| 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
| ||
|
Message-ID: <1746748.uPZFXamDiM@tachyon.chronox.de> Date: Fri, 27 Mar 2015 23:49:13 +0100 From: Stephan Mueller <smueller@...onox.de> To: 'Herbert Xu <herbert@...dor.apana.org.au> Cc: linux-crypto@...r.kernel.org, linux-kernel@...r.kernel.org Subject: [PATCH v2 00/20] crypto: restrict usage of helper ciphers Hi, Based on the discussion in the thread [1], a flag is added to the kernel crypto API to allow ciphers to be marked as internal. The patch set is tested in FIPS and non-FIPS mode. In addition, the enforcement that the helper cipher of __driver-gcm-aes-aesni cannot be loaded, but the wrapper of rfc4106-gcm-aesni can be used is tested to demonstrate that the patch works. The testing also shows that of__driver-gcm-aes-aesni is subject to the testmgr self test an can therefore be used in FIPS mode. All cipher implementation whose definition has a cra_priority of 0 as well as the ciphers that are wrapped by cryptd and mcryptd are marked as internal ciphers to prevent them from being called by users. The testing also includes the invocation of normal crypto operations from user space via AF_ALG and libkcapi showing that all of them work unaffected. [1] http://comments.gmane.org/gmane.linux.kernel.cryptoapi/13705 Changes v2: * Overhaul enforcement of the internal flag as suggested by Herbert: a cipher marked as internal can only be invoked if the caller instantiates it with the internal flag set in the type and mask field. * The overhaul implies that cryptd and mcryptd instances are marked as internal if the underlying cipher is marked as internal. * The overhaul implies that the testmgr must try to allocate a cipher again with the internal flag in case the first allocation failed with -ENOENT. * Mark internal cipher in arch/x86/crypto/sha-mb/sha1_mb.c. Stephan Mueller (20): crypto: prevent helper ciphers from being used crypto: testmgr to use CRYPTO_ALG_INTERNAL crypto: cryptd to process CRYPTO_ALG_INTERNAL crypto: /proc/crypto: identify internal ciphers crypto: mark AES-NI helper ciphers crypto: mark ghash clmulni helper ciphers crypto: mark GHASH ARMv8 vmull.p64 helper ciphers crypto: mark AES-NI Camellia helper ciphers crypto: mark CAST5 helper ciphers crypto: mark AVX Camellia helper ciphers crypto: mark CAST6 helper ciphers crypto: mark Serpent AVX2 helper ciphers crypto: mark Serpent AVX helper ciphers crypto: mark Serpent SSE2 helper ciphers crypto: mark Twofish AVX helper ciphers crypto: mark NEON bit sliced AES helper ciphers crypto: mark ARMv8 AES helper ciphers crypto: mark 64 bit ARMv8 AES helper ciphers crypto: mcryptd to process CRYPTO_ALG_INTERNAL crypto: mark Multi buffer SHA1 helper cipher arch/arm/crypto/aes-ce-glue.c | 12 +++++--- arch/arm/crypto/aesbs-glue.c | 9 ++++-- arch/arm/crypto/ghash-ce-glue.c | 6 ++-- arch/arm64/crypto/aes-glue.c | 12 +++++--- arch/x86/crypto/aesni-intel_glue.c | 23 +++++++++----- arch/x86/crypto/camellia_aesni_avx2_glue.c | 15 ++++++--- arch/x86/crypto/camellia_aesni_avx_glue.c | 15 ++++++--- arch/x86/crypto/cast5_avx_glue.c | 9 ++++-- arch/x86/crypto/cast6_avx_glue.c | 15 ++++++--- arch/x86/crypto/ghash-clmulni-intel_glue.c | 7 +++-- arch/x86/crypto/serpent_avx2_glue.c | 15 ++++++--- arch/x86/crypto/serpent_avx_glue.c | 15 ++++++--- arch/x86/crypto/serpent_sse2_glue.c | 15 ++++++--- arch/x86/crypto/sha-mb/sha1_mb.c | 7 +++-- arch/x86/crypto/twofish_avx_glue.c | 15 ++++++--- crypto/ablk_helper.c | 3 +- crypto/api.c | 6 ++++ crypto/cryptd.c | 49 +++++++++++++++++++++++++----- crypto/mcryptd.c | 25 +++++++++++++-- crypto/proc.c | 3 ++ crypto/testmgr.c | 22 ++++++++++++++ include/linux/crypto.h | 6 ++++ 22 files changed, 230 insertions(+), 74 deletions(-) -- 2.1.0 -- 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