kernel.git/arch/x86/include/asm/crypto, branch linux-3.11.y

Revert "crypto: twofish - add AVX2/x86_64 assembler implementation of twofish cipher"

2013-06-21T06:44:29Z

This reverts commit cf1521a1a5e21fd1e79a458605c4282fbfbbeee2. Instruction (vpgatherdd) that this implementation relied on turned out to be slow performer on real hardware (i5-4570). The previous 8-way twofish/AVX implementation is therefore faster and this implementation should be removed. Converting this implementation to use the same method as in twofish/AVX for table look-ups would give additional ~3% speed up vs twofish/AVX, but would hardly be worth of the added code and binary size. Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu

Revert "crypto: blowfish - add AVX2/x86_64 implementation of blowfish cipher"

2013-06-21T06:44:28Z

This reverts commit 604880107010a1e5794552d184cd5471ea31b973. Instruction (vpgatherdd) that this implementation relied on turned out to be slow performer on real hardware (i5-4570). The previous 4-way blowfish implementation is therefore faster and this implementation should be removed. Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu

crypto: camellia - add AVX2/AES-NI/x86_64 assembler implementation of camellia cipher

2013-04-25T13:09:07Z

Patch adds AVX2/AES-NI/x86-64 implementation of Camellia cipher, requiring 32 parallel blocks for input (512 bytes). Compared to AVX implementation, this version is extended to use the 256-bit wide YMM registers. For AES-NI instructions data is split to two 128-bit registers and merged afterwards. Even with this additional handling, performance should be higher compared to the AES-NI/AVX implementation. Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu

crypto: serpent - add AVX2/x86_64 assembler implementation of serpent cipher

2013-04-25T13:09:07Z

Patch adds AVX2/x86-64 implementation of Serpent cipher, requiring 16 parallel blocks for input (256 bytes). Implementation is based on the AVX implementation and extends to use the 256-bit wide YMM registers. Since serpent does not use table look-ups, this implementation should be close to two times faster than the AVX implementation. Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu

crypto: twofish - add AVX2/x86_64 assembler implementation of twofish cipher

2013-04-25T13:09:05Z

Patch adds AVX2/x86-64 implementation of Twofish cipher, requiring 16 parallel blocks for input (256 bytes). Table look-ups are performed using vpgatherdd instruction directly from vector registers and thus should be faster than earlier implementations. Implementation also uses 256-bit wide YMM registers, which should give additional speed up compared to the AVX implementation. Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu

crypto: blowfish - add AVX2/x86_64 implementation of blowfish cipher

2013-04-25T13:09:04Z

Patch adds AVX2/x86-64 implementation of Blowfish cipher, requiring 32 parallel blocks for input (256 bytes). Table look-ups are performed using vpgatherdd instruction directly from vector registers and thus should be faster than earlier implementations. Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu

crypto: x86 - add more optimized XTS-mode for serpent-avx

2013-04-25T13:01:51Z

This patch adds AVX optimized XTS-mode helper functions/macros and converts serpent-avx to use the new facilities. Benefits are slightly improved speed and reduced stack usage as use of temporary IV-array is avoided. tcrypt results, with Intel i5-2450M: enc dec 16B 1.00x 1.00x 64B 1.00x 1.00x 256B 1.04x 1.06x 1024B 1.09x 1.09x 8192B 1.10x 1.09x Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu

crypto: camellia-x86_64 - share common functions and move structures and function definitions to header file

2012-11-09T09:32:31Z

Prepare camellia-x86_64 functions to be reused from AVX/AESNI implementation module. Signed-off-by: Jussi Kivilinna Acked-by: David S. Miller Signed-off-by: Herbert Xu

crypto: serpent/avx - avoid using temporary stack buffers

2012-10-24T13:10:55Z

Introduce new assembler functions to avoid use temporary stack buffers in glue code. This also allows use of vector instructions for xoring output in CTR and CBC modes and construction of IVs for CTR mode. ECB mode sees ~0.5% decrease in speed because added one extra function call. CBC mode decryption and CTR mode benefit from vector operations and gain ~3%. Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu

crypto: x86/glue_helper - use le128 instead of u128 for CTR mode

2012-10-24T13:10:54Z

'u128' currently used for CTR mode is on little-endian 'long long' swapped and would require extra swap operations by SSE/AVX code. Use of le128 instead of u128 allows IV calculations to be done with vector registers easier. Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu