[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20190507170039.GB1399@sol.localdomain>
Date: Tue, 7 May 2019 10:00:41 -0700
From: Eric Biggers <ebiggers@...nel.org>
To: Kees Cook <keescook@...omium.org>
Cc: Herbert Xu <herbert@...dor.apana.org.au>,
Joao Moreira <jmoreira@...e.de>,
Ingo Molnar <mingo@...hat.com>,
Thomas Gleixner <tglx@...utronix.de>,
Borislav Petkov <bp@...en8.de>, x86@...nel.org,
linux-crypto@...r.kernel.org, linux-kernel@...r.kernel.org,
kernel-hardening@...ts.openwall.com
Subject: Re: [PATCH v3 0/7] crypto: x86: Fix indirect function call casts
On Tue, May 07, 2019 at 09:13:14AM -0700, Kees Cook wrote:
> It is possible to indirectly invoke functions with prototypes that do
> not match those of the respectively used function pointers by using void
> types or casts. This feature is frequently used as a way of relaxing
> function invocation, making it possible that different data structures
> are passed to different functions through the same pointer.
>
> Despite the benefits, this can lead to a situation where functions with a
> given prototype are invoked by pointers with a different prototype. This
> is undesirable as it may prevent the use of heuristics such as prototype
> matching-based Control-Flow Integrity, which can be used to prevent
> ROP-based attacks.
>
> One way of fixing this situation is through the use of inline helper
> functions with prototypes that match the one in the respective invoking
> pointer.
>
> Given the above, the current efforts to improve the Linux security,
> and the upcoming kernel support to compilers with CFI features, this
> creates macros to be used to build the needed function definitions,
> to be used in camellia, cast6, serpent, twofish, and aesni.
So why not change the function prototypes to be compatible with common_glue_*_t
instead, rather than wrapping them with another layer of functions? Is it
because indirect calls into asm code won't be allowed with CFI?
>
> -Kees (and Joao)
>
> v3:
> - no longer RFC
> - consolidate macros into glue_helper.h
> - include aesni which was using casts as well
> - remove XTS_TWEAK_CAST while we're at it
>
> v2:
> - update cast macros for clarity
>
> v1:
> - initial prototype
>
> Joao Moreira (4):
> crypto: x86/crypto: Use new glue function macros
This one should be "x86/serpent", not "x86/crypto".
> crypto: x86/camellia: Use new glue function macros
> crypto: x86/twofish: Use new glue function macros
> crypto: x86/cast6: Use new glue function macros
>
> Kees Cook (3):
> crypto: x86/glue_helper: Add static inline function glue macros
> crypto: x86/aesni: Use new glue function macros
> crypto: x86/glue_helper: Remove function prototype cast helpers
>
> arch/x86/crypto/aesni-intel_glue.c | 31 ++++-----
> arch/x86/crypto/camellia_aesni_avx2_glue.c | 73 +++++++++-------------
> arch/x86/crypto/camellia_aesni_avx_glue.c | 63 +++++++------------
> arch/x86/crypto/camellia_glue.c | 21 +++----
> arch/x86/crypto/cast6_avx_glue.c | 65 +++++++++----------
> arch/x86/crypto/serpent_avx2_glue.c | 65 +++++++++----------
> arch/x86/crypto/serpent_avx_glue.c | 58 ++++++-----------
> arch/x86/crypto/serpent_sse2_glue.c | 27 +++++---
> arch/x86/crypto/twofish_avx_glue.c | 71 ++++++++-------------
> arch/x86/crypto/twofish_glue_3way.c | 28 ++++-----
> arch/x86/include/asm/crypto/camellia.h | 64 ++++++-------------
> arch/x86/include/asm/crypto/glue_helper.h | 34 ++++++++--
> arch/x86/include/asm/crypto/serpent-avx.h | 28 ++++-----
> arch/x86/include/asm/crypto/twofish.h | 22 ++++---
> include/crypto/xts.h | 2 -
> 15 files changed, 283 insertions(+), 369 deletions(-)
>
> --
> 2.17.1
>
Powered by blists - more mailing lists