linux-kernel - Re: [kernel-hardening] Re: [PATCH] lkdtm: add test for executing .rodata

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: <CAGXu5jKFB6b_fdK4Pg19cyGcpfjmN4SdJR3638T-7QXxntA+8A@mail.gmail.com>
Date:	Tue, 23 Feb 2016 12:51:44 -0800
From:	Kees Cook <keescook@...omium.org>
To:	David Brown <david.brown@...aro.org>
Cc:	"kernel-hardening@...ts.openwall.com" 
	<kernel-hardening@...ts.openwall.com>,
	PaX Team <pageexec@...email.hu>,
	Ard Biesheuvel <ard.biesheuvel@...aro.org>,
	Laura Abbott <labbott@...hat.com>,
	Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
	Mark Rutland <mark.rutland@....com>,
	Jeremy Linton <jeremy.linton@....com>,
	Arnd Bergmann <arnd@...db.de>,
	LKML <linux-kernel@...r.kernel.org>
Subject: Re: [kernel-hardening] Re: [PATCH] lkdtm: add test for executing .rodata

On Tue, Feb 23, 2016 at 12:31 PM, David Brown <david.brown@...aro.org> wrote:
> On Mon, Feb 22, 2016 at 12:46:28PM -0800, Kees Cook wrote:
>>
>> On Thu, Feb 18, 2016 at 1:27 PM, PaX Team <pageexec@...email.hu> wrote:
>>>
>>> On 18 Feb 2016 at 12:34, Ard Biesheuvel wrote:
>>>
>>>> However, that does not fix the issue Kees is trying to solve, where a
>>>> .rodata section is emitted with the "x" bit set, which causes the
>>>> linker to complain:
>>>>
>>>> /tmp/cc50ffWw.s: Assembler messages:
>>>> /tmp/cc50ffWw.s:2: Warning: setting incorrect section attributes for
>>>> .rodata.text
>>>
>>>
>>> in that case why not use a top-level asm statement to set the section
>>> and its attributes (and compile the file with fno-toplevel-reorder)?
>>
>>
>> GCC really wants to declare the section. :(
>
>
> Why not then just use its mechanism to set the section on the code?

It's noted in the quoted section above. :) The problem is that gcc emits:

.section ".rodata","ax",@progbits

But we need the flags to be "a" not "ax". (And I'd note that even "@"
is arch-specific. % is used when @ is a comment start character.)

objcopy can get me the "a", but it seems to require renaming lkdtm.c,
which I really don't want to do.

-Kees

>
> static void do_nothing_rodata(void)
>         __attribute__((section(".rodata")))
> {
>         return;
> }
>
> David



-- 
Kees Cook
Chrome OS & Brillo Security