[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAGXu5jKiZkzYHOF92Kc7s0QyCEnBHx4fU-V4WpqdeL6APr2=8w@mail.gmail.com>
Date: Fri, 20 Feb 2015 10:38:28 -0800
From: Kees Cook <keescook@...omium.org>
To: Alexander Kuleshov <kuleshovmail@...il.com>
Cc: "H. Peter Anvin" <hpa@...or.com>,
Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...hat.com>,
LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] x86/compressed: remove unnecessary check
On Fri, Feb 20, 2015 at 9:55 AM, Alexander Kuleshov
<kuleshovmail@...il.com> wrote:
> region.start is entry->addr, but we already did check for case when
> entry->addr + entry->size < minimum, so no need to check that
> region.start is less than minimum.
Nope, that's not correct. Those are separate checks. For example, a
region 20..130 would be seen as:
start = 20
size = 110
With minimum = 50, the first check eliminates anything with size <= 30
(since the region would be entirely (20...50) below the minimum).
The second check (the one you're suggesting to remove) raises the
bottom of the region up to the minimum:
start = 50
And later on ("Reduce size by any delta from the original address."),
the size is also adjusted to notice any changes:
size = 80
Now the region is 50...130: the bottom is cut off to adapt to the
minimum location.
Is there some situation you found where this logic isn't working?
-Kees
>
> Signed-off-by: Alexander Kuleshov <kuleshovmail@...il.com>
> ---
> arch/x86/boot/compressed/aslr.c | 6 +-----
> 1 file changed, 1 insertion(+), 5 deletions(-)
>
> diff --git a/arch/x86/boot/compressed/aslr.c b/arch/x86/boot/compressed/aslr.c
> index bb13763..13424a9 100644
> --- a/arch/x86/boot/compressed/aslr.c
> +++ b/arch/x86/boot/compressed/aslr.c
> @@ -249,11 +249,7 @@ static void process_e820_entry(struct e820entry *entry,
>
> region.start = entry->addr;
> region.size = entry->size;
> -
> - /* Potentially raise address to minimum location. */
> - if (region.start < minimum)
> - region.start = minimum;
> -
> +
> /* Potentially raise address to meet alignment requirements. */
> region.start = ALIGN(region.start, CONFIG_PHYSICAL_ALIGN);
>
> --
> 2.3.0.80.g18d0fec
>
--
Kees Cook
Chrome OS Security
--
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