From a482b081a2d4d74d16bc9ea8779f9f6055f95852 Mon Sep 17 00:00:00 2001 From: Sonic Zhang Date: Fri, 25 May 2012 17:54:13 +0800 Subject: crypto: testmgr - Add new test cases for Blackfin CRC crypto driver Signed-off-by: Sonic Zhang Acked-by: Mike Frysinger Signed-off-by: Herbert Xu --- crypto/tcrypt.c | 3 ++ crypto/testmgr.c | 9 ++++++ crypto/testmgr.h | 90 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 102 insertions(+) (limited to 'crypto') diff --git a/crypto/tcrypt.c b/crypto/tcrypt.c index 8f147bff098..750cce44bad 100644 --- a/crypto/tcrypt.c +++ b/crypto/tcrypt.c @@ -1192,6 +1192,9 @@ static int do_test(int m) case 109: ret += tcrypt_test("vmac(aes)"); break; + case 110: + ret += tcrypt_test("hmac(crc32)"); + break; case 150: ret += tcrypt_test("ansi_cprng"); diff --git a/crypto/testmgr.c b/crypto/testmgr.c index 5674878ff6c..eb6d20f8ec5 100644 --- a/crypto/testmgr.c +++ b/crypto/testmgr.c @@ -2219,6 +2219,15 @@ static const struct alg_test_desc alg_test_descs[] = { .count = GHASH_TEST_VECTORS } } + }, { + .alg = "hmac(crc32)", + .test = alg_test_hash, + .suite = { + .hash = { + .vecs = bfin_crc_tv_template, + .count = BFIN_CRC_TEST_VECTORS + } + } }, { .alg = "hmac(md5)", .test = alg_test_hash, diff --git a/crypto/testmgr.h b/crypto/testmgr.h index 36e5a8ee0e1..34a9d511dee 100644 --- a/crypto/testmgr.h +++ b/crypto/testmgr.h @@ -14858,4 +14858,94 @@ static struct hash_testvec crc32c_tv_template[] = { }, }; +/* + * Blakcifn CRC test vectors + */ +#define BFIN_CRC_TEST_VECTORS 6 + +static struct hash_testvec bfin_crc_tv_template[] = { + { + .psize = 0, + .digest = "\x00\x00\x00\x00", + }, + { + .key = "\x87\xa9\xcb\xed", + .ksize = 4, + .psize = 0, + .digest = "\x87\xa9\xcb\xed", + }, + { + .key = "\xff\xff\xff\xff", + .ksize = 4, + .plaintext = "\x01\x02\x03\x04\x05\x06\x07\x08" + "\x09\x0a\x0b\x0c\x0d\x0e\x0f\x10" + "\x11\x12\x13\x14\x15\x16\x17\x18" + "\x19\x1a\x1b\x1c\x1d\x1e\x1f\x20" + "\x21\x22\x23\x24\x25\x26\x27\x28", + .psize = 40, + .digest = "\x84\x0c\x8d\xa2", + }, + { + .key = "\xff\xff\xff\xff", + .ksize = 4, + .plaintext = "\x01\x02\x03\x04\x05\x06\x07\x08" + "\x09\x0a\x0b\x0c\x0d\x0e\x0f\x10" + "\x11\x12\x13\x14\x15\x16\x17\x18" + "\x19\x1a\x1b\x1c\x1d\x1e\x1f\x20" + "\x21\x22\x23\x24\x25\x26", + .psize = 38, + .digest = "\x8c\x58\xec\xb7", + }, + { + .key = "\xff\xff\xff\xff", + .ksize = 4, + .plaintext = "\x01\x02\x03\x04\x05\x06\x07\x08" + "\x09\x0a\x0b\x0c\x0d\x0e\x0f\x10" + "\x11\x12\x13\x14\x15\x16\x17\x18" + "\x19\x1a\x1b\x1c\x1d\x1e\x1f\x20" + "\x21\x22\x23\x24\x25\x26\x27", + .psize = 39, + .digest = "\xdc\x50\x28\x7b", + }, + { + .key = "\xff\xff\xff\xff", + .ksize = 4, + .plaintext = "\x01\x02\x03\x04\x05\x06\x07\x08" + "\x09\x0a\x0b\x0c\x0d\x0e\x0f\x10" + "\x11\x12\x13\x14\x15\x16\x17\x18" + "\x19\x1a\x1b\x1c\x1d\x1e\x1f\x20" + "\x21\x22\x23\x24\x25\x26\x27\x28" + "\x29\x2a\x2b\x2c\x2d\x2e\x2f\x30" + "\x31\x32\x33\x34\x35\x36\x37\x38" + "\x39\x3a\x3b\x3c\x3d\x3e\x3f\x40" + "\x41\x42\x43\x44\x45\x46\x47\x48" + "\x49\x4a\x4b\x4c\x4d\x4e\x4f\x50" + "\x51\x52\x53\x54\x55\x56\x57\x58" + "\x59\x5a\x5b\x5c\x5d\x5e\x5f\x60" + "\x61\x62\x63\x64\x65\x66\x67\x68" + "\x69\x6a\x6b\x6c\x6d\x6e\x6f\x70" + "\x71\x72\x73\x74\x75\x76\x77\x78" + "\x79\x7a\x7b\x7c\x7d\x7e\x7f\x80" + "\x81\x82\x83\x84\x85\x86\x87\x88" + "\x89\x8a\x8b\x8c\x8d\x8e\x8f\x90" + "\x91\x92\x93\x94\x95\x96\x97\x98" + "\x99\x9a\x9b\x9c\x9d\x9e\x9f\xa0" + "\xa1\xa2\xa3\xa4\xa5\xa6\xa7\xa8" + "\xa9\xaa\xab\xac\xad\xae\xaf\xb0" + "\xb1\xb2\xb3\xb4\xb5\xb6\xb7\xb8" + "\xb9\xba\xbb\xbc\xbd\xbe\xbf\xc0" + "\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8" + "\xc9\xca\xcb\xcc\xcd\xce\xcf\xd0" + "\xd1\xd2\xd3\xd4\xd5\xd6\xd7\xd8" + "\xd9\xda\xdb\xdc\xdd\xde\xdf\xe0" + "\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8" + "\xe9\xea\xeb\xec\xed\xee\xef\xf0", + .psize = 240, + .digest = "\x10\x19\x4a\x5c", + .np = 2, + .tap = { 31, 209 } + }, + +}; + #endif /* _CRYPTO_TESTMGR_H */ -- cgit v1.2.3 From 107778b592576c0c8e8d2ca7a2aa5415a4908223 Mon Sep 17 00:00:00 2001 From: Johannes Goetzfried Date: Mon, 28 May 2012 15:54:24 +0200 Subject: crypto: twofish - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Twofish block cipher. The implementation processes eight blocks in parallel (two 4 block chunk AVX operations). The table-lookups are done in general-purpose registers. For small blocksizes the 3way-parallel functions from the twofish-x86_64-3way module are called. A good performance increase is provided for blocksizes greater or equal to 128B. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) twofish-avx-x86_64 vs. twofish-x86_64-3way 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.96x 0.97x 1.00x 0.95x 0.97x 0.97x 0.96x 0.95x 0.95x 0.98x 64B 0.99x 0.99x 1.00x 0.99x 0.98x 0.98x 0.99x 0.98x 0.99x 0.98x 256B 1.20x 1.21x 1.00x 1.19x 1.15x 1.14x 1.19x 1.20x 1.18x 1.19x 1024B 1.29x 1.30x 1.00x 1.28x 1.23x 1.24x 1.26x 1.28x 1.26x 1.27x 8192B 1.31x 1.32x 1.00x 1.31x 1.25x 1.25x 1.28x 1.29x 1.28x 1.30x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 0.96x 0.96x 1.00x 0.96x 0.97x 0.98x 0.95x 0.95x 0.95x 0.96x 64B 1.00x 0.99x 1.00x 0.98x 0.98x 1.01x 0.98x 0.98x 0.98x 0.98x 256B 1.20x 1.21x 1.00x 1.21x 1.15x 1.15x 1.19x 1.20x 1.18x 1.19x 1024B 1.29x 1.30x 1.00x 1.28x 1.23x 1.23x 1.26x 1.27x 1.26x 1.27x 8192B 1.31x 1.33x 1.00x 1.31x 1.26x 1.26x 1.29x 1.29x 1.28x 1.30x twofish-avx-x86_64 vs aes-asm (8kB block): 128bit 256bit ecb-enc 1.19x 1.63x ecb-dec 1.18x 1.62x cbc-enc 0.75x 1.03x cbc-dec 1.23x 1.67x ctr-enc 1.24x 1.65x ctr-dec 1.24x 1.65x lrw-enc 1.15x 1.53x lrw-dec 1.14x 1.52x xts-enc 1.16x 1.56x xts-dec 1.16x 1.56x Signed-off-by: Johannes Goetzfried Signed-off-by: Herbert Xu --- arch/x86/crypto/Makefile | 2 + arch/x86/crypto/twofish-avx-x86_64-asm_64.S | 301 ++++++++ arch/x86/crypto/twofish_avx_glue.c | 1086 +++++++++++++++++++++++++++ arch/x86/crypto/twofish_glue_3way.c | 2 + crypto/Kconfig | 24 + crypto/tcrypt.c | 23 + crypto/testmgr.c | 60 ++ 7 files changed, 1498 insertions(+) create mode 100644 arch/x86/crypto/twofish-avx-x86_64-asm_64.S create mode 100644 arch/x86/crypto/twofish_avx_glue.c (limited to 'crypto') diff --git a/arch/x86/crypto/Makefile b/arch/x86/crypto/Makefile index 479f95a744f..3420feef0c7 100644 --- a/arch/x86/crypto/Makefile +++ b/arch/x86/crypto/Makefile @@ -12,6 +12,7 @@ obj-$(CONFIG_CRYPTO_CAMELLIA_X86_64) += camellia-x86_64.o obj-$(CONFIG_CRYPTO_BLOWFISH_X86_64) += blowfish-x86_64.o obj-$(CONFIG_CRYPTO_TWOFISH_X86_64) += twofish-x86_64.o obj-$(CONFIG_CRYPTO_TWOFISH_X86_64_3WAY) += twofish-x86_64-3way.o +obj-$(CONFIG_CRYPTO_TWOFISH_AVX_X86_64) += twofish-avx-x86_64.o obj-$(CONFIG_CRYPTO_SALSA20_X86_64) += salsa20-x86_64.o obj-$(CONFIG_CRYPTO_SERPENT_SSE2_X86_64) += serpent-sse2-x86_64.o obj-$(CONFIG_CRYPTO_AES_NI_INTEL) += aesni-intel.o @@ -30,6 +31,7 @@ camellia-x86_64-y := camellia-x86_64-asm_64.o camellia_glue.o blowfish-x86_64-y := blowfish-x86_64-asm_64.o blowfish_glue.o twofish-x86_64-y := twofish-x86_64-asm_64.o twofish_glue.o twofish-x86_64-3way-y := twofish-x86_64-asm_64-3way.o twofish_glue_3way.o +twofish-avx-x86_64-y := twofish-avx-x86_64-asm_64.o twofish_avx_glue.o salsa20-x86_64-y := salsa20-x86_64-asm_64.o salsa20_glue.o serpent-sse2-x86_64-y := serpent-sse2-x86_64-asm_64.o serpent_sse2_glue.o diff --git a/arch/x86/crypto/twofish-avx-x86_64-asm_64.S b/arch/x86/crypto/twofish-avx-x86_64-asm_64.S new file mode 100644 index 00000000000..fc31b89ba4c --- /dev/null +++ b/arch/x86/crypto/twofish-avx-x86_64-asm_64.S @@ -0,0 +1,301 @@ +/* + * Twofish Cipher 8-way parallel algorithm (AVX/x86_64) + * + * Copyright (C) 2012 Johannes Goetzfried + * + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 + * USA + * + */ + +.file "twofish-avx-x86_64-asm_64.S" +.text + +/* structure of crypto context */ +#define s0 0 +#define s1 1024 +#define s2 2048 +#define s3 3072 +#define w 4096 +#define k 4128 + +/********************************************************************** + 8-way AVX twofish + **********************************************************************/ +#define CTX %rdi + +#define RA1 %xmm0 +#define RB1 %xmm1 +#define RC1 %xmm2 +#define RD1 %xmm3 + +#define RA2 %xmm4 +#define RB2 %xmm5 +#define RC2 %xmm6 +#define RD2 %xmm7 + +#define RX %xmm8 +#define RY %xmm9 + +#define RK1 %xmm10 +#define RK2 %xmm11 + +#define RID1 %rax +#define RID1b %al +#define RID2 %rbx +#define RID2b %bl + +#define RGI1 %rdx +#define RGI1bl %dl +#define RGI1bh %dh +#define RGI2 %rcx +#define RGI2bl %cl +#define RGI2bh %ch + +#define RGS1 %r8 +#define RGS1d %r8d +#define RGS2 %r9 +#define RGS2d %r9d +#define RGS3 %r10 +#define RGS3d %r10d + + +#define lookup_32bit(t0, t1, t2, t3, src, dst) \ + movb src ## bl, RID1b; \ + movb src ## bh, RID2b; \ + movl t0(CTX, RID1, 4), dst ## d; \ + xorl t1(CTX, RID2, 4), dst ## d; \ + shrq $16, src; \ + movb src ## bl, RID1b; \ + movb src ## bh, RID2b; \ + xorl t2(CTX, RID1, 4), dst ## d; \ + xorl t3(CTX, RID2, 4), dst ## d; + +#define G(a, x, t0, t1, t2, t3) \ + vmovq a, RGI1; \ + vpsrldq $8, a, x; \ + vmovq x, RGI2; \ + \ + lookup_32bit(t0, t1, t2, t3, RGI1, RGS1); \ + shrq $16, RGI1; \ + lookup_32bit(t0, t1, t2, t3, RGI1, RGS2); \ + shlq $32, RGS2; \ + orq RGS1, RGS2; \ + \ + lookup_32bit(t0, t1, t2, t3, RGI2, RGS1); \ + shrq $16, RGI2; \ + lookup_32bit(t0, t1, t2, t3, RGI2, RGS3); \ + shlq $32, RGS3; \ + orq RGS1, RGS3; \ + \ + vmovq RGS2, x; \ + vpinsrq $1, RGS3, x, x; + +#define encround(a, b, c, d, x, y) \ + G(a, x, s0, s1, s2, s3); \ + G(b, y, s1, s2, s3, s0); \ + vpaddd x, y, x; \ + vpaddd y, x, y; \ + vpaddd x, RK1, x; \ + vpaddd y, RK2, y; \ + vpxor x, c, c; \ + vpsrld $1, c, x; \ + vpslld $(32 - 1), c, c; \ + vpor c, x, c; \ + vpslld $1, d, x; \ + vpsrld $(32 - 1), d, d; \ + vpor d, x, d; \ + vpxor d, y, d; + +#define decround(a, b, c, d, x, y) \ + G(a, x, s0, s1, s2, s3); \ + G(b, y, s1, s2, s3, s0); \ + vpaddd x, y, x; \ + vpaddd y, x, y; \ + vpaddd y, RK2, y; \ + vpxor d, y, d; \ + vpsrld $1, d, y; \ + vpslld $(32 - 1), d, d; \ + vpor d, y, d; \ + vpslld $1, c, y; \ + vpsrld $(32 - 1), c, c; \ + vpor c, y, c; \ + vpaddd x, RK1, x; \ + vpxor x, c, c; + +#define encrypt_round(n, a, b, c, d) \ + vbroadcastss (k+4*(2*(n)))(CTX), RK1; \ + vbroadcastss (k+4*(2*(n)+1))(CTX), RK2; \ + encround(a ## 1, b ## 1, c ## 1, d ## 1, RX, RY); \ + encround(a ## 2, b ## 2, c ## 2, d ## 2, RX, RY); + +#define decrypt_round(n, a, b, c, d) \ + vbroadcastss (k+4*(2*(n)))(CTX), RK1; \ + vbroadcastss (k+4*(2*(n)+1))(CTX), RK2; \ + decround(a ## 1, b ## 1, c ## 1, d ## 1, RX, RY); \ + decround(a ## 2, b ## 2, c ## 2, d ## 2, RX, RY); + +#define encrypt_cycle(n) \ + encrypt_round((2*n), RA, RB, RC, RD); \ + encrypt_round(((2*n) + 1), RC, RD, RA, RB); + +#define decrypt_cycle(n) \ + decrypt_round(((2*n) + 1), RC, RD, RA, RB); \ + decrypt_round((2*n), RA, RB, RC, RD); + + +#define transpose_4x4(x0, x1, x2, x3, t0, t1, t2) \ + vpunpckldq x1, x0, t0; \ + vpunpckhdq x1, x0, t2; \ + vpunpckldq x3, x2, t1; \ + vpunpckhdq x3, x2, x3; \ + \ + vpunpcklqdq t1, t0, x0; \ + vpunpckhqdq t1, t0, x1; \ + vpunpcklqdq x3, t2, x2; \ + vpunpckhqdq x3, t2, x3; + +#define inpack_blocks(in, x0, x1, x2, x3, wkey, t0, t1, t2) \ + vpxor (0*4*4)(in), wkey, x0; \ + vpxor (1*4*4)(in), wkey, x1; \ + vpxor (2*4*4)(in), wkey, x2; \ + vpxor (3*4*4)(in), wkey, x3; \ + \ + transpose_4x4(x0, x1, x2, x3, t0, t1, t2) + +#define outunpack_blocks(out, x0, x1, x2, x3, wkey, t0, t1, t2) \ + transpose_4x4(x0, x1, x2, x3, t0, t1, t2) \ + \ + vpxor x0, wkey, x0; \ + vmovdqu x0, (0*4*4)(out); \ + vpxor x1, wkey, x1; \ + vmovdqu x1, (1*4*4)(out); \ + vpxor x2, wkey, x2; \ + vmovdqu x2, (2*4*4)(out); \ + vpxor x3, wkey, x3; \ + vmovdqu x3, (3*4*4)(out); + +#define outunpack_xor_blocks(out, x0, x1, x2, x3, wkey, t0, t1, t2) \ + transpose_4x4(x0, x1, x2, x3, t0, t1, t2) \ + \ + vpxor x0, wkey, x0; \ + vpxor (0*4*4)(out), x0, x0; \ + vmovdqu x0, (0*4*4)(out); \ + vpxor x1, wkey, x1; \ + vpxor (1*4*4)(out), x1, x1; \ + vmovdqu x1, (1*4*4)(out); \ + vpxor x2, wkey, x2; \ + vpxor (2*4*4)(out), x2, x2; \ + vmovdqu x2, (2*4*4)(out); \ + vpxor x3, wkey, x3; \ + vpxor (3*4*4)(out), x3, x3; \ + vmovdqu x3, (3*4*4)(out); + +.align 8 +.global __twofish_enc_blk_8way +.type __twofish_enc_blk_8way,@function; + +__twofish_enc_blk_8way: + /* input: + * %rdi: ctx, CTX + * %rsi: dst + * %rdx: src + * %rcx: bool, if true: xor output + */ + + pushq %rbx; + pushq %rcx; + + vmovdqu w(CTX), RK1; + + leaq (4*4*4)(%rdx), %rax; + inpack_blocks(%rdx, RA1, RB1, RC1, RD1, RK1, RX, RY, RK2); + inpack_blocks(%rax, RA2, RB2, RC2, RD2, RK1, RX, RY, RK2); + + xorq RID1, RID1; + xorq RID2, RID2; + + encrypt_cycle(0); + encrypt_cycle(1); + encrypt_cycle(2); + encrypt_cycle(3); + encrypt_cycle(4); + encrypt_cycle(5); + encrypt_cycle(6); + encrypt_cycle(7); + + vmovdqu (w+4*4)(CTX), RK1; + + popq %rcx; + popq %rbx; + + leaq (4*4*4)(%rsi), %rax; + leaq (4*4*4)(%rax), %rdx; + + testb %cl, %cl; + jnz __enc_xor8; + + outunpack_blocks(%rsi, RC1, RD1, RA1, RB1, RK1, RX, RY, RK2); + outunpack_blocks(%rax, RC2, RD2, RA2, RB2, RK1, RX, RY, RK2); + + ret; + +__enc_xor8: + outunpack_xor_blocks(%rsi, RC1, RD1, RA1, RB1, RK1, RX, RY, RK2); + outunpack_xor_blocks(%rax, RC2, RD2, RA2, RB2, RK1, RX, RY, RK2); + + ret; + +.align 8 +.global twofish_dec_blk_8way +.type twofish_dec_blk_8way,@function; + +twofish_dec_blk_8way: + /* input: + * %rdi: ctx, CTX + * %rsi: dst + * %rdx: src + */ + + pushq %rbx; + + vmovdqu (w+4*4)(CTX), RK1; + + leaq (4*4*4)(%rdx), %rax; + inpack_blocks(%rdx, RC1, RD1, RA1, RB1, RK1, RX, RY, RK2); + inpack_blocks(%rax, RC2, RD2, RA2, RB2, RK1, RX, RY, RK2); + + xorq RID1, RID1; + xorq RID2, RID2; + + decrypt_cycle(7); + decrypt_cycle(6); + decrypt_cycle(5); + decrypt_cycle(4); + decrypt_cycle(3); + decrypt_cycle(2); + decrypt_cycle(1); + decrypt_cycle(0); + + vmovdqu (w)(CTX), RK1; + + popq %rbx; + + leaq (4*4*4)(%rsi), %rax; + outunpack_blocks(%rsi, RA1, RB1, RC1, RD1, RK1, RX, RY, RK2); + outunpack_blocks(%rax, RA2, RB2, RC2, RD2, RK1, RX, RY, RK2); + + ret; diff --git a/arch/x86/crypto/twofish_avx_glue.c b/arch/x86/crypto/twofish_avx_glue.c new file mode 100644 index 00000000000..599f19e4bef --- /dev/null +++ b/arch/x86/crypto/twofish_avx_glue.c @@ -0,0 +1,1086 @@ +/* + * Glue Code for AVX assembler version of Twofish Cipher + * + * Copyright (C) 2012 Johannes Goetzfried + * + * + * Glue code based on serpent_sse2_glue.c by: + * Copyright (C) 2011 Jussi Kivilinna + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 + * USA + * + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +#define TWOFISH_PARALLEL_BLOCKS 8 + +/* regular block cipher functions from twofish_x86_64 module */ +asmlinkage void twofish_enc_blk(struct twofish_ctx *ctx, u8 *dst, + const u8 *src); +asmlinkage void twofish_dec_blk(struct twofish_ctx *ctx, u8 *dst, + const u8 *src); + +/* 3-way parallel cipher functions from twofish_x86_64-3way module */ +asmlinkage void __twofish_enc_blk_3way(struct twofish_ctx *ctx, u8 *dst, + const u8 *src, bool xor); +asmlinkage void twofish_dec_blk_3way(struct twofish_ctx *ctx, u8 *dst, + const u8 *src); + +static inline void twofish_enc_blk_3way(struct twofish_ctx *ctx, u8 *dst, + const u8 *src) +{ + __twofish_enc_blk_3way(ctx, dst, src, false); +} + +static inline void twofish_enc_blk_3way_xor(struct twofish_ctx *ctx, u8 *dst, + const u8 *src) +{ + __twofish_enc_blk_3way(ctx, dst, src, true); +} + +/* 8-way parallel cipher functions */ +asmlinkage void __twofish_enc_blk_8way(struct twofish_ctx *ctx, u8 *dst, + const u8 *src, bool xor); +asmlinkage void twofish_dec_blk_8way(struct twofish_ctx *ctx, u8 *dst, + const u8 *src); + +static inline void twofish_enc_blk_xway(struct twofish_ctx *ctx, u8 *dst, + const u8 *src) +{ + __twofish_enc_blk_8way(ctx, dst, src, false); +} + +static inline void twofish_enc_blk_xway_xor(struct twofish_ctx *ctx, u8 *dst, + const u8 *src) +{ + __twofish_enc_blk_8way(ctx, dst, src, true); +} + +static inline void twofish_dec_blk_xway(struct twofish_ctx *ctx, u8 *dst, + const u8 *src) +{ + twofish_dec_blk_8way(ctx, dst, src); +} + + + +struct async_twofish_ctx { + struct cryptd_ablkcipher *cryptd_tfm; +}; + +static inline bool twofish_fpu_begin(bool fpu_enabled, unsigned int nbytes) +{ + if (fpu_enabled) + return true; + + /* AVX is only used when chunk to be processed is large enough, so + * do not enable FPU until it is necessary. + */ + if (nbytes < TF_BLOCK_SIZE * TWOFISH_PARALLEL_BLOCKS) + return false; + + kernel_fpu_begin(); + return true; +} + +static inline void twofish_fpu_end(bool fpu_enabled) +{ + if (fpu_enabled) + kernel_fpu_end(); +} + +static int ecb_crypt(struct blkcipher_desc *desc, struct blkcipher_walk *walk, + bool enc) +{ + bool fpu_enabled = false; + struct twofish_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + const unsigned int bsize = TF_BLOCK_SIZE; + unsigned int nbytes; + int err; + + err = blkcipher_walk_virt(desc, walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + + while ((nbytes = walk->nbytes)) { + u8 *wsrc = walk->src.virt.addr; + u8 *wdst = walk->dst.virt.addr; + + fpu_enabled = twofish_fpu_begin(fpu_enabled, nbytes); + + /* Process multi-block batch */ + if (nbytes >= bsize * TWOFISH_PARALLEL_BLOCKS) { + do { + if (enc) + twofish_enc_blk_xway(ctx, wdst, wsrc); + else + twofish_dec_blk_xway(ctx, wdst, wsrc); + + wsrc += bsize * TWOFISH_PARALLEL_BLOCKS; + wdst += bsize * TWOFISH_PARALLEL_BLOCKS; + nbytes -= bsize * TWOFISH_PARALLEL_BLOCKS; + } while (nbytes >= bsize * TWOFISH_PARALLEL_BLOCKS); + + if (nbytes < bsize) + goto done; + } + + /* Process three block batch */ + if (nbytes >= bsize * 3) { + do { + if (enc) + twofish_enc_blk_3way(ctx, wdst, wsrc); + else + twofish_dec_blk_3way(ctx, wdst, wsrc); + + wsrc += bsize * 3; + wdst += bsize * 3; + nbytes -= bsize * 3; + } while (nbytes >= bsize * 3); + + if (nbytes < bsize) + goto done; + } + + /* Handle leftovers */ + do { + if (enc) + twofish_enc_blk(ctx, wdst, wsrc); + else + twofish_dec_blk(ctx, wdst, wsrc); + + wsrc += bsize; + wdst += bsize; + nbytes -= bsize; + } while (nbytes >= bsize); + +done: + err = blkcipher_walk_done(desc, walk, nbytes); + } + + twofish_fpu_end(fpu_enabled); + return err; +} + +static int ecb_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct blkcipher_walk walk; + + blkcipher_walk_init(&walk, dst, src, nbytes); + return ecb_crypt(desc, &walk, true); +} + +static int ecb_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct blkcipher_walk walk; + + blkcipher_walk_init(&walk, dst, src, nbytes); + return ecb_crypt(desc, &walk, false); +} + +static unsigned int __cbc_encrypt(struct blkcipher_desc *desc, + struct blkcipher_walk *walk) +{ + struct twofish_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + const unsigned int bsize = TF_BLOCK_SIZE; + unsigned int nbytes = walk->nbytes; + u128 *src = (u128 *)walk->src.virt.addr; + u128 *dst = (u128 *)walk->dst.virt.addr; + u128 *iv = (u128 *)walk->iv; + + do { + u128_xor(dst, src, iv); + twofish_enc_blk(ctx, (u8 *)dst, (u8 *)dst); + iv = dst; + + src += 1; + dst += 1; + nbytes -= bsize; + } while (nbytes >= bsize); + + u128_xor((u128 *)walk->iv, (u128 *)walk->iv, iv); + return nbytes; +} + +static int cbc_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + + while ((nbytes = walk.nbytes)) { + nbytes = __cbc_encrypt(desc, &walk); + err = blkcipher_walk_done(desc, &walk, nbytes); + } + + return err; +} + +static unsigned int __cbc_decrypt(struct blkcipher_desc *desc, + struct blkcipher_walk *walk) +{ + struct twofish_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + const unsigned int bsize = TF_BLOCK_SIZE; + unsigned int nbytes = walk->nbytes; + u128 *src = (u128 *)walk->src.virt.addr; + u128 *dst = (u128 *)walk->dst.virt.addr; + u128 ivs[TWOFISH_PARALLEL_BLOCKS - 1]; + u128 last_iv; + int i; + + /* Start of the last block. */ + src += nbytes / bsize - 1; + dst += nbytes / bsize - 1; + + last_iv = *src; + + /* Process multi-block batch */ + if (nbytes >= bsize * TWOFISH_PARALLEL_BLOCKS) { + do { + nbytes -= bsize * (TWOFISH_PARALLEL_BLOCKS - 1); + src -= TWOFISH_PARALLEL_BLOCKS - 1; + dst -= TWOFISH_PARALLEL_BLOCKS - 1; + + for (i = 0; i < TWOFISH_PARALLEL_BLOCKS - 1; i++) + ivs[i] = src[i]; + + twofish_dec_blk_xway(ctx, (u8 *)dst, (u8 *)src); + + for (i = 0; i < TWOFISH_PARALLEL_BLOCKS - 1; i++) + u128_xor(dst + (i + 1), dst + (i + 1), ivs + i); + + nbytes -= bsize; + if (nbytes < bsize) + goto done; + + u128_xor(dst, dst, src - 1); + src -= 1; + dst -= 1; + } while (nbytes >= bsize * TWOFISH_PARALLEL_BLOCKS); + + if (nbytes < bsize) + goto done; + } + + /* Process three block batch */ + if (nbytes >= bsize * 3) { + do { + nbytes -= bsize * (3 - 1); + src -= 3 - 1; + dst -= 3 - 1; + + ivs[0] = src[0]; + ivs[1] = src[1]; + + twofish_dec_blk_3way(ctx, (u8 *)dst, (u8 *)src); + + u128_xor(dst + 1, dst + 1, ivs + 0); + u128_xor(dst + 2, dst + 2, ivs + 1); + + nbytes -= bsize; + if (nbytes < bsize) + goto done; + + u128_xor(dst, dst, src - 1); + src -= 1; + dst -= 1; + } while (nbytes >= bsize * 3); + + if (nbytes < bsize) + goto done; + } + + /* Handle leftovers */ + for (;;) { + twofish_dec_blk(ctx, (u8 *)dst, (u8 *)src); + + nbytes -= bsize; + if (nbytes < bsize) + break; + + u128_xor(dst, dst, src - 1); + src -= 1; + dst -= 1; + } + +done: + u128_xor(dst, dst, (u128 *)walk->iv); + *(u128 *)walk->iv = last_iv; + + return nbytes; +} + +static int cbc_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + bool fpu_enabled = false; + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + + while ((nbytes = walk.nbytes)) { + fpu_enabled = twofish_fpu_begin(fpu_enabled, nbytes); + nbytes = __cbc_decrypt(desc, &walk); + err = blkcipher_walk_done(desc, &walk, nbytes); + } + + twofish_fpu_end(fpu_enabled); + return err; +} + +static inline void u128_to_be128(be128 *dst, const u128 *src) +{ + dst->a = cpu_to_be64(src->a); + dst->b = cpu_to_be64(src->b); +} + +static inline void be128_to_u128(u128 *dst, const be128 *src) +{ + dst->a = be64_to_cpu(src->a); + dst->b = be64_to_cpu(src->b); +} + +static inline void u128_inc(u128 *i) +{ + i->b++; + if (!i->b) + i->a++; +} + +static void ctr_crypt_final(struct blkcipher_desc *desc, + struct blkcipher_walk *walk) +{ + struct twofish_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + u8 *ctrblk = walk->iv; + u8 keystream[TF_BLOCK_SIZE]; + u8 *src = walk->src.virt.addr; + u8 *dst = walk->dst.virt.addr; + unsigned int nbytes = walk->nbytes; + + twofish_enc_blk(ctx, keystream, ctrblk); + crypto_xor(keystream, src, nbytes); + memcpy(dst, keystream, nbytes); + + crypto_inc(ctrblk, TF_BLOCK_SIZE); +} + +static unsigned int __ctr_crypt(struct blkcipher_desc *desc, + struct blkcipher_walk *walk) +{ + struct twofish_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + const unsigned int bsize = TF_BLOCK_SIZE; + unsigned int nbytes = walk->nbytes; + u128 *src = (u128 *)walk->src.virt.addr; + u128 *dst = (u128 *)walk->dst.virt.addr; + u128 ctrblk; + be128 ctrblocks[TWOFISH_PARALLEL_BLOCKS]; + int i; + + be128_to_u128(&ctrblk, (be128 *)walk->iv); + + /* Process multi-block batch */ + if (nbytes >= bsize * TWOFISH_PARALLEL_BLOCKS) { + do { + /* create ctrblks for parallel encrypt */ + for (i = 0; i < TWOFISH_PARALLEL_BLOCKS; i++) { + if (dst != src) + dst[i] = src[i]; + + u128_to_be128(&ctrblocks[i], &ctrblk); + u128_inc(&ctrblk); + } + + twofish_enc_blk_xway_xor(ctx, (u8 *)dst, + (u8 *)ctrblocks); + + src += TWOFISH_PARALLEL_BLOCKS; + dst += TWOFISH_PARALLEL_BLOCKS; + nbytes -= bsize * TWOFISH_PARALLEL_BLOCKS; + } while (nbytes >= bsize * TWOFISH_PARALLEL_BLOCKS); + + if (nbytes < bsize) + goto done; + } + + /* Process three block batch */ + if (nbytes >= bsize * 3) { + do { + if (dst != src) { + dst[0] = src[0]; + dst[1] = src[1]; + dst[2] = src[2]; + } + + /* create ctrblks for parallel encrypt */ + u128_to_be128(&ctrblocks[0], &ctrblk); + u128_inc(&ctrblk); + u128_to_be128(&ctrblocks[1], &ctrblk); + u128_inc(&ctrblk); + u128_to_be128(&ctrblocks[2], &ctrblk); + u128_inc(&ctrblk); + + twofish_enc_blk_3way_xor(ctx, (u8 *)dst, + (u8 *)ctrblocks); + + src += 3; + dst += 3; + nbytes -= bsize * 3; + } while (nbytes >= bsize * 3); + + if (nbytes < bsize) + goto done; + } + + /* Handle leftovers */ + do { + if (dst != src) + *dst = *src; + + u128_to_be128(&ctrblocks[0], &ctrblk); + u128_inc(&ctrblk); + + twofish_enc_blk(ctx, (u8 *)ctrblocks, (u8 *)ctrblocks); + u128_xor(dst, dst, (u128 *)ctrblocks); + + src += 1; + dst += 1; + nbytes -= bsize; + } while (nbytes >= bsize); + +done: + u128_to_be128((be128 *)walk->iv, &ctrblk); + return nbytes; +} + +static int ctr_crypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + bool fpu_enabled = false; + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt_block(desc, &walk, TF_BLOCK_SIZE); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + + while ((nbytes = walk.nbytes) >= TF_BLOCK_SIZE) { + fpu_enabled = twofish_fpu_begin(fpu_enabled, nbytes); + nbytes = __ctr_crypt(desc, &walk); + err = blkcipher_walk_done(desc, &walk, nbytes); + } + + twofish_fpu_end(fpu_enabled); + + if (walk.nbytes) { + ctr_crypt_final(desc, &walk); + err = blkcipher_walk_done(desc, &walk, 0); + } + + return err; +} + +struct crypt_priv { + struct twofish_ctx *ctx; + bool fpu_enabled; +}; + +static void encrypt_callback(void *priv, u8 *srcdst, unsigned int nbytes) +{ + const unsigned int bsize = TF_BLOCK_SIZE; + struct crypt_priv *ctx = priv; + int i; + + ctx->fpu_enabled = twofish_fpu_begin(ctx->fpu_enabled, nbytes); + + if (nbytes == bsize * TWOFISH_PARALLEL_BLOCKS) { + twofish_enc_blk_xway(ctx->ctx, srcdst, srcdst); + return; + } + + for (i = 0; i < nbytes / (bsize * 3); i++, srcdst += bsize * 3) + twofish_enc_blk_3way(ctx->ctx, srcdst, srcdst); + + nbytes %= bsize * 3; + + for (i = 0; i < nbytes / bsize; i++, srcdst += bsize) + twofish_enc_blk(ctx->ctx, srcdst, srcdst); +} + +static void decrypt_callback(void *priv, u8 *srcdst, unsigned int nbytes) +{ + const unsigned int bsize = TF_BLOCK_SIZE; + struct crypt_priv *ctx = priv; + int i; + + ctx->fpu_enabled = twofish_fpu_begin(ctx->fpu_enabled, nbytes); + + if (nbytes == bsize * TWOFISH_PARALLEL_BLOCKS) { + twofish_dec_blk_xway(ctx->ctx, srcdst, srcdst); + return; + } + + for (i = 0; i < nbytes / (bsize * 3); i++, srcdst += bsize * 3) + twofish_dec_blk_3way(ctx->ctx, srcdst, srcdst); + + nbytes %= bsize * 3; + + for (i = 0; i < nbytes / bsize; i++, srcdst += bsize) + twofish_dec_blk(ctx->ctx, srcdst, srcdst); +} + +struct twofish_lrw_ctx { + struct lrw_table_ctx lrw_table; + struct twofish_ctx twofish_ctx; +}; + +static int lrw_twofish_setkey(struct crypto_tfm *tfm, const u8 *key, + unsigned int keylen) +{ + struct twofish_lrw_ctx *ctx = crypto_tfm_ctx(tfm); + int err; + + err = __twofish_setkey(&ctx->twofish_ctx, key, + keylen - TF_BLOCK_SIZE, &tfm->crt_flags); + if (err) + return err; + + return lrw_init_table(&ctx->lrw_table, key + keylen - + TF_BLOCK_SIZE); +} + +static int lrw_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct twofish_lrw_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + be128 buf[TWOFISH_PARALLEL_BLOCKS]; + struct crypt_priv crypt_ctx = { + .ctx = &ctx->twofish_ctx, + .fpu_enabled = false, + }; + struct lrw_crypt_req req = { + .tbuf = buf, + .tbuflen = sizeof(buf), + + .table_ctx = &ctx->lrw_table, + .crypt_ctx = &crypt_ctx, + .crypt_fn = encrypt_callback, + }; + int ret; + + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + ret = lrw_crypt(desc, dst, src, nbytes, &req); + twofish_fpu_end(crypt_ctx.fpu_enabled); + + return ret; +} + +static int lrw_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct twofish_lrw_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + be128 buf[TWOFISH_PARALLEL_BLOCKS]; + struct crypt_priv crypt_ctx = { + .ctx = &ctx->twofish_ctx, + .fpu_enabled = false, + }; + struct lrw_crypt_req req = { + .tbuf = buf, + .tbuflen = sizeof(buf), + + .table_ctx = &ctx->lrw_table, + .crypt_ctx = &crypt_ctx, + .crypt_fn = decrypt_callback, + }; + int ret; + + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + ret = lrw_crypt(desc, dst, src, nbytes, &req); + twofish_fpu_end(crypt_ctx.fpu_enabled); + + return ret; +} + +static void lrw_exit_tfm(struct crypto_tfm *tfm) +{ + struct twofish_lrw_ctx *ctx = crypto_tfm_ctx(tfm); + + lrw_free_table(&ctx->lrw_table); +} + +struct twofish_xts_ctx { + struct twofish_ctx tweak_ctx; + struct twofish_ctx crypt_ctx; +}; + +static int xts_twofish_setkey(struct crypto_tfm *tfm, const u8 *key, + unsigned int keylen) +{ + struct twofish_xts_ctx *ctx = crypto_tfm_ctx(tfm); + u32 *flags = &tfm->crt_flags; + int err; + + /* key consists of keys of equal size concatenated, therefore + * the length must be even + */ + if (keylen % 2) { + *flags |= CRYPTO_TFM_RES_BAD_KEY_LEN; + return -EINVAL; + } + + /* first half of xts-key is for crypt */ + err = __twofish_setkey(&ctx->crypt_ctx, key, keylen / 2, flags); + if (err) + return err; + + /* second half of xts-key is for tweak */ + return __twofish_setkey(&ctx->tweak_ctx, + key + keylen / 2, keylen / 2, flags); +} + +static int xts_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct twofish_xts_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + be128 buf[TWOFISH_PARALLEL_BLOCKS]; + struct crypt_priv crypt_ctx = { + .ctx = &ctx->crypt_ctx, + .fpu_enabled = false, + }; + struct xts_crypt_req req = { + .tbuf = buf, + .tbuflen = sizeof(buf), + + .tweak_ctx = &ctx->tweak_ctx, + .tweak_fn = XTS_TWEAK_CAST(twofish_enc_blk), + .crypt_ctx = &crypt_ctx, + .crypt_fn = encrypt_callback, + }; + int ret; + + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + ret = xts_crypt(desc, dst, src, nbytes, &req); + twofish_fpu_end(crypt_ctx.fpu_enabled); + + return ret; +} + +static int xts_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct twofish_xts_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + be128 buf[TWOFISH_PARALLEL_BLOCKS]; + struct crypt_priv crypt_ctx = { + .ctx = &ctx->crypt_ctx, + .fpu_enabled = false, + }; + struct xts_crypt_req req = { + .tbuf = buf, + .tbuflen = sizeof(buf), + + .tweak_ctx = &ctx->tweak_ctx, + .tweak_fn = XTS_TWEAK_CAST(twofish_enc_blk), + .crypt_ctx = &crypt_ctx, + .crypt_fn = decrypt_callback, + }; + int ret; + + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + ret = xts_crypt(desc, dst, src, nbytes, &req); + twofish_fpu_end(crypt_ctx.fpu_enabled); + + return ret; +} + +static int ablk_set_key(struct crypto_ablkcipher *tfm, const u8 *key, + unsigned int key_len) +{ + struct async_twofish_ctx *ctx = crypto_ablkcipher_ctx(tfm); + struct crypto_ablkcipher *child = &ctx->cryptd_tfm->base; + int err; + + crypto_ablkcipher_clear_flags(child, CRYPTO_TFM_REQ_MASK); + crypto_ablkcipher_set_flags(child, crypto_ablkcipher_get_flags(tfm) + & CRYPTO_TFM_REQ_MASK); + err = crypto_ablkcipher_setkey(child, key, key_len); + crypto_ablkcipher_set_flags(tfm, crypto_ablkcipher_get_flags(child) + & CRYPTO_TFM_RES_MASK); + return err; +} + +static int __ablk_encrypt(struct ablkcipher_request *req) +{ + struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); + struct async_twofish_ctx *ctx = crypto_ablkcipher_ctx(tfm); + struct blkcipher_desc desc; + + desc.tfm = cryptd_ablkcipher_child(ctx->cryptd_tfm); + desc.info = req->info; + desc.flags = 0; + + return crypto_blkcipher_crt(desc.tfm)->encrypt( + &desc, req->dst, req->src, req->nbytes); +} + +static int ablk_encrypt(struct ablkcipher_request *req) +{ + struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); + struct async_twofish_ctx *ctx = crypto_ablkcipher_ctx(tfm); + + if (!irq_fpu_usable()) { + struct ablkcipher_request *cryptd_req = + ablkcipher_request_ctx(req); + + memcpy(cryptd_req, req, sizeof(*req)); + ablkcipher_request_set_tfm(cryptd_req, &ctx->cryptd_tfm->base); + + return crypto_ablkcipher_encrypt(cryptd_req); + } else { + return __ablk_encrypt(req); + } +} + +static int ablk_decrypt(struct ablkcipher_request *req) +{ + struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); + struct async_twofish_ctx *ctx = crypto_ablkcipher_ctx(tfm); + + if (!irq_fpu_usable()) { + struct ablkcipher_request *cryptd_req = + ablkcipher_request_ctx(req); + + memcpy(cryptd_req, req, sizeof(*req)); + ablkcipher_request_set_tfm(cryptd_req, &ctx->cryptd_tfm->base); + + return crypto_ablkcipher_decrypt(cryptd_req); + } else { + struct blkcipher_desc desc; + + desc.tfm = cryptd_ablkcipher_child(ctx->cryptd_tfm); + desc.info = req->info; + desc.flags = 0; + + return crypto_blkcipher_crt(desc.tfm)->decrypt( + &desc, req->dst, req->src, req->nbytes); + } +} + +static void ablk_exit(struct crypto_tfm *tfm) +{ + struct async_twofish_ctx *ctx = crypto_tfm_ctx(tfm); + + cryptd_free_ablkcipher(ctx->cryptd_tfm); +} + +static int ablk_init(struct crypto_tfm *tfm) +{ + struct async_twofish_ctx *ctx = crypto_tfm_ctx(tfm); + struct cryptd_ablkcipher *cryptd_tfm; + char drv_name[CRYPTO_MAX_ALG_NAME]; + + snprintf(drv_name, sizeof(drv_name), "__driver-%s", + crypto_tfm_alg_driver_name(tfm)); + + cryptd_tfm = cryptd_alloc_ablkcipher(drv_name, 0, 0); + if (IS_ERR(cryptd_tfm)) + return PTR_ERR(cryptd_tfm); + + ctx->cryptd_tfm = cryptd_tfm; + tfm->crt_ablkcipher.reqsize = sizeof(struct ablkcipher_request) + + crypto_ablkcipher_reqsize(&cryptd_tfm->base); + + return 0; +} + +static struct crypto_alg twofish_algs[10] = { { + .cra_name = "__ecb-twofish-avx", + .cra_driver_name = "__driver-ecb-twofish-avx", + .cra_priority = 0, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = TF_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct twofish_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(twofish_algs[0].cra_list), + .cra_u = { + .blkcipher = { + .min_keysize = TF_MIN_KEY_SIZE, + .max_keysize = TF_MAX_KEY_SIZE, + .setkey = twofish_setkey, + .encrypt = ecb_encrypt, + .decrypt = ecb_decrypt, + }, + }, +}, { + .cra_name = "__cbc-twofish-avx", + .cra_driver_name = "__driver-cbc-twofish-avx", + .cra_priority = 0, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = TF_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct twofish_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(twofish_algs[1].cra_list), + .cra_u = { + .blkcipher = { + .min_keysize = TF_MIN_KEY_SIZE, + .max_keysize = TF_MAX_KEY_SIZE, + .setkey = twofish_setkey, + .encrypt = cbc_encrypt, + .decrypt = cbc_decrypt, + }, + }, +}, { + .cra_name = "__ctr-twofish-avx", + .cra_driver_name = "__driver-ctr-twofish-avx", + .cra_priority = 0, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = 1, + .cra_ctxsize = sizeof(struct twofish_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(twofish_algs[2].cra_list), + .cra_u = { + .blkcipher = { + .min_keysize = TF_MIN_KEY_SIZE, + .max_keysize = TF_MAX_KEY_SIZE, + .ivsize = TF_BLOCK_SIZE, + .setkey = twofish_setkey, + .encrypt = ctr_crypt, + .decrypt = ctr_crypt, + }, + }, +}, { + .cra_name = "__lrw-twofish-avx", + .cra_driver_name = "__driver-lrw-twofish-avx", + .cra_priority = 0, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = TF_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct twofish_lrw_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(twofish_algs[3].cra_list), + .cra_exit = lrw_exit_tfm, + .cra_u = { + .blkcipher = { + .min_keysize = TF_MIN_KEY_SIZE + + TF_BLOCK_SIZE, + .max_keysize = TF_MAX_KEY_SIZE + + TF_BLOCK_SIZE, + .ivsize = TF_BLOCK_SIZE, + .setkey = lrw_twofish_setkey, + .encrypt = lrw_encrypt, + .decrypt = lrw_decrypt, + }, + }, +}, { + .cra_name = "__xts-twofish-avx", + .cra_driver_name = "__driver-xts-twofish-avx", + .cra_priority = 0, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = TF_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct twofish_xts_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(twofish_algs[4].cra_list), + .cra_u = { + .blkcipher = { + .min_keysize = TF_MIN_KEY_SIZE * 2, + .max_keysize = TF_MAX_KEY_SIZE * 2, + .ivsize = TF_BLOCK_SIZE, + .setkey = xts_twofish_setkey, + .encrypt = xts_encrypt, + .decrypt = xts_decrypt, + }, + }, +}, { + .cra_name = "ecb(twofish)", + .cra_driver_name = "ecb-twofish-avx", + .cra_priority = 400, + .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, + .cra_blocksize = TF_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct async_twofish_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_ablkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(twofish_algs[5].cra_list), + .cra_init = ablk_init, + .cra_exit = ablk_exit, + .cra_u = { + .ablkcipher = { + .min_keysize = TF_MIN_KEY_SIZE, + .max_keysize = TF_MAX_KEY_SIZE, + .setkey = ablk_set_key, + .encrypt = ablk_encrypt, + .decrypt = ablk_decrypt, + }, + }, +}, { + .cra_name = "cbc(twofish)", + .cra_driver_name = "cbc-twofish-avx", + .cra_priority = 400, + .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, + .cra_blocksize = TF_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct async_twofish_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_ablkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(twofish_algs[6].cra_list), + .cra_init = ablk_init, + .cra_exit = ablk_exit, + .cra_u = { + .ablkcipher = { + .min_keysize = TF_MIN_KEY_SIZE, + .max_keysize = TF_MAX_KEY_SIZE, + .ivsize = TF_BLOCK_SIZE, + .setkey = ablk_set_key, + .encrypt = __ablk_encrypt, + .decrypt = ablk_decrypt, + }, + }, +}, { + .cra_name = "ctr(twofish)", + .cra_driver_name = "ctr-twofish-avx", + .cra_priority = 400, + .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, + .cra_blocksize = 1, + .cra_ctxsize = sizeof(struct async_twofish_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_ablkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(twofish_algs[7].cra_list), + .cra_init = ablk_init, + .cra_exit = ablk_exit, + .cra_u = { + .ablkcipher = { + .min_keysize = TF_MIN_KEY_SIZE, + .max_keysize = TF_MAX_KEY_SIZE, + .ivsize = TF_BLOCK_SIZE, + .setkey = ablk_set_key, + .encrypt = ablk_encrypt, + .decrypt = ablk_encrypt, + .geniv = "chainiv", + }, + }, +}, { + .cra_name = "lrw(twofish)", + .cra_driver_name = "lrw-twofish-avx", + .cra_priority = 400, + .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, + .cra_blocksize = TF_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct async_twofish_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_ablkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(twofish_algs[8].cra_list), + .cra_init = ablk_init, + .cra_exit = ablk_exit, + .cra_u = { + .ablkcipher = { + .min_keysize = TF_MIN_KEY_SIZE + + TF_BLOCK_SIZE, + .max_keysize = TF_MAX_KEY_SIZE + + TF_BLOCK_SIZE, + .ivsize = TF_BLOCK_SIZE, + .setkey = ablk_set_key, + .encrypt = ablk_encrypt, + .decrypt = ablk_decrypt, + }, + }, +}, { + .cra_name = "xts(twofish)", + .cra_driver_name = "xts-twofish-avx", + .cra_priority = 400, + .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, + .cra_blocksize = TF_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct async_twofish_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_ablkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(twofish_algs[9].cra_list), + .cra_init = ablk_init, + .cra_exit = ablk_exit, + .cra_u = { + .ablkcipher = { + .min_keysize = TF_MIN_KEY_SIZE * 2, + .max_keysize = TF_MAX_KEY_SIZE * 2, + .ivsize = TF_BLOCK_SIZE, + .setkey = ablk_set_key, + .encrypt = ablk_encrypt, + .decrypt = ablk_decrypt, + }, + }, +} }; + +static int __init twofish_init(void) +{ + u64 xcr0; + + if (!cpu_has_avx || !cpu_has_osxsave) { + printk(KERN_INFO "AVX instructions are not detected.\n"); + return -ENODEV; + } + + xcr0 = xgetbv(XCR_XFEATURE_ENABLED_MASK); + if ((xcr0 & (XSTATE_SSE | XSTATE_YMM)) != (XSTATE_SSE | XSTATE_YMM)) { + printk(KERN_INFO "AVX detected but unusable.\n"); + return -ENODEV; + } + + return crypto_register_algs(twofish_algs, ARRAY_SIZE(twofish_algs)); +} + +static void __exit twofish_exit(void) +{ + crypto_unregister_algs(twofish_algs, ARRAY_SIZE(twofish_algs)); +} + +module_init(twofish_init); +module_exit(twofish_exit); + +MODULE_DESCRIPTION("Twofish Cipher Algorithm, AVX optimized"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("twofish"); diff --git a/arch/x86/crypto/twofish_glue_3way.c b/arch/x86/crypto/twofish_glue_3way.c index 922ab24cce3..77e4e55a266 100644 --- a/arch/x86/crypto/twofish_glue_3way.c +++ b/arch/x86/crypto/twofish_glue_3way.c @@ -45,8 +45,10 @@ asmlinkage void twofish_dec_blk(struct twofish_ctx *ctx, u8 *dst, /* 3-way parallel cipher functions */ asmlinkage void __twofish_enc_blk_3way(struct twofish_ctx *ctx, u8 *dst, const u8 *src, bool xor); +EXPORT_SYMBOL_GPL(__twofish_enc_blk_3way); asmlinkage void twofish_dec_blk_3way(struct twofish_ctx *ctx, u8 *dst, const u8 *src); +EXPORT_SYMBOL_GPL(twofish_dec_blk_3way); static inline void twofish_enc_blk_3way(struct twofish_ctx *ctx, u8 *dst, const u8 *src) diff --git a/crypto/Kconfig b/crypto/Kconfig index 8e84225c096..e00a4e49e01 100644 --- a/crypto/Kconfig +++ b/crypto/Kconfig @@ -913,6 +913,30 @@ config CRYPTO_TWOFISH_X86_64_3WAY See also: +config CRYPTO_TWOFISH_AVX_X86_64 + tristate "Twofish cipher algorithm (x86_64/AVX)" + depends on X86 && 64BIT + select CRYPTO_ALGAPI + select CRYPTO_CRYPTD + select CRYPTO_TWOFISH_COMMON + select CRYPTO_TWOFISH_X86_64 + select CRYPTO_TWOFISH_X86_64_3WAY + select CRYPTO_LRW + select CRYPTO_XTS + help + Twofish cipher algorithm (x86_64/AVX). + + Twofish was submitted as an AES (Advanced Encryption Standard) + candidate cipher by researchers at CounterPane Systems. It is a + 16 round block cipher supporting key sizes of 128, 192, and 256 + bits. + + This module provides the Twofish cipher algorithm that processes + eight blocks parallel using the AVX Instruction Set. + + See also: + + comment "Compression" config CRYPTO_DEFLATE diff --git a/crypto/tcrypt.c b/crypto/tcrypt.c index 750cce44bad..2af879786e7 100644 --- a/crypto/tcrypt.c +++ b/crypto/tcrypt.c @@ -1566,6 +1566,29 @@ static int do_test(int m) speed_template_32_64); break; + case 504: + test_acipher_speed("ecb(twofish)", ENCRYPT, sec, NULL, 0, + speed_template_16_24_32); + test_acipher_speed("ecb(twofish)", DECRYPT, sec, NULL, 0, + speed_template_16_24_32); + test_acipher_speed("cbc(twofish)", ENCRYPT, sec, NULL, 0, + speed_template_16_24_32); + test_acipher_speed("cbc(twofish)", DECRYPT, sec, NULL, 0, + speed_template_16_24_32); + test_acipher_speed("ctr(twofish)", ENCRYPT, sec, NULL, 0, + speed_template_16_24_32); + test_acipher_speed("ctr(twofish)", DECRYPT, sec, NULL, 0, + speed_template_16_24_32); + test_acipher_speed("lrw(twofish)", ENCRYPT, sec, NULL, 0, + speed_template_32_40_48); + test_acipher_speed("lrw(twofish)", DECRYPT, sec, NULL, 0, + speed_template_32_40_48); + test_acipher_speed("xts(twofish)", ENCRYPT, sec, NULL, 0, + speed_template_32_48_64); + test_acipher_speed("xts(twofish)", DECRYPT, sec, NULL, 0, + speed_template_32_48_64); + break; + case 1000: test_available(); break; diff --git a/crypto/testmgr.c b/crypto/testmgr.c index eb6d20f8ec5..73b3ec6fe1a 100644 --- a/crypto/testmgr.c +++ b/crypto/testmgr.c @@ -1548,6 +1548,21 @@ static const struct alg_test_desc alg_test_descs[] = { } } } + }, { + .alg = "__cbc-twofish-avx", + .test = alg_test_null, + .suite = { + .cipher = { + .enc = { + .vecs = NULL, + .count = 0 + }, + .dec = { + .vecs = NULL, + .count = 0 + } + } + } }, { .alg = "__driver-cbc-aes-aesni", .test = alg_test_null, @@ -1578,6 +1593,21 @@ static const struct alg_test_desc alg_test_descs[] = { } } } + }, { + .alg = "__driver-cbc-twofish-avx", + .test = alg_test_null, + .suite = { + .cipher = { + .enc = { + .vecs = NULL, + .count = 0 + }, + .dec = { + .vecs = NULL, + .count = 0 + } + } + } }, { .alg = "__driver-ecb-aes-aesni", .test = alg_test_null, @@ -1608,6 +1638,21 @@ static const struct alg_test_desc alg_test_descs[] = { } } } + }, { + .alg = "__driver-ecb-twofish-avx", + .test = alg_test_null, + .suite = { + .cipher = { + .enc = { + .vecs = NULL, + .count = 0 + }, + .dec = { + .vecs = NULL, + .count = 0 + } + } + } }, { .alg = "__ghash-pclmulqdqni", .test = alg_test_null, @@ -1805,6 +1850,21 @@ static const struct alg_test_desc alg_test_descs[] = { } } } + }, { + .alg = "cryptd(__driver-ecb-twofish-avx)", + .test = alg_test_null, + .suite = { + .cipher = { + .enc = { + .vecs = NULL, + .count = 0 + }, + .dec = { + .vecs = NULL, + .count = 0 + } + } + } }, { .alg = "cryptd(__ghash-pclmulqdqni)", .test = alg_test_null, -- cgit v1.2.3 From 4da7de4d8be7d18559c56bca446b1161a3b63acc Mon Sep 17 00:00:00 2001 From: Johannes Goetzfried Date: Mon, 28 May 2012 15:55:38 +0200 Subject: crypto: testmgr - expand twofish test vectors The AVX implementation of the twofish cipher processes 8 blocks parallel, so we need to make test vectors larger to check parallel code paths. Test vectors are also large enough to deal with 16 block parallel implementations which may occur in the future. Signed-off-by: Johannes Goetzfried Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu --- crypto/testmgr.h | 928 +++++++++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 896 insertions(+), 32 deletions(-) (limited to 'crypto') diff --git a/crypto/testmgr.h b/crypto/testmgr.h index 34a9d511dee..4d84fe48df0 100644 --- a/crypto/testmgr.h +++ b/crypto/testmgr.h @@ -2765,8 +2765,62 @@ static struct cipher_testvec tf_enc_tv_template[] = { "\x1E\x92\x29\xC0\x34\xCB\x62\xF9" "\x6D\x04\x9B\x0F\xA6\x3D\xD4\x48" "\xDF\x76\x0D\x81\x18\xAF\x23\xBA" - "\x51\xE8\x5C\xF3\x8A\x21\x95\x2C", - .ilen = 64, + "\x51\xE8\x5C\xF3\x8A\x21\x95\x2C" + "\xC3\x37\xCE\x65\xFC\x70\x07\x9E" + "\x12\xA9\x40\xD7\x4B\xE2\x79\x10" + "\x84\x1B\xB2\x26\xBD\x54\xEB\x5F" + "\xF6\x8D\x01\x98\x2F\xC6\x3A\xD1" + "\x68\xFF\x73\x0A\xA1\x15\xAC\x43" + "\xDA\x4E\xE5\x7C\x13\x87\x1E\xB5" + "\x29\xC0\x57\xEE\x62\xF9\x90\x04" + "\x9B\x32\xC9\x3D\xD4\x6B\x02\x76" + "\x0D\xA4\x18\xAF\x46\xDD\x51\xE8" + "\x7F\x16\x8A\x21\xB8\x2C\xC3\x5A" + "\xF1\x65\xFC\x93\x07\x9E\x35\xCC" + "\x40\xD7\x6E\x05\x79\x10\xA7\x1B" + "\xB2\x49\xE0\x54\xEB\x82\x19\x8D" + "\x24\xBB\x2F\xC6\x5D\xF4\x68\xFF" + "\x96\x0A\xA1\x38\xCF\x43\xDA\x71" + "\x08\x7C\x13\xAA\x1E\xB5\x4C\xE3" + "\x57\xEE\x85\x1C\x90\x27\xBE\x32" + "\xC9\x60\xF7\x6B\x02\x99\x0D\xA4" + "\x3B\xD2\x46\xDD\x74\x0B\x7F\x16" + "\xAD\x21\xB8\x4F\xE6\x5A\xF1\x88" + "\x1F\x93\x2A\xC1\x35\xCC\x63\xFA" + "\x6E\x05\x9C\x10\xA7\x3E\xD5\x49" + "\xE0\x77\x0E\x82\x19\xB0\x24\xBB" + "\x52\xE9\x5D\xF4\x8B\x22\x96\x2D" + "\xC4\x38\xCF\x66\xFD\x71\x08\x9F" + "\x13\xAA\x41\xD8\x4C\xE3\x7A\x11" + "\x85\x1C\xB3\x27\xBE\x55\xEC\x60" + "\xF7\x8E\x02\x99\x30\xC7\x3B\xD2" + "\x69\x00\x74\x0B\xA2\x16\xAD\x44" + "\xDB\x4F\xE6\x7D\x14\x88\x1F\xB6" + "\x2A\xC1\x58\xEF\x63\xFA\x91\x05" + "\x9C\x33\xCA\x3E\xD5\x6C\x03\x77" + "\x0E\xA5\x19\xB0\x47\xDE\x52\xE9" + "\x80\x17\x8B\x22\xB9\x2D\xC4\x5B" + "\xF2\x66\xFD\x94\x08\x9F\x36\xCD" + "\x41\xD8\x6F\x06\x7A\x11\xA8\x1C" + "\xB3\x4A\xE1\x55\xEC\x83\x1A\x8E" + "\x25\xBC\x30\xC7\x5E\xF5\x69\x00" + "\x97\x0B\xA2\x39\xD0\x44\xDB\x72" + "\x09\x7D\x14\xAB\x1F\xB6\x4D\xE4" + "\x58\xEF\x86\x1D\x91\x28\xBF\x33" + "\xCA\x61\xF8\x6C\x03\x9A\x0E\xA5" + "\x3C\xD3\x47\xDE\x75\x0C\x80\x17" + "\xAE\x22\xB9\x50\xE7\x5B\xF2\x89" + "\x20\x94\x2B\xC2\x36\xCD\x64\xFB" + "\x6F\x06\x9D\x11\xA8\x3F\xD6\x4A" + "\xE1\x78\x0F\x83\x1A\xB1\x25\xBC" + "\x53\xEA\x5E\xF5\x8C\x00\x97\x2E" + "\xC5\x39\xD0\x67\xFE\x72\x09\xA0" + "\x14\xAB\x42\xD9\x4D\xE4\x7B\x12" + "\x86\x1D\xB4\x28\xBF\x56\xED\x61" + "\xF8\x8F\x03\x9A\x31\xC8\x3C\xD3" + "\x6A\x01\x75\x0C\xA3\x17\xAE\x45" + "\xDC\x50\xE7\x7E\x15\x89\x20\xB7", + .ilen = 496, .result = "\x88\xCB\x1E\xC2\xAF\x8A\x97\xFF" "\xF6\x90\x46\x9C\x4A\x0F\x08\xDC" "\xDE\xAB\xAD\xFA\xFC\xA8\xC2\x3D" @@ -2774,8 +2828,62 @@ static struct cipher_testvec tf_enc_tv_template[] = { "\x34\x9E\xB6\x08\xB2\xDD\xA8\xF5" "\xDF\xFA\xC7\xE8\x09\x50\x76\x08" "\xA2\xB6\x6A\x59\xC0\x2B\x6D\x05" - "\x89\xF6\x82\xF0\xD3\xDB\x06\x02", - .rlen = 64, + "\x89\xF6\x82\xF0\xD3\xDB\x06\x02" + "\xB5\x11\x5C\x5E\x79\x1A\xAC\x43" + "\x5C\xC0\x30\x4B\x6B\x16\xA1\x40" + "\x80\x27\x88\xBA\x2C\x74\x42\xE0" + "\x1B\xA5\x85\x08\xB9\xE6\x22\x7A" + "\x36\x3B\x0D\x9F\xA0\x22\x6C\x2A" + "\x91\x75\x47\xBC\x67\x21\x4E\xF9" + "\xEA\xFF\xD9\xD5\xC0\xFC\x9E\x2C" + "\x3E\xAD\xC6\x61\x0E\x93\x7A\x22" + "\x09\xC8\x8D\xC1\x8E\xB4\x8B\x5C" + "\xC6\x24\x42\xB8\x23\x66\x80\xA9" + "\x32\x0B\x7A\x29\xBF\xB3\x0B\x63" + "\x43\x27\x13\xA9\xBE\xEB\xBD\xF3" + "\x33\x62\x70\xE2\x1B\x86\x7A\xA1" + "\x51\x4A\x16\xFE\x29\x63\x7E\xD0" + "\x7A\xA4\x6E\x2C\xF8\xC1\xDB\xE8" + "\xCB\x4D\xD2\x8C\x04\x14\xB4\x66" + "\x41\xB7\x3A\x96\x16\x7C\x1D\x5B" + "\xB6\x41\x42\x64\x43\xEE\x6E\x7C" + "\x8B\xAF\x01\x9C\xA4\x6E\x75\x8F" + "\xDE\x10\x9F\xA6\xE7\xD6\x44\x97" + "\x66\xA3\x96\x0F\x1C\x25\x60\xF5" + "\x3C\x2E\x32\x69\x0E\x82\xFF\x27" + "\x0F\xB5\x06\xDA\xD8\x31\x15\x6C" + "\xDF\x18\x6C\x87\xF5\x3B\x11\x9A" + "\x1B\x42\x1F\x5B\x29\x19\x96\x13" + "\x68\x2E\x5E\x08\x1C\x8F\x32\x4B" + "\x81\x77\x6D\xF4\xA0\x01\x42\xEC" + "\xDD\x5B\xFD\x3A\x8E\x6A\x14\xFB" + "\x83\x54\xDF\x0F\x86\xB7\xEA\x40" + "\x46\x39\xF7\x2A\x89\x8D\x4E\x96" + "\x5F\x5F\x6D\x76\xC6\x13\x9D\x3D" + "\x1D\x5F\x0C\x7D\xE2\xBC\xC2\x16" + "\x16\xBE\x89\x3E\xB0\x61\xA2\x5D" + "\xAF\xD1\x40\x5F\x1A\xB8\x26\x41" + "\xC6\xBD\x36\xEF\xED\x29\x50\x6D" + "\x10\xEF\x26\xE8\xA8\x93\x11\x3F" + "\x2D\x1F\x88\x20\x77\x45\xF5\x66" + "\x08\xB9\xF1\xEF\xB1\x93\xA8\x81" + "\x65\xC5\xCD\x3E\x8C\x06\x60\x2C" + "\xB2\x10\x7A\xCA\x05\x25\x59\xDB" + "\xC7\x28\xF5\x20\x35\x52\x9E\x62" + "\xF8\x88\x24\x1C\x4D\x84\x12\x39" + "\x39\xE4\x2E\xF4\xD4\x9D\x2B\xBC" + "\x87\x66\xE6\xC0\x6B\x31\x9A\x66" + "\x03\xDC\x95\xD8\x6B\xD0\x30\x8F" + "\xDF\x8F\x8D\xFA\xEC\x1F\x08\xBD" + "\xA3\x63\xE2\x71\x4F\x03\x94\x87" + "\x50\xDF\x15\x1F\xED\x3A\xA3\x7F" + "\x1F\x2A\xB5\xA1\x69\xAC\x4B\x0D" + "\x84\x9B\x2A\xE9\x55\xDD\x46\x91" + "\x15\x33\xF3\x2B\x9B\x46\x97\x00" + "\xF0\x29\xD8\x59\x5D\x33\x37\xF9" + "\x58\x33\x9B\x78\xC7\x58\x48\x6B" + "\x2C\x75\x64\xC4\xCA\xC1\x7E\xD5", + .rlen = 496, }, }; @@ -2822,8 +2930,62 @@ static struct cipher_testvec tf_dec_tv_template[] = { "\x34\x9E\xB6\x08\xB2\xDD\xA8\xF5" "\xDF\xFA\xC7\xE8\x09\x50\x76\x08" "\xA2\xB6\x6A\x59\xC0\x2B\x6D\x05" - "\x89\xF6\x82\xF0\xD3\xDB\x06\x02", - .ilen = 64, + "\x89\xF6\x82\xF0\xD3\xDB\x06\x02" + "\xB5\x11\x5C\x5E\x79\x1A\xAC\x43" + "\x5C\xC0\x30\x4B\x6B\x16\xA1\x40" + "\x80\x27\x88\xBA\x2C\x74\x42\xE0" + "\x1B\xA5\x85\x08\xB9\xE6\x22\x7A" + "\x36\x3B\x0D\x9F\xA0\x22\x6C\x2A" + "\x91\x75\x47\xBC\x67\x21\x4E\xF9" + "\xEA\xFF\xD9\xD5\xC0\xFC\x9E\x2C" + "\x3E\xAD\xC6\x61\x0E\x93\x7A\x22" + "\x09\xC8\x8D\xC1\x8E\xB4\x8B\x5C" + "\xC6\x24\x42\xB8\x23\x66\x80\xA9" + "\x32\x0B\x7A\x29\xBF\xB3\x0B\x63" + "\x43\x27\x13\xA9\xBE\xEB\xBD\xF3" + "\x33\x62\x70\xE2\x1B\x86\x7A\xA1" + "\x51\x4A\x16\xFE\x29\x63\x7E\xD0" + "\x7A\xA4\x6E\x2C\xF8\xC1\xDB\xE8" + "\xCB\x4D\xD2\x8C\x04\x14\xB4\x66" + "\x41\xB7\x3A\x96\x16\x7C\x1D\x5B" + "\xB6\x41\x42\x64\x43\xEE\x6E\x7C" + "\x8B\xAF\x01\x9C\xA4\x6E\x75\x8F" + "\xDE\x10\x9F\xA6\xE7\xD6\x44\x97" + "\x66\xA3\x96\x0F\x1C\x25\x60\xF5" + "\x3C\x2E\x32\x69\x0E\x82\xFF\x27" + "\x0F\xB5\x06\xDA\xD8\x31\x15\x6C" + "\xDF\x18\x6C\x87\xF5\x3B\x11\x9A" + "\x1B\x42\x1F\x5B\x29\x19\x96\x13" + "\x68\x2E\x5E\x08\x1C\x8F\x32\x4B" + "\x81\x77\x6D\xF4\xA0\x01\x42\xEC" + "\xDD\x5B\xFD\x3A\x8E\x6A\x14\xFB" + "\x83\x54\xDF\x0F\x86\xB7\xEA\x40" + "\x46\x39\xF7\x2A\x89\x8D\x4E\x96" + "\x5F\x5F\x6D\x76\xC6\x13\x9D\x3D" + "\x1D\x5F\x0C\x7D\xE2\xBC\xC2\x16" + "\x16\xBE\x89\x3E\xB0\x61\xA2\x5D" + "\xAF\xD1\x40\x5F\x1A\xB8\x26\x41" + "\xC6\xBD\x36\xEF\xED\x29\x50\x6D" + "\x10\xEF\x26\xE8\xA8\x93\x11\x3F" + "\x2D\x1F\x88\x20\x77\x45\xF5\x66" + "\x08\xB9\xF1\xEF\xB1\x93\xA8\x81" + "\x65\xC5\xCD\x3E\x8C\x06\x60\x2C" + "\xB2\x10\x7A\xCA\x05\x25\x59\xDB" + "\xC7\x28\xF5\x20\x35\x52\x9E\x62" + "\xF8\x88\x24\x1C\x4D\x84\x12\x39" + "\x39\xE4\x2E\xF4\xD4\x9D\x2B\xBC" + "\x87\x66\xE6\xC0\x6B\x31\x9A\x66" + "\x03\xDC\x95\xD8\x6B\xD0\x30\x8F" + "\xDF\x8F\x8D\xFA\xEC\x1F\x08\xBD" + "\xA3\x63\xE2\x71\x4F\x03\x94\x87" + "\x50\xDF\x15\x1F\xED\x3A\xA3\x7F" + "\x1F\x2A\xB5\xA1\x69\xAC\x4B\x0D" + "\x84\x9B\x2A\xE9\x55\xDD\x46\x91" + "\x15\x33\xF3\x2B\x9B\x46\x97\x00" + "\xF0\x29\xD8\x59\x5D\x33\x37\xF9" + "\x58\x33\x9B\x78\xC7\x58\x48\x6B" + "\x2C\x75\x64\xC4\xCA\xC1\x7E\xD5", + .ilen = 496, .result = "\x56\xED\x84\x1B\x8F\x26\xBD\x31" "\xC8\x5F\xF6\x6A\x01\x98\x0C\xA3" "\x3A\xD1\x45\xDC\x73\x0A\x7E\x15" @@ -2831,8 +2993,62 @@ static struct cipher_testvec tf_dec_tv_template[] = { "\x1E\x92\x29\xC0\x34\xCB\x62\xF9" "\x6D\x04\x9B\x0F\xA6\x3D\xD4\x48" "\xDF\x76\x0D\x81\x18\xAF\x23\xBA" - "\x51\xE8\x5C\xF3\x8A\x21\x95\x2C", - .rlen = 64, + "\x51\xE8\x5C\xF3\x8A\x21\x95\x2C" + "\xC3\x37\xCE\x65\xFC\x70\x07\x9E" + "\x12\xA9\x40\xD7\x4B\xE2\x79\x10" + "\x84\x1B\xB2\x26\xBD\x54\xEB\x5F" + "\xF6\x8D\x01\x98\x2F\xC6\x3A\xD1" + "\x68\xFF\x73\x0A\xA1\x15\xAC\x43" + "\xDA\x4E\xE5\x7C\x13\x87\x1E\xB5" + "\x29\xC0\x57\xEE\x62\xF9\x90\x04" + "\x9B\x32\xC9\x3D\xD4\x6B\x02\x76" + "\x0D\xA4\x18\xAF\x46\xDD\x51\xE8" + "\x7F\x16\x8A\x21\xB8\x2C\xC3\x5A" + "\xF1\x65\xFC\x93\x07\x9E\x35\xCC" + "\x40\xD7\x6E\x05\x79\x10\xA7\x1B" + "\xB2\x49\xE0\x54\xEB\x82\x19\x8D" + "\x24\xBB\x2F\xC6\x5D\xF4\x68\xFF" + "\x96\x0A\xA1\x38\xCF\x43\xDA\x71" + "\x08\x7C\x13\xAA\x1E\xB5\x4C\xE3" + "\x57\xEE\x85\x1C\x90\x27\xBE\x32" + "\xC9\x60\xF7\x6B\x02\x99\x0D\xA4" + "\x3B\xD2\x46\xDD\x74\x0B\x7F\x16" + "\xAD\x21\xB8\x4F\xE6\x5A\xF1\x88" + "\x1F\x93\x2A\xC1\x35\xCC\x63\xFA" + "\x6E\x05\x9C\x10\xA7\x3E\xD5\x49" + "\xE0\x77\x0E\x82\x19\xB0\x24\xBB" + "\x52\xE9\x5D\xF4\x8B\x22\x96\x2D" + "\xC4\x38\xCF\x66\xFD\x71\x08\x9F" + "\x13\xAA\x41\xD8\x4C\xE3\x7A\x11" + "\x85\x1C\xB3\x27\xBE\x55\xEC\x60" + "\xF7\x8E\x02\x99\x30\xC7\x3B\xD2" + "\x69\x00\x74\x0B\xA2\x16\xAD\x44" + "\xDB\x4F\xE6\x7D\x14\x88\x1F\xB6" + "\x2A\xC1\x58\xEF\x63\xFA\x91\x05" + "\x9C\x33\xCA\x3E\xD5\x6C\x03\x77" + "\x0E\xA5\x19\xB0\x47\xDE\x52\xE9" + "\x80\x17\x8B\x22\xB9\x2D\xC4\x5B" + "\xF2\x66\xFD\x94\x08\x9F\x36\xCD" + "\x41\xD8\x6F\x06\x7A\x11\xA8\x1C" + "\xB3\x4A\xE1\x55\xEC\x83\x1A\x8E" + "\x25\xBC\x30\xC7\x5E\xF5\x69\x00" + "\x97\x0B\xA2\x39\xD0\x44\xDB\x72" + "\x09\x7D\x14\xAB\x1F\xB6\x4D\xE4" + "\x58\xEF\x86\x1D\x91\x28\xBF\x33" + "\xCA\x61\xF8\x6C\x03\x9A\x0E\xA5" + "\x3C\xD3\x47\xDE\x75\x0C\x80\x17" + "\xAE\x22\xB9\x50\xE7\x5B\xF2\x89" + "\x20\x94\x2B\xC2\x36\xCD\x64\xFB" + "\x6F\x06\x9D\x11\xA8\x3F\xD6\x4A" + "\xE1\x78\x0F\x83\x1A\xB1\x25\xBC" + "\x53\xEA\x5E\xF5\x8C\x00\x97\x2E" + "\xC5\x39\xD0\x67\xFE\x72\x09\xA0" + "\x14\xAB\x42\xD9\x4D\xE4\x7B\x12" + "\x86\x1D\xB4\x28\xBF\x56\xED\x61" + "\xF8\x8F\x03\x9A\x31\xC8\x3C\xD3" + "\x6A\x01\x75\x0C\xA3\x17\xAE\x45" + "\xDC\x50\xE7\x7E\x15\x89\x20\xB7", + .rlen = 496, }, }; @@ -2894,8 +3110,62 @@ static struct cipher_testvec tf_cbc_enc_tv_template[] = { "\x1E\x92\x29\xC0\x34\xCB\x62\xF9" "\x6D\x04\x9B\x0F\xA6\x3D\xD4\x48" "\xDF\x76\x0D\x81\x18\xAF\x23\xBA" - "\x51\xE8\x5C\xF3\x8A\x21\x95\x2C", - .ilen = 64, + "\x51\xE8\x5C\xF3\x8A\x21\x95\x2C" + "\xC3\x37\xCE\x65\xFC\x70\x07\x9E" + "\x12\xA9\x40\xD7\x4B\xE2\x79\x10" + "\x84\x1B\xB2\x26\xBD\x54\xEB\x5F" + "\xF6\x8D\x01\x98\x2F\xC6\x3A\xD1" + "\x68\xFF\x73\x0A\xA1\x15\xAC\x43" + "\xDA\x4E\xE5\x7C\x13\x87\x1E\xB5" + "\x29\xC0\x57\xEE\x62\xF9\x90\x04" + "\x9B\x32\xC9\x3D\xD4\x6B\x02\x76" + "\x0D\xA4\x18\xAF\x46\xDD\x51\xE8" + "\x7F\x16\x8A\x21\xB8\x2C\xC3\x5A" + "\xF1\x65\xFC\x93\x07\x9E\x35\xCC" + "\x40\xD7\x6E\x05\x79\x10\xA7\x1B" + "\xB2\x49\xE0\x54\xEB\x82\x19\x8D" + "\x24\xBB\x2F\xC6\x5D\xF4\x68\xFF" + "\x96\x0A\xA1\x38\xCF\x43\xDA\x71" + "\x08\x7C\x13\xAA\x1E\xB5\x4C\xE3" + "\x57\xEE\x85\x1C\x90\x27\xBE\x32" + "\xC9\x60\xF7\x6B\x02\x99\x0D\xA4" + "\x3B\xD2\x46\xDD\x74\x0B\x7F\x16" + "\xAD\x21\xB8\x4F\xE6\x5A\xF1\x88" + "\x1F\x93\x2A\xC1\x35\xCC\x63\xFA" + "\x6E\x05\x9C\x10\xA7\x3E\xD5\x49" + "\xE0\x77\x0E\x82\x19\xB0\x24\xBB" + "\x52\xE9\x5D\xF4\x8B\x22\x96\x2D" + "\xC4\x38\xCF\x66\xFD\x71\x08\x9F" + "\x13\xAA\x41\xD8\x4C\xE3\x7A\x11" + "\x85\x1C\xB3\x27\xBE\x55\xEC\x60" + "\xF7\x8E\x02\x99\x30\xC7\x3B\xD2" + "\x69\x00\x74\x0B\xA2\x16\xAD\x44" + "\xDB\x4F\xE6\x7D\x14\x88\x1F\xB6" + "\x2A\xC1\x58\xEF\x63\xFA\x91\x05" + "\x9C\x33\xCA\x3E\xD5\x6C\x03\x77" + "\x0E\xA5\x19\xB0\x47\xDE\x52\xE9" + "\x80\x17\x8B\x22\xB9\x2D\xC4\x5B" + "\xF2\x66\xFD\x94\x08\x9F\x36\xCD" + "\x41\xD8\x6F\x06\x7A\x11\xA8\x1C" + "\xB3\x4A\xE1\x55\xEC\x83\x1A\x8E" + "\x25\xBC\x30\xC7\x5E\xF5\x69\x00" + "\x97\x0B\xA2\x39\xD0\x44\xDB\x72" + "\x09\x7D\x14\xAB\x1F\xB6\x4D\xE4" + "\x58\xEF\x86\x1D\x91\x28\xBF\x33" + "\xCA\x61\xF8\x6C\x03\x9A\x0E\xA5" + "\x3C\xD3\x47\xDE\x75\x0C\x80\x17" + "\xAE\x22\xB9\x50\xE7\x5B\xF2\x89" + "\x20\x94\x2B\xC2\x36\xCD\x64\xFB" + "\x6F\x06\x9D\x11\xA8\x3F\xD6\x4A" + "\xE1\x78\x0F\x83\x1A\xB1\x25\xBC" + "\x53\xEA\x5E\xF5\x8C\x00\x97\x2E" + "\xC5\x39\xD0\x67\xFE\x72\x09\xA0" + "\x14\xAB\x42\xD9\x4D\xE4\x7B\x12" + "\x86\x1D\xB4\x28\xBF\x56\xED\x61" + "\xF8\x8F\x03\x9A\x31\xC8\x3C\xD3" + "\x6A\x01\x75\x0C\xA3\x17\xAE\x45" + "\xDC\x50\xE7\x7E\x15\x89\x20\xB7", + .ilen = 496, .result = "\xC8\xFF\xF2\x53\xA6\x27\x09\xD1" "\x33\x38\xC2\xC0\x0C\x14\x7E\xB5" "\x26\x1B\x05\x0C\x05\x12\x3F\xC0" @@ -2903,8 +3173,62 @@ static struct cipher_testvec tf_cbc_enc_tv_template[] = { "\x3D\x32\xDF\xDA\x56\x00\x6E\xEE" "\x5B\x2A\x72\x9D\xC2\x4D\x19\xBC" "\x8C\x53\xFA\x87\x6F\xDD\x81\xA3" - "\xB1\xD3\x44\x65\xDF\xE7\x63\x38", - .rlen = 64, + "\xB1\xD3\x44\x65\xDF\xE7\x63\x38" + "\x4A\xFC\xDC\xEC\x3F\x26\x8E\xB8" + "\x43\xFC\xFE\x18\xB5\x11\x6D\x31" + "\x81\x8B\x0D\x75\xF6\x80\xEC\x84" + "\x04\xB9\xE6\x09\x63\xED\x39\xDB" + "\xC3\xF6\x14\xD6\x6E\x5E\x8B\xBD" + "\x3E\xFA\xD7\x98\x50\x6F\xD9\x63" + "\x02\xCD\x0D\x39\x4B\x0D\xEC\x80" + "\xE3\x6A\x17\xF4\xCC\xAD\xFF\x68" + "\x45\xDD\xC8\x83\x1D\x41\x96\x0D" + "\x91\x2E\x05\xD3\x59\x82\xE0\x43" + "\x90\x4F\xB9\xF7\xAD\x6B\x2E\xAF" + "\xA7\x84\x00\x53\xCD\x6F\xD1\x0C" + "\x4E\xF9\x5A\x23\xFB\xCA\xC7\xD3" + "\xA9\xAA\x9D\xB2\x3F\x66\xF1\xAC" + "\x25\x21\x8F\xF7\xEF\xF2\x6A\xDF" + "\xE8\xDA\x75\x1A\x8A\xF1\xDD\x38" + "\x1F\xF9\x3D\x68\x4A\xBB\x9E\x34" + "\x1F\x66\x1F\x9C\x2B\x54\xFF\x60" + "\x7F\x29\x4B\x55\x80\x8F\x4E\xA7" + "\xA6\x9A\x0A\xD9\x0D\x19\x00\xF8" + "\x1F\xBC\x0C\x40\x6B\xEC\x99\x25" + "\x94\x70\x74\x0E\x1D\xC5\xBC\x12" + "\xF3\x42\xBE\x95\xBF\xFB\x4E\x55" + "\x9A\xB9\xCE\x14\x16\x5B\xDC\xD3" + "\x75\x42\x62\x04\x31\x1F\x95\x7C" + "\x66\x1A\x97\xDC\x2F\x40\x5C\x39" + "\x78\xE6\x02\xDB\x49\xE1\xC6\x47" + "\xC2\x78\x9A\xBB\xF3\xBE\xCB\x93" + "\xD8\xB8\xE8\xBB\x8C\xB3\x9B\xA7" + "\xC2\x89\xF3\x91\x88\x83\x3D\xF0" + "\x29\xA2\xCD\xB5\x79\x16\xC2\x40" + "\x11\x03\x8E\x9C\xFD\xC9\x43\xC4" + "\xC2\x19\xF0\x4A\x32\xEF\x0C\x2B" + "\xD3\x2B\xE9\xD4\x4C\xDE\x95\xCF" + "\x04\x03\xD3\x2C\x7F\x82\xC8\xFA" + "\x0F\xD8\x7A\x39\x7B\x01\x41\x9C" + "\x78\xB6\xC9\xBF\xF9\x78\x57\x88" + "\xB1\xA5\xE1\xE0\xD9\x16\xD4\xC8" + "\xEE\xC4\xBE\x7B\x55\x59\x00\x48" + "\x1B\xBC\x14\xFA\x2A\x9D\xC9\x1C" + "\xFB\x28\x3F\x95\xDD\xB7\xD6\xCE" + "\x3A\x7F\x09\x0C\x0E\x69\x30\x7D" + "\xBC\x68\x9C\x91\x2A\x59\x57\x04" + "\xED\x1A\x1E\x00\xB1\x85\x92\x04" + "\x28\x8C\x0C\x3C\xC1\xD5\x12\xF7" + "\x4C\x3E\xB0\xE7\x86\x62\x68\x91" + "\xFC\xC4\xE2\xCE\xA6\xDC\x5E\x93" + "\x5D\x8D\x8C\x68\xB3\xB2\xB9\x64" + "\x16\xB8\xC8\x6F\xD8\xEE\x21\xBD" + "\xAC\x18\x0C\x7D\x0D\x05\xAB\xF1" + "\xFA\xDD\xE2\x48\xDF\x4C\x02\x39" + "\x69\xA1\x62\xBD\x49\x3A\x9D\x91" + "\x30\x70\x56\xA4\x37\xDD\x7C\xC0" + "\x0A\xA3\x30\x10\x26\x25\x41\x2C", + .rlen = 496, }, }; @@ -2966,8 +3290,62 @@ static struct cipher_testvec tf_cbc_dec_tv_template[] = { "\x3D\x32\xDF\xDA\x56\x00\x6E\xEE" "\x5B\x2A\x72\x9D\xC2\x4D\x19\xBC" "\x8C\x53\xFA\x87\x6F\xDD\x81\xA3" - "\xB1\xD3\x44\x65\xDF\xE7\x63\x38", - .ilen = 64, + "\xB1\xD3\x44\x65\xDF\xE7\x63\x38" + "\x4A\xFC\xDC\xEC\x3F\x26\x8E\xB8" + "\x43\xFC\xFE\x18\xB5\x11\x6D\x31" + "\x81\x8B\x0D\x75\xF6\x80\xEC\x84" + "\x04\xB9\xE6\x09\x63\xED\x39\xDB" + "\xC3\xF6\x14\xD6\x6E\x5E\x8B\xBD" + "\x3E\xFA\xD7\x98\x50\x6F\xD9\x63" + "\x02\xCD\x0D\x39\x4B\x0D\xEC\x80" + "\xE3\x6A\x17\xF4\xCC\xAD\xFF\x68" + "\x45\xDD\xC8\x83\x1D\x41\x96\x0D" + "\x91\x2E\x05\xD3\x59\x82\xE0\x43" + "\x90\x4F\xB9\xF7\xAD\x6B\x2E\xAF" + "\xA7\x84\x00\x53\xCD\x6F\xD1\x0C" + "\x4E\xF9\x5A\x23\xFB\xCA\xC7\xD3" + "\xA9\xAA\x9D\xB2\x3F\x66\xF1\xAC" + "\x25\x21\x8F\xF7\xEF\xF2\x6A\xDF" + "\xE8\xDA\x75\x1A\x8A\xF1\xDD\x38" + "\x1F\xF9\x3D\x68\x4A\xBB\x9E\x34" + "\x1F\x66\x1F\x9C\x2B\x54\xFF\x60" + "\x7F\x29\x4B\x55\x80\x8F\x4E\xA7" + "\xA6\x9A\x0A\xD9\x0D\x19\x00\xF8" + "\x1F\xBC\x0C\x40\x6B\xEC\x99\x25" + "\x94\x70\x74\x0E\x1D\xC5\xBC\x12" + "\xF3\x42\xBE\x95\xBF\xFB\x4E\x55" + "\x9A\xB9\xCE\x14\x16\x5B\xDC\xD3" + "\x75\x42\x62\x04\x31\x1F\x95\x7C" + "\x66\x1A\x97\xDC\x2F\x40\x5C\x39" + "\x78\xE6\x02\xDB\x49\xE1\xC6\x47" + "\xC2\x78\x9A\xBB\xF3\xBE\xCB\x93" + "\xD8\xB8\xE8\xBB\x8C\xB3\x9B\xA7" + "\xC2\x89\xF3\x91\x88\x83\x3D\xF0" + "\x29\xA2\xCD\xB5\x79\x16\xC2\x40" + "\x11\x03\x8E\x9C\xFD\xC9\x43\xC4" + "\xC2\x19\xF0\x4A\x32\xEF\x0C\x2B" + "\xD3\x2B\xE9\xD4\x4C\xDE\x95\xCF" + "\x04\x03\xD3\x2C\x7F\x82\xC8\xFA" + "\x0F\xD8\x7A\x39\x7B\x01\x41\x9C" + "\x78\xB6\xC9\xBF\xF9\x78\x57\x88" + "\xB1\xA5\xE1\xE0\xD9\x16\xD4\xC8" + "\xEE\xC4\xBE\x7B\x55\x59\x00\x48" + "\x1B\xBC\x14\xFA\x2A\x9D\xC9\x1C" + "\xFB\x28\x3F\x95\xDD\xB7\xD6\xCE" + "\x3A\x7F\x09\x0C\x0E\x69\x30\x7D" + "\xBC\x68\x9C\x91\x2A\x59\x57\x04" + "\xED\x1A\x1E\x00\xB1\x85\x92\x04" + "\x28\x8C\x0C\x3C\xC1\xD5\x12\xF7" + "\x4C\x3E\xB0\xE7\x86\x62\x68\x91" + "\xFC\xC4\xE2\xCE\xA6\xDC\x5E\x93" + "\x5D\x8D\x8C\x68\xB3\xB2\xB9\x64" + "\x16\xB8\xC8\x6F\xD8\xEE\x21\xBD" + "\xAC\x18\x0C\x7D\x0D\x05\xAB\xF1" + "\xFA\xDD\xE2\x48\xDF\x4C\x02\x39" + "\x69\xA1\x62\xBD\x49\x3A\x9D\x91" + "\x30\x70\x56\xA4\x37\xDD\x7C\xC0" + "\x0A\xA3\x30\x10\x26\x25\x41\x2C", + .ilen = 496, .result = "\x56\xED\x84\x1B\x8F\x26\xBD\x31" "\xC8\x5F\xF6\x6A\x01\x98\x0C\xA3" "\x3A\xD1\x45\xDC\x73\x0A\x7E\x15" @@ -2975,8 +3353,62 @@ static struct cipher_testvec tf_cbc_dec_tv_template[] = { "\x1E\x92\x29\xC0\x34\xCB\x62\xF9" "\x6D\x04\x9B\x0F\xA6\x3D\xD4\x48" "\xDF\x76\x0D\x81\x18\xAF\x23\xBA" - "\x51\xE8\x5C\xF3\x8A\x21\x95\x2C", - .rlen = 64, + "\x51\xE8\x5C\xF3\x8A\x21\x95\x2C" + "\xC3\x37\xCE\x65\xFC\x70\x07\x9E" + "\x12\xA9\x40\xD7\x4B\xE2\x79\x10" + "\x84\x1B\xB2\x26\xBD\x54\xEB\x5F" + "\xF6\x8D\x01\x98\x2F\xC6\x3A\xD1" + "\x68\xFF\x73\x0A\xA1\x15\xAC\x43" + "\xDA\x4E\xE5\x7C\x13\x87\x1E\xB5" + "\x29\xC0\x57\xEE\x62\xF9\x90\x04" + "\x9B\x32\xC9\x3D\xD4\x6B\x02\x76" + "\x0D\xA4\x18\xAF\x46\xDD\x51\xE8" + "\x7F\x16\x8A\x21\xB8\x2C\xC3\x5A" + "\xF1\x65\xFC\x93\x07\x9E\x35\xCC" + "\x40\xD7\x6E\x05\x79\x10\xA7\x1B" + "\xB2\x49\xE0\x54\xEB\x82\x19\x8D" + "\x24\xBB\x2F\xC6\x5D\xF4\x68\xFF" + "\x96\x0A\xA1\x38\xCF\x43\xDA\x71" + "\x08\x7C\x13\xAA\x1E\xB5\x4C\xE3" + "\x57\xEE\x85\x1C\x90\x27\xBE\x32" + "\xC9\x60\xF7\x6B\x02\x99\x0D\xA4" + "\x3B\xD2\x46\xDD\x74\x0B\x7F\x16" + "\xAD\x21\xB8\x4F\xE6\x5A\xF1\x88" + "\x1F\x93\x2A\xC1\x35\xCC\x63\xFA" + "\x6E\x05\x9C\x10\xA7\x3E\xD5\x49" + "\xE0\x77\x0E\x82\x19\xB0\x24\xBB" + "\x52\xE9\x5D\xF4\x8B\x22\x96\x2D" + "\xC4\x38\xCF\x66\xFD\x71\x08\x9F" + "\x13\xAA\x41\xD8\x4C\xE3\x7A\x11" + "\x85\x1C\xB3\x27\xBE\x55\xEC\x60" + "\xF7\x8E\x02\x99\x30\xC7\x3B\xD2" + "\x69\x00\x74\x0B\xA2\x16\xAD\x44" + "\xDB\x4F\xE6\x7D\x14\x88\x1F\xB6" + "\x2A\xC1\x58\xEF\x63\xFA\x91\x05" + "\x9C\x33\xCA\x3E\xD5\x6C\x03\x77" + "\x0E\xA5\x19\xB0\x47\xDE\x52\xE9" + "\x80\x17\x8B\x22\xB9\x2D\xC4\x5B" + "\xF2\x66\xFD\x94\x08\x9F\x36\xCD" + "\x41\xD8\x6F\x06\x7A\x11\xA8\x1C" + "\xB3\x4A\xE1\x55\xEC\x83\x1A\x8E" + "\x25\xBC\x30\xC7\x5E\xF5\x69\x00" + "\x97\x0B\xA2\x39\xD0\x44\xDB\x72" + "\x09\x7D\x14\xAB\x1F\xB6\x4D\xE4" + "\x58\xEF\x86\x1D\x91\x28\xBF\x33" + "\xCA\x61\xF8\x6C\x03\x9A\x0E\xA5" + "\x3C\xD3\x47\xDE\x75\x0C\x80\x17" + "\xAE\x22\xB9\x50\xE7\x5B\xF2\x89" + "\x20\x94\x2B\xC2\x36\xCD\x64\xFB" + "\x6F\x06\x9D\x11\xA8\x3F\xD6\x4A" + "\xE1\x78\x0F\x83\x1A\xB1\x25\xBC" + "\x53\xEA\x5E\xF5\x8C\x00\x97\x2E" + "\xC5\x39\xD0\x67\xFE\x72\x09\xA0" + "\x14\xAB\x42\xD9\x4D\xE4\x7B\x12" + "\x86\x1D\xB4\x28\xBF\x56\xED\x61" + "\xF8\x8F\x03\x9A\x31\xC8\x3C\xD3" + "\x6A\x01\x75\x0C\xA3\x17\xAE\x45" + "\xDC\x50\xE7\x7E\x15\x89\x20\xB7", + .rlen = 496, }, }; @@ -2996,8 +3428,62 @@ static struct cipher_testvec tf_ctr_enc_tv_template[] = { "\x1E\x92\x29\xC0\x34\xCB\x62\xF9" "\x6D\x04\x9B\x0F\xA6\x3D\xD4\x48" "\xDF\x76\x0D\x81\x18\xAF\x23\xBA" - "\x51\xE8\x5C\xF3\x8A\x21\x95\x2C", - .ilen = 64, + "\x51\xE8\x5C\xF3\x8A\x21\x95\x2C" + "\xC3\x37\xCE\x65\xFC\x70\x07\x9E" + "\x12\xA9\x40\xD7\x4B\xE2\x79\x10" + "\x84\x1B\xB2\x26\xBD\x54\xEB\x5F" + "\xF6\x8D\x01\x98\x2F\xC6\x3A\xD1" + "\x68\xFF\x73\x0A\xA1\x15\xAC\x43" + "\xDA\x4E\xE5\x7C\x13\x87\x1E\xB5" + "\x29\xC0\x57\xEE\x62\xF9\x90\x04" + "\x9B\x32\xC9\x3D\xD4\x6B\x02\x76" + "\x0D\xA4\x18\xAF\x46\xDD\x51\xE8" + "\x7F\x16\x8A\x21\xB8\x2C\xC3\x5A" + "\xF1\x65\xFC\x93\x07\x9E\x35\xCC" + "\x40\xD7\x6E\x05\x79\x10\xA7\x1B" + "\xB2\x49\xE0\x54\xEB\x82\x19\x8D" + "\x24\xBB\x2F\xC6\x5D\xF4\x68\xFF" + "\x96\x0A\xA1\x38\xCF\x43\xDA\x71" + "\x08\x7C\x13\xAA\x1E\xB5\x4C\xE3" + "\x57\xEE\x85\x1C\x90\x27\xBE\x32" + "\xC9\x60\xF7\x6B\x02\x99\x0D\xA4" + "\x3B\xD2\x46\xDD\x74\x0B\x7F\x16" + "\xAD\x21\xB8\x4F\xE6\x5A\xF1\x88" + "\x1F\x93\x2A\xC1\x35\xCC\x63\xFA" + "\x6E\x05\x9C\x10\xA7\x3E\xD5\x49" + "\xE0\x77\x0E\x82\x19\xB0\x24\xBB" + "\x52\xE9\x5D\xF4\x8B\x22\x96\x2D" + "\xC4\x38\xCF\x66\xFD\x71\x08\x9F" + "\x13\xAA\x41\xD8\x4C\xE3\x7A\x11" + "\x85\x1C\xB3\x27\xBE\x55\xEC\x60" + "\xF7\x8E\x02\x99\x30\xC7\x3B\xD2" + "\x69\x00\x74\x0B\xA2\x16\xAD\x44" + "\xDB\x4F\xE6\x7D\x14\x88\x1F\xB6" + "\x2A\xC1\x58\xEF\x63\xFA\x91\x05" + "\x9C\x33\xCA\x3E\xD5\x6C\x03\x77" + "\x0E\xA5\x19\xB0\x47\xDE\x52\xE9" + "\x80\x17\x8B\x22\xB9\x2D\xC4\x5B" + "\xF2\x66\xFD\x94\x08\x9F\x36\xCD" + "\x41\xD8\x6F\x06\x7A\x11\xA8\x1C" + "\xB3\x4A\xE1\x55\xEC\x83\x1A\x8E" + "\x25\xBC\x30\xC7\x5E\xF5\x69\x00" + "\x97\x0B\xA2\x39\xD0\x44\xDB\x72" + "\x09\x7D\x14\xAB\x1F\xB6\x4D\xE4" + "\x58\xEF\x86\x1D\x91\x28\xBF\x33" + "\xCA\x61\xF8\x6C\x03\x9A\x0E\xA5" + "\x3C\xD3\x47\xDE\x75\x0C\x80\x17" + "\xAE\x22\xB9\x50\xE7\x5B\xF2\x89" + "\x20\x94\x2B\xC2\x36\xCD\x64\xFB" + "\x6F\x06\x9D\x11\xA8\x3F\xD6\x4A" + "\xE1\x78\x0F\x83\x1A\xB1\x25\xBC" + "\x53\xEA\x5E\xF5\x8C\x00\x97\x2E" + "\xC5\x39\xD0\x67\xFE\x72\x09\xA0" + "\x14\xAB\x42\xD9\x4D\xE4\x7B\x12" + "\x86\x1D\xB4\x28\xBF\x56\xED\x61" + "\xF8\x8F\x03\x9A\x31\xC8\x3C\xD3" + "\x6A\x01\x75\x0C\xA3\x17\xAE\x45" + "\xDC\x50\xE7\x7E\x15\x89\x20\xB7", + .ilen = 496, .result = "\xDF\xDD\x69\xFA\xB0\x2E\xFD\xFE" "\x70\x9E\xC5\x4B\xC9\xD4\xA1\x30" "\x26\x9B\x89\xA1\xEE\x43\xE0\x52" @@ -3005,8 +3491,62 @@ static struct cipher_testvec tf_ctr_enc_tv_template[] = { "\x9F\x8D\x40\x9F\x24\xFD\x92\xA0" "\xBC\x8F\x35\xDD\x67\x38\xD8\xAA" "\xCF\xF8\x48\xCA\xFB\xE4\x5C\x60" - "\x01\x41\x21\x12\x38\xAB\x52\x4F", - .rlen = 64, + "\x01\x41\x21\x12\x38\xAB\x52\x4F" + "\xA8\x57\x20\xE0\x21\x6A\x17\x0D" + "\x0E\xF9\x8E\x49\x42\x00\x3C\x94" + "\x14\xC0\xD0\x8D\x8A\x98\xEB\x29" + "\xEC\xAE\x96\x44\xC0\x3C\x48\xDC" + "\x29\x35\x25\x2F\xE7\x11\x6C\x68" + "\xC8\x67\x0A\x2F\xF4\x07\xBE\xF9" + "\x2C\x31\x87\x40\xAB\xB2\xB6\xFA" + "\xD2\xC9\x6D\x5C\x50\xE9\xE6\x7E" + "\xE3\x0A\xD2\xD5\x6D\x8D\x64\x9E" + "\x70\xCE\x03\x76\xDD\xE0\xF0\x8C" + "\x84\x86\x8B\x6A\xFE\xC7\xF9\x69" + "\x2E\xFE\xFC\xC2\xC4\x1A\x55\x58" + "\xB3\xBE\xE2\x7E\xED\x39\x42\x6C" + "\xB4\x42\x97\x9A\xEC\xE1\x0A\x06" + "\x02\xC5\x03\x9D\xC4\x48\x15\x66" + "\x35\x6A\xC2\xC9\xA2\x26\x30\xBB" + "\xDB\x2D\xC8\x08\x2B\xA0\x29\x1A" + "\x23\x61\x48\xEA\x80\x04\x27\xAA" + "\x69\x49\xE8\xE8\x4A\x83\x6B\x5A" + "\xCA\x7C\xD3\xB1\xB5\x0B\xCC\x23" + "\x74\x1F\xA9\x87\xCD\xED\xC0\x2D" + "\xBF\xEB\xCF\x16\x2D\x2A\x2E\x1D" + "\x96\xBA\x36\x11\x45\x41\xDA\xCE" + "\xA4\x48\x80\x8B\x06\xF4\x98\x89" + "\x8B\x23\x08\x53\xF4\xD4\x5A\x24" + "\x8B\xF8\x43\x73\xD1\xEE\xC4\xB0" + "\xF8\xFE\x09\x0C\x75\x05\x38\x0B" + "\x7C\x81\xDE\x9D\xE4\x61\x37\x63" + "\x63\xAD\x12\xD2\x04\xB9\xCE\x45" + "\x5A\x1A\x6E\xB3\x78\x2A\xA4\x74" + "\x86\xD0\xE3\xFF\xDA\x38\x9C\xB5" + "\xB8\xB1\xDB\x38\x2F\xC5\x6A\xB4" + "\xEB\x6E\x96\xE8\x43\x80\xB5\x51" + "\x61\x2D\x48\xAA\x07\x65\x11\x8C" + "\x48\xE3\x90\x7E\x78\x3A\xEC\x97" + "\x05\x3D\x84\xE7\x90\x2B\xAA\xBD" + "\x83\x29\x0E\x1A\x81\x73\x7B\xE0" + "\x7A\x01\x4A\x37\x3B\x77\x7F\x8D" + "\x49\xA4\x2F\x6E\xBE\x68\x99\x08" + "\x99\xAA\x4C\x12\x04\xAE\x1F\x77" + "\x35\x88\xF1\x65\x06\x0A\x0B\x4D" + "\x47\xF9\x50\x38\x5D\x71\xF9\x6E" + "\xDE\xEC\x61\x35\x2C\x4C\x96\x50" + "\xE8\x28\x93\x9C\x7E\x01\xC6\x04" + "\xB2\xD6\xBC\x6C\x17\xEB\xC1\x7D" + "\x11\xE9\x43\x83\x76\xAA\x53\x37" + "\x0C\x1D\x39\x89\x53\x72\x09\x7E" + "\xD9\x85\x16\x04\xA5\x2C\x05\x6F" + "\x17\x0C\x6E\x66\xAA\x84\xA7\xD9" + "\xE2\xD9\xC4\xEB\x43\x3E\xB1\x8D" + "\x7C\x36\xC7\x71\x70\x9C\x10\xD8" + "\xE8\x47\x2A\x4D\xFD\xA1\xBC\xE3" + "\xB9\x32\xE2\xC1\x82\xAC\xFE\xCC" + "\xC5\xC9\x7F\x9E\xCF\x33\x7A\xDF", + .rlen = 496, }, { /* Generated with Crypto++ */ .key = "\x85\x62\x3F\x1C\xF9\xD6\x1C\xF9" "\xD6\xB3\x90\x6D\x4A\x90\x6D\x4A" @@ -3023,8 +3563,62 @@ static struct cipher_testvec tf_ctr_enc_tv_template[] = { "\x6D\x04\x9B\x0F\xA6\x3D\xD4\x48" "\xDF\x76\x0D\x81\x18\xAF\x23\xBA" "\x51\xE8\x5C\xF3\x8A\x21\x95\x2C" - "\xC3\x37\xCE", - .ilen = 67, + "\xC3\x37\xCE\x65\xFC\x70\x07\x9E" + "\x12\xA9\x40\xD7\x4B\xE2\x79\x10" + "\x84\x1B\xB2\x26\xBD\x54\xEB\x5F" + "\xF6\x8D\x01\x98\x2F\xC6\x3A\xD1" + "\x68\xFF\x73\x0A\xA1\x15\xAC\x43" + "\xDA\x4E\xE5\x7C\x13\x87\x1E\xB5" + "\x29\xC0\x57\xEE\x62\xF9\x90\x04" + "\x9B\x32\xC9\x3D\xD4\x6B\x02\x76" + "\x0D\xA4\x18\xAF\x46\xDD\x51\xE8" + "\x7F\x16\x8A\x21\xB8\x2C\xC3\x5A" + "\xF1\x65\xFC\x93\x07\x9E\x35\xCC" + "\x40\xD7\x6E\x05\x79\x10\xA7\x1B" + "\xB2\x49\xE0\x54\xEB\x82\x19\x8D" + "\x24\xBB\x2F\xC6\x5D\xF4\x68\xFF" + "\x96\x0A\xA1\x38\xCF\x43\xDA\x71" + "\x08\x7C\x13\xAA\x1E\xB5\x4C\xE3" + "\x57\xEE\x85\x1C\x90\x27\xBE\x32" + "\xC9\x60\xF7\x6B\x02\x99\x0D\xA4" + "\x3B\xD2\x46\xDD\x74\x0B\x7F\x16" + "\xAD\x21\xB8\x4F\xE6\x5A\xF1\x88" + "\x1F\x93\x2A\xC1\x35\xCC\x63\xFA" + "\x6E\x05\x9C\x10\xA7\x3E\xD5\x49" + "\xE0\x77\x0E\x82\x19\xB0\x24\xBB" + "\x52\xE9\x5D\xF4\x8B\x22\x96\x2D" + "\xC4\x38\xCF\x66\xFD\x71\x08\x9F" + "\x13\xAA\x41\xD8\x4C\xE3\x7A\x11" + "\x85\x1C\xB3\x27\xBE\x55\xEC\x60" + "\xF7\x8E\x02\x99\x30\xC7\x3B\xD2" + "\x69\x00\x74\x0B\xA2\x16\xAD\x44" + "\xDB\x4F\xE6\x7D\x14\x88\x1F\xB6" + "\x2A\xC1\x58\xEF\x63\xFA\x91\x05" + "\x9C\x33\xCA\x3E\xD5\x6C\x03\x77" + "\x0E\xA5\x19\xB0\x47\xDE\x52\xE9" + "\x80\x17\x8B\x22\xB9\x2D\xC4\x5B" + "\xF2\x66\xFD\x94\x08\x9F\x36\xCD" + "\x41\xD8\x6F\x06\x7A\x11\xA8\x1C" + "\xB3\x4A\xE1\x55\xEC\x83\x1A\x8E" + "\x25\xBC\x30\xC7\x5E\xF5\x69\x00" + "\x97\x0B\xA2\x39\xD0\x44\xDB\x72" + "\x09\x7D\x14\xAB\x1F\xB6\x4D\xE4" + "\x58\xEF\x86\x1D\x91\x28\xBF\x33" + "\xCA\x61\xF8\x6C\x03\x9A\x0E\xA5" + "\x3C\xD3\x47\xDE\x75\x0C\x80\x17" + "\xAE\x22\xB9\x50\xE7\x5B\xF2\x89" + "\x20\x94\x2B\xC2\x36\xCD\x64\xFB" + "\x6F\x06\x9D\x11\xA8\x3F\xD6\x4A" + "\xE1\x78\x0F\x83\x1A\xB1\x25\xBC" + "\x53\xEA\x5E\xF5\x8C\x00\x97\x2E" + "\xC5\x39\xD0\x67\xFE\x72\x09\xA0" + "\x14\xAB\x42\xD9\x4D\xE4\x7B\x12" + "\x86\x1D\xB4\x28\xBF\x56\xED\x61" + "\xF8\x8F\x03\x9A\x31\xC8\x3C\xD3" + "\x6A\x01\x75\x0C\xA3\x17\xAE\x45" + "\xDC\x50\xE7\x7E\x15\x89\x20\xB7" + "\x2B\xC2\x59", + .ilen = 499, .result = "\xDF\xDD\x69\xFA\xB0\x2E\xFD\xFE" "\x70\x9E\xC5\x4B\xC9\xD4\xA1\x30" "\x26\x9B\x89\xA1\xEE\x43\xE0\x52" @@ -3033,8 +3627,62 @@ static struct cipher_testvec tf_ctr_enc_tv_template[] = { "\xBC\x8F\x35\xDD\x67\x38\xD8\xAA" "\xCF\xF8\x48\xCA\xFB\xE4\x5C\x60" "\x01\x41\x21\x12\x38\xAB\x52\x4F" - "\xA8\x57\x20", - .rlen = 67, + "\xA8\x57\x20\xE0\x21\x6A\x17\x0D" + "\x0E\xF9\x8E\x49\x42\x00\x3C\x94" + "\x14\xC0\xD0\x8D\x8A\x98\xEB\x29" + "\xEC\xAE\x96\x44\xC0\x3C\x48\xDC" + "\x29\x35\x25\x2F\xE7\x11\x6C\x68" + "\xC8\x67\x0A\x2F\xF4\x07\xBE\xF9" + "\x2C\x31\x87\x40\xAB\xB2\xB6\xFA" + "\xD2\xC9\x6D\x5C\x50\xE9\xE6\x7E" + "\xE3\x0A\xD2\xD5\x6D\x8D\x64\x9E" + "\x70\xCE\x03\x76\xDD\xE0\xF0\x8C" + "\x84\x86\x8B\x6A\xFE\xC7\xF9\x69" + "\x2E\xFE\xFC\xC2\xC4\x1A\x55\x58" + "\xB3\xBE\xE2\x7E\xED\x39\x42\x6C" + "\xB4\x42\x97\x9A\xEC\xE1\x0A\x06" + "\x02\xC5\x03\x9D\xC4\x48\x15\x66" + "\x35\x6A\xC2\xC9\xA2\x26\x30\xBB" + "\xDB\x2D\xC8\x08\x2B\xA0\x29\x1A" + "\x23\x61\x48\xEA\x80\x04\x27\xAA" + "\x69\x49\xE8\xE8\x4A\x83\x6B\x5A" + "\xCA\x7C\xD3\xB1\xB5\x0B\xCC\x23" + "\x74\x1F\xA9\x87\xCD\xED\xC0\x2D" + "\xBF\xEB\xCF\x16\x2D\x2A\x2E\x1D" + "\x96\xBA\x36\x11\x45\x41\xDA\xCE" + "\xA4\x48\x80\x8B\x06\xF4\x98\x89" + "\x8B\x23\x08\x53\xF4\xD4\x5A\x24" + "\x8B\xF8\x43\x73\xD1\xEE\xC4\xB0" + "\xF8\xFE\x09\x0C\x75\x05\x38\x0B" + "\x7C\x81\xDE\x9D\xE4\x61\x37\x63" + "\x63\xAD\x12\xD2\x04\xB9\xCE\x45" + "\x5A\x1A\x6E\xB3\x78\x2A\xA4\x74" + "\x86\xD0\xE3\xFF\xDA\x38\x9C\xB5" + "\xB8\xB1\xDB\x38\x2F\xC5\x6A\xB4" + "\xEB\x6E\x96\xE8\x43\x80\xB5\x51" + "\x61\x2D\x48\xAA\x07\x65\x11\x8C" + "\x48\xE3\x90\x7E\x78\x3A\xEC\x97" + "\x05\x3D\x84\xE7\x90\x2B\xAA\xBD" + "\x83\x29\x0E\x1A\x81\x73\x7B\xE0" + "\x7A\x01\x4A\x37\x3B\x77\x7F\x8D" + "\x49\xA4\x2F\x6E\xBE\x68\x99\x08" + "\x99\xAA\x4C\x12\x04\xAE\x1F\x77" + "\x35\x88\xF1\x65\x06\x0A\x0B\x4D" + "\x47\xF9\x50\x38\x5D\x71\xF9\x6E" + "\xDE\xEC\x61\x35\x2C\x4C\x96\x50" + "\xE8\x28\x93\x9C\x7E\x01\xC6\x04" + "\xB2\xD6\xBC\x6C\x17\xEB\xC1\x7D" + "\x11\xE9\x43\x83\x76\xAA\x53\x37" + "\x0C\x1D\x39\x89\x53\x72\x09\x7E" + "\xD9\x85\x16\x04\xA5\x2C\x05\x6F" + "\x17\x0C\x6E\x66\xAA\x84\xA7\xD9" + "\xE2\xD9\xC4\xEB\x43\x3E\xB1\x8D" + "\x7C\x36\xC7\x71\x70\x9C\x10\xD8" + "\xE8\x47\x2A\x4D\xFD\xA1\xBC\xE3" + "\xB9\x32\xE2\xC1\x82\xAC\xFE\xCC" + "\xC5\xC9\x7F\x9E\xCF\x33\x7A\xDF" + "\x6C\x82\x9D", + .rlen = 499, }, }; @@ -3054,8 +3702,62 @@ static struct cipher_testvec tf_ctr_dec_tv_template[] = { "\x9F\x8D\x40\x9F\x24\xFD\x92\xA0" "\xBC\x8F\x35\xDD\x67\x38\xD8\xAA" "\xCF\xF8\x48\xCA\xFB\xE4\x5C\x60" - "\x01\x41\x21\x12\x38\xAB\x52\x4F", - .ilen = 