From: Andy Lutomirski <luto@amacapital.net>
To: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Nadav Amit <nadav.amit@gmail.com>,
Kees Cook <keescook@chromium.org>,
Josh Poimboeuf <jpoimboe@redhat.com>,
Borislav Petkov <bp@alien8.de>, X86 ML <x86@kernel.org>,
"kernel-hardening@lists.openwall.com"
<kernel-hardening@lists.openwall.com>,
Brian Gerst <brgerst@gmail.com>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
Linus Torvalds <torvalds@linux-foundation.org>
Subject: [kernel-hardening] Re: [PATCH 00/13] Virtually mapped stacks with guard pages (x86, core)
Date: Sun, 19 Jun 2016 23:01:48 -0700 [thread overview]
Message-ID: <CALCETrUgpbTXnvkmLNZz9sSq8fgz1g8eoGv=1Lif2dhyB6GrSg@mail.gmail.com> (raw)
In-Reply-To: <20160620055836.GA3266@osiris>
On Sun, Jun 19, 2016 at 10:58 PM, Heiko Carstens
<heiko.carstens@de.ibm.com> wrote:
> On Fri, Jun 17, 2016 at 10:38:24AM -0700, Andy Lutomirski wrote:
>> > A disassembly looks like this (r15 is the stackpointer):
>> >
>> > 0000000000000670 <setup_arch>:
>> > 670: eb 6f f0 48 00 24 stmg %r6,%r15,72(%r15)
>> > 676: c0 d0 00 00 00 00 larl %r13,676 <setup_arch+0x6>
>> > 67c: a7 f1 3f 80 tmll %r15,16256 <--- test if enough space left
>> > 680: b9 04 00 ef lgr %r14,%r15
>> > 684: a7 84 00 01 je 686 <setup_arch+0x16> <--- branch to illegal op
>> > 688: e3 f0 ff 90 ff 71 lay %r15,-112(%r15)
>> >
>> > The branch jumps actually into the branch instruction itself since the 0001
>> > part of the "je" instruction is an illegal instruction.
>> >
>> > This catches at least wild stack overflows because of two many functions
>> > being called.
>> >
>> > Of course it doesn't catch wild accesses outside the stack because e.g. the
>> > index into an array on the stack is wrong.
>> >
>> > The runtime overhead is within noise ratio, therefore we have this always
>> > enabled.
>> >
>>
>> Neat! What exactly does tmll do? I assume this works by checking the
>> low bits of the stack pointer.
>>
>> x86_64 would have to do:
>>
>> movl %esp, %r11d
>> shll %r11d, $18
>> cmpl %r11d, <threshold>
>> jg error
>>
>> Or similar. I think the cmpl could be eliminated if the threshold
>> were a power of two by simply testing the low bits of the stack
>> pointer.
>
> The tmll instruction tests if any of the higher bits within the 16k
> stackframe address are set. In this specific case that would be bits 7-15
> (mask 0x3f80). If no bit would be set we know that only up to 128 bytes
> would be left on the stack, and thus trigger an exception.
>
> This check does of course only work if a 16k stack is also 16k aligned,
> which is always the case.
>
Oh, interesting. How do you handle the case of a single function that
uses more than 128 bytes of stack?
