[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20061019202354.GI26530@redhat.com>
Date: Thu, 19 Oct 2006 16:23:54 -0400
From: Dave Jones <davej@...hat.com>
To: Marcus Meissner <meissner@...e.de>
Cc: linux-kernel@...r.kernel.org, akpm@...l.org
Subject: Re: [PATCH] binfmt_elf: randomize PIE binaries
On Thu, Oct 19, 2006 at 04:35:47PM +0200, Marcus Meissner wrote:
> --- linux-2.6.18/fs/binfmt_elf.c.xx 2006-10-19 11:21:49.000000000 +0200
> +++ linux-2.6.18/fs/binfmt_elf.c 2006-10-19 11:24:58.000000000 +0200
> @@ -856,7 +856,12 @@ static int load_elf_binary(struct linux_
> * 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. */
> - load_bias = ELF_PAGESTART(ELF_ET_DYN_BASE - vaddr);
> + if (current->flags & PF_RANDOMIZE)
> + load_bias = randomize_range(0, ELF_ET_DYN_BASE,
> + vaddr);
> + else
> + load_bias = ELF_ET_DYN_BASE - vaddr;
> + load_bias = ELF_PAGESTART(load_bias);
> }
>
> error = elf_map(bprm->file, load_bias + vaddr, elf_ppnt,
vaddr seems odd to be using as the len for randomize_range doesn't it ?
Should that be randomize_range(0, ELF_ET_DYN_BASE, ELF_ET_DYN_BASE - vaddr);
maybe ?
Also, when we get to the elf_map(), we add vaddr back again, but in the
randomize_range case, we never subtracted it. hmm?
Dave
--
http://www.codemonkey.org.uk
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists