark Wielaard  <mark@klomp.org>
>    	* dwarf_getlocation.c (__libdw_intern_expression): Handle
> diff --git a/libdw/libdwP.h b/libdw/libdwP.h
> index 78c0013..e092d8e 100644
> --- a/libdw/libdwP.h
> +++ b/libdw/libdwP.h
> @@ -643,7 +643,8 @@ __libdw_in_section (Dwarf *dbg, int sec_index,
>     if (data == NULL)
>       return false;
>     if (unlikely (addr < data->d_buf)
> -      || unlikely (data->d_size - (addr - data->d_buf) < size))
> +      || unlikely (data->d_size < size)
> +      || unlikely ((size_t)(addr - data->d_buf) > data->d_size -
> size))
>       {
>         __libdw_seterrno (DWARF_E_INVALID_OFFSET);
>         return false;

The transformation seems correct. But if we can overflow/underflow
here, do we have the same problem in __libdw_offset_in_section where we
  check data->d_size - offset < size, with offset a Dwarf_Off?

Thanks,

Mark