--Andy
--
Andy Lutomirski
AMA Capital Management, LLC
next prev parent reply other threads:[~2016-06-20 6:01 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-06-16 0:28 [kernel-hardening] [PATCH 00/13] Virtually mapped stacks with guard pages (x86, core) Andy Lutomirski
2016-06-16 0:28 ` [kernel-hardening] [PATCH 01/13] x86/mm/hotplug: Don't remove PGD entries in remove_pagetable() Andy Lutomirski
2016-06-16 0:28 ` [kernel-hardening] [PATCH 02/13] x86/cpa: In populate_pgd, don't set the pgd entry until it's populated Andy Lutomirski
2016-06-16 0:28 ` [kernel-hardening] [PATCH 03/13] x86/cpa: Warn if kernel_unmap_pages_in_pgd is used inappropriately Andy Lutomirski
2016-06-16 0:28 ` [kernel-hardening] [PATCH 04/13] mm: Track NR_KERNEL_STACK in pages instead of number of stacks Andy Lutomirski
2016-06-16 11:10 ` [kernel-hardening] " Vladimir Davydov
2016-06-16 17:21 ` Andy Lutomirski
2016-06-16 19:20 ` Andy Lutomirski
2016-06-16 15:33 ` Josh Poimboeuf
2016-06-16 17:39 ` Andy Lutomirski
2016-06-16 19:39 ` Josh Poimboeuf
2016-06-16 0:28 ` [kernel-hardening] [PATCH 05/13] mm: Move memcg stack accounting to account_kernel_stack Andy Lutomirski
2016-06-16 0:28 ` [kernel-hardening] [PATCH 06/13] fork: Add generic vmalloced stack support Andy Lutomirski
2016-06-16 17:25 ` [kernel-hardening] " Kees Cook
2016-06-16 17:37 ` Andy Lutomirski
2016-06-16 0:28 ` [kernel-hardening] [PATCH 07/13] x86/die: Don't try to recover from an OOPS on a non-default stack Andy Lutomirski
2016-06-16 0:28 ` [kernel-hardening] [PATCH 08/13] x86/dumpstack: When OOPSing, rewind the stack before do_exit Andy Lutomirski
2016-06-16 17:50 ` [kernel-hardening] " Josh Poimboeuf
2016-06-16 17:57 ` Andy Lutomirski
2016-06-16 0:28 ` [kernel-hardening] [PATCH 09/13] x86/dumpstack: When dumping stack bytes due to OOPS, start with regs->sp Andy Lutomirski
2016-06-16 11:56 ` [kernel-hardening] " Borislav Petkov
2016-06-16 0:28 ` [kernel-hardening] [PATCH 10/13] x86/dumpstack: Try harder to get a call trace on stack overflow Andy Lutomirski
2016-06-16 18:16 ` [kernel-hardening] " Josh Poimboeuf
2016-06-16 18:22 ` Andy Lutomirski
2016-06-16 18:33 ` Josh Poimboeuf
2016-06-16 18:37 ` Andy Lutomirski
2016-06-16 18:54 ` Josh Poimboeuf
2016-06-16 0:28 ` [kernel-hardening] [PATCH 11/13] x86/dumpstack/64: Handle faults when printing the "Stack:" part of an OOPS Andy Lutomirski
2016-06-16 0:28 ` [kernel-hardening] [PATCH 12/13] x86/mm/64: Enable vmapped stacks Andy Lutomirski
2016-06-16 4:17 ` [kernel-hardening] " Mika Penttilä
2016-06-16 5:33 ` Andy Lutomirski
2016-06-16 13:11 ` Rik van Riel
2016-06-16 0:28 ` [kernel-hardening] [PATCH 13/13] x86/mm: Improve stack-overflow #PF handling Andy Lutomirski
2016-06-16 6:05 ` [kernel-hardening] Re: [PATCH 00/13] Virtually mapped stacks with guard pages (x86, core) Heiko Carstens
2016-06-16 17:50 ` Andy Lutomirski
2016-06-16 18:14 ` Andy Lutomirski
2016-06-16 21:27 ` Andy Lutomirski
2016-06-17 3:58 ` Andy Lutomirski
2016-06-17 7:27 ` Heiko Carstens
2016-06-17 17:38 ` Andy Lutomirski
2016-06-20 5:58 ` Heiko Carstens
2016-06-20 6:01 ` Andy Lutomirski [this message]
2016-06-20 7:07 ` Heiko Carstens
2016-06-16 17:24 ` Kees Cook
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CALCETrUgpbTXnvkmLNZz9sSq8fgz1g8eoGv=1Lif2dhyB6GrSg@mail.gmail.com' \
--to=luto@amacapital.net \
--cc=bp@alien8.de \
--cc=brgerst@gmail.com \
--cc=heiko.carstens@de.ibm.com \
--cc=jpoimboe@redhat.com \
--cc=keescook@chromium.org \
--cc=kernel-hardening@lists.openwall.com \
--cc=linux-kernel@vger.kernel.org \
--cc=nadav.amit@gmail.com \
--cc=torvalds@linux-foundation.org \
--cc=x86@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).