crypto: aes - add generic time invariant AES cipher

Lookup table based AES is sensitive to timing attacks, which is due to
the fact that such table lookups are data dependent, and the fact that
8 KB worth of tables covers a significant number of cachelines on any
architecture, resulting in an exploitable correlation between the key
and the processing time for known plaintexts.

For network facing algorithms such as CTR, CCM or GCM, this presents a
security risk, which is why arch specific AES ports are typically time
invariant, either through the use of special instructions, or by using
SIMD algorithms that don't rely on table lookups.

For generic code, this is difficult to achieve without losing too much
performance, but we can improve the situation significantly by switching
to an implementation that only needs 256 bytes of table data (the actual
S-box itself), which can be prefetched at the start of each block to
eliminate data dependent latencies.

This code encrypts at ~25 cycles per byte on ARM Cortex-A57 (while the
ordinary generic AES driver manages 18 cycles per byte on this
hardware). Decryption is substantially slower.

Signed-off-by: Ard Biesheuvel <[email protected]>
Signed-off-by: Herbert Xu <[email protected]>

crypto/Kconfig

@@ -895,6 +895,23 @@ config CRYPTO_AES
 
 	  See <http://csrc.nist.gov/CryptoToolkit/aes/> for more information.
 
+config CRYPTO_AES_TI
+	tristate "Fixed time AES cipher"
+	select CRYPTO_ALGAPI
+	help
+	  This is a generic implementation of AES that attempts to eliminate
+	  data dependent latencies as much as possible without affecting
+	  performance too much. It is intended for use by the generic CCM
+	  and GCM drivers, and other CTR or CMAC/XCBC based modes that rely
+	  solely on encryption (although decryption is supported as well, but
+	  with a more dramatic performance hit)
+
+	  Instead of using 16 lookup tables of 1 KB each, (8 for encryption and
+	  8 for decryption), this implementation only uses just two S-boxes of
+	  256 bytes each, and attempts to eliminate data dependent latencies by
+	  prefetching the entire table into the cache at the start of each
+	  block.
+
 config CRYPTO_AES_586
 	tristate "AES cipher algorithms (i586)"
 	depends on (X86 || UML_X86) && !64BIT

crypto/Makefile

@@ -99,6 +99,7 @@ obj-$(CONFIG_CRYPTO_TWOFISH) += twofish_generic.o
 obj-$(CONFIG_CRYPTO_TWOFISH_COMMON) += twofish_common.o
 obj-$(CONFIG_CRYPTO_SERPENT) += serpent_generic.o
 obj-$(CONFIG_CRYPTO_AES) += aes_generic.o
+obj-$(CONFIG_CRYPTO_AES_TI) += aes_ti.o
 obj-$(CONFIG_CRYPTO_CAMELLIA) += camellia_generic.o
 obj-$(CONFIG_CRYPTO_CAST_COMMON) += cast_common.o
 obj-$(CONFIG_CRYPTO_CAST5) += cast5_generic.o