64, + "\x01\x41\x21\x12\x38\xAB\x52\x4F" + "\xA8\x57\x20\xE0\x21\x6A\x17\x0D" + "\x0E\xF9\x8E\x49\x42\x00\x3C\x94" + "\x14\xC0\xD0\x8D\x8A\x98\xEB\x29" + "\xEC\xAE\x96\x44\xC0\x3C\x48\xDC" + "\x29\x35\x25\x2F\xE7\x11\x6C\x68" + "\xC8\x67\x0A\x2F\xF4\x07\xBE\xF9" + "\x2C\x31\x87\x40\xAB\xB2\xB6\xFA" + "\xD2\xC9\x6D\x5C\x50\xE9\xE6\x7E" + "\xE3\x0A\xD2\xD5\x6D\x8D\x64\x9E" + "\x70\xCE\x03\x76\xDD\xE0\xF0\x8C" + "\x84\x86\x8B\x6A\xFE\xC7\xF9\x69" + "\x2E\xFE\xFC\xC2\xC4\x1A\x55\x58" + "\xB3\xBE\xE2\x7E\xED\x39\x42\x6C" + "\xB4\x42\x97\x9A\xEC\xE1\x0A\x06" + "\x02\xC5\x03\x9D\xC4\x48\x15\x66" + "\x35\x6A\xC2\xC9\xA2\x26\x30\xBB" + "\xDB\x2D\xC8\x08\x2B\xA0\x29\x1A" + "\x23\x61\x48\xEA\x80\x04\x27\xAA" + "\x69\x49\xE8\xE8\x4A\x83\x6B\x5A" + "\xCA\x7C\xD3\xB1\xB5\x0B\xCC\x23" + "\x74\x1F\xA9\x87\xCD\xED\xC0\x2D" + "\xBF\xEB\xCF\x16\x2D\x2A\x2E\x1D" + "\x96\xBA\x36\x11\x45\x41\xDA\xCE" + "\xA4\x48\x80\x8B\x06\xF4\x98\x89" + "\x8B\x23\x08\x53\xF4\xD4\x5A\x24" + "\x8B\xF8\x43\x73\xD1\xEE\xC4\xB0" + "\xF8\xFE\x09\x0C\x75\x05\x38\x0B" + "\x7C\x81\xDE\x9D\xE4\x61\x37\x63" + "\x63\xAD\x12\xD2\x04\xB9\xCE\x45" + "\x5A\x1A\x6E\xB3\x78\x2A\xA4\x74" + "\x86\xD0\xE3\xFF\xDA\x38\x9C\xB5" + "\xB8\xB1\xDB\x38\x2F\xC5\x6A\xB4" + "\xEB\x6E\x96\xE8\x43\x80\xB5\x51" + "\x61\x2D\x48\xAA\x07\x65\x11\x8C" + "\x48\xE3\x90\x7E\x78\x3A\xEC\x97" + "\x05\x3D\x84\xE7\x90\x2B\xAA\xBD" + "\x83\x29\x0E\x1A\x81\x73\x7B\xE0" + "\x7A\x01\x4A\x37\x3B\x77\x7F\x8D" + "\x49\xA4\x2F\x6E\xBE\x68\x99\x08" + "\x99\xAA\x4C\x12\x04\xAE\x1F\x77" + "\x35\x88\xF1\x65\x06\x0A\x0B\x4D" + "\x47\xF9\x50\x38\x5D\x71\xF9\x6E" + "\xDE\xEC\x61\x35\x2C\x4C\x96\x50" + "\xE8\x28\x93\x9C\x7E\x01\xC6\x04" + "\xB2\xD6\xBC\x6C\x17\xEB\xC1\x7D" + "\x11\xE9\x43\x83\x76\xAA\x53\x37" + "\x0C\x1D\x39\x89\x53\x72\x09\x7E" + "\xD9\x85\x16\x04\xA5\x2C\x05\x6F" + "\x17\x0C\x6E\x66\xAA\x84\xA7\xD9" + "\xE2\xD9\xC4\xEB\x43\x3E\xB1\x8D" + "\x7C\x36\xC7\x71\x70\x9C\x10\xD8" + "\xE8\x47\x2A\x4D\xFD\xA1\xBC\xE3" + "\xB9\x32\xE2\xC1\x82\xAC\xFE\xCC" + "\xC5\xC9\x7F\x9E\xCF\x33\x7A\xDF", + .ilen = 496, .result = "\x56\xED\x84\x1B\x8F\x26\xBD\x31" "\xC8\x5F\xF6\x6A\x01\x98\x0C\xA3" "\x3A\xD1\x45\xDC\x73\x0A\x7E\x15" @@ -3063,8 +3765,62 @@ static struct cipher_testvec tf_ctr_dec_tv_template[] = { "\x1E\x92\x29\xC0\x34\xCB\x62\xF9" "\x6D\x04\x9B\x0F\xA6\x3D\xD4\x48" "\xDF\x76\x0D\x81\x18\xAF\x23\xBA" - "\x51\xE8\x5C\xF3\x8A\x21\x95\x2C", - .rlen = 64, + "\x51\xE8\x5C\xF3\x8A\x21\x95\x2C" + "\xC3\x37\xCE\x65\xFC\x70\x07\x9E" + "\x12\xA9\x40\xD7\x4B\xE2\x79\x10" + "\x84\x1B\xB2\x26\xBD\x54\xEB\x5F" + "\xF6\x8D\x01\x98\x2F\xC6\x3A\xD1" + "\x68\xFF\x73\x0A\xA1\x15\xAC\x43" + "\xDA\x4E\xE5\x7C\x13\x87\x1E\xB5" + "\x29\xC0\x57\xEE\x62\xF9\x90\x04" + "\x9B\x32\xC9\x3D\xD4\x6B\x02\x76" + "\x0D\xA4\x18\xAF\x46\xDD\x51\xE8" + "\x7F\x16\x8A\x21\xB8\x2C\xC3\x5A" + "\xF1\x65\xFC\x93\x07\x9E\x35\xCC" + "\x40\xD7\x6E\x05\x79\x10\xA7\x1B" + "\xB2\x49\xE0\x54\xEB\x82\x19\x8D" + "\x24\xBB\x2F\xC6\x5D\xF4\x68\xFF" + "\x96\x0A\xA1\x38\xCF\x43\xDA\x71" + "\x08\x7C\x13\xAA\x1E\xB5\x4C\xE3" + "\x57\xEE\x85\x1C\x90\x27\xBE\x32" + "\xC9\x60\xF7\x6B\x02\x99\x0D\xA4" + "\x3B\xD2\x46\xDD\x74\x0B\x7F\x16" + "\xAD\x21\xB8\x4F\xE6\x5A\xF1\x88" + "\x1F\x93\x2A\xC1\x35\xCC\x63\xFA" + "\x6E\x05\x9C\x10\xA7\x3E\xD5\x49" + "\xE0\x77\x0E\x82\x19\xB0\x24\xBB" + "\x52\xE9\x5D\xF4\x8B\x22\x96\x2D" + "\xC4\x38\xCF\x66\xFD\x71\x08\x9F" + "\x13\xAA\x41\xD8\x4C\xE3\x7A\x11" + "\x85\x1C\xB3\x27\xBE\x55\xEC\x60" + "\xF7\x8E\x02\x99\x30\xC7\x3B\xD2" + "\x69\x00\x74\x0B\xA2\x16\xAD\x44" + "\xDB\x4F\xE6\x7D\x14\x88\x1F\xB6" + "\x2A\xC1\x58\xEF\x63\xFA\x91\x05" + "\x9C\x33\xCA\x3E\xD5\x6C\x03\x77" + "\x0E\xA5\x19\xB0\x47\xDE\x52\xE9" + "\x80\x17\x8B\x22\xB9\x2D\xC4\x5B" + "\xF2\x66\xFD\x94\x08\x9F\x36\xCD" + "\x41\xD8\x6F\x06\x7A\x11\xA8\x1C" + "\xB3\x4A\xE1\x55\xEC\x83\x1A\x8E" + "\x25\xBC\x30\xC7\x5E\xF5\x69\x00" + "\x97\x0B\xA2\x39\xD0\x44\xDB\x72" + "\x09\x7D\x14\xAB\x1F\xB6\x4D\xE4" + "\x58\xEF\x86\x1D\x91\x28\xBF\x33" + "\xCA\x61\xF8\x6C\x03\x9A\x0E\xA5" + "\x3C\xD3\x47\xDE\x75\x0C\x80\x17" + "\xAE\x22\xB9\x50\xE7\x5B\xF2\x89" + "\x20\x94\x2B\xC2\x36\xCD\x64\xFB" + "\x6F\x06\x9D\x11\xA8\x3F\xD6\x4A" + "\xE1\x78\x0F\x83\x1A\xB1\x25\xBC" + "\x53\xEA\x5E\xF5\x8C\x00\x97\x2E" + "\xC5\x39\xD0\x67\xFE\x72\x09\xA0" + "\x14\xAB\x42\xD9\x4D\xE4\x7B\x12" + "\x86\x1D\xB4\x28\xBF\x56\xED\x61" + "\xF8\x8F\x03\x9A\x31\xC8\x3C\xD3" + "\x6A\x01\x75\x0C\xA3\x17\xAE\x45" + "\xDC\x50\xE7\x7E\x15\x89\x20\xB7", + .rlen = 496, }, { /* Generated with Crypto++ */ .key = "\x85\x62\x3F\x1C\xF9\xD6\x1C\xF9" "\xD6\xB3\x90\x6D\x4A\x90\x6D\x4A" @@ -3081,8 +3837,62 @@ static struct cipher_testvec tf_ctr_dec_tv_template[] = { "\xBC\x8F\x35\xDD\x67\x38\xD8\xAA" "\xCF\xF8\x48\xCA\xFB\xE4\x5C\x60" "\x01\x41\x21\x12\x38\xAB\x52\x4F" - "\xA8\x57\x20", - .ilen = 67, + "\xA8\x57\x20\xE0\x21\x6A\x17\x0D" + "\x0E\xF9\x8E\x49\x42\x00\x3C\x94" + "\x14\xC0\xD0\x8D\x8A\x98\xEB\x29" + "\xEC\xAE\x96\x44\xC0\x3C\x48\xDC" + "\x29\x35\x25\x2F\xE7\x11\x6C\x68" + "\xC8\x67\x0A\x2F\xF4\x07\xBE\xF9" + "\x2C\x31\x87\x40\xAB\xB2\xB6\xFA" + "\xD2\xC9\x6D\x5C\x50\xE9\xE6\x7E" + "\xE3\x0A\xD2\xD5\x6D\x8D\x64\x9E" + "\x70\xCE\x03\x76\xDD\xE0\xF0\x8C" + "\x84\x86\x8B\x6A\xFE\xC7\xF9\x69" + "\x2E\xFE\xFC\xC2\xC4\x1A\x55\x58" + "\xB3\xBE\xE2\x7E\xED\x39\x42\x6C" + "\xB4\x42\x97\x9A\xEC\xE1\x0A\x06" + "\x02\xC5\x03\x9D\xC4\x48\x15\x66" + "\x35\x6A\xC2\xC9\xA2\x26\x30\xBB" + "\xDB\x2D\xC8\x08\x2B\xA0\x29\x1A" + "\x23\x61\x48\xEA\x80\x04\x27\xAA" + "\x69\x49\xE8\xE8\x4A\x83\x6B\x5A" + "\xCA\x7C\xD3\xB1\xB5\x0B\xCC\x23" + "\x74\x1F\xA9\x87\xCD\xED\xC0\x2D" + "\xBF\xEB\xCF\x16\x2D\x2A\x2E\x1D" + "\x96\xBA\x36\x11\x45\x41\xDA\xCE" + "\xA4\x48\x80\x8B\x06\xF4\x98\x89" + "\x8B\x23\x08\x53\xF4\xD4\x5A\x24" + "\x8B\xF8\x43\x73\xD1\xEE\xC4\xB0" + "\xF8\xFE\x09\x0C\x75\x05\x38\x0B" + "\x7C\x81\xDE\x9D\xE4\x61\x37\x63" + "\x63\xAD\x12\xD2\x04\xB9\xCE\x45" + "\x5A\x1A\x6E\xB3\x78\x2A\xA4\x74" + "\x86\xD0\xE3\xFF\xDA\x38\x9C\xB5" + "\xB8\xB1\xDB\x38\x2F\xC5\x6A\xB4" + "\xEB\x6E\x96\xE8\x43\x80\xB5\x51" + "\x61\x2D\x48\xAA\x07\x65\x11\x8C" + "\x48\xE3\x90\x7E\x78\x3A\xEC\x97" + "\x05\x3D\x84\xE7\x90\x2B\xAA\xBD" + "\x83\x29\x0E\x1A\x81\x73\x7B\xE0" + "\x7A\x01\x4A\x37\x3B\x77\x7F\x8D" + "\x49\xA4\x2F\x6E\xBE\x68\x99\x08" + "\x99\xAA\x4C\x12\x04\xAE\x1F\x77" + "\x35\x88\xF1\x65\x06\x0A\x0B\x4D" + "\x47\xF9\x50\x38\x5D\x71\xF9\x6E" + "\xDE\xEC\x61\x35\x2C\x4C\x96\x50" + "\xE8\x28\x93\x9C\x7E\x01\xC6\x04" + "\xB2\xD6\xBC\x6C\x17\xEB\xC1\x7D" + "\x11\xE9\x43\x83\x76\xAA\x53\x37" + "\x0C\x1D\x39\x89\x53\x72\x09\x7E" + "\xD9\x85\x16\x04\xA5\x2C\x05\x6F" + "\x17\x0C\x6E\x66\xAA\x84\xA7\xD9" + "\xE2\xD9\xC4\xEB\x43\x3E\xB1\x8D" + "\x7C\x36\xC7\x71\x70\x9C\x10\xD8" + "\xE8\x47\x2A\x4D\xFD\xA1\xBC\xE3" + "\xB9\x32\xE2\xC1\x82\xAC\xFE\xCC" + "\xC5\xC9\x7F\x9E\xCF\x33\x7A\xDF" + "\x6C\x82\x9D", + .ilen = 499, .result = "\x56\xED\x84\x1B\x8F\x26\xBD\x31" "\xC8\x5F\xF6\x6A\x01\x98\x0C\xA3" "\x3A\xD1\x45\xDC\x73\x0A\x7E\x15" @@ -3091,8 +3901,62 @@ static struct cipher_testvec tf_ctr_dec_tv_template[] = { "\x6D\x04\x9B\x0F\xA6\x3D\xD4\x48" "\xDF\x76\x0D\x81\x18\xAF\x23\xBA" "\x51\xE8\x5C\xF3\x8A\x21\x95\x2C" - "\xC3\x37\xCE", - .rlen = 67, + "\xC3\x37\xCE\x65\xFC\x70\x07\x9E" + "\x12\xA9\x40\xD7\x4B\xE2\x79\x10" + "\x84\x1B\xB2\x26\xBD\x54\xEB\x5F" + "\xF6\x8D\x01\x98\x2F\xC6\x3A\xD1" + "\x68\xFF\x73\x0A\xA1\x15\xAC\x43" + "\xDA\x4E\xE5\x7C\x13\x87\x1E\xB5" + "\x29\xC0\x57\xEE\x62\xF9\x90\x04" + "\x9B\x32\xC9\x3D\xD4\x6B\x02\x76" + "\x0D\xA4\x18\xAF\x46\xDD\x51\xE8" + "\x7F\x16\x8A\x21\xB8\x2C\xC3\x5A" + "\xF1\x65\xFC\x93\x07\x9E\x35\xCC" + "\x40\xD7\x6E\x05\x79\x10\xA7\x1B" + "\xB2\x49\xE0\x54\xEB\x82\x19\x8D" + "\x24\xBB\x2F\xC6\x5D\xF4\x68\xFF" + "\x96\x0A\xA1\x38\xCF\x43\xDA\x71" + "\x08\x7C\x13\xAA\x1E\xB5\x4C\xE3" + "\x57\xEE\x85\x1C\x90\x27\xBE\x32" + "\xC9\x60\xF7\x6B\x02\x99\x0D\xA4" + "\x3B\xD2\x46\xDD\x74\x0B\x7F\x16" + "\xAD\x21\xB8\x4F\xE6\x5A\xF1\x88" + "\x1F\x93\x2A\xC1\x35\xCC\x63\xFA" + "\x6E\x05\x9C\x10\xA7\x3E\xD5\x49" + "\xE0\x77\x0E\x82\x19\xB0\x24\xBB" + "\x52\xE9\x5D\xF4\x8B\x22\x96\x2D" + "\xC4\x38\xCF\x66\xFD\x71\x08\x9F" + "\x13\xAA\x41\xD8\x4C\xE3\x7A\x11" + "\x85\x1C\xB3\x27\xBE\x55\xEC\x60" + "\xF7\x8E\x02\x99\x30\xC7\x3B\xD2" + "\x69\x00\x74\x0B\xA2\x16\xAD\x44" + "\xDB\x4F\xE6\x7D\x14\x88\x1F\xB6" + "\x2A\xC1\x58\xEF\x63\xFA\x91\x05" + "\x9C\x33\xCA\x3E\xD5\x6C\x03\x77" + "\x0E\xA5\x19\xB0\x47\xDE\x52\xE9" + "\x80\x17\x8B\x22\xB9\x2D\xC4\x5B" + "\xF2\x66\xFD\x94\x08\x9F\x36\xCD" + "\x41\xD8\x6F\x06\x7A\x11\xA8\x1C" + "\xB3\x4A\xE1\x55\xEC\x83\x1A\x8E" + "\x25\xBC\x30\xC7\x5E\xF5\x69\x00" + "\x97\x0B\xA2\x39\xD0\x44\xDB\x72" + "\x09\x7D\x14\xAB\x1F\xB6\x4D\xE4" + "\x58\xEF\x86\x1D\x91\x28\xBF\x33" + "\xCA\x61\xF8\x6C\x03\x9A\x0E\xA5" + "\x3C\xD3\x47\xDE\x75\x0C\x80\x17" + "\xAE\x22\xB9\x50\xE7\x5B\xF2\x89" + "\x20\x94\x2B\xC2\x36\xCD\x64\xFB" + "\x6F\x06\x9D\x11\xA8\x3F\xD6\x4A" + "\xE1\x78\x0F\x83\x1A\xB1\x25\xBC" + "\x53\xEA\x5E\xF5\x8C\x00\x97\x2E" + "\xC5\x39\xD0\x67\xFE\x72\x09\xA0" + "\x14\xAB\x42\xD9\x4D\xE4\x7B\x12" + "\x86\x1D\xB4\x28\xBF\x56\xED\x61" + "\xF8\x8F\x03\x9A\x31\xC8\x3C\xD3" + "\x6A\x01\x75\x0C\xA3\x17\xAE\x45" + "\xDC\x50\xE7\x7E\x15\x89\x20\xB7" + "\x2B\xC2\x59", + .rlen = 499, }, }; -- cgit v1.2.3 From 7efe4076725aeb01722445b56613681aa492c8d6 Mon Sep 17 00:00:00 2001 From: Johannes Goetzfried Date: Tue, 12 Jun 2012 16:47:43 +0800 Subject: crypto: serpent - add x86_64/avx assembler implementation This patch adds a x86_64/avx assembler implementation of the Serpent block cipher. The implementation is very similar to the sse2 implementation and processes eight blocks in parallel. Because of the new non-destructive three operand syntax all move-instructions can be removed and therefore a little performance increase is provided. Patch has been tested with tcrypt and automated filesystem tests. Tcrypt benchmark results: Intel Core i5-2500 CPU (fam:6, model:42, step:7) serpent-avx-x86_64 vs. serpent-sse2-x86_64 128bit key: (lrw:256bit) (xts:256bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 1.03x 1.01x 1.01x 1.01x 1.00x 1.00x 1.00x 1.00x 1.00x 1.01x 64B 1.00x 1.00x 1.00x 1.00x 1.00x 0.99x 1.00x 1.01x 1.00x 1.00x 256B 1.05x 1.03x 1.00x 1.02x 1.05x 1.06x 1.05x 1.02x 1.05x 1.02x 1024B 1.05x 1.02x 1.00x 1.02x 1.05x 1.06x 1.05x 1.03x 1.05x 1.02x 8192B 1.05x 1.02x 1.00x 1.02x 1.06x 1.06x 1.04x 1.03x 1.04x 1.02x 256bit key: (lrw:384bit) (xts:512bit) size ecb-enc ecb-dec cbc-enc cbc-dec ctr-enc ctr-dec lrw-enc lrw-dec xts-enc xts-dec 16B 1.01x 1.00x 1.01x 1.01x 1.00x 1.00x 0.99x 1.03x 1.01x 1.01x 64B 1.00x 1.00x 1.00x 1.00x 1.00x 1.00x 1.00x 1.01x 1.00x 1.02x 256B 1.05x 1.02x 1.00x 1.02x 1.05x 1.02x 1.04x 1.05x 1.05x 1.02x 1024B 1.06x 1.02x 1.00x 1.02x 1.07x 1.06x 1.05x 1.04x 1.05x 1.02x 8192B 1.05x 1.02x 1.00x 1.02x 1.06x 1.06x 1.04x 1.05x 1.05x 1.02x serpent-avx-x86_64 vs aes-asm (8kB block): 128bit 256bit ecb-enc 1.26x 1.73x ecb-dec 1.20x 1.64x cbc-enc 0.33x 0.45x cbc-dec 1.24x 1.67x ctr-enc 1.32x 1.76x ctr-dec 1.32x 1.76x lrw-enc 1.20x 1.60x lrw-dec 1.15x 1.54x xts-enc 1.22x 1.64x xts-dec 1.17x 1.57x Signed-off-by: Johannes Goetzfried Signed-off-by: Herbert Xu --- arch/x86/crypto/Makefile | 2 + arch/x86/crypto/serpent-avx-x86_64-asm_64.S | 704 +++++++++++++++++++++ arch/x86/crypto/serpent_avx_glue.c | 949 ++++++++++++++++++++++++++++ crypto/Kconfig | 20 + crypto/testmgr.c | 60 ++ 5 files changed, 1735 insertions(+) create mode 100644 arch/x86/crypto/serpent-avx-x86_64-asm_64.S create mode 100644 arch/x86/crypto/serpent_avx_glue.c (limited to 'crypto') diff --git a/arch/x86/crypto/Makefile b/arch/x86/crypto/Makefile index 3420feef0c7..83caa4b948c 100644 --- a/arch/x86/crypto/Makefile +++ b/arch/x86/crypto/Makefile @@ -15,6 +15,7 @@ obj-$(CONFIG_CRYPTO_TWOFISH_X86_64_3WAY) += twofish-x86_64-3way.o obj-$(CONFIG_CRYPTO_TWOFISH_AVX_X86_64) += twofish-avx-x86_64.o obj-$(CONFIG_CRYPTO_SALSA20_X86_64) += salsa20-x86_64.o obj-$(CONFIG_CRYPTO_SERPENT_SSE2_X86_64) += serpent-sse2-x86_64.o +obj-$(CONFIG_CRYPTO_SERPENT_AVX_X86_64) += serpent-avx-x86_64.o obj-$(CONFIG_CRYPTO_AES_NI_INTEL) += aesni-intel.o obj-$(CONFIG_CRYPTO_GHASH_CLMUL_NI_INTEL) += ghash-clmulni-intel.o @@ -34,6 +35,7 @@ twofish-x86_64-3way-y := twofish-x86_64-asm_64-3way.o twofish_glue_3way.o twofish-avx-x86_64-y := twofish-avx-x86_64-asm_64.o twofish_avx_glue.o salsa20-x86_64-y := salsa20-x86_64-asm_64.o salsa20_glue.o serpent-sse2-x86_64-y := serpent-sse2-x86_64-asm_64.o serpent_sse2_glue.o +serpent-avx-x86_64-y := serpent-avx-x86_64-asm_64.o serpent_avx_glue.o aesni-intel-y := aesni-intel_asm.o aesni-intel_glue.o fpu.o ghash-clmulni-intel-y := ghash-clmulni-intel_asm.o ghash-clmulni-intel_glue.o diff --git a/arch/x86/crypto/serpent-avx-x86_64-asm_64.S b/arch/x86/crypto/serpent-avx-x86_64-asm_64.S new file mode 100644 index 00000000000..0ed47a124ba --- /dev/null +++ b/arch/x86/crypto/serpent-avx-x86_64-asm_64.S @@ -0,0 +1,704 @@ +/* + * Serpent Cipher 8-way parallel algorithm (x86_64/AVX) + * + * Copyright (C) 2012 Johannes Goetzfried + * + * + * Based on arch/x86/crypto/serpent-sse2-x86_64-asm_64.S by + * Copyright (C) 2011 Jussi Kivilinna + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 + * USA + * + */ + +.file "serpent-avx-x86_64-asm_64.S" +.text + +#define CTX %rdi + +/********************************************************************** + 8-way AVX serpent + **********************************************************************/ +#define RA1 %xmm0 +#define RB1 %xmm1 +#define RC1 %xmm2 +#define RD1 %xmm3 +#define RE1 %xmm4 + +#define tp %xmm5 + +#define RA2 %xmm6 +#define RB2 %xmm7 +#define RC2 %xmm8 +#define RD2 %xmm9 +#define RE2 %xmm10 + +#define RNOT %xmm11 + +#define RK0 %xmm12 +#define RK1 %xmm13 +#define RK2 %xmm14 +#define RK3 %xmm15 + + +#define S0_1(x0, x1, x2, x3, x4) \ + vpor x0, x3, tp; \ + vpxor x3, x0, x0; \ + vpxor x2, x3, x4; \ + vpxor RNOT, x4, x4; \ + vpxor x1, tp, x3; \ + vpand x0, x1, x1; \ + vpxor x4, x1, x1; \ + vpxor x0, x2, x2; +#define S0_2(x0, x1, x2, x3, x4) \ + vpxor x3, x0, x0; \ + vpor x0, x4, x4; \ + vpxor x2, x0, x0; \ + vpand x1, x2, x2; \ + vpxor x2, x3, x3; \ + vpxor RNOT, x1, x1; \ + vpxor x4, x2, x2; \ + vpxor x2, x1, x1; + +#define S1_1(x0, x1, x2, x3, x4) \ + vpxor x0, x1, tp; \ + vpxor x3, x0, x0; \ + vpxor RNOT, x3, x3; \ + vpand tp, x1, x4; \ + vpor tp, x0, x0; \ + vpxor x2, x3, x3; \ + vpxor x3, x0, x0; \ + vpxor x3, tp, x1; +#define S1_2(x0, x1, x2, x3, x4) \ + vpxor x4, x3, x3; \ + vpor x4, x1, x1; \ + vpxor x2, x4, x4; \ + vpand x0, x2, x2; \ + vpxor x1, x2, x2; \ + vpor x0, x1, x1; \ + vpxor RNOT, x0, x0; \ + vpxor x2, x0, x0; \ + vpxor x1, x4, x4; + +#define S2_1(x0, x1, x2, x3, x4) \ + vpxor RNOT, x3, x3; \ + vpxor x0, x1, x1; \ + vpand x2, x0, tp; \ + vpxor x3, tp, tp; \ + vpor x0, x3, x3; \ + vpxor x1, x2, x2; \ + vpxor x1, x3, x3; \ + vpand tp, x1, x1; +#define S2_2(x0, x1, x2, x3, x4) \ + vpxor x2, tp, tp; \ + vpand x3, x2, x2; \ + vpor x1, x3, x3; \ + vpxor RNOT, tp, tp; \ + vpxor tp, x3, x3; \ + vpxor tp, x0, x4; \ + vpxor x2, tp, x0; \ + vpor x2, x1, x1; + +#define S3_1(x0, x1, x2, x3, x4) \ + vpxor x3, x1, tp; \ + vpor x0, x3, x3; \ + vpand x0, x1, x4; \ + vpxor x2, x0, x0; \ + vpxor tp, x2, x2; \ + vpand x3, tp, x1; \ + vpxor x3, x2, x2; \ + vpor x4, x0, x0; \ + vpxor x3, x4, x4; +#define S3_2(x0, x1, x2, x3, x4) \ + vpxor x0, x1, x1; \ + vpand x3, x0, x0; \ + vpand x4, x3, x3; \ + vpxor x2, x3, x3; \ + vpor x1, x4, x4; \ + vpand x1, x2, x2; \ + vpxor x3, x4, x4; \ + vpxor x3, x0, x0; \ + vpxor x2, x3, x3; + +#define S4_1(x0, x1, x2, x3, x4) \ + vpand x0, x3, tp; \ + vpxor x3, x0, x0; \ + vpxor x2, tp, tp; \ + vpor x3, x2, x2; \ + vpxor x1, x0, x0; \ + vpxor tp, x3, x4; \ + vpor x0, x2, x2; \ + vpxor x1, x2, x2; +#define S4_2(x0, x1, x2, x3, x4) \ + vpand x0, x1, x1; \ + vpxor x4, x1, x1; \ + vpand x2, x4, x4; \ + vpxor tp, x2, x2; \ + vpxor x0, x4, x4; \ + vpor x1, tp, x3; \ + vpxor RNOT, x1, x1; \ + vpxor x0, x3, x3; + +#define S5_1(x0, x1, x2, x3, x4) \ + vpor x0, x1, tp; \ + vpxor tp, x2, x2; \ + vpxor RNOT, x3, x3; \ + vpxor x0, x1, x4; \ + vpxor x2, x0, x0; \ + vpand x4, tp, x1; \ + vpor x3, x4, x4; \ + vpxor x0, x4, x4; +#define S5_2(x0, x1, x2, x3, x4) \ + vpand x3, x0, x0; \ + vpxor x3, x1, x1; \ + vpxor x2, x3, x3; \ + vpxor x1, x0, x0; \ + vpand x4, x2, x2; \ + vpxor x2, x1, x1; \ + vpand x0, x2, x2; \ + vpxor x2, x3, x3; + +#define S6_1(x0, x1, x2, x3, x4) \ + vpxor x0, x3, x3; \ + vpxor x2, x1, tp; \ + vpxor x0, x2, x2; \ + vpand x3, x0, x0; \ + vpor x3, tp, tp; \ + vpxor RNOT, x1, x4; \ + vpxor tp, x0, x0; \ + vpxor x2, tp, x1; +#define S6_2(x0, x1, x2, x3, x4) \ + vpxor x4, x3, x3; \ + vpxor x0, x4, x4; \ + vpand x0, x2, x2; \ + vpxor x1, x4, x4; \ + vpxor x3, x2, x2; \ + vpand x1, x3, x3; \ + vpxor x0, x3, x3; \ + vpxor x2, x1, x1; + +#define S7_1(x0, x1, x2, x3, x4) \ + vpxor RNOT, x1, tp; \ + vpxor RNOT, x0, x0; \ + vpand x2, tp, x1; \ + vpxor x3, x1, x1; \ + vpor tp, x3, x3; \ + vpxor x2, tp, x4; \ + vpxor x3, x2, x2; \ + vpxor x0, x3, x3; \ + vpor x1, x0, x0; +#define S7_2(x0, x1, x2, x3, x4) \ + vpand x0, x2, x2; \ + vpxor x4, x0, x0; \ + vpxor x3, x4, x4; \ + vpand x0, x3, x3; \ + vpxor x1, x4, x4; \ + vpxor x4, x2, x2; \ + vpxor x1, x3, x3; \ + vpor x0, x4, x4; \ + vpxor x1, x4, x4; + +#define SI0_1(x0, x1, x2, x3, x4) \ + vpxor x0, x1, x1; \ + vpor x1, x3, tp; \ + vpxor x1, x3, x4; \ + vpxor RNOT, x0, x0; \ + vpxor tp, x2, x2; \ + vpxor x0, tp, x3; \ + vpand x1, x0, x0; \ + vpxor x2, x0, x0; +#define SI0_2(x0, x1, x2, x3, x4) \ + vpand x3, x2, x2; \ + vpxor x4, x3, x3; \ + vpxor x3, x2, x2; \ + vpxor x3, x1, x1; \ + vpand x0, x3, x3; \ + vpxor x0, x1, x1; \ + vpxor x2, x0, x0; \ + vpxor x3, x4, x4; + +#define SI1_1(x0, x1, x2, x3, x4) \ + vpxor x3, x1, x1; \ + vpxor x2, x0, tp; \ + vpxor RNOT, x2, x2; \ + vpor x1, x0, x4; \ + vpxor x3, x4, x4; \ + vpand x1, x3, x3; \ + vpxor x2, x1, x1; \ + vpand x4, x2, x2; +#define SI1_2(x0, x1, x2, x3, x4) \ + vpxor x1, x4, x4; \ + vpor x3, x1, x1; \ + vpxor tp, x3, x3; \ + vpxor tp, x2, x2; \ + vpor x4, tp, x0; \ + vpxor x4, x2, x2; \ + vpxor x0, x1, x1; \ + vpxor x1, x4, x4; + +#define SI2_1(x0, x1, x2, x3, x4) \ + vpxor x1, x2, x2; \ + vpxor RNOT, x3, tp; \ + vpor x2, tp, tp; \ + vpxor x3, x2, x2; \ + vpxor x0, x3, x4; \ + vpxor x1, tp, x3; \ + vpor x2, x1, x1; \ + vpxor x0, x2, x2; +#define SI2_2(x0, x1, x2, x3, x4) \ + vpxor x4, x1, x1; \ + vpor x3, x4, x4; \ + vpxor x3, x2, x2; \ + vpxor x2, x4, x4; \ + vpand x1, x2, x2; \ + vpxor x3, x2, x2; \ + vpxor x4, x3, x3; \ + vpxor x0, x4, x4; + +#define SI3_1(x0, x1, x2, x3, x4) \ + vpxor x1, x2, x2; \ + vpand x2, x1, tp; \ + vpxor x0, tp, tp; \ + vpor x1, x0, x0; \ + vpxor x3, x1, x4; \ + vpxor x3, x0, x0; \ + vpor tp, x3, x3; \ + vpxor x2, tp, x1; +#define SI3_2(x0, x1, x2, x3, x4) \ + vpxor x3, x1, x1; \ + vpxor x2, x0, x0; \ + vpxor x3, x2, x2; \ + vpand x1, x3, x3; \ + vpxor x0, x1, x1; \ + vpand x2, x0, x0; \ + vpxor x3, x4, x4; \ + vpxor x0, x3, x3; \ + vpxor x1, x0, x0; + +#define SI4_1(x0, x1, x2, x3, x4) \ + vpxor x3, x2, x2; \ + vpand x1, x0, tp; \ + vpxor x2, tp, tp; \ + vpor x3, x2, x2; \ + vpxor RNOT, x0, x4; \ + vpxor tp, x1, x1; \ + vpxor x2, tp, x0; \ + vpand x4, x2, x2; +#define SI4_2(x0, x1, x2, x3, x4) \ + vpxor x0, x2, x2; \ + vpor x4, x0, x0; \ + vpxor x3, x0, x0; \ + vpand x2, x3, x3; \ + vpxor x3, x4, x4; \ + vpxor x1, x3, x3; \ + vpand x0, x1, x1; \ + vpxor x1, x4, x4; \ + vpxor x3, x0, x0; + +#define SI5_1(x0, x1, x2, x3, x4) \ + vpor x2, x1, tp; \ + vpxor x1, x2, x2; \ + vpxor x3, tp, tp; \ + vpand x1, x3, x3; \ + vpxor x3, x2, x2; \ + vpor x0, x3, x3; \ + vpxor RNOT, x0, x0; \ + vpxor x2, x3, x3; \ + vpor x0, x2, x2; +#define SI5_2(x0, x1, x2, x3, x4) \ + vpxor tp, x1, x4; \ + vpxor x4, x2, x2; \ + vpand x0, x4, x4; \ + vpxor tp, x0, x0; \ + vpxor x3, tp, x1; \ + vpand x2, x0, x0; \ + vpxor x3, x2, x2; \ + vpxor x2, x0, x0; \ + vpxor x4, x2, x2; \ + vpxor x3, x4, x4; + +#define SI6_1(x0, x1, x2, x3, x4) \ + vpxor x2, x0, x0; \ + vpand x3, x0, tp; \ + vpxor x3, x2, x2; \ + vpxor x2, tp, tp; \ + vpxor x1, x3, x3; \ + vpor x0, x2, x2; \ + vpxor x3, x2, x2; \ + vpand tp, x3, x3; +#define SI6_2(x0, x1, x2, x3, x4) \ + vpxor RNOT, tp, tp; \ + vpxor x1, x3, x3; \ + vpand x2, x1, x1; \ + vpxor tp, x0, x4; \ + vpxor x4, x3, x3; \ + vpxor x2, x4, x4; \ + vpxor x1, tp, x0; \ + vpxor x0, x2, x2; + +#define SI7_1(x0, x1, x2, x3, x4) \ + vpand x0, x3, tp; \ + vpxor x2, x0, x0; \ + vpor x3, x2, x2; \ + vpxor x1, x3, x4; \ + vpxor RNOT, x0, x0; \ + vpor tp, x1, x1; \ + vpxor x0, x4, x4; \ + vpand x2, x0, x0; \ + vpxor x1, x0, x0; +#define SI7_2(x0, x1, x2, x3, x4) \ + vpand x2, x1, x1; \ + vpxor x2, tp, x3; \ + vpxor x3, x4, x4; \ + vpand x3, x2, x2; \ + vpor x0, x3, x3; \ + vpxor x4, x1, x1; \ + vpxor x4, x3, x3; \ + vpand x0, x4, x4; \ + vpxor x2, x4, x4; + +#define get_key(i, j, t) \ + vbroadcastss (4*(i)+(j))*4(CTX), t; + +#define K2(x0, x1, x2, x3, x4, i) \ + get_key(i, 0, RK0); \ + get_key(i, 1, RK1); \ + get_key(i, 2, RK2); \ + get_key(i, 3, RK3); \ + vpxor RK0, x0 ## 1, x0 ## 1; \ + vpxor RK1, x1 ## 1, x1 ## 1; \ + vpxor RK2, x2 ## 1, x2 ## 1; \ + vpxor RK3, x3 ## 1, x3 ## 1; \ + vpxor RK0, x0 ## 2, x0 ## 2; \ + vpxor RK1, x1 ## 2, x1 ## 2; \ + vpxor RK2, x2 ## 2, x2 ## 2; \ + vpxor RK3, x3 ## 2, x3 ## 2; + +#define LK2(x0, x1, x2, x3, x4, i) \ + vpslld $13, x0 ## 1, x4 ## 1; \ + vpsrld $(32 - 13), x0 ## 1, x0 ## 1; \ + vpor x4 ## 1, x0 ## 1, x0 ## 1; \ + vpxor x0 ## 1, x1 ## 1, x1 ## 1; \ + vpslld $3, x2 ## 1, x4 ## 1; \ + vpsrld $(32 - 3), x2 ## 1, x2 ## 1; \ + vpor x4 ## 1, x2 ## 1, x2 ## 1; \ + vpxor x2 ## 1, x1 ## 1, x1 ## 1; \ + vpslld $13, x0 ## 2, x4 ## 2; \ + vpsrld $(32 - 13), x0 ## 2, x0 ## 2; \ + vpor x4 ## 2, x0 ## 2, x0 ## 2; \ + vpxor x0 ## 2, x1 ## 2, x1 ## 2; \ + vpslld $3, x2 ## 2, x4 ## 2; \ + vpsrld $(32 - 3), x2 ## 2, x2 ## 2; \ + vpor x4 ## 2, x2 ## 2, x2 ## 2; \ + vpxor x2 ## 2, x1 ## 2, x1 ## 2; \ + vpslld $1, x1 ## 1, x4 ## 1; \ + vpsrld $(32 - 1), x1 ## 1, x1 ## 1; \ + vpor x4 ## 1, x1 ## 1, x1 ## 1; \ + vpslld $3, x0 ## 1, x4 ## 1; \ + vpxor x2 ## 1, x3 ## 1, x3 ## 1; \ + vpxor x4 ## 1, x3 ## 1, x3 ## 1; \ + get_key(i, 1, RK1); \ + vpslld $1, x1 ## 2, x4 ## 2; \ + vpsrld $(32 - 1), x1 ## 2, x1 ## 2; \ + vpor x4 ## 2, x1 ## 2, x1 ## 2; \ + vpslld $3, x0 ## 2, x4 ## 2; \ + vpxor x2 ## 2, x3 ## 2, x3 ## 2; \ + vpxor x4 ## 2, x3 ## 2, x3 ## 2; \ + get_key(i, 3, RK3); \ + vpslld $7, x3 ## 1, x4 ## 1; \ + vpsrld $(32 - 7), x3 ## 1, x3 ## 1; \ + vpor x4 ## 1, x3 ## 1, x3 ## 1; \ + vpslld $7, x1 ## 1, x4 ## 1; \ + vpxor x1 ## 1, x0 ## 1, x0 ## 1; \ + vpxor x3 ## 1, x0 ## 1, x0 ## 1; \ + vpxor x3 ## 1, x2 ## 1, x2 ## 1; \ + vpxor x4 ## 1, x2 ## 1, x2 ## 1; \ + get_key(i, 0, RK0); \ + vpslld $7, x3 ## 2, x4 ## 2; \ + vpsrld $(32 - 7), x3 ## 2, x3 ## 2; \ + vpor x4 ## 2, x3 ## 2, x3 ## 2; \ + vpslld $7, x1 ## 2, x4 ## 2; \ + vpxor x1 ## 2, x0 ## 2, x0 ## 2; \ + vpxor x3 ## 2, x0 ## 2, x0 ## 2; \ + vpxor x3 ## 2, x2 ## 2, x2 ## 2; \ + vpxor x4 ## 2, x2 ## 2, x2 ## 2; \ + get_key(i, 2, RK2); \ + vpxor RK1, x1 ## 1, x1 ## 1; \ + vpxor RK3, x3 ## 1, x3 ## 1; \ + vpslld $5, x0 ## 1, x4 ## 1; \ + vpsrld $(32 - 5), x0 ## 1, x0 ## 1; \ + vpor x4 ## 1, x0 ## 1, x0 ## 1; \ + vpslld $22, x2 ## 1, x4 ## 1; \ + vpsrld $(32 - 22), x2 ## 1, x2 ## 1; \ + vpor x4 ## 1, x2 ## 1, x2 ## 1; \ + vpxor RK0, x0 ## 1, x0 ## 1; \ + vpxor RK2, x2 ## 1, x2 ## 1; \ + vpxor RK1, x1 ## 2, x1 ## 2; \ + vpxor RK3, x3 ## 2, x3 ## 2; \ + vpslld $5, x0 ## 2, x4 ## 2; \ + vpsrld $(32 - 5), x0 ## 2, x0 ## 2; \ + vpor x4 ## 2, x0 ## 2, x0 ## 2; \ + vpslld $22, x2 ## 2, x4 ## 2; \ + vpsrld $(32 - 22), x2 ## 2, x2 ## 2; \ + vpor x4 ## 2, x2 ## 2, x2 ## 2; \ + vpxor RK0, x0 ## 2, x0 ## 2; \ + vpxor RK2, x2 ## 2, x2 ## 2; + +#define KL2(x0, x1, x2, x3, x4, i) \ + vpxor RK0, x0 ## 1, x0 ## 1; \ + vpxor RK2, x2 ## 1, x2 ## 1; \ + vpsrld $5, x0 ## 1, x4 ## 1; \ + vpslld $(32 - 5), x0 ## 1, x0 ## 1; \ + vpor x4 ## 1, x0 ## 1, x0 ## 1; \ + vpxor RK3, x3 ## 1, x3 ## 1; \ + vpxor RK1, x1 ## 1, x1 ## 1; \ + vpsrld $22, x2 ## 1, x4 ## 1; \ + vpslld $(32 - 22), x2 ## 1, x2 ## 1; \ + vpor x4 ## 1, x2 ## 1, x2 ## 1; \ + vpxor x3 ## 1, x2 ## 1, x2 ## 1; \ + vpxor RK0, x0 ## 2, x0 ## 2; \ + vpxor RK2, x2 ## 2, x2 ## 2; \ + vpsrld $5, x0 ## 2, x4 ## 2; \ + vpslld $(32 - 5), x0 ## 2, x0 ## 2; \ + vpor x4 ## 2, x0 ## 2, x0 ## 2; \ + vpxor RK3, x3 ## 2, x3 ## 2; \ + vpxor RK1, x1 ## 2, x1 ## 2; \ + vpsrld $22, x2 ## 2, x4 ## 2; \ + vpslld $(32 - 22), x2 ## 2, x2 ## 2; \ + vpor x4 ## 2, x2 ## 2, x2 ## 2; \ + vpxor x3 ## 2, x2 ## 2, x2 ## 2; \ + vpxor x3 ## 1, x0 ## 1, x0 ## 1; \ + vpslld $7, x1 ## 1, x4 ## 1; \ + vpxor x1 ## 1, x0 ## 1, x0 ## 1; \ + vpxor x4 ## 1, x2 ## 1, x2 ## 1; \ + vpsrld $1, x1 ## 1, x4 ## 1; \ + vpslld $(32 - 1), x1 ## 1, x1 ## 1; \ + vpor x4 ## 1, x1 ## 1, x1 ## 1; \ + vpxor x3 ## 2, x0 ## 2, x0 ## 2; \ + vpslld $7, x1 ## 2, x4 ## 2; \ + vpxor x1 ## 2, x0 ## 2, x0 ## 2; \ + vpxor x4 ## 2, x2 ## 2, x2 ## 2; \ + vpsrld $1, x1 ## 2, x4 ## 2; \ + vpslld $(32 - 1), x1 ## 2, x1 ## 2; \ + vpor x4 ## 2, x1 ## 2, x1 ## 2; \ + vpsrld $7, x3 ## 1, x4 ## 1; \ + vpslld $(32 - 7), x3 ## 1, x3 ## 1; \ + vpor x4 ## 1, x3 ## 1, x3 ## 1; \ + vpxor x0 ## 1, x1 ## 1, x1 ## 1; \ + vpslld $3, x0 ## 1, x4 ## 1; \ + vpxor x4 ## 1, x3 ## 1, x3 ## 1; \ + vpsrld $7, x3 ## 2, x4 ## 2; \ + vpslld $(32 - 7), x3 ## 2, x3 ## 2; \ + vpor x4 ## 2, x3 ## 2, x3 ## 2; \ + vpxor x0 ## 2, x1 ## 2, x1 ## 2; \ + vpslld $3, x0 ## 2, x4 ## 2; \ + vpxor x4 ## 2, x3 ## 2, x3 ## 2; \ + vpsrld $13, x0 ## 1, x4 ## 1; \ + vpslld $(32 - 13), x0 ## 1, x0 ## 1; \ + vpor x4 ## 1, x0 ## 1, x0 ## 1; \ + vpxor x2 ## 1, x1 ## 1, x1 ## 1; \ + vpxor x2 ## 1, x3 ## 1, x3 ## 1; \ + vpsrld $3, x2 ## 1, x4 ## 1; \ + vpslld $(32 - 3), x2 ## 1, x2 ## 1; \ + vpor x4 ## 1, x2 ## 1, x2 ## 1; \ + vpsrld $13, x0 ## 2, x4 ## 2; \ + vpslld $(32 - 13), x0 ## 2, x0 ## 2; \ + vpor x4 ## 2, x0 ## 2, x0 ## 2; \ + vpxor x2 ## 2, x1 ## 2, x1 ## 2; \ + vpxor x2 ## 2, x3 ## 2, x3 ## 2; \ + vpsrld $3, x2 ## 2, x4 ## 2; \ + vpslld $(32 - 3), x2 ## 2, x2 ## 2; \ + vpor x4 ## 2, x2 ## 2, x2 ## 2; + +#define S(SBOX, x0, x1, x2, x3, x4) \ + SBOX ## _1(x0 ## 1, x1 ## 1, x2 ## 1, x3 ## 1, x4 ## 1); \ + SBOX ## _2(x0 ## 1, x1 ## 1, x2 ## 1, x3 ## 1, x4 ## 1); \ + SBOX ## _1(x0 ## 2, x1 ## 2, x2 ## 2, x3 ## 2, x4 ## 2); \ + SBOX ## _2(x0 ## 2, x1 ## 2, x2 ## 2, x3 ## 2, x4 ## 2); + +#define SP(SBOX, x0, x1, x2, x3, x4, i) \ + get_key(i, 0, RK0); \ + SBOX ## _1(x0 ## 1, x1 ## 1, x2 ## 1, x3 ## 1, x4 ## 1); \ + get_key(i, 2, RK2); \ + SBOX ## _2(x0 ## 1, x1 ## 1, x2 ## 1, x3 ## 1, x4 ## 1); \ + get_key(i, 3, RK3); \ + SBOX ## _1(x0 ## 2, x1 ## 2, x2 ## 2, x3 ## 2, x4 ## 2); \ + get_key(i, 1, RK1); \ + SBOX ## _2(x0 ## 2, x1 ## 2, x2 ## 2, x3 ## 2, x4 ## 2); \ + +#define transpose_4x4(x0, x1, x2, x3, t0, t1, t2) \ + vpunpckldq x1, x0, t0; \ + vpunpckhdq x1, x0, t2; \ + vpunpckldq x3, x2, t1; \ + vpunpckhdq x3, x2, x3; \ + \ + vpunpcklqdq t1, t0, x0; \ + vpunpckhqdq t1, t0, x1; \ + vpunpcklqdq x3, t2, x2; \ + vpunpckhqdq x3, t2, x3; + +#define read_blocks(in, x0, x1, x2, x3, t0, t1, t2) \ + vmovdqu (0*4*4)(in), x0; \ + vmovdqu (1*4*4)(in), x1; \ + vmovdqu (2*4*4)(in), x2; \ + vmovdqu (3*4*4)(in), x3; \ + \ + transpose_4x4(x0, x1, x2, x3, t0, t1, t2) + +#define write_blocks(out, x0, x1, x2, x3, t0, t1, t2) \ + transpose_4x4(x0, x1, x2, x3, t0, t1, t2) \ + \ + vmovdqu x0, (0*4*4)(out); \ + vmovdqu x1, (1*4*4)(out); \ + vmovdqu x2, (2*4*4)(out); \ + vmovdqu x3, (3*4*4)(out); + +#define xor_blocks(out, x0, x1, x2, x3, t0, t1, t2) \ + transpose_4x4(x0, x1, x2, x3, t0, t1, t2) \ + \ + vpxor (0*4*4)(out), x0, x0; \ + vmovdqu x0, (0*4*4)(out); \ + vpxor (1*4*4)(out), x1, x1; \ + vmovdqu x1, (1*4*4)(out); \ + vpxor (2*4*4)(out), x2, x2; \ + vmovdqu x2, (2*4*4)(out); \ + vpxor (3*4*4)(out), x3, x3; \ + vmovdqu x3, (3*4*4)(out); + +.align 8 +.global __serpent_enc_blk_8way +.type __serpent_enc_blk_8way,@function; + +__serpent_enc_blk_8way: + /* input: + * %rdi: ctx, CTX + * %rsi: dst + * %rdx: src + * %rcx: bool, if true: xor output + */ + + vpcmpeqd RNOT, RNOT, RNOT; + + leaq (4*4*4)(%rdx), %rax; + read_blocks(%rdx, RA1, RB1, RC1, RD1, RK0, RK1, RK2); + read_blocks(%rax, RA2, RB2, RC2, RD2, RK0, RK1, RK2); + + K2(RA, RB, RC, RD, RE, 0); + S(S0, RA, RB, RC, RD, RE); LK2(RC, RB, RD, RA, RE, 1); + S(S1, RC, RB, RD, RA, RE); LK2(RE, RD, RA, RC, RB, 2); + S(S2, RE, RD, RA, RC, RB); LK2(RB, RD, RE, RC, RA, 3); + S(S3, RB, RD, RE, RC, RA); LK2(RC, RA, RD, RB, RE, 4); + S(S4, RC, RA, RD, RB, RE); LK2(RA, RD, RB, RE, RC, 5); + S(S5, RA, RD, RB, RE, RC); LK2(RC, RA, RD, RE, RB, 6); + S(S6, RC, RA, RD, RE, RB); LK2(RD, RB, RA, RE, RC, 7); + S(S7, RD, RB, RA, RE, RC); LK2(RC, RA, RE, RD, RB, 8); + S(S0, RC, RA, RE, RD, RB); LK2(RE, RA, RD, RC, RB, 9); + S(S1, RE, RA, RD, RC, RB); LK2(RB, RD, RC, RE, RA, 10); + S(S2, RB, RD, RC, RE, RA); LK2(RA, RD, RB, RE, RC, 11); + S(S3, RA, RD, RB, RE, RC); LK2(RE, RC, RD, RA, RB, 12); + S(S4, RE, RC, RD, RA, RB); LK2(RC, RD, RA, RB, RE, 13); + S(S5, RC, RD, RA, RB, RE); LK2(RE, RC, RD, RB, RA, 14); + S(S6, RE, RC, RD, RB, RA); LK2(RD, RA, RC, RB, RE, 15); + S(S7, RD, RA, RC, RB, RE); LK2(RE, RC, RB, RD, RA, 16); + S(S0, RE, RC, RB, RD, RA); LK2(RB, RC, RD, RE, RA, 17); + S(S1, RB, RC, RD, RE, RA); LK2(RA, RD, RE, RB, RC, 18); + S(S2, RA, RD, RE, RB, RC); LK2(RC, RD, RA, RB, RE, 19); + S(S3, RC, RD, RA, RB, RE); LK2(RB, RE, RD, RC, RA, 20); + S(S4, RB, RE, RD, RC, RA); LK2(RE, RD, RC, RA, RB, 21); + S(S5, RE, RD, RC, RA, RB); LK2(RB, RE, RD, RA, RC, 22); + S(S6, RB, RE, RD, RA, RC); LK2(RD, RC, RE, RA, RB, 23); + S(S7, RD, RC, RE, RA, RB); LK2(RB, RE, RA, RD, RC, 24); + S(S0, RB, RE, RA, RD, RC); LK2(RA, RE, RD, RB, RC, 25); + S(S1, RA, RE, RD, RB, RC); LK2(RC, RD, RB, RA, RE, 26); + S(S2, RC, RD, RB, RA, RE); LK2(RE, RD, RC, RA, RB, 27); + S(S3, RE, RD, RC, RA, RB); LK2(RA, RB, RD, RE, RC, 28); + S(S4, RA, RB, RD, RE, RC); LK2(RB, RD, RE, RC, RA, 29); + S(S5, RB, RD, RE, RC, RA); LK2(RA, RB, RD, RC, RE, 30); + S(S6, RA, RB, RD, RC, RE); LK2(RD, RE, RB, RC, RA, 31); + S(S7, RD, RE, RB, RC, RA); K2(RA, RB, RC, RD, RE, 32); + + leaq (4*4*4)(%rsi), %rax; + + testb %cl, %cl; + jnz __enc_xor8; + + write_blocks(%rsi, RA1, RB1, RC1, RD1, RK0, RK1, RK2); + write_blocks(%rax, RA2, RB2, RC2, RD2, RK0, RK1, RK2); + + ret; + +__enc_xor8: + xor_blocks(%rsi, RA1, RB1, RC1, RD1, RK0, RK1, RK2); + xor_blocks(%rax, RA2, RB2, RC2, RD2, RK0, RK1, RK2); + + ret; + +.align 8 +.global serpent_dec_blk_8way +.type serpent_dec_blk_8way,@function; + +serpent_dec_blk_8way: + /* input: + * %rdi: ctx, CTX + * %rsi: dst + * %rdx: src + */ + + vpcmpeqd RNOT, RNOT, RNOT; + + leaq (4*4*4)(%rdx), %rax; + read_blocks(%rdx, RA1, RB1, RC1, RD1, RK0, RK1, RK2); + read_blocks(%rax, RA2, RB2, RC2, RD2, RK0, RK1, RK2); + + K2(RA, RB, RC, RD, RE, 32); + SP(SI7, RA, RB, RC, RD, RE, 31); KL2(RB, RD, RA, RE, RC, 31); + SP(SI6, RB, RD, RA, RE, RC, 30); KL2(RA, RC, RE, RB, RD, 30); + SP(SI5, RA, RC, RE, RB, RD, 29); KL2(RC, RD, RA, RE, RB, 29); + SP(SI4, RC, RD, RA, RE, RB, 28); KL2(RC, RA, RB, RE, RD, 28); + SP(SI3, RC, RA, RB, RE, RD, 27); KL2(RB, RC, RD, RE, RA, 27); + SP(SI2, RB, RC, RD, RE, RA, 26); KL2(RC, RA, RE, RD, RB, 26); + SP(SI1, RC, RA, RE, RD, RB, 25); KL2(RB, RA, RE, RD, RC, 25); + SP(SI0, RB, RA, RE, RD, RC, 24); KL2(RE, RC, RA, RB, RD, 24); + SP(SI7, RE, RC, RA, RB, RD, 23); KL2(RC, RB, RE, RD, RA, 23); + SP(SI6, RC, RB, RE, RD, RA, 22); KL2(RE, RA, RD, RC, RB, 22); + SP(SI5, RE, RA, RD, RC, RB, 21); KL2(RA, RB, RE, RD, RC, 21); + SP(SI4, RA, RB, RE, RD, RC, 20); KL2(RA, RE, RC, RD, RB, 20); + SP(SI3, RA, RE, RC, RD, RB, 19); KL2(RC, RA, RB, RD, RE, 19); + SP(SI2, RC, RA, RB, RD, RE, 18); KL2(RA, RE, RD, RB, RC, 18); + SP(SI1, RA, RE, RD, RB, RC, 17); KL2(RC, RE, RD, RB, RA, 17); + SP(SI0, RC, RE, RD, RB, RA, 16); KL2(RD, RA, RE, RC, RB, 16); + SP(SI7, RD, RA, RE, RC, RB, 15); KL2(RA, RC, RD, RB, RE, 15); + SP(SI6, RA, RC, RD, RB, RE, 14); KL2(RD, RE, RB, RA, RC, 14); + SP(SI5, RD, RE, RB, RA, RC, 13); KL2(RE, RC, RD, RB, RA, 13); + SP(SI4, RE, RC, RD, RB, RA, 12); KL2(RE, RD, RA, RB, RC, 12); + SP(SI3, RE, RD, RA, RB, RC, 11); KL2(RA, RE, RC, RB, RD, 11); + SP(SI2, RA, RE, RC, RB, RD, 10); KL2(RE, RD, RB, RC, RA, 10); + SP(SI1, RE, RD, RB, RC, RA, 9); KL2(RA, RD, RB, RC, RE, 9); + SP(SI0, RA, RD, RB, RC, RE, 8); KL2(RB, RE, RD, RA, RC, 8); + SP(SI7, RB, RE, RD, RA, RC, 7); KL2(RE, RA, RB, RC, RD, 7); + SP(SI6, RE, RA, RB, RC, RD, 6); KL2(RB, RD, RC, RE, RA, 6); + SP(SI5, RB, RD, RC, RE, RA, 5); KL2(RD, RA, RB, RC, RE, 5); + SP(SI4, RD, RA, RB, RC, RE, 4); KL2(RD, RB, RE, RC, RA, 4); + SP(SI3, RD, RB, RE, RC, RA, 3); KL2(RE, RD, RA, RC, RB, 3); + SP(SI2, RE, RD, RA, RC, RB, 2); KL2(RD, RB, RC, RA, RE, 2); + SP(SI1, RD, RB, RC, RA, RE, 1); KL2(RE, RB, RC, RA, RD, 1); + S(SI0, RE, RB, RC, RA, RD); K2(RC, RD, RB, RE, RA, 0); + + leaq (4*4*4)(%rsi), %rax; + write_blocks(%rsi, RC1, RD1, RB1, RE1, RK0, RK1, RK2); + write_blocks(%rax, RC2, RD2, RB2, RE2, RK0, RK1, RK2); + + ret; diff --git a/arch/x86/crypto/serpent_avx_glue.c b/arch/x86/crypto/serpent_avx_glue.c new file mode 100644 index 00000000000..0dc7a26535e --- /dev/null +++ b/arch/x86/crypto/serpent_avx_glue.c @@ -0,0 +1,949 @@ +/* + * Glue Code for AVX assembler versions of Serpent Cipher + * + * Copyright (C) 2012 Johannes Goetzfried + * + * + * Glue code based on serpent_sse2_glue.c by: + * Copyright (C) 2011 Jussi Kivilinna + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 + * USA + * + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +struct async_serpent_ctx { + struct cryptd_ablkcipher *cryptd_tfm; +}; + +static inline bool serpent_fpu_begin(bool fpu_enabled, unsigned int nbytes) +{ + if (fpu_enabled) + return true; + + /* AVX is only used when chunk to be processed is large enough, so + * do not enable FPU until it is necessary. + */ + if (nbytes < SERPENT_BLOCK_SIZE * SERPENT_PARALLEL_BLOCKS) + return false; + + kernel_fpu_begin(); + return true; +} + +static inline void serpent_fpu_end(bool fpu_enabled) +{ + if (fpu_enabled) + kernel_fpu_end(); +} + +static int ecb_crypt(struct blkcipher_desc *desc, struct blkcipher_walk *walk, + bool enc) +{ + bool fpu_enabled = false; + struct serpent_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + const unsigned int bsize = SERPENT_BLOCK_SIZE; + unsigned int nbytes; + int err; + + err = blkcipher_walk_virt(desc, walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + + while ((nbytes = walk->nbytes)) { + u8 *wsrc = walk->src.virt.addr; + u8 *wdst = walk->dst.virt.addr; + + fpu_enabled = serpent_fpu_begin(fpu_enabled, nbytes); + + /* Process multi-block batch */ + if (nbytes >= bsize * SERPENT_PARALLEL_BLOCKS) { + do { + if (enc) + serpent_enc_blk_xway(ctx, wdst, wsrc); + else + serpent_dec_blk_xway(ctx, wdst, wsrc); + + wsrc += bsize * SERPENT_PARALLEL_BLOCKS; + wdst += bsize * SERPENT_PARALLEL_BLOCKS; + nbytes -= bsize * SERPENT_PARALLEL_BLOCKS; + } while (nbytes >= bsize * SERPENT_PARALLEL_BLOCKS); + + if (nbytes < bsize) + goto done; + } + + /* Handle leftovers */ + do { + if (enc) + __serpent_encrypt(ctx, wdst, wsrc); + else + __serpent_decrypt(ctx, wdst, wsrc); + + wsrc += bsize; + wdst += bsize; + nbytes -= bsize; + } while (nbytes >= bsize); + +done: + err = blkcipher_walk_done(desc, walk, nbytes); + } + + serpent_fpu_end(fpu_enabled); + return err; +} + +static int ecb_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct blkcipher_walk walk; + + blkcipher_walk_init(&walk, dst, src, nbytes); + return ecb_crypt(desc, &walk, true); +} + +static int ecb_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct blkcipher_walk walk; + + blkcipher_walk_init(&walk, dst, src, nbytes); + return ecb_crypt(desc, &walk, false); +} + +static unsigned int __cbc_encrypt(struct blkcipher_desc *desc, + struct blkcipher_walk *walk) +{ + struct serpent_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + const unsigned int bsize = SERPENT_BLOCK_SIZE; + unsigned int nbytes = walk->nbytes; + u128 *src = (u128 *)walk->src.virt.addr; + u128 *dst = (u128 *)walk->dst.virt.addr; + u128 *iv = (u128 *)walk->iv; + + do { + u128_xor(dst, src, iv); + __serpent_encrypt(ctx, (u8 *)dst, (u8 *)dst); + iv = dst; + + src += 1; + dst += 1; + nbytes -= bsize; + } while (nbytes >= bsize); + + u128_xor((u128 *)walk->iv, (u128 *)walk->iv, iv); + return nbytes; +} + +static int cbc_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + + while ((nbytes = walk.nbytes)) { + nbytes = __cbc_encrypt(desc, &walk); + err = blkcipher_walk_done(desc, &walk, nbytes); + } + + return err; +} + +static unsigned int __cbc_decrypt(struct blkcipher_desc *desc, + struct blkcipher_walk *walk) +{ + struct serpent_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + const unsigned int bsize = SERPENT_BLOCK_SIZE; + unsigned int nbytes = walk->nbytes; + u128 *src = (u128 *)walk->src.virt.addr; + u128 *dst = (u128 *)walk->dst.virt.addr; + u128 ivs[SERPENT_PARALLEL_BLOCKS - 1]; + u128 last_iv; + int i; + + /* Start of the last block. */ + src += nbytes / bsize - 1; + dst += nbytes / bsize - 1; + + last_iv = *src; + + /* Process multi-block batch */ + if (nbytes >= bsize * SERPENT_PARALLEL_BLOCKS) { + do { + nbytes -= bsize * (SERPENT_PARALLEL_BLOCKS - 1); + src -= SERPENT_PARALLEL_BLOCKS - 1; + dst -= SERPENT_PARALLEL_BLOCKS - 1; + + for (i = 0; i < SERPENT_PARALLEL_BLOCKS - 1; i++) + ivs[i] = src[i]; + + serpent_dec_blk_xway(ctx, (u8 *)dst, (u8 *)src); + + for (i = 0; i < SERPENT_PARALLEL_BLOCKS - 1; i++) + u128_xor(dst + (i + 1), dst + (i + 1), ivs + i); + + nbytes -= bsize; + if (nbytes < bsize) + goto done; + + u128_xor(dst, dst, src - 1); + src -= 1; + dst -= 1; + } while (nbytes >= bsize * SERPENT_PARALLEL_BLOCKS); + + if (nbytes < bsize) + goto done; + } + + /* Handle leftovers */ + for (;;) { + __serpent_decrypt(ctx, (u8 *)dst, (u8 *)src); + + nbytes -= bsize; + if (nbytes < bsize) + break; + + u128_xor(dst, dst, src - 1); + src -= 1; + dst -= 1; + } + +done: + u128_xor(dst, dst, (u128 *)walk->iv); + *(u128 *)walk->iv = last_iv; + + return nbytes; +} + +static int cbc_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + bool fpu_enabled = false; + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + + while ((nbytes = walk.nbytes)) { + fpu_enabled = serpent_fpu_begin(fpu_enabled, nbytes); + nbytes = __cbc_decrypt(desc, &walk); + err = blkcipher_walk_done(desc, &walk, nbytes); + } + + serpent_fpu_end(fpu_enabled); + return err; +} + +static inline void u128_to_be128(be128 *dst, const u128 *src) +{ + dst->a = cpu_to_be64(src->a); + dst->b = cpu_to_be64(src->b); +} + +static inline void be128_to_u128(u128 *dst, const be128 *src) +{ + dst->a = be64_to_cpu(src->a); + dst->b = be64_to_cpu(src->b); +} + +static inline void u128_inc(u128 *i) +{ + i->b++; + if (!i->b) + i->a++; +} + +static void ctr_crypt_final(struct blkcipher_desc *desc, + struct blkcipher_walk *walk) +{ + struct serpent_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + u8 *ctrblk = walk->iv; + u8 keystream[SERPENT_BLOCK_SIZE]; + u8 *src = walk->src.virt.addr; + u8 *dst = walk->dst.virt.addr; + unsigned int nbytes = walk->nbytes; + + __serpent_encrypt(ctx, keystream, ctrblk); + crypto_xor(keystream, src, nbytes); + memcpy(dst, keystream, nbytes); + + crypto_inc(ctrblk, SERPENT_BLOCK_SIZE); +} + +static unsigned int __ctr_crypt(struct blkcipher_desc *desc, + struct blkcipher_walk *walk) +{ + struct serpent_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + const unsigned int bsize = SERPENT_BLOCK_SIZE; + unsigned int nbytes = walk->nbytes; + u128 *src = (u128 *)walk->src.virt.addr; + u128 *dst = (u128 *)walk->dst.virt.addr; + u128 ctrblk; + be128 ctrblocks[SERPENT_PARALLEL_BLOCKS]; + int i; + + be128_to_u128(&ctrblk, (be128 *)walk->iv); + + /* Process multi-block batch */ + if (nbytes >= bsize * SERPENT_PARALLEL_BLOCKS) { + do { + /* create ctrblks for parallel encrypt */ + for (i = 0; i < SERPENT_PARALLEL_BLOCKS; i++) { + if (dst != src) + dst[i] = src[i]; + + u128_to_be128(&ctrblocks[i], &ctrblk); + u128_inc(&ctrblk); + } + + serpent_enc_blk_xway_xor(ctx, (u8 *)dst, + (u8 *)ctrblocks); + + src += SERPENT_PARALLEL_BLOCKS; + dst += SERPENT_PARALLEL_BLOCKS; + nbytes -= bsize * SERPENT_PARALLEL_BLOCKS; + } while (nbytes >= bsize * SERPENT_PARALLEL_BLOCKS); + + if (nbytes < bsize) + goto done; + } + + /* Handle leftovers */ + do { + if (dst != src) + *dst = *src; + + u128_to_be128(&ctrblocks[0], &ctrblk); + u128_inc(&ctrblk); + + __serpent_encrypt(ctx, (u8 *)ctrblocks, (u8 *)ctrblocks); + u128_xor(dst, dst, (u128 *)ctrblocks); + + src += 1; + dst += 1; + nbytes -= bsize; + } while (nbytes >= bsize); + +done: + u128_to_be128((be128 *)walk->iv, &ctrblk); + return nbytes; +} + +static int ctr_crypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + bool fpu_enabled = false; + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt_block(desc, &walk, SERPENT_BLOCK_SIZE); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + + while ((nbytes = walk.nbytes) >= SERPENT_BLOCK_SIZE) { + fpu_enabled = serpent_fpu_begin(fpu_enabled, nbytes); + nbytes = __ctr_crypt(desc, &walk); + err = blkcipher_walk_done(desc, &walk, nbytes); + } + + serpent_fpu_end(fpu_enabled); + + if (walk.nbytes) { + ctr_crypt_final(desc, &walk); + err = blkcipher_walk_done(desc, &walk, 0); + } + + return err; +} + +struct crypt_priv { + struct serpent_ctx *ctx; + bool fpu_enabled; +}; + +static void encrypt_callback(void *priv, u8 *srcdst, unsigned int nbytes) +{ + const unsigned int bsize = SERPENT_BLOCK_SIZE; + struct crypt_priv *ctx = priv; + int i; + + ctx->fpu_enabled = serpent_fpu_begin(ctx->fpu_enabled, nbytes); + + if (nbytes == bsize * SERPENT_PARALLEL_BLOCKS) { + serpent_enc_blk_xway(ctx->ctx, srcdst, srcdst); + return; + } + + for (i = 0; i < nbytes / bsize; i++, srcdst += bsize) + __serpent_encrypt(ctx->ctx, srcdst, srcdst); +} + +static void decrypt_callback(void *priv, u8 *srcdst, unsigned int nbytes) +{ + const unsigned int bsize = SERPENT_BLOCK_SIZE; + struct crypt_priv *ctx = priv; + int i; + + ctx->fpu_enabled = serpent_fpu_begin(ctx->fpu_enabled, nbytes); + + if (nbytes == bsize * SERPENT_PARALLEL_BLOCKS) { + serpent_dec_blk_xway(ctx->ctx, srcdst, srcdst); + return; + } + + for (i = 0; i < nbytes / bsize; i++, srcdst += bsize) + __serpent_decrypt(ctx->ctx, srcdst, srcdst); +} + +struct serpent_lrw_ctx { + struct lrw_table_ctx lrw_table; + struct serpent_ctx serpent_ctx; +}; + +static int lrw_serpent_setkey(struct crypto_tfm *tfm, const u8 *key, + unsigned int keylen) +{ + struct serpent_lrw_ctx *ctx = crypto_tfm_ctx(tfm); + int err; + + err = __serpent_setkey(&ctx->serpent_ctx, key, keylen - + SERPENT_BLOCK_SIZE); + if (err) + return err; + + return lrw_init_table(&ctx->lrw_table, key + keylen - + SERPENT_BLOCK_SIZE); +} + +static int lrw_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct serpent_lrw_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + be128 buf[SERPENT_PARALLEL_BLOCKS]; + struct crypt_priv crypt_ctx = { + .ctx = &ctx->serpent_ctx, + .fpu_enabled = false, + }; + struct lrw_crypt_req req = { + .tbuf = buf, + .tbuflen = sizeof(buf), + + .table_ctx = &ctx->lrw_table, + .crypt_ctx = &crypt_ctx, + .crypt_fn = encrypt_callback, + }; + int ret; + + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + ret = lrw_crypt(desc, dst, src, nbytes, &req); + serpent_fpu_end(crypt_ctx.fpu_enabled); + + return ret; +} + +static int lrw_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct serpent_lrw_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + be128 buf[SERPENT_PARALLEL_BLOCKS]; + struct crypt_priv crypt_ctx = { + .ctx = &ctx->serpent_ctx, + .fpu_enabled = false, + }; + struct lrw_crypt_req req = { + .tbuf = buf, + .tbuflen = sizeof(buf), + + .table_ctx = &ctx->lrw_table, + .crypt_ctx = &crypt_ctx, + .crypt_fn = decrypt_callback, + }; + int ret; + + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + ret = lrw_crypt(desc, dst, src, nbytes, &req); + serpent_fpu_end(crypt_ctx.fpu_enabled); + + return ret; +} + +static void lrw_exit_tfm(struct crypto_tfm *tfm) +{ + struct serpent_lrw_ctx *ctx = crypto_tfm_ctx(tfm); + + lrw_free_table(&ctx->lrw_table); +} + +struct serpent_xts_ctx { + struct serpent_ctx tweak_ctx; + struct serpent_ctx crypt_ctx; +}; + +static int xts_serpent_setkey(struct crypto_tfm *tfm, const u8 *key, + unsigned int keylen) +{ + struct serpent_xts_ctx *ctx = crypto_tfm_ctx(tfm); + u32 *flags = &tfm->crt_flags; + int err; + + /* key consists of keys of equal size concatenated, therefore + * the length must be even + */ + if (keylen % 2) { + *flags |= CRYPTO_TFM_RES_BAD_KEY_LEN; + return -EINVAL; + } + + /* first half of xts-key is for crypt */ + err = __serpent_setkey(&ctx->crypt_ctx, key, keylen / 2); + if (err) + return err; + + /* second half of xts-key is for tweak */ + return __serpent_setkey(&ctx->tweak_ctx, key + keylen / 2, keylen / 2); +} + +static int xts_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct serpent_xts_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + be128 buf[SERPENT_PARALLEL_BLOCKS]; + struct crypt_priv crypt_ctx = { + .ctx = &ctx->crypt_ctx, + .fpu_enabled = false, + }; + struct xts_crypt_req req = { + .tbuf = buf, + .tbuflen = sizeof(buf), + + .tweak_ctx = &ctx->tweak_ctx, + .tweak_fn = XTS_TWEAK_CAST(__serpent_encrypt), + .crypt_ctx = &crypt_ctx, + .crypt_fn = encrypt_callback, + }; + int ret; + + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + ret = xts_crypt(desc, dst, src, nbytes, &req); + serpent_fpu_end(crypt_ctx.fpu_enabled); + + return ret; +} + +static int xts_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct serpent_xts_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + be128 buf[SERPENT_PARALLEL_BLOCKS]; + struct crypt_priv crypt_ctx = { + .ctx = &ctx->crypt_ctx, + .fpu_enabled = false, + }; + struct xts_crypt_req req = { + .tbuf = buf, + .tbuflen = sizeof(buf), + + .tweak_ctx = &ctx->tweak_ctx, + .tweak_fn = XTS_TWEAK_CAST(__serpent_encrypt), + .crypt_ctx = &crypt_ctx, + .crypt_fn = decrypt_callback, + }; + int ret; + + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + ret = xts_crypt(desc, dst, src, nbytes, &req); + serpent_fpu_end(crypt_ctx.fpu_enabled); + + return ret; +} + +static int ablk_set_key(struct crypto_ablkcipher *tfm, const u8 *key, + unsigned int key_len) +{ + struct async_serpent_ctx *ctx = crypto_ablkcipher_ctx(tfm); + struct crypto_ablkcipher *child = &ctx->cryptd_tfm->base; + int err; + + crypto_ablkcipher_clear_flags(child, CRYPTO_TFM_REQ_MASK); + crypto_ablkcipher_set_flags(child, crypto_ablkcipher_get_flags(tfm) + & CRYPTO_TFM_REQ_MASK); + err = crypto_ablkcipher_setkey(child, key, key_len); + crypto_ablkcipher_set_flags(tfm, crypto_ablkcipher_get_flags(child) + & CRYPTO_TFM_RES_MASK); + return err; +} + +static int __ablk_encrypt(struct ablkcipher_request *req) +{ + struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); + struct async_serpent_ctx *ctx = crypto_ablkcipher_ctx(tfm); + struct blkcipher_desc desc; + + desc.tfm = cryptd_ablkcipher_child(ctx->cryptd_tfm); + desc.info = req->info; + desc.flags = 0; + + return crypto_blkcipher_crt(desc.tfm)->encrypt( + &desc, req->dst, req->src, req->nbytes); +} + +static int ablk_encrypt(struct ablkcipher_request *req) +{ + struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); + struct async_serpent_ctx *ctx = crypto_ablkcipher_ctx(tfm); + + if (!irq_fpu_usable()) { + struct ablkcipher_request *cryptd_req = + ablkcipher_request_ctx(req); + + memcpy(cryptd_req, req, sizeof(*req)); + ablkcipher_request_set_tfm(cryptd_req, &ctx->cryptd_tfm->base); + + return crypto_ablkcipher_encrypt(cryptd_req); + } else { + return __ablk_encrypt(req); + } +} + +static int ablk_decrypt(struct ablkcipher_request *req) +{ + struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); + struct async_serpent_ctx *ctx = crypto_ablkcipher_ctx(tfm); + + if (!irq_fpu_usable()) { + struct ablkcipher_request *cryptd_req = + ablkcipher_request_ctx(req); + + memcpy(cryptd_req, req, sizeof(*req)); + ablkcipher_request_set_tfm(cryptd_req, &ctx->cryptd_tfm->base); + + return crypto_ablkcipher_decrypt(cryptd_req); + } else { + struct blkcipher_desc desc; + + desc.tfm = cryptd_ablkcipher_child(ctx->cryptd_tfm); + desc.info = req->info; + desc.flags = 0; + + return crypto_blkcipher_crt(desc.tfm)->decrypt( + &desc, req->dst, req->src, req->nbytes); + } +} + +static void ablk_exit(struct crypto_tfm *tfm) +{ + struct async_serpent_ctx *ctx = crypto_tfm_ctx(tfm); + + cryptd_free_ablkcipher(ctx->cryptd_tfm); +} + +static int ablk_init(struct crypto_tfm *tfm) +{ + struct async_serpent_ctx *ctx = crypto_tfm_ctx(tfm); + struct cryptd_ablkcipher *cryptd_tfm; + char drv_name[CRYPTO_MAX_ALG_NAME]; + + snprintf(drv_name, sizeof(drv_name), "__driver-%s", + crypto_tfm_alg_driver_name(tfm)); + + cryptd_tfm = cryptd_alloc_ablkcipher(drv_name, 0, 0); + if (IS_ERR(cryptd_tfm)) + return PTR_ERR(cryptd_tfm); + + ctx->cryptd_tfm = cryptd_tfm; + tfm->crt_ablkcipher.reqsize = sizeof(struct ablkcipher_request) + + crypto_ablkcipher_reqsize(&cryptd_tfm->base); + + return 0; +} + +static struct crypto_alg serpent_algs[10] = { { + .cra_name = "__ecb-serpent-avx", + .cra_driver_name = "__driver-ecb-serpent-avx", + .cra_priority = 0, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = SERPENT_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct serpent_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(serpent_algs[0].cra_list), + .cra_u = { + .blkcipher = { + .min_keysize = SERPENT_MIN_KEY_SIZE, + .max_keysize = SERPENT_MAX_KEY_SIZE, + .setkey = serpent_setkey, + .encrypt = ecb_encrypt, + .decrypt = ecb_decrypt, + }, + }, +}, { + .cra_name = "__cbc-serpent-avx", + .cra_driver_name = "__driver-cbc-serpent-avx", + .cra_priority = 0, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = SERPENT_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct serpent_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(serpent_algs[1].cra_list), + .cra_u = { + .blkcipher = { + .min_keysize = SERPENT_MIN_KEY_SIZE, + .max_keysize = SERPENT_MAX_KEY_SIZE, + .setkey = serpent_setkey, + .encrypt = cbc_encrypt, + .decrypt = cbc_decrypt, + }, + }, +}, { + .cra_name = "__ctr-serpent-avx", + .cra_driver_name = "__driver-ctr-serpent-avx", + .cra_priority = 0, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = 1, + .cra_ctxsize = sizeof(struct serpent_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(serpent_algs[2].cra_list), + .cra_u = { + .blkcipher = { + .min_keysize = SERPENT_MIN_KEY_SIZE, + .max_keysize = SERPENT_MAX_KEY_SIZE, + .ivsize = SERPENT_BLOCK_SIZE, + .setkey = serpent_setkey, + .encrypt = ctr_crypt, + .decrypt = ctr_crypt, + }, + }, +}, { + .cra_name = "__lrw-serpent-avx", + .cra_driver_name = "__driver-lrw-serpent-avx", + .cra_priority = 0, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = SERPENT_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct serpent_lrw_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(serpent_algs[3].cra_list), + .cra_exit = lrw_exit_tfm, + .cra_u = { + .blkcipher = { + .min_keysize = SERPENT_MIN_KEY_SIZE + + SERPENT_BLOCK_SIZE, + .max_keysize = SERPENT_MAX_KEY_SIZE + + SERPENT_BLOCK_SIZE, + .ivsize = SERPENT_BLOCK_SIZE, + .setkey = lrw_serpent_setkey, + .encrypt = lrw_encrypt, + .decrypt = lrw_decrypt, + }, + }, +}, { + .cra_name = "__xts-serpent-avx", + .cra_driver_name = "__driver-xts-serpent-avx", + .cra_priority = 0, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = SERPENT_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct serpent_xts_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(serpent_algs[4].cra_list), + .cra_u = { + .blkcipher = { + .min_keysize = SERPENT_MIN_KEY_SIZE * 2, + .max_keysize = SERPENT_MAX_KEY_SIZE * 2, + .ivsize = SERPENT_BLOCK_SIZE, + .setkey = xts_serpent_setkey, + .encrypt = xts_encrypt, + .decrypt = xts_decrypt, + }, + }, +}, { + .cra_name = "ecb(serpent)", + .cra_driver_name = "ecb-serpent-avx", + .cra_priority = 500, + .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, + .cra_blocksize = SERPENT_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct async_serpent_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_ablkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(serpent_algs[5].cra_list), + .cra_init = ablk_init, + .cra_exit = ablk_exit, + .cra_u = { + .ablkcipher = { + .min_keysize = SERPENT_MIN_KEY_SIZE, + .max_keysize = SERPENT_MAX_KEY_SIZE, + .setkey = ablk_set_key, + .encrypt = ablk_encrypt, + .decrypt = ablk_decrypt, + }, + }, +}, { + .cra_name = "cbc(serpent)", + .cra_driver_name = "cbc-serpent-avx", + .cra_priority = 500, + .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, + .cra_blocksize = SERPENT_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct async_serpent_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_ablkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(serpent_algs[6].cra_list), + .cra_init = ablk_init, + .cra_exit = ablk_exit, + .cra_u = { + .ablkcipher = { + .min_keysize = SERPENT_MIN_KEY_SIZE, + .max_keysize = SERPENT_MAX_KEY_SIZE, + .ivsize = SERPENT_BLOCK_SIZE, + .setkey = ablk_set_key, + .encrypt = __ablk_encrypt, + .decrypt = ablk_decrypt, + }, + }, +}, { + .cra_name = "ctr(serpent)", + .cra_driver_name = "ctr-serpent-avx", + .cra_priority = 500, + .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, + .cra_blocksize = 1, + .cra_ctxsize = sizeof(struct async_serpent_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_ablkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(serpent_algs[7].cra_list), + .cra_init = ablk_init, + .cra_exit = ablk_exit, + .cra_u = { + .ablkcipher = { + .min_keysize = SERPENT_MIN_KEY_SIZE, + .max_keysize = SERPENT_MAX_KEY_SIZE, + .ivsize = SERPENT_BLOCK_SIZE, + .setkey = ablk_set_key, + .encrypt = ablk_encrypt, + .decrypt = ablk_encrypt, + .geniv = "chainiv", + }, + }, +}, { + .cra_name = "lrw(serpent)", + .cra_driver_name = "lrw-serpent-avx", + .cra_priority = 500, + .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, + .cra_blocksize = SERPENT_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct async_serpent_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_ablkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(serpent_algs[8].cra_list), + .cra_init = ablk_init, + .cra_exit = ablk_exit, + .cra_u = { + .ablkcipher = { + .min_keysize = SERPENT_MIN_KEY_SIZE + + SERPENT_BLOCK_SIZE, + .max_keysize = SERPENT_MAX_KEY_SIZE + + SERPENT_BLOCK_SIZE, + .ivsize = SERPENT_BLOCK_SIZE, + .setkey = ablk_set_key, + .encrypt = ablk_encrypt, + .decrypt = ablk_decrypt, + }, + }, +}, { + .cra_name = "xts(serpent)", + .cra_driver_name = "xts-serpent-avx", + .cra_priority = 500, + .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, + .cra_blocksize = SERPENT_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct async_serpent_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_ablkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(serpent_algs[9].cra_list), + .cra_init = ablk_init, + .cra_exit = ablk_exit, + .cra_u = { + .ablkcipher = { + .min_keysize = SERPENT_MIN_KEY_SIZE * 2, + .max_keysize = SERPENT_MAX_KEY_SIZE * 2, + .ivsize = SERPENT_BLOCK_SIZE, + .setkey = ablk_set_key, + .encrypt = ablk_encrypt, + .decrypt = ablk_decrypt, + }, + }, +} }; + +static int __init serpent_init(void) +{ + u64 xcr0; + + if (!cpu_has_avx || !cpu_has_osxsave) { + printk(KERN_INFO "AVX instructions are not detected.\n"); + return -ENODEV; + } + + xcr0 = xgetbv(XCR_XFEATURE_ENABLED_MASK); + if ((xcr0 & (XSTATE_SSE | XSTATE_YMM)) != (XSTATE_SSE | XSTATE_YMM)) { + printk(KERN_INFO "AVX detected but unusable.\n"); + return -ENODEV; + } + + return crypto_register_algs(serpent_algs, ARRAY_SIZE(serpent_algs)); +} + +static void __exit serpent_exit(void) +{ + crypto_unregister_algs(serpent_algs, ARRAY_SIZE(serpent_algs)); +} + +module_init(serpent_init); +module_exit(serpent_exit); + +MODULE_DESCRIPTION("Serpent Cipher Algorithm, AVX optimized"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("serpent"); diff --git a/crypto/Kconfig b/crypto/Kconfig index e00a4e49e01..2c1c2dfcc02 100644 --- a/crypto/Kconfig +++ b/crypto/Kconfig @@ -821,6 +821,26 @@ config CRYPTO_SERPENT_SSE2_586 See also: +config CRYPTO_SERPENT_AVX_X86_64 + tristate "Serpent cipher algorithm (x86_64/AVX)" + depends on X86 && 64BIT + select CRYPTO_ALGAPI + select CRYPTO_CRYPTD + select CRYPTO_SERPENT + select CRYPTO_LRW + select CRYPTO_XTS + help + Serpent cipher algorithm, by Anderson, Biham & Knudsen. + + Keys are allowed to be from 0 to 256 bits in length, in steps + of 8 bits. + + This module provides the Serpent cipher algorithm that processes + eight blocks parallel using the AVX instruction set. + + See also: + + config CRYPTO_TEA tristate "TEA, XTEA and XETA cipher algorithms" select CRYPTO_ALGAPI diff --git a/crypto/testmgr.c b/crypto/testmgr.c index 73b3ec6fe1a..36748a5996e 100644 --- a/crypto/testmgr.c +++ b/crypto/testmgr.c @@ -1534,6 +1534,21 @@ static int alg_test_null(const struct alg_test_desc *desc, /* Please keep this list sorted by algorithm name. */ static const struct alg_test_desc alg_test_descs[] = { { + .alg = "__cbc-serpent-avx", + .test = alg_test_null, + .suite = { + .cipher = { + .enc = { + .vecs = NULL, + .count = 0 + }, + .dec = { + .vecs = NULL, + .count = 0 + } + } + } + }, { .alg = "__cbc-serpent-sse2", .test = alg_test_null, .suite = { @@ -1578,6 +1593,21 @@ static const struct alg_test_desc alg_test_descs[] = { } } } + }, { + .alg = "__driver-cbc-serpent-avx", + .test = alg_test_null, + .suite = { + .cipher = { + .enc = { + .vecs = NULL, + .count = 0 + }, + .dec = { + .vecs = NULL, + .count = 0 + } + } + } }, { .alg = "__driver-cbc-serpent-sse2", .test = alg_test_null, @@ -1623,6 +1653,21 @@ static const struct alg_test_desc alg_test_descs[] = { } } } + }, { + .alg = "__driver-ecb-serpent-avx", + .test = alg_test_null, + .suite = { + .cipher = { + .enc = { + .vecs = NULL, + .count = 0 + }, + .dec = { + .vecs = NULL, + .count = 0 + } + } + } }, { .alg = "__driver-ecb-serpent-sse2", .test = alg_test_null, @@ -1835,6 +1880,21 @@ static const struct alg_test_desc alg_test_descs[] = { } } } + }, { + .alg = "cryptd(__driver-ecb-serpent-avx)", + .test = alg_test_null, + .suite = { + .cipher = { + .enc = { + .vecs = NULL, + .count = 0 + }, + .dec = { + .vecs = NULL, + .count = 0 + } + } + } }, { .alg = "cryptd(__driver-ecb-serpent-sse2)", .test = alg_test_null, -- cgit v1.2.3 From 31b4cd2907322bd56e4090168e4ce8f8dcd17820 Mon Sep 17 00:00:00 2001 From: Jussi Kivilinna Date: Tue, 12 Jun 2012 16:52:04 +0800 Subject: crypto: testmgr - add ecb(arc4) speed tests Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu --- crypto/tcrypt.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'crypto') diff --git a/crypto/tcrypt.c b/crypto/tcrypt.c index 2af879786e7..58ad4fe5d20 100644 --- a/crypto/tcrypt.c +++ b/crypto/tcrypt.c @@ -1342,6 +1342,11 @@ static int do_test(int m) speed_template_32_64); break; + case 208: + test_cipher_speed("ecb(arc4)", ENCRYPT, sec, NULL, 0, + speed_template_8); + break; + case 300: /* fall through */ @@ -1589,6 +1594,11 @@ static int do_test(int m) speed_template_32_48_64); break; + case 505: + test_acipher_speed("ecb(arc4)", ENCRYPT, sec, NULL, 0, + speed_template_8); + break; + case 1000: test_available(); break; -- cgit v1.2.3 From ce6dd368984068c8938e8d3fba8e292ef63cee97 Mon Sep 17 00:00:00 2001 From: Jussi Kivilinna Date: Sat, 9 Jun 2012 18:25:40 +0300 Subject: crypto: arc4 - improve performance by adding ecb(arc4) Currently arc4.c provides simple one-byte blocksize cipher which is wrapped by ecb() module, giving function call overhead on every encrypted byte. This patch adds ecb(arc4) directly into arc4.c for higher performance. tcrypt results (speed ratios: new/old): AMD Phenom II, x86-64 : x2.7 Intel Core 2, x86-64 : x1.9 Intel Atom N260, i386 : x1.4 Cc: Jon Oberheide Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu --- crypto/arc4.c | 109 ++++++++++++++++++++++++++++++++++++++++++++++------------ 1 file changed, 87 insertions(+), 22 deletions(-) (limited to 'crypto') diff --git a/crypto/arc4.c b/crypto/arc4.c index 0d12a96da1d..07913fc52c4 100644 --- a/crypto/arc4.c +++ b/crypto/arc4.c @@ -11,9 +11,11 @@ * (at your option) any later version. * */ + #include #include #include +#include #define ARC4_MIN_KEY_SIZE 1 #define ARC4_MAX_KEY_SIZE 256 @@ -48,51 +50,114 @@ static int arc4_set_key(struct crypto_tfm *tfm, const u8 *in_key, return 0; } -static void arc4_crypt(struct crypto_tfm *tfm, u8 *out, const u8 *in) +static void arc4_crypt(struct arc4_ctx *ctx, u8 *out, const u8 *in, + unsigned int len) { - struct arc4_ctx *ctx = crypto_tfm_ctx(tfm); - u8 *const S = ctx->S; - u8 x = ctx->x; - u8 y = ctx->y; - u8 a, b; + u8 x, y, a, b; + u8 ty, ta, tb; + + if (len == 0) + return; + + x = ctx->x; + y = ctx->y; a = S[x]; y = (y + a) & 0xff; b = S[y]; - S[x] = b; - S[y] = a; - x = (x + 1) & 0xff; - *out++ = *in ^ S[(a + b) & 0xff]; + + do { + S[y] = a; + a = (a + b) & 0xff; + S[x] = b; + x = (x + 1) & 0xff; + ta = S[x]; + ty = (y + ta) & 0xff; + tb = S[ty]; + *out++ = *in++ ^ S[a]; + if (--len == 0) + break; + y = ty; + a = ta; + b = tb; + } while (true); ctx->x = x; ctx->y = y; } -static struct crypto_alg arc4_alg = { +static void arc4_crypt_one(struct crypto_tfm *tfm, u8 *out, const u8 *in) +{ + arc4_crypt(crypto_tfm_ctx(tfm), out, in, 1); +} + +static int ecb_arc4_crypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct arc4_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + + err = blkcipher_walk_virt(desc, &walk); + + while (walk.nbytes > 0) { + u8 *wsrc = walk.src.virt.addr; + u8 *wdst = walk.dst.virt.addr; + + arc4_crypt(ctx, wdst, wsrc, walk.nbytes); + + err = blkcipher_walk_done(desc, &walk, 0); + } + + return err; +} + +static struct crypto_alg arc4_algs[2] = { { .cra_name = "arc4", .cra_flags = CRYPTO_ALG_TYPE_CIPHER, .cra_blocksize = ARC4_BLOCK_SIZE, .cra_ctxsize = sizeof(struct arc4_ctx), .cra_module = THIS_MODULE, - .cra_list = LIST_HEAD_INIT(arc4_alg.cra_list), - .cra_u = { .cipher = { - .cia_min_keysize = ARC4_MIN_KEY_SIZE, - .cia_max_keysize = ARC4_MAX_KEY_SIZE, - .cia_setkey = arc4_set_key, - .cia_encrypt = arc4_crypt, - .cia_decrypt = arc4_crypt } } -}; + .cra_u = { + .cipher = { + .cia_min_keysize = ARC4_MIN_KEY_SIZE, + .cia_max_keysize = ARC4_MAX_KEY_SIZE, + .cia_setkey = arc4_set_key, + .cia_encrypt = arc4_crypt_one, + .cia_decrypt = arc4_crypt_one, + }, + }, +}, { + .cra_name = "ecb(arc4)", + .cra_priority = 100, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = ARC4_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct arc4_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_u = { + .blkcipher = { + .min_keysize = ARC4_MIN_KEY_SIZE, + .max_keysize = ARC4_MAX_KEY_SIZE, + .setkey = arc4_set_key, + .encrypt = ecb_arc4_crypt, + .decrypt = ecb_arc4_crypt, + }, + }, +} }; static int __init arc4_init(void) { - return crypto_register_alg(&arc4_alg); + return crypto_register_algs(arc4_algs, ARRAY_SIZE(arc4_algs)); } - static void __exit arc4_exit(void) { - crypto_unregister_alg(&arc4_alg); + crypto_unregister_algs(arc4_algs, ARRAY_SIZE(arc4_algs)); } module_init(arc4_init); -- cgit v1.2.3 From d366db605c8c4a9878589bc4a87e55f6063184ac Mon Sep 17 00:00:00 2001 From: Jussi Kivilinna Date: Sat, 9 Jun 2012 18:25:46 +0300 Subject: crypto: arc4 - improve performance by using u32 for ctx and variables This patch changes u8 in struct arc4_ctx and variables to u32 (as AMD seems to have problem with u8 array). Below are tcrypt results of old 1-byte block cipher versus ecb(arc4) with u8 and ecb(arc4) with u32. tcrypt results, x86-64 (speed ratios: new-u32/old, new-u8/old): u32 u8 AMD Phenom II : x3.6 x2.7 Intel Core 2 : x2.0 x1.9 tcrypt results, i386 (speed ratios: new-u32/old, new-u8/old): u32 u8 Intel Atom N260 : x1.5 x1.4 Cc: Jon Oberheide Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu --- crypto/arc4.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'crypto') diff --git a/crypto/arc4.c b/crypto/arc4.c index 07913fc52c4..5a772c3657d 100644 --- a/crypto/arc4.c +++ b/crypto/arc4.c @@ -22,8 +22,8 @@ #define ARC4_BLOCK_SIZE 1 struct arc4_ctx { - u8 S[256]; - u8 x, y; + u32 S[256]; + u32 x, y; }; static int arc4_set_key(struct crypto_tfm *tfm, const u8 *in_key, @@ -39,7 +39,7 @@ static int arc4_set_key(struct crypto_tfm *tfm, const u8 *in_key, ctx->S[i] = i; for (i = 0; i < 256; i++) { - u8 a = ctx->S[i]; + u32 a = ctx->S[i]; j = (j + in_key[k] + a) & 0xff; ctx->S[i] = ctx->S[j]; ctx->S[j] = a; @@ -53,9 +53,9 @@ static int arc4_set_key(struct crypto_tfm *tfm, const u8 *in_key, static void arc4_crypt(struct arc4_ctx *ctx, u8 *out, const u8 *in, unsigned int len) { - u8 *const S = ctx->S; - u8 x, y, a, b; - u8 ty, ta, tb; + u32 *const S = ctx->S; + u32 x, y, a, b; + u32 ty, ta, tb; if (len == 0) return; -- cgit v1.2.3 From 398710379f516012c52d2ae396a9ba919bd6a7ab Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Fri, 22 Jun 2012 20:08:29 +0800 Subject: crypto: algapi - Move larval completion into algboss It has been observed that sometimes the crypto allocation code will get stuck for 60 seconds or multiples thereof. This is usually caused by an algorithm failing to pass the self-test. If an algorithm fails to be constructed, we will immediately notify all larval waiters. However, if it succeeds in construction, but then fails the self-test, we won't notify anyone at all. This patch fixes this by merging the notification in the case where the algorithm fails to be constructed with that of the the case where it pases the self-test. This way regardless of what happens, we'll give the larval waiters an answer. Signed-off-by: Herbert Xu --- crypto/algapi.c | 17 ----------------- crypto/algboss.c | 17 +++++++++-------- crypto/internal.h | 1 - 3 files changed, 9 insertions(+), 26 deletions(-) (limited to 'crypto') diff --git a/crypto/algapi.c b/crypto/algapi.c index 056571b8544..c3b9bfeeb7f 100644 --- a/crypto/algapi.c +++ b/crypto/algapi.c @@ -24,22 +24,6 @@ static LIST_HEAD(crypto_template_list); -void crypto_larval_error(const char *name, u32 type, u32 mask) -{ - struct crypto_alg *alg; - - alg = crypto_alg_lookup(name, type, mask); - - if (alg) { - if (crypto_is_larval(alg)) { - struct crypto_larval *larval = (void *)alg; - complete_all(&larval->completion); - } - crypto_mod_put(alg); - } -} -EXPORT_SYMBOL_GPL(crypto_larval_error); - static inline int crypto_set_driver_name(struct crypto_alg *alg) { static const char suffix[] = "-generic"; @@ -295,7 +279,6 @@ found: continue; larval->adult = alg; - complete_all(&larval->completion); continue; } diff --git a/crypto/algboss.c b/crypto/algboss.c index 791d194958f..f97027e7d99 100644 --- a/crypto/algboss.c +++ b/crypto/algboss.c @@ -11,6 +11,7 @@ */ #include +#include #include #include #include @@ -47,6 +48,8 @@ struct cryptomgr_param { char larval[CRYPTO_MAX_ALG_NAME]; char template[CRYPTO_MAX_ALG_NAME]; + struct completion *completion; + u32 otype; u32 omask; }; @@ -66,7 +69,7 @@ static int cryptomgr_probe(void *data) tmpl = crypto_lookup_template(param->template); if (!tmpl) - goto err; + goto out; do { if (tmpl->create) { @@ -83,16 +86,10 @@ static int cryptomgr_probe(void *data) crypto_tmpl_put(tmpl); - if (err) - goto err; - out: + complete(param->completion); kfree(param); module_put_and_exit(0); - -err: - crypto_larval_error(param->larval, param->otype, param->omask); - goto out; } static int cryptomgr_schedule_probe(struct crypto_larval *larval) @@ -192,10 +189,14 @@ static int cryptomgr_schedule_probe(struct crypto_larval *larval) memcpy(param->larval, larval->alg.cra_name, CRYPTO_MAX_ALG_NAME); + param->completion = &larval->completion; + thread = kthread_run(cryptomgr_probe, param, "cryptomgr_probe"); if (IS_ERR(thread)) goto err_free_param; + wait_for_completion_interruptible(&larval->completion); + return NOTIFY_STOP; err_free_param: diff --git a/crypto/internal.h b/crypto/internal.h index b865ca1a861..9ebedae3fb5 100644 --- a/crypto/internal.h +++ b/crypto/internal.h @@ -83,7 +83,6 @@ void crypto_exit_compress_ops(struct crypto_tfm *tfm); struct crypto_larval *crypto_larval_alloc(const char *name, u32 type, u32 mask); void crypto_larval_kill(struct crypto_alg *alg); struct crypto_alg *crypto_larval_lookup(const char *name, u32 type, u32 mask); -void crypto_larval_error(const char *name, u32 type, u32 mask); void crypto_alg_tested(const char *name, int err); void crypto_remove_spawns(struct crypto_alg *alg, struct list_head *list, -- cgit v1.2.3 From ffaf9156320a077ebf9c5b9a5cf987689dc1a6b3 Mon Sep 17 00:00:00 2001 From: Jussi Kivilinna Date: Mon, 18 Jun 2012 14:06:58 +0300 Subject: crypto: ablk_helper - move ablk_* functions from serpent-sse2/avx glue code to shared module Move ablk-* functions to separate module to share common code between cipher implementations. Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu --- arch/x86/crypto/Makefile | 2 + arch/x86/crypto/ablk_helper.c | 150 ++++++++++++++++++++++++++++++ arch/x86/crypto/serpent_avx_glue.c | 115 ++--------------------- arch/x86/crypto/serpent_sse2_glue.c | 115 ++--------------------- arch/x86/include/asm/crypto/ablk_helper.h | 29 ++++++ crypto/Kconfig | 8 ++ 6 files changed, 201 insertions(+), 218 deletions(-) create mode 100644 arch/x86/crypto/ablk_helper.c create mode 100644 arch/x86/include/asm/crypto/ablk_helper.h (limited to 'crypto') diff --git a/arch/x86/crypto/Makefile b/arch/x86/crypto/Makefile index 83caa4b948c..ad746916f91 100644 --- a/arch/x86/crypto/Makefile +++ b/arch/x86/crypto/Makefile @@ -2,6 +2,8 @@ # Arch-specific CryptoAPI modules. # +obj-$(CONFIG_CRYPTO_ABLK_HELPER_X86) += ablk_helper.o + obj-$(CONFIG_CRYPTO_AES_586) += aes-i586.o obj-$(CONFIG_CRYPTO_TWOFISH_586) += twofish-i586.o obj-$(CONFIG_CRYPTO_SALSA20_586) += salsa20-i586.o diff --git a/arch/x86/crypto/ablk_helper.c b/arch/x86/crypto/ablk_helper.c new file mode 100644 index 00000000000..284ca3bc238 --- /dev/null +++ b/arch/x86/crypto/ablk_helper.c @@ -0,0 +1,150 @@ +/* + * Shared async block cipher helpers + * + * Copyright (c) 2012 Jussi Kivilinna + * + * Based on aesni-intel_glue.c by: + * Copyright (C) 2008, Intel Corp. + * Author: Huang Ying + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 + * USA + * + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +int ablk_set_key(struct crypto_ablkcipher *tfm, const u8 *key, + unsigned int key_len) +{ + struct async_helper_ctx *ctx = crypto_ablkcipher_ctx(tfm); + struct crypto_ablkcipher *child = &ctx->cryptd_tfm->base; + int err; + + crypto_ablkcipher_clear_flags(child, CRYPTO_TFM_REQ_MASK); + crypto_ablkcipher_set_flags(child, crypto_ablkcipher_get_flags(tfm) + & CRYPTO_TFM_REQ_MASK); + err = crypto_ablkcipher_setkey(child, key, key_len); + crypto_ablkcipher_set_flags(tfm, crypto_ablkcipher_get_flags(child) + & CRYPTO_TFM_RES_MASK); + return err; +} +EXPORT_SYMBOL_GPL(ablk_set_key); + +int __ablk_encrypt(struct ablkcipher_request *req) +{ + struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); + struct async_helper_ctx *ctx = crypto_ablkcipher_ctx(tfm); + struct blkcipher_desc desc; + + desc.tfm = cryptd_ablkcipher_child(ctx->cryptd_tfm); + desc.info = req->info; + desc.flags = 0; + + return crypto_blkcipher_crt(desc.tfm)->encrypt( + &desc, req->dst, req->src, req->nbytes); +} +EXPORT_SYMBOL_GPL(__ablk_encrypt); + +int ablk_encrypt(struct ablkcipher_request *req) +{ + struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); + struct async_helper_ctx *ctx = crypto_ablkcipher_ctx(tfm); + + if (!irq_fpu_usable()) { + struct ablkcipher_request *cryptd_req = + ablkcipher_request_ctx(req); + + memcpy(cryptd_req, req, sizeof(*req)); + ablkcipher_request_set_tfm(cryptd_req, &ctx->cryptd_tfm->base); + + return crypto_ablkcipher_encrypt(cryptd_req); + } else { + return __ablk_encrypt(req); + } +} +EXPORT_SYMBOL_GPL(ablk_encrypt); + +int ablk_decrypt(struct ablkcipher_request *req) +{ + struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); + struct async_helper_ctx *ctx = crypto_ablkcipher_ctx(tfm); + + if (!irq_fpu_usable()) { + struct ablkcipher_request *cryptd_req = + ablkcipher_request_ctx(req); + + memcpy(cryptd_req, req, sizeof(*req)); + ablkcipher_request_set_tfm(cryptd_req, &ctx->cryptd_tfm->base); + + return crypto_ablkcipher_decrypt(cryptd_req); + } else { + struct blkcipher_desc desc; + + desc.tfm = cryptd_ablkcipher_child(ctx->cryptd_tfm); + desc.info = req->info; + desc.flags = 0; + + return crypto_blkcipher_crt(desc.tfm)->decrypt( + &desc, req->dst, req->src, req->nbytes); + } +} +EXPORT_SYMBOL_GPL(ablk_decrypt); + +void ablk_exit(struct crypto_tfm *tfm) +{ + struct async_helper_ctx *ctx = crypto_tfm_ctx(tfm); + + cryptd_free_ablkcipher(ctx->cryptd_tfm); +} +EXPORT_SYMBOL_GPL(ablk_exit); + +void ablk_init_common(struct crypto_tfm *tfm, + struct cryptd_ablkcipher *cryptd_tfm) +{ + struct async_helper_ctx *ctx = crypto_tfm_ctx(tfm); + + ctx->cryptd_tfm = cryptd_tfm; + tfm->crt_ablkcipher.reqsize = sizeof(struct ablkcipher_request) + + crypto_ablkcipher_reqsize(&cryptd_tfm->base); +} +EXPORT_SYMBOL_GPL(ablk_init_common); + +int ablk_init(struct crypto_tfm *tfm) +{ + struct cryptd_ablkcipher *cryptd_tfm; + char drv_name[CRYPTO_MAX_ALG_NAME]; + + snprintf(drv_name, sizeof(drv_name), "__driver-%s", + crypto_tfm_alg_driver_name(tfm)); + + cryptd_tfm = cryptd_alloc_ablkcipher(drv_name, 0, 0); + if (IS_ERR(cryptd_tfm)) + return PTR_ERR(cryptd_tfm); + + ablk_init_common(tfm, cryptd_tfm); + + return 0; +} +EXPORT_SYMBOL_GPL(ablk_init); + +MODULE_LICENSE("GPL"); diff --git a/arch/x86/crypto/serpent_avx_glue.c b/arch/x86/crypto/serpent_avx_glue.c index dd81bab4f11..31eb567cc89 100644 --- a/arch/x86/crypto/serpent_avx_glue.c +++ b/arch/x86/crypto/serpent_avx_glue.c @@ -40,14 +40,11 @@ #include #include #include +#include #include #include #include -struct async_serpent_ctx { - struct cryptd_ablkcipher *cryptd_tfm; -}; - static inline bool serpent_fpu_begin(bool fpu_enabled, unsigned int nbytes) { if (fpu_enabled) @@ -593,106 +590,6 @@ static int xts_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, return ret; } -static int ablk_set_key(struct crypto_ablkcipher *tfm, const u8 *key, - unsigned int key_len) -{ - struct async_serpent_ctx *ctx = crypto_ablkcipher_ctx(tfm); - struct crypto_ablkcipher *child = &ctx->cryptd_tfm->base; - int err; - - crypto_ablkcipher_clear_flags(child, CRYPTO_TFM_REQ_MASK); - crypto_ablkcipher_set_flags(child, crypto_ablkcipher_get_flags(tfm) - & CRYPTO_TFM_REQ_MASK); - err = crypto_ablkcipher_setkey(child, key, key_len); - crypto_ablkcipher_set_flags(tfm, crypto_ablkcipher_get_flags(child) - & CRYPTO_TFM_RES_MASK); - return err; -} - -static int __ablk_encrypt(struct ablkcipher_request *req) -{ - struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); - struct async_serpent_ctx *ctx = crypto_ablkcipher_ctx(tfm); - struct blkcipher_desc desc; - - desc.tfm = cryptd_ablkcipher_child(ctx->cryptd_tfm); - desc.info = req->info; - desc.flags = 0; - - return crypto_blkcipher_crt(desc.tfm)->encrypt( - &desc, req->dst, req->src, req->nbytes); -} - -static int ablk_encrypt(struct ablkcipher_request *req) -{ - struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); - struct async_serpent_ctx *ctx = crypto_ablkcipher_ctx(tfm); - - if (!irq_fpu_usable()) { - struct ablkcipher_request *cryptd_req = - ablkcipher_request_ctx(req); - - memcpy(cryptd_req, req, sizeof(*req)); - ablkcipher_request_set_tfm(cryptd_req, &ctx->cryptd_tfm->base); - - return crypto_ablkcipher_encrypt(cryptd_req); - } else { - return __ablk_encrypt(req); - } -} - -static int ablk_decrypt(struct ablkcipher_request *req) -{ - struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); - struct async_serpent_ctx *ctx = crypto_ablkcipher_ctx(tfm); - - if (!irq_fpu_usable()) { - struct ablkcipher_request *cryptd_req = - ablkcipher_request_ctx(req); - - memcpy(cryptd_req, req, sizeof(*req)); - ablkcipher_request_set_tfm(cryptd_req, &ctx->cryptd_tfm->base); - - return crypto_ablkcipher_decrypt(cryptd_req); - } else { - struct blkcipher_desc desc; - - desc.tfm = cryptd_ablkcipher_child(ctx->cryptd_tfm); - desc.info = req->info; - desc.flags = 0; - - return crypto_blkcipher_crt(desc.tfm)->decrypt( - &desc, req->dst, req->src, req->nbytes); - } -} - -static void ablk_exit(struct crypto_tfm *tfm) -{ - struct async_serpent_ctx *ctx = crypto_tfm_ctx(tfm); - - cryptd_free_ablkcipher(ctx->cryptd_tfm); -} - -static int ablk_init(struct crypto_tfm *tfm) -{ - struct async_serpent_ctx *ctx = crypto_tfm_ctx(tfm); - struct cryptd_ablkcipher *cryptd_tfm; - char drv_name[CRYPTO_MAX_ALG_NAME]; - - snprintf(drv_name, sizeof(drv_name), "__driver-%s", - crypto_tfm_alg_driver_name(tfm)); - - cryptd_tfm = cryptd_alloc_ablkcipher(drv_name, 0, 0); - if (IS_ERR(cryptd_tfm)) - return PTR_ERR(cryptd_tfm); - - ctx->cryptd_tfm = cryptd_tfm; - tfm->crt_ablkcipher.reqsize = sizeof(struct ablkcipher_request) + - crypto_ablkcipher_reqsize(&cryptd_tfm->base); - - return 0; -} - static struct crypto_alg serpent_algs[10] = { { .cra_name = "__ecb-serpent-avx", .cra_driver_name = "__driver-ecb-serpent-avx", @@ -805,7 +702,7 @@ static struct crypto_alg serpent_algs[10] = { { .cra_priority = 500, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = SERPENT_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct async_serpent_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, @@ -827,7 +724,7 @@ static struct crypto_alg serpent_algs[10] = { { .cra_priority = 500, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = SERPENT_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct async_serpent_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, @@ -850,7 +747,7 @@ static struct crypto_alg serpent_algs[10] = { { .cra_priority = 500, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = 1, - .cra_ctxsize = sizeof(struct async_serpent_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, @@ -874,7 +771,7 @@ static struct crypto_alg serpent_algs[10] = { { .cra_priority = 500, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = SERPENT_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct async_serpent_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, @@ -899,7 +796,7 @@ static struct crypto_alg serpent_algs[10] = { { .cra_priority = 500, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = SERPENT_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct async_serpent_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, diff --git a/arch/x86/crypto/serpent_sse2_glue.c b/arch/x86/crypto/serpent_sse2_glue.c index deecd25c129..805c91fda7a 100644 --- a/arch/x86/crypto/serpent_sse2_glue.c +++ b/arch/x86/crypto/serpent_sse2_glue.c @@ -43,14 +43,11 @@ #include #include #include +#include #include #include #include -struct async_serpent_ctx { - struct cryptd_ablkcipher *cryptd_tfm; -}; - static inline bool serpent_fpu_begin(bool fpu_enabled, unsigned int nbytes) { if (fpu_enabled) @@ -596,106 +593,6 @@ static int xts_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, return ret; } -static int ablk_set_key(struct crypto_ablkcipher *tfm, const u8 *key, - unsigned int key_len) -{ - struct async_serpent_ctx *ctx = crypto_ablkcipher_ctx(tfm); - struct crypto_ablkcipher *child = &ctx->cryptd_tfm->base; - int err; - - crypto_ablkcipher_clear_flags(child, CRYPTO_TFM_REQ_MASK); - crypto_ablkcipher_set_flags(child, crypto_ablkcipher_get_flags(tfm) - & CRYPTO_TFM_REQ_MASK); - err = crypto_ablkcipher_setkey(child, key, key_len); - crypto_ablkcipher_set_flags(tfm, crypto_ablkcipher_get_flags(child) - & CRYPTO_TFM_RES_MASK); - return err; -} - -static int __ablk_encrypt(struct ablkcipher_request *req) -{ - struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); - struct async_serpent_ctx *ctx = crypto_ablkcipher_ctx(tfm); - struct blkcipher_desc desc; - - desc.tfm = cryptd_ablkcipher_child(ctx->cryptd_tfm); - desc.info = req->info; - desc.flags = 0; - - return crypto_blkcipher_crt(desc.tfm)->encrypt( - &desc, req->dst, req->src, req->nbytes); -} - -static int ablk_encrypt(struct ablkcipher_request *req) -{ - struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); - struct async_serpent_ctx *ctx = crypto_ablkcipher_ctx(tfm); - - if (!irq_fpu_usable()) { - struct ablkcipher_request *cryptd_req = - ablkcipher_request_ctx(req); - - memcpy(cryptd_req, req, sizeof(*req)); - ablkcipher_request_set_tfm(cryptd_req, &ctx->cryptd_tfm->base); - - return crypto_ablkcipher_encrypt(cryptd_req); - } else { - return __ablk_encrypt(req); - } -} - -static int ablk_decrypt(struct ablkcipher_request *req) -{ - struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); - struct async_serpent_ctx *ctx = crypto_ablkcipher_ctx(tfm); - - if (!irq_fpu_usable()) { - struct ablkcipher_request *cryptd_req = - ablkcipher_request_ctx(req); - - memcpy(cryptd_req, req, sizeof(*req)); - ablkcipher_request_set_tfm(cryptd_req, &ctx->cryptd_tfm->base); - - return crypto_ablkcipher_decrypt(cryptd_req); - } else { - struct blkcipher_desc desc; - - desc.tfm = cryptd_ablkcipher_child(ctx->cryptd_tfm); - desc.info = req->info; - desc.flags = 0; - - return crypto_blkcipher_crt(desc.tfm)->decrypt( - &desc, req->dst, req->src, req->nbytes); - } -} - -static void ablk_exit(struct crypto_tfm *tfm) -{ - struct async_serpent_ctx *ctx = crypto_tfm_ctx(tfm); - - cryptd_free_ablkcipher(ctx->cryptd_tfm); -} - -static int ablk_init(struct crypto_tfm *tfm) -{ - struct async_serpent_ctx *ctx = crypto_tfm_ctx(tfm); - struct cryptd_ablkcipher *cryptd_tfm; - char drv_name[CRYPTO_MAX_ALG_NAME]; - - snprintf(drv_name, sizeof(drv_name), "__driver-%s", - crypto_tfm_alg_driver_name(tfm)); - - cryptd_tfm = cryptd_alloc_ablkcipher(drv_name, 0, 0); - if (IS_ERR(cryptd_tfm)) - return PTR_ERR(cryptd_tfm); - - ctx->cryptd_tfm = cryptd_tfm; - tfm->crt_ablkcipher.reqsize = sizeof(struct ablkcipher_request) + - crypto_ablkcipher_reqsize(&cryptd_tfm->base); - - return 0; -} - static struct crypto_alg serpent_algs[10] = { { .cra_name = "__ecb-serpent-sse2", .cra_driver_name = "__driver-ecb-serpent-sse2", @@ -808,7 +705,7 @@ static struct crypto_alg serpent_algs[10] = { { .cra_priority = 400, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = SERPENT_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct async_serpent_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, @@ -830,7 +727,7 @@ static struct crypto_alg serpent_algs[10] = { { .cra_priority = 400, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = SERPENT_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct async_serpent_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, @@ -853,7 +750,7 @@ static struct crypto_alg serpent_algs[10] = { { .cra_priority = 400, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = 1, - .cra_ctxsize = sizeof(struct async_serpent_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, @@ -877,7 +774,7 @@ static struct crypto_alg serpent_algs[10] = { { .cra_priority = 400, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = SERPENT_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct async_serpent_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, @@ -902,7 +799,7 @@ static struct crypto_alg serpent_algs[10] = { { .cra_priority = 400, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = SERPENT_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct async_serpent_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, diff --git a/arch/x86/include/asm/crypto/ablk_helper.h b/arch/x86/include/asm/crypto/ablk_helper.h new file mode 100644 index 00000000000..6d6b37c6f5d --- /dev/null +++ b/arch/x86/include/asm/crypto/ablk_helper.h @@ -0,0 +1,29 @@ +/* + * Shared async block cipher helpers + */ + +#ifndef _CRYPTO_ABLK_HELPER_H +#define _CRYPTO_ABLK_HELPER_H + +#include +#include +#include + +struct async_helper_ctx { + struct cryptd_ablkcipher *cryptd_tfm; +}; + +extern int ablk_set_key(struct crypto_ablkcipher *tfm, const u8 *key, + unsigned int key_len); + +extern int __ablk_encrypt(struct ablkcipher_request *req); + +extern int ablk_encrypt(struct ablkcipher_request *req); + +extern int ablk_decrypt(struct ablkcipher_request *req); + +extern void ablk_exit(struct crypto_tfm *tfm); + +extern int ablk_init(struct crypto_tfm *tfm); + +#endif /* _CRYPTO_ABLK_HELPER_H */ diff --git a/crypto/Kconfig b/crypto/Kconfig index 2c1c2dfcc02..8e9145c07d8 100644 --- a/crypto/Kconfig +++ b/crypto/Kconfig @@ -174,6 +174,11 @@ config CRYPTO_TEST help Quick & dirty crypto test module. +config CRYPTO_ABLK_HELPER_X86 + tristate + depends on X86 + select CRYPTO_CRYPTD + comment "Authenticated Encryption with Associated Data" config CRYPTO_CCM @@ -786,6 +791,7 @@ config CRYPTO_SERPENT_SSE2_X86_64 depends on X86 && 64BIT select CRYPTO_ALGAPI select CRYPTO_CRYPTD + select CRYPTO_ABLK_HELPER_X86 select CRYPTO_SERPENT select CRYPTO_LRW select CRYPTO_XTS @@ -806,6 +812,7 @@ config CRYPTO_SERPENT_SSE2_586 depends on X86 && !64BIT select CRYPTO_ALGAPI select CRYPTO_CRYPTD + select CRYPTO_ABLK_HELPER_X86 select CRYPTO_SERPENT select CRYPTO_LRW select CRYPTO_XTS @@ -826,6 +833,7 @@ config CRYPTO_SERPENT_AVX_X86_64 depends on X86 && 64BIT select CRYPTO_ALGAPI select CRYPTO_CRYPTD + select CRYPTO_ABLK_HELPER_X86 select CRYPTO_SERPENT select CRYPTO_LRW select CRYPTO_XTS -- cgit v1.2.3 From 30a04008827b58c4aafbd1d6a27d6b6ed239e993 Mon Sep 17 00:00:00 2001 From: Jussi Kivilinna Date: Mon, 18 Jun 2012 14:07:03 +0300 Subject: crypto: twofish-avx - change to use shared ablk_* functions Remove duplicate ablk_* functions and make use of ablk_helper module instead. Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu --- arch/x86/crypto/twofish_avx_glue.c | 116 ++----------------------------------- crypto/Kconfig | 1 + 2 files changed, 7 insertions(+), 110 deletions(-) (limited to 'crypto') diff --git a/arch/x86/crypto/twofish_avx_glue.c b/arch/x86/crypto/twofish_avx_glue.c index 599f19e4bef..cabe058eba1 100644 --- a/arch/x86/crypto/twofish_avx_glue.c +++ b/arch/x86/crypto/twofish_avx_glue.c @@ -39,6 +39,7 @@ #include #include #include +#include #include #include #include @@ -95,11 +96,6 @@ static inline void twofish_dec_blk_xway(struct twofish_ctx *ctx, u8 *dst, } - -struct async_twofish_ctx { - struct cryptd_ablkcipher *cryptd_tfm; -}; - static inline bool twofish_fpu_begin(bool fpu_enabled, unsigned int nbytes) { if (fpu_enabled) @@ -730,106 +726,6 @@ static int xts_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, return ret; } -static int ablk_set_key(struct crypto_ablkcipher *tfm, const u8 *key, - unsigned int key_len) -{ - struct async_twofish_ctx *ctx = crypto_ablkcipher_ctx(tfm); - struct crypto_ablkcipher *child = &ctx->cryptd_tfm->base; - int err; - - crypto_ablkcipher_clear_flags(child, CRYPTO_TFM_REQ_MASK); - crypto_ablkcipher_set_flags(child, crypto_ablkcipher_get_flags(tfm) - & CRYPTO_TFM_REQ_MASK); - err = crypto_ablkcipher_setkey(child, key, key_len); - crypto_ablkcipher_set_flags(tfm, crypto_ablkcipher_get_flags(child) - & CRYPTO_TFM_RES_MASK); - return err; -} - -static int __ablk_encrypt(struct ablkcipher_request *req) -{ - struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); - struct async_twofish_ctx *ctx = crypto_ablkcipher_ctx(tfm); - struct blkcipher_desc desc; - - desc.tfm = cryptd_ablkcipher_child(ctx->cryptd_tfm); - desc.info = req->info; - desc.flags = 0; - - return crypto_blkcipher_crt(desc.tfm)->encrypt( - &desc, req->dst, req->src, req->nbytes); -} - -static int ablk_encrypt(struct ablkcipher_request *req) -{ - struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); - struct async_twofish_ctx *ctx = crypto_ablkcipher_ctx(tfm); - - if (!irq_fpu_usable()) { - struct ablkcipher_request *cryptd_req = - ablkcipher_request_ctx(req); - - memcpy(cryptd_req, req, sizeof(*req)); - ablkcipher_request_set_tfm(cryptd_req, &ctx->cryptd_tfm->base); - - return crypto_ablkcipher_encrypt(cryptd_req); - } else { - return __ablk_encrypt(req); - } -} - -static int ablk_decrypt(struct ablkcipher_request *req) -{ - struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); - struct async_twofish_ctx *ctx = crypto_ablkcipher_ctx(tfm); - - if (!irq_fpu_usable()) { - struct ablkcipher_request *cryptd_req = - ablkcipher_request_ctx(req); - - memcpy(cryptd_req, req, sizeof(*req)); - ablkcipher_request_set_tfm(cryptd_req, &ctx->cryptd_tfm->base); - - return crypto_ablkcipher_decrypt(cryptd_req); - } else { - struct blkcipher_desc desc; - - desc.tfm = cryptd_ablkcipher_child(ctx->cryptd_tfm); - desc.info = req->info; - desc.flags = 0; - - return crypto_blkcipher_crt(desc.tfm)->decrypt( - &desc, req->dst, req->src, req->nbytes); - } -} - -static void ablk_exit(struct crypto_tfm *tfm) -{ - struct async_twofish_ctx *ctx = crypto_tfm_ctx(tfm); - - cryptd_free_ablkcipher(ctx->cryptd_tfm); -} - -static int ablk_init(struct crypto_tfm *tfm) -{ - struct async_twofish_ctx *ctx = crypto_tfm_ctx(tfm); - struct cryptd_ablkcipher *cryptd_tfm; - char drv_name[CRYPTO_MAX_ALG_NAME]; - - snprintf(drv_name, sizeof(drv_name), "__driver-%s", - crypto_tfm_alg_driver_name(tfm)); - - cryptd_tfm = cryptd_alloc_ablkcipher(drv_name, 0, 0); - if (IS_ERR(cryptd_tfm)) - return PTR_ERR(cryptd_tfm); - - ctx->cryptd_tfm = cryptd_tfm; - tfm->crt_ablkcipher.reqsize = sizeof(struct ablkcipher_request) + - crypto_ablkcipher_reqsize(&cryptd_tfm->base); - - return 0; -} - static struct crypto_alg twofish_algs[10] = { { .cra_name = "__ecb-twofish-avx", .cra_driver_name = "__driver-ecb-twofish-avx", @@ -942,7 +838,7 @@ static struct crypto_alg twofish_algs[10] = { { .cra_priority = 400, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = TF_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct async_twofish_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, @@ -964,7 +860,7 @@ static struct crypto_alg twofish_algs[10] = { { .cra_priority = 400, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = TF_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct async_twofish_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, @@ -987,7 +883,7 @@ static struct crypto_alg twofish_algs[10] = { { .cra_priority = 400, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = 1, - .cra_ctxsize = sizeof(struct async_twofish_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, @@ -1011,7 +907,7 @@ static struct crypto_alg twofish_algs[10] = { { .cra_priority = 400, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = TF_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct async_twofish_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, @@ -1036,7 +932,7 @@ static struct crypto_alg twofish_algs[10] = { { .cra_priority = 400, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = TF_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct async_twofish_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, diff --git a/crypto/Kconfig b/crypto/Kconfig index 8e9145c07d8..24b929e30bf 100644 --- a/crypto/Kconfig +++ b/crypto/Kconfig @@ -946,6 +946,7 @@ config CRYPTO_TWOFISH_AVX_X86_64 depends on X86 && 64BIT select CRYPTO_ALGAPI select CRYPTO_CRYPTD + select CRYPTO_ABLK_HELPER_X86 select CRYPTO_TWOFISH_COMMON select CRYPTO_TWOFISH_X86_64 select CRYPTO_TWOFISH_X86_64_3WAY -- cgit v1.2.3 From a9629d7142ea22567eaa999232d8a31a7493665a Mon Sep 17 00:00:00 2001 From: Jussi Kivilinna Date: Mon, 18 Jun 2012 14:07:08 +0300 Subject: crypto: aes_ni - change to use shared ablk_* functions Remove duplicate ablk_* functions and make use of ablk_helper module instead. Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu --- arch/x86/crypto/ablk_helper.c | 19 +++--- arch/x86/crypto/aesni-intel_glue.c | 100 +++--------------------------- arch/x86/include/asm/crypto/ablk_helper.h | 2 + crypto/Kconfig | 1 + 4 files changed, 20 insertions(+), 102 deletions(-) (limited to 'crypto') diff --git a/arch/x86/crypto/ablk_helper.c b/arch/x86/crypto/ablk_helper.c index 284ca3bc238..43282fe04a8 100644 --- a/arch/x86/crypto/ablk_helper.c +++ b/arch/x86/crypto/ablk_helper.c @@ -118,32 +118,31 @@ void ablk_exit(struct crypto_tfm *tfm) } EXPORT_SYMBOL_GPL(ablk_exit); -void ablk_init_common(struct crypto_tfm *tfm, - struct cryptd_ablkcipher *cryptd_tfm) +int ablk_init_common(struct crypto_tfm *tfm, const char *drv_name) { struct async_helper_ctx *ctx = crypto_tfm_ctx(tfm); + struct cryptd_ablkcipher *cryptd_tfm; + + cryptd_tfm = cryptd_alloc_ablkcipher(drv_name, 0, 0); + if (IS_ERR(cryptd_tfm)) + return PTR_ERR(cryptd_tfm); ctx->cryptd_tfm = cryptd_tfm; tfm->crt_ablkcipher.reqsize = sizeof(struct ablkcipher_request) + crypto_ablkcipher_reqsize(&cryptd_tfm->base); + + return 0; } EXPORT_SYMBOL_GPL(ablk_init_common); int ablk_init(struct crypto_tfm *tfm) { - struct cryptd_ablkcipher *cryptd_tfm; char drv_name[CRYPTO_MAX_ALG_NAME]; snprintf(drv_name, sizeof(drv_name), "__driver-%s", crypto_tfm_alg_driver_name(tfm)); - cryptd_tfm = cryptd_alloc_ablkcipher(drv_name, 0, 0); - if (IS_ERR(cryptd_tfm)) - return PTR_ERR(cryptd_tfm); - - ablk_init_common(tfm, cryptd_tfm); - - return 0; + return ablk_init_common(tfm, drv_name); } EXPORT_SYMBOL_GPL(ablk_init); diff --git a/arch/x86/crypto/aesni-intel_glue.c b/arch/x86/crypto/aesni-intel_glue.c index ac7f5cd019e..7c9d54d8dc4 100644 --- a/arch/x86/crypto/aesni-intel_glue.c +++ b/arch/x86/crypto/aesni-intel_glue.c @@ -31,6 +31,7 @@ #include #include #include +#include #include #include #include @@ -52,10 +53,6 @@ #define HAS_XTS #endif -struct async_aes_ctx { - struct cryptd_ablkcipher *cryptd_tfm; -}; - /* This data is stored at the end of the crypto_tfm struct. * It's a type of per "session" data storage location. * This needs to be 16 byte aligned. @@ -377,87 +374,6 @@ static int ctr_crypt(struct blkcipher_desc *desc, } #endif -static int ablk_set_key(struct crypto_ablkcipher *tfm, const u8 *key, - unsigned int key_len) -{ - struct async_aes_ctx *ctx = crypto_ablkcipher_ctx(tfm); - struct crypto_ablkcipher *child = &ctx->cryptd_tfm->base; - int err; - - crypto_ablkcipher_clear_flags(child, CRYPTO_TFM_REQ_MASK); - crypto_ablkcipher_set_flags(child, crypto_ablkcipher_get_flags(tfm) - & CRYPTO_TFM_REQ_MASK); - err = crypto_ablkcipher_setkey(child, key, key_len); - crypto_ablkcipher_set_flags(tfm, crypto_ablkcipher_get_flags(child) - & CRYPTO_TFM_RES_MASK); - return err; -} - -static int ablk_encrypt(struct ablkcipher_request *req) -{ - struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); - struct async_aes_ctx *ctx = crypto_ablkcipher_ctx(tfm); - - if (!irq_fpu_usable()) { - struct ablkcipher_request *cryptd_req = - ablkcipher_request_ctx(req); - memcpy(cryptd_req, req, sizeof(*req)); - ablkcipher_request_set_tfm(cryptd_req, &ctx->cryptd_tfm->base); - return crypto_ablkcipher_encrypt(cryptd_req); - } else { - struct blkcipher_desc desc; - desc.tfm = cryptd_ablkcipher_child(ctx->cryptd_tfm); - desc.info = req->info; - desc.flags = 0; - return crypto_blkcipher_crt(desc.tfm)->encrypt( - &desc, req->dst, req->src, req->nbytes); - } -} - -static int ablk_decrypt(struct ablkcipher_request *req) -{ - struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); - struct async_aes_ctx *ctx = crypto_ablkcipher_ctx(tfm); - - if (!irq_fpu_usable()) { - struct ablkcipher_request *cryptd_req = - ablkcipher_request_ctx(req); - memcpy(cryptd_req, req, sizeof(*req)); - ablkcipher_request_set_tfm(cryptd_req, &ctx->cryptd_tfm->base); - return crypto_ablkcipher_decrypt(cryptd_req); - } else { - struct blkcipher_desc desc; - desc.tfm = cryptd_ablkcipher_child(ctx->cryptd_tfm); - desc.info = req->info; - desc.flags = 0; - return crypto_blkcipher_crt(desc.tfm)->decrypt( - &desc, req->dst, req->src, req->nbytes); - } -} - -static void ablk_exit(struct crypto_tfm *tfm) -{ - struct async_aes_ctx *ctx = crypto_tfm_ctx(tfm); - - cryptd_free_ablkcipher(ctx->cryptd_tfm); -} - -static int ablk_init_common(struct crypto_tfm *tfm, const char *drv_name) -{ - struct async_aes_ctx *ctx = crypto_tfm_ctx(tfm); - struct cryptd_ablkcipher *cryptd_tfm; - - cryptd_tfm = cryptd_alloc_ablkcipher(drv_name, 0, 0); - if (IS_ERR(cryptd_tfm)) - return PTR_ERR(cryptd_tfm); - - ctx->cryptd_tfm = cryptd_tfm; - tfm->crt_ablkcipher.reqsize = sizeof(struct ablkcipher_request) + - crypto_ablkcipher_reqsize(&cryptd_tfm->base); - - return 0; -} - static int ablk_ecb_init(struct crypto_tfm *tfm) { return ablk_init_common(tfm, "__driver-ecb-aes-aesni"); @@ -968,7 +884,7 @@ static struct crypto_alg aesni_algs[] = { { .cra_priority = 400, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = AES_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct async_aes_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, @@ -989,7 +905,7 @@ static struct crypto_alg aesni_algs[] = { { .cra_priority = 400, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = AES_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct async_aes_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, @@ -1033,7 +949,7 @@ static struct crypto_alg aesni_algs[] = { { .cra_priority = 400, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = 1, - .cra_ctxsize = sizeof(struct async_aes_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, @@ -1098,7 +1014,7 @@ static struct crypto_alg aesni_algs[] = { { .cra_priority = 400, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = 1, - .cra_ctxsize = sizeof(struct async_aes_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, @@ -1126,7 +1042,7 @@ static struct crypto_alg aesni_algs[] = { { .cra_priority = 400, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = AES_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct async_aes_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, @@ -1150,7 +1066,7 @@ static struct crypto_alg aesni_algs[] = { { .cra_priority = 400, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = AES_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct async_aes_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, @@ -1174,7 +1090,7 @@ static struct crypto_alg aesni_algs[] = { { .cra_priority = 400, .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, .cra_blocksize = AES_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct async_aes_ctx), + .cra_ctxsize = sizeof(struct async_helper_ctx), .cra_alignmask = 0, .cra_type = &crypto_ablkcipher_type, .cra_module = THIS_MODULE, diff --git a/arch/x86/include/asm/crypto/ablk_helper.h b/arch/x86/include/asm/crypto/ablk_helper.h index 6d6b37c6f5d..4f93df50c23 100644 --- a/arch/x86/include/asm/crypto/ablk_helper.h +++ b/arch/x86/include/asm/crypto/ablk_helper.h @@ -24,6 +24,8 @@ extern int ablk_decrypt(struct ablkcipher_request *req); extern void ablk_exit(struct crypto_tfm *tfm); +extern int ablk_init_common(struct crypto_tfm *tfm, const char *drv_name); + extern int ablk_init(struct crypto_tfm *tfm); #endif /* _CRYPTO_ABLK_HELPER_H */ diff --git a/crypto/Kconfig b/crypto/Kconfig index 24b929e30bf..471cc468e18 100644 --- a/crypto/Kconfig +++ b/crypto/Kconfig @@ -557,6 +557,7 @@ config CRYPTO_AES_NI_INTEL select CRYPTO_AES_X86_64 if 64BIT select CRYPTO_AES_586 if !64BIT select CRYPTO_CRYPTD + select CRYPTO_ABLK_HELPER_X86 select CRYPTO_ALGAPI help Use Intel AES-NI instructions for AES algorithm. -- cgit v1.2.3 From 596d875052dac6bf084f0c3a3e946fb4709b727b Mon Sep 17 00:00:00 2001 From: Jussi Kivilinna Date: Mon, 18 Jun 2012 14:07:19 +0300 Subject: crypto: serpent-sse2 - split generic glue code to new helper module Now that serpent-sse2 glue code has been made generic, it can be split to separate module. Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu --- arch/x86/crypto/Makefile | 1 + arch/x86/crypto/glue_helper.c | 307 ++++++++++++++++++++++++++ arch/x86/crypto/serpent_sse2_glue.c | 352 +----------------------------- arch/x86/include/asm/crypto/glue_helper.h | 115 ++++++++++ crypto/Kconfig | 7 + 5 files changed, 431 insertions(+), 351 deletions(-) create mode 100644 arch/x86/crypto/glue_helper.c create mode 100644 arch/x86/include/asm/crypto/glue_helper.h (limited to 'crypto') diff --git a/arch/x86/crypto/Makefile b/arch/x86/crypto/Makefile index ad746916f91..e908e5de82d 100644 --- a/arch/x86/crypto/Makefile +++ b/arch/x86/crypto/Makefile @@ -3,6 +3,7 @@ # obj-$(CONFIG_CRYPTO_ABLK_HELPER_X86) += ablk_helper.o +obj-$(CONFIG_CRYPTO_GLUE_HELPER_X86) += glue_helper.o obj-$(CONFIG_CRYPTO_AES_586) += aes-i586.o obj-$(CONFIG_CRYPTO_TWOFISH_586) += twofish-i586.o diff --git a/arch/x86/crypto/glue_helper.c b/arch/x86/crypto/glue_helper.c new file mode 100644 index 00000000000..4854f0f31e4 --- /dev/null +++ b/arch/x86/crypto/glue_helper.c @@ -0,0 +1,307 @@ +/* + * Shared glue code for 128bit block ciphers + * + * Copyright (c) 2012 Jussi Kivilinna + * + * CBC & ECB parts based on code (crypto/cbc.c,ecb.c) by: + * Copyright (c) 2006 Herbert Xu + * CTR part based on code (crypto/ctr.c) by: + * (C) Copyright IBM Corp. 2007 - Joy Latten + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 + * USA + * + */ + +#include +#include +#include +#include +#include +#include + +static int __glue_ecb_crypt_128bit(const struct common_glue_ctx *gctx, + struct blkcipher_desc *desc, + struct blkcipher_walk *walk) +{ + void *ctx = crypto_blkcipher_ctx(desc->tfm); + const unsigned int bsize = 128 / 8; + unsigned int nbytes, i, func_bytes; + bool fpu_enabled = false; + int err; + + err = blkcipher_walk_virt(desc, walk); + + while ((nbytes = walk->nbytes)) { + u8 *wsrc = walk->src.virt.addr; + u8 *wdst = walk->dst.virt.addr; + + fpu_enabled = glue_fpu_begin(bsize, gctx->fpu_blocks_limit, + desc, fpu_enabled, nbytes); + + for (i = 0; i < gctx->num_funcs; i++) { + func_bytes = bsize * gctx->funcs[i].num_blocks; + + /* Process multi-block batch */ + if (nbytes >= func_bytes) { + do { + gctx->funcs[i].fn_u.ecb(ctx, wdst, + wsrc); + + wsrc += func_bytes; + wdst += func_bytes; + nbytes -= func_bytes; + } while (nbytes >= func_bytes); + + if (nbytes < bsize) + goto done; + } + } + +done: + err = blkcipher_walk_done(desc, walk, nbytes); + } + + glue_fpu_end(fpu_enabled); + return err; +} + +int glue_ecb_crypt_128bit(const struct common_glue_ctx *gctx, + struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct blkcipher_walk walk; + + blkcipher_walk_init(&walk, dst, src, nbytes); + return __glue_ecb_crypt_128bit(gctx, desc, &walk); +} +EXPORT_SYMBOL_GPL(glue_ecb_crypt_128bit); + +static unsigned int __glue_cbc_encrypt_128bit(const common_glue_func_t fn, + struct blkcipher_desc *desc, + struct blkcipher_walk *walk) +{ + void *ctx = crypto_blkcipher_ctx(desc->tfm); + const unsigned int bsize = 128 / 8; + unsigned int nbytes = walk->nbytes; + u128 *src = (u128 *)walk->src.virt.addr; + u128 *dst = (u128 *)walk->dst.virt.addr; + u128 *iv = (u128 *)walk->iv; + + do { + u128_xor(dst, src, iv); + fn(ctx, (u8 *)dst, (u8 *)dst); + iv = dst; + + src += 1; + dst += 1; + nbytes -= bsize; + } while (nbytes >= bsize); + + u128_xor((u128 *)walk->iv, (u128 *)walk->iv, iv); + return nbytes; +} + +int glue_cbc_encrypt_128bit(const common_glue_func_t fn, + struct blkcipher_desc *desc, + struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + + while ((nbytes = walk.nbytes)) { + nbytes = __glue_cbc_encrypt_128bit(fn, desc, &walk); + err = blkcipher_walk_done(desc, &walk, nbytes); + } + + return err; +} +EXPORT_SYMBOL_GPL(glue_cbc_encrypt_128bit); + +static unsigned int +__glue_cbc_decrypt_128bit(const struct common_glue_ctx *gctx, + struct blkcipher_desc *desc, + struct blkcipher_walk *walk) +{ + void *ctx = crypto_blkcipher_ctx(desc->tfm); + const unsigned int bsize = 128 / 8; + unsigned int nbytes = walk->nbytes; + u128 *src = (u128 *)walk->src.virt.addr; + u128 *dst = (u128 *)walk->dst.virt.addr; + u128 last_iv; + unsigned int num_blocks, func_bytes; + unsigned int i; + + /* Start of the last block. */ + src += nbytes / bsize - 1; + dst += nbytes / bsize - 1; + + last_iv = *src; + + for (i = 0; i < gctx->num_funcs; i++) { + num_blocks = gctx->funcs[i].num_blocks; + func_bytes = bsize * num_blocks; + + /* Process multi-block batch */ + if (nbytes >= func_bytes) { + do { + nbytes -= func_bytes - bsize; + src -= num_blocks - 1; + dst -= num_blocks - 1; + + gctx->funcs[i].fn_u.cbc(ctx, dst, src); + + nbytes -= bsize; + if (nbytes < bsize) + goto done; + + u128_xor(dst, dst, src - 1); + src -= 1; + dst -= 1; + } while (nbytes >= func_bytes); + + if (nbytes < bsize) + goto done; + } + } + +done: + u128_xor(dst, dst, (u128 *)walk->iv); + *(u128 *)walk->iv = last_iv; + + return nbytes; +} + +int glue_cbc_decrypt_128bit(const struct common_glue_ctx *gctx, + struct blkcipher_desc *desc, + struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + const unsigned int bsize = 128 / 8; + bool fpu_enabled = false; + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + + while ((nbytes = walk.nbytes)) { + fpu_enabled = glue_fpu_begin(bsize, gctx->fpu_blocks_limit, + desc, fpu_enabled, nbytes); + nbytes = __glue_cbc_decrypt_128bit(gctx, desc, &walk); + err = blkcipher_walk_done(desc, &walk, nbytes); + } + + glue_fpu_end(fpu_enabled); + return err; +} +EXPORT_SYMBOL_GPL(glue_cbc_decrypt_128bit); + +static void glue_ctr_crypt_final_128bit(const common_glue_ctr_func_t fn_ctr, + struct blkcipher_desc *desc, + struct blkcipher_walk *walk) +{ + void *ctx = crypto_blkcipher_ctx(desc->tfm); + u8 *src = (u8 *)walk->src.virt.addr; + u8 *dst = (u8 *)walk->dst.virt.addr; + unsigned int nbytes = walk->nbytes; + u128 ctrblk; + u128 tmp; + + be128_to_u128(&ctrblk, (be128 *)walk->iv); + + memcpy(&tmp, src, nbytes); + fn_ctr(ctx, &tmp, &tmp, &ctrblk); + memcpy(dst, &tmp, nbytes); + + u128_to_be128((be128 *)walk->iv, &ctrblk); +} +EXPORT_SYMBOL_GPL(glue_ctr_crypt_final_128bit); + +static unsigned int __glue_ctr_crypt_128bit(const struct common_glue_ctx *gctx, + struct blkcipher_desc *desc, + struct blkcipher_walk *walk) +{ + const unsigned int bsize = 128 / 8; + void *ctx = crypto_blkcipher_ctx(desc->tfm); + unsigned int nbytes = walk->nbytes; + u128 *src = (u128 *)walk->src.virt.addr; + u128 *dst = (u128 *)walk->dst.virt.addr; + u128 ctrblk; + unsigned int num_blocks, func_bytes; + unsigned int i; + + be128_to_u128(&ctrblk, (be128 *)walk->iv); + + /* Process multi-block batch */ + for (i = 0; i < gctx->num_funcs; i++) { + num_blocks = gctx->funcs[i].num_blocks; + func_bytes = bsize * num_blocks; + + if (nbytes >= func_bytes) { + do { + gctx->funcs[i].fn_u.ctr(ctx, dst, src, &ctrblk); + + src += num_blocks; + dst += num_blocks; + nbytes -= func_bytes; + } while (nbytes >= func_bytes); + + if (nbytes < bsize) + goto done; + } + } + +done: + u128_to_be128((be128 *)walk->iv, &ctrblk); + return nbytes; +} + +int glue_ctr_crypt_128bit(const struct common_glue_ctx *gctx, + struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + const unsigned int bsize = 128 / 8; + bool fpu_enabled = false; + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt_block(desc, &walk, bsize); + + while ((nbytes = walk.nbytes) >= bsize) { + fpu_enabled = glue_fpu_begin(bsize, gctx->fpu_blocks_limit, + desc, fpu_enabled, nbytes); + nbytes = __glue_ctr_crypt_128bit(gctx, desc, &walk); + err = blkcipher_walk_done(desc, &walk, nbytes); + } + + glue_fpu_end(fpu_enabled); + + if (walk.nbytes) { + glue_ctr_crypt_final_128bit( + gctx->funcs[gctx->num_funcs - 1].fn_u.ctr, desc, &walk); + err = blkcipher_walk_done(desc, &walk, 0); + } + + return err; +} +EXPORT_SYMBOL_GPL(glue_ctr_crypt_128bit); + +MODULE_LICENSE("GPL"); diff --git a/arch/x86/crypto/serpent_sse2_glue.c b/arch/x86/crypto/serpent_sse2_glue.c index 8c86239010a..49a32eedf0c 100644 --- a/arch/x86/crypto/serpent_sse2_glue.c +++ b/arch/x86/crypto/serpent_sse2_glue.c @@ -41,359 +41,9 @@ #include #include #include -#include #include #include -#include -#include -#include - -typedef void (*common_glue_func_t)(void *ctx, u8 *dst, const u8 *src); -typedef void (*common_glue_cbc_func_t)(void *ctx, u128 *dst, const u128 *src); -typedef void (*common_glue_ctr_func_t)(void *ctx, u128 *dst, const u128 *src, - u128 *iv); - -#define GLUE_FUNC_CAST(fn) ((common_glue_func_t)(fn)) -#define GLUE_CBC_FUNC_CAST(fn) ((common_glue_cbc_func_t)(fn)) -#define GLUE_CTR_FUNC_CAST(fn) ((common_glue_ctr_func_t)(fn)) - -struct common_glue_func_entry { - unsigned int num_blocks; /* number of blocks that @fn will process */ - union { - common_glue_func_t ecb; - common_glue_cbc_func_t cbc; - common_glue_ctr_func_t ctr; - } fn_u; -}; - -struct common_glue_ctx { - unsigned int num_funcs; - int fpu_blocks_limit; /* -1 means fpu not needed at all */ - - /* - * First funcs entry must have largest num_blocks and last funcs entry - * must have num_blocks == 1! - */ - struct common_glue_func_entry funcs[]; -}; - -static inline bool glue_fpu_begin(unsigned int bsize, int fpu_blocks_limit, - struct blkcipher_desc *desc, - bool fpu_enabled, unsigned int nbytes) -{ - if (likely(fpu_blocks_limit < 0)) - return false; - - if (fpu_enabled) - return true; - - /* - * Vector-registers are only used when chunk to be processed is large - * enough, so do not enable FPU until it is necessary. - */ - if (nbytes < bsize * (unsigned int)fpu_blocks_limit) - return false; - - if (desc) { - /* prevent sleeping if FPU is in use */ - desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; - } - - kernel_fpu_begin(); - return true; -} - -static inline void glue_fpu_end(bool fpu_enabled) -{ - if (fpu_enabled) - kernel_fpu_end(); -} - -static int __glue_ecb_crypt_128bit(const struct common_glue_ctx *gctx, - struct blkcipher_desc *desc, - struct blkcipher_walk *walk) -{ - void *ctx = crypto_blkcipher_ctx(desc->tfm); - const unsigned int bsize = 128 / 8; - unsigned int nbytes, i, func_bytes; - bool fpu_enabled = false; - int err; - - err = blkcipher_walk_virt(desc, walk); - - while ((nbytes = walk->nbytes)) { - u8 *wsrc = walk->src.virt.addr; - u8 *wdst = walk->dst.virt.addr; - - fpu_enabled = glue_fpu_begin(bsize, gctx->fpu_blocks_limit, - desc, fpu_enabled, nbytes); - - for (i = 0; i < gctx->num_funcs; i++) { - func_bytes = bsize * gctx->funcs[i].num_blocks; - - /* Process multi-block batch */ - if (nbytes >= func_bytes) { - do { - gctx->funcs[i].fn_u.ecb(ctx, wdst, - wsrc); - - wsrc += func_bytes; - wdst += func_bytes; - nbytes -= func_bytes; - } while (nbytes >= func_bytes); - - if (nbytes < bsize) - goto done; - } - } - -done: - err = blkcipher_walk_done(desc, walk, nbytes); - } - - glue_fpu_end(fpu_enabled); - return err; -} - -int glue_ecb_crypt_128bit(const struct common_glue_ctx *gctx, - struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) -{ - struct blkcipher_walk walk; - - blkcipher_walk_init(&walk, dst, src, nbytes); - return __glue_ecb_crypt_128bit(gctx, desc, &walk); -} - -static unsigned int __glue_cbc_encrypt_128bit(const common_glue_func_t fn, - struct blkcipher_desc *desc, - struct blkcipher_walk *walk) -{ - void *ctx = crypto_blkcipher_ctx(desc->tfm); - const unsigned int bsize = 128 / 8; - unsigned int nbytes = walk->nbytes; - u128 *src = (u128 *)walk->src.virt.addr; - u128 *dst = (u128 *)walk->dst.virt.addr; - u128 *iv = (u128 *)walk->iv; - - do { - u128_xor(dst, src, iv); - fn(ctx, (u8 *)dst, (u8 *)dst); - iv = dst; - - src += 1; - dst += 1; - nbytes -= bsize; - } while (nbytes >= bsize); - - u128_xor((u128 *)walk->iv, (u128 *)walk->iv, iv); - return nbytes; -} - -int glue_cbc_encrypt_128bit(const common_glue_func_t fn, - struct blkcipher_desc *desc, - struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) -{ - struct blkcipher_walk walk; - int err; - - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt(desc, &walk); - - while ((nbytes = walk.nbytes)) { - nbytes = __glue_cbc_encrypt_128bit(fn, desc, &walk); - err = blkcipher_walk_done(desc, &walk, nbytes); - } - - return err; -} - -static unsigned int -__glue_cbc_decrypt_128bit(const struct common_glue_ctx *gctx, - struct blkcipher_desc *desc, - struct blkcipher_walk *walk) -{ - void *ctx = crypto_blkcipher_ctx(desc->tfm); - const unsigned int bsize = 128 / 8; - unsigned int nbytes = walk->nbytes; - u128 *src = (u128 *)walk->src.virt.addr; - u128 *dst = (u128 *)walk->dst.virt.addr; - u128 last_iv; - unsigned int num_blocks, func_bytes; - unsigned int i; - - /* Start of the last block. */ - src += nbytes / bsize - 1; - dst += nbytes / bsize - 1; - - last_iv = *src; - - for (i = 0; i < gctx->num_funcs; i++) { - num_blocks = gctx->funcs[i].num_blocks; - func_bytes = bsize * num_blocks; - - /* Process multi-block batch */ - if (nbytes >= func_bytes) { - do { - nbytes -= func_bytes - bsize; - src -= num_blocks - 1; - dst -= num_blocks - 1; - - gctx->funcs[i].fn_u.cbc(ctx, dst, src); - - nbytes -= bsize; - if (nbytes < bsize) - goto done; - - u128_xor(dst, dst, src - 1); - src -= 1; - dst -= 1; - } while (nbytes >= func_bytes); - - if (nbytes < bsize) - goto done; - } - } - -done: - u128_xor(dst, dst, (u128 *)walk->iv); - *(u128 *)walk->iv = last_iv; - - return nbytes; -} - -int glue_cbc_decrypt_128bit(const struct common_glue_ctx *gctx, - struct blkcipher_desc *desc, - struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) -{ - const unsigned int bsize = 128 / 8; - bool fpu_enabled = false; - struct blkcipher_walk walk; - int err; - - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt(desc, &walk); - - while ((nbytes = walk.nbytes)) { - fpu_enabled = glue_fpu_begin(bsize, gctx->fpu_blocks_limit, - desc, fpu_enabled, nbytes); - nbytes = __glue_cbc_decrypt_128bit(gctx, desc, &walk); - err = blkcipher_walk_done(desc, &walk, nbytes); - } - - glue_fpu_end(fpu_enabled); - return err; -} - -static inline void u128_to_be128(be128 *dst, const u128 *src) -{ - dst->a = cpu_to_be64(src->a); - dst->b = cpu_to_be64(src->b); -} - -static inline void be128_to_u128(u128 *dst, const be128 *src) -{ - dst->a = be64_to_cpu(src->a); - dst->b = be64_to_cpu(src->b); -} - -static inline void u128_inc(u128 *i) -{ - i->b++; - if (!i->b) - i->a++; -} - -static void glue_ctr_crypt_final_128bit(const common_glue_ctr_func_t fn_ctr, - struct blkcipher_desc *desc, - struct blkcipher_walk *walk) -{ - void *ctx = crypto_blkcipher_ctx(desc->tfm); - u8 *src = (u8 *)walk->src.virt.addr; - u8 *dst = (u8 *)walk->dst.virt.addr; - unsigned int nbytes = walk->nbytes; - u128 ctrblk; - u128 tmp; - - be128_to_u128(&ctrblk, (be128 *)walk->iv); - - memcpy(&tmp, src, nbytes); - fn_ctr(ctx, &tmp, &tmp, &ctrblk); - memcpy(dst, &tmp, nbytes); - - u128_to_be128((be128 *)walk->iv, &ctrblk); -} - -static unsigned int __glue_ctr_crypt_128bit(const struct common_glue_ctx *gctx, - struct blkcipher_desc *desc, - struct blkcipher_walk *walk) -{ - const unsigned int bsize = 128 / 8; - void *ctx = crypto_blkcipher_ctx(desc->tfm); - unsigned int nbytes = walk->nbytes; - u128 *src = (u128 *)walk->src.virt.addr; - u128 *dst = (u128 *)walk->dst.virt.addr; - u128 ctrblk; - unsigned int num_blocks, func_bytes; - unsigned int i; - - be128_to_u128(&ctrblk, (be128 *)walk->iv); - - /* Process multi-block batch */ - for (i = 0; i < gctx->num_funcs; i++) { - num_blocks = gctx->funcs[i].num_blocks; - func_bytes = bsize * num_blocks; - - if (nbytes >= func_bytes) { - do { - gctx->funcs[i].fn_u.ctr(ctx, dst, src, &ctrblk); - - src += num_blocks; - dst += num_blocks; - nbytes -= func_bytes; - } while (nbytes >= func_bytes); - - if (nbytes < bsize) - goto done; - } - } - -done: - u128_to_be128((be128 *)walk->iv, &ctrblk); - return nbytes; -} - -int glue_ctr_crypt_128bit(const struct common_glue_ctx *gctx, - struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) -{ - const unsigned int bsize = 128 / 8; - bool fpu_enabled = false; - struct blkcipher_walk walk; - int err; - - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt_block(desc, &walk, bsize); - - while ((nbytes = walk.nbytes) >= bsize) { - fpu_enabled = glue_fpu_begin(bsize, gctx->fpu_blocks_limit, - desc, fpu_enabled, nbytes); - nbytes = __glue_ctr_crypt_128bit(gctx, desc, &walk); - err = blkcipher_walk_done(desc, &walk, nbytes); - } - - glue_fpu_end(fpu_enabled); - - if (walk.nbytes) { - glue_ctr_crypt_final_128bit( - gctx->funcs[gctx->num_funcs - 1].fn_u.ctr, desc, &walk); - err = blkcipher_walk_done(desc, &walk, 0); - } - - return err; -} +#include static void serpent_decrypt_cbc_xway(void *ctx, u128 *dst, const u128 *src) { diff --git a/arch/x86/include/asm/crypto/glue_helper.h b/arch/x86/include/asm/crypto/glue_helper.h new file mode 100644 index 00000000000..3e408bddc96 --- /dev/null +++ b/arch/x86/include/asm/crypto/glue_helper.h @@ -0,0 +1,115 @@ +/* + * Shared glue code for 128bit block ciphers + */ + +#ifndef _CRYPTO_GLUE_HELPER_H +#define _CRYPTO_GLUE_HELPER_H + +#include +#include +#include +#include + +typedef void (*common_glue_func_t)(void *ctx, u8 *dst, const u8 *src); +typedef void (*common_glue_cbc_func_t)(void *ctx, u128 *dst, const u128 *src); +typedef void (*common_glue_ctr_func_t)(void *ctx, u128 *dst, const u128 *src, + u128 *iv); + +#define GLUE_FUNC_CAST(fn) ((common_glue_func_t)(fn)) +#define GLUE_CBC_FUNC_CAST(fn) ((common_glue_cbc_func_t)(fn)) +#define GLUE_CTR_FUNC_CAST(fn) ((common_glue_ctr_func_t)(fn)) + +struct common_glue_func_entry { + unsigned int num_blocks; /* number of blocks that @fn will process */ + union { + common_glue_func_t ecb; + common_glue_cbc_func_t cbc; + common_glue_ctr_func_t ctr; + } fn_u; +}; + +struct common_glue_ctx { + unsigned int num_funcs; + int fpu_blocks_limit; /* -1 means fpu not needed at all */ + + /* + * First funcs entry must have largest num_blocks and last funcs entry + * must have num_blocks == 1! + */ + struct common_glue_func_entry funcs[]; +}; + +static inline bool glue_fpu_begin(unsigned int bsize, int fpu_blocks_limit, + struct blkcipher_desc *desc, + bool fpu_enabled, unsigned int nbytes) +{ + if (likely(fpu_blocks_limit < 0)) + return false; + + if (fpu_enabled) + return true; + + /* + * Vector-registers are only used when chunk to be processed is large + * enough, so do not enable FPU until it is necessary. + */ + if (nbytes < bsize * (unsigned int)fpu_blocks_limit) + return false; + + if (desc) { + /* prevent sleeping if FPU is in use */ + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + } + + kernel_fpu_begin(); + return true; +} + +static inline void glue_fpu_end(bool fpu_enabled) +{ + if (fpu_enabled) + kernel_fpu_end(); +} + +static inline void u128_to_be128(be128 *dst, const u128 *src) +{ + dst->a = cpu_to_be64(src->a); + dst->b = cpu_to_be64(src->b); +} + +static inline void be128_to_u128(u128 *dst, const be128 *src) +{ + dst->a = be64_to_cpu(src->a); + dst->b = be64_to_cpu(src->b); +} + +static inline void u128_inc(u128 *i) +{ + i->b++; + if (!i->b) + i->a++; +} + +extern int glue_ecb_crypt_128bit(const struct common_glue_ctx *gctx, + struct blkcipher_desc *desc, + struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes); + +extern int glue_cbc_encrypt_128bit(const common_glue_func_t fn, + struct blkcipher_desc *desc, + struct scatterlist *dst, + struct scatterlist *src, + unsigned int nbytes); + +extern int glue_cbc_decrypt_128bit(const struct common_glue_ctx *gctx, + struct blkcipher_desc *desc, + struct scatterlist *dst, + struct scatterlist *src, + unsigned int nbytes); + +extern int glue_ctr_crypt_128bit(const struct common_glue_ctx *gctx, + struct blkcipher_desc *desc, + struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes); + +#endif /* _CRYPTO_GLUE_HELPER_H */ diff --git a/crypto/Kconfig b/crypto/Kconfig index 471cc468e18..92b46970c85 100644 --- a/crypto/Kconfig +++ b/crypto/Kconfig @@ -179,6 +179,11 @@ config CRYPTO_ABLK_HELPER_X86 depends on X86 select CRYPTO_CRYPTD +config CRYPTO_GLUE_HELPER_X86 + tristate + depends on X86 + select CRYPTO_ALGAPI + comment "Authenticated Encryption with Associated Data" config CRYPTO_CCM @@ -793,6 +798,7 @@ config CRYPTO_SERPENT_SSE2_X86_64 select CRYPTO_ALGAPI select CRYPTO_CRYPTD select CRYPTO_ABLK_HELPER_X86 + select CRYPTO_GLUE_HELPER_X86 select CRYPTO_SERPENT select CRYPTO_LRW select CRYPTO_XTS @@ -814,6 +820,7 @@ config CRYPTO_SERPENT_SSE2_586 select CRYPTO_ALGAPI select CRYPTO_CRYPTD select CRYPTO_ABLK_HELPER_X86 + select CRYPTO_GLUE_HELPER_X86 select CRYPTO_SERPENT select CRYPTO_LRW select CRYPTO_XTS -- cgit v1.2.3 From 1d0debbd4671a8d302a11837a126d5f87db16bdc Mon Sep 17 00:00:00 2001 From: Jussi Kivilinna Date: Mon, 18 Jun 2012 14:07:24 +0300 Subject: crypto: serpent-avx: remove duplicated glue code and use shared glue code from glue_helper Now that shared glue code is available, convert serpent-avx to use it. Cc: Johannes Goetzfried Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu --- arch/x86/crypto/serpent_avx_glue.c | 398 +++++++++---------------------------- crypto/Kconfig | 1 + 2 files changed, 95 insertions(+), 304 deletions(-) (limited to 'crypto') diff --git a/arch/x86/crypto/serpent_avx_glue.c b/arch/x86/crypto/serpent_avx_glue.c index 31eb567cc89..c61b91aa42a 100644 --- a/arch/x86/crypto/serpent_avx_glue.c +++ b/arch/x86/crypto/serpent_avx_glue.c @@ -36,357 +36,147 @@ #include #include #include -#include #include #include #include #include -#include -#include -#include +#include -static inline bool serpent_fpu_begin(bool fpu_enabled, unsigned int nbytes) +static void serpent_decrypt_cbc_xway(void *ctx, u128 *dst, const u128 *src) { - if (fpu_enabled) - return true; - - /* AVX is only used when chunk to be processed is large enough, so - * do not enable FPU until it is necessary. - */ - if (nbytes < SERPENT_BLOCK_SIZE * SERPENT_PARALLEL_BLOCKS) - return false; - - kernel_fpu_begin(); - return true; -} - -static inline void serpent_fpu_end(bool fpu_enabled) -{ - if (fpu_enabled) - kernel_fpu_end(); -} - -static int ecb_crypt(struct blkcipher_desc *desc, struct blkcipher_walk *walk, - bool enc) -{ - bool fpu_enabled = false; - struct serpent_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - const unsigned int bsize = SERPENT_BLOCK_SIZE; - unsigned int nbytes; - int err; + u128 ivs[SERPENT_PARALLEL_BLOCKS - 1]; + unsigned int j; - err = blkcipher_walk_virt(desc, walk); - desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + for (j = 0; j < SERPENT_PARALLEL_BLOCKS - 1; j++) + ivs[j] = src[j]; - while ((nbytes = walk->nbytes)) { - u8 *wsrc = walk->src.virt.addr; - u8 *wdst = walk->dst.virt.addr; - - fpu_enabled = serpent_fpu_begin(fpu_enabled, nbytes); - - /* Process multi-block batch */ - if (nbytes >= bsize * SERPENT_PARALLEL_BLOCKS) { - do { - if (enc) - serpent_enc_blk_xway(ctx, wdst, wsrc); - else - serpent_dec_blk_xway(ctx, wdst, wsrc); - - wsrc += bsize * SERPENT_PARALLEL_BLOCKS; - wdst += bsize * SERPENT_PARALLEL_BLOCKS; - nbytes -= bsize * SERPENT_PARALLEL_BLOCKS; - } while (nbytes >= bsize * SERPENT_PARALLEL_BLOCKS); - - if (nbytes < bsize) - goto done; - } - - /* Handle leftovers */ - do { - if (enc) - __serpent_encrypt(ctx, wdst, wsrc); - else - __serpent_decrypt(ctx, wdst, wsrc); - - wsrc += bsize; - wdst += bsize; - nbytes -= bsize; - } while (nbytes >= bsize); - -done: - err = blkcipher_walk_done(desc, walk, nbytes); - } + serpent_dec_blk_xway(ctx, (u8 *)dst, (u8 *)src); - serpent_fpu_end(fpu_enabled); - return err; + for (j = 0; j < SERPENT_PARALLEL_BLOCKS - 1; j++) + u128_xor(dst + (j + 1), dst + (j + 1), ivs + j); } -static int ecb_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) +static void serpent_crypt_ctr(void *ctx, u128 *dst, const u128 *src, u128 *iv) { - struct blkcipher_walk walk; - - blkcipher_walk_init(&walk, dst, src, nbytes); - return ecb_crypt(desc, &walk, true); -} + be128 ctrblk; -static int ecb_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) -{ - struct blkcipher_walk walk; + u128_to_be128(&ctrblk, iv); + u128_inc(iv); - blkcipher_walk_init(&walk, dst, src, nbytes); - return ecb_crypt(desc, &walk, false); + __serpent_encrypt(ctx, (u8 *)&ctrblk, (u8 *)&ctrblk); + u128_xor(dst, src, (u128 *)&ctrblk); } -static unsigned int __cbc_encrypt(struct blkcipher_desc *desc, - struct blkcipher_walk *walk) +static void serpent_crypt_ctr_xway(void *ctx, u128 *dst, const u128 *src, + u128 *iv) { - struct serpent_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - const unsigned int bsize = SERPENT_BLOCK_SIZE; - unsigned int nbytes = walk->nbytes; - u128 *src = (u128 *)walk->src.virt.addr; - u128 *dst = (u128 *)walk->dst.virt.addr; - u128 *iv = (u128 *)walk->iv; - - do { - u128_xor(dst, src, iv); - __serpent_encrypt(ctx, (u8 *)dst, (u8 *)dst); - iv = dst; - - src += 1; - dst += 1; - nbytes -= bsize; - } while (nbytes >= bsize); - - u128_xor((u128 *)walk->iv, (u128 *)walk->iv, iv); - return nbytes; -} + be128 ctrblks[SERPENT_PARALLEL_BLOCKS]; + unsigned int i; -static int cbc_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) -{ - struct blkcipher_walk walk; - int err; - - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt(desc, &walk); + for (i = 0; i < SERPENT_PARALLEL_BLOCKS; i++) { + if (dst != src) + dst[i] = src[i]; - while ((nbytes = walk.nbytes)) { - nbytes = __cbc_encrypt(desc, &walk); - err = blkcipher_walk_done(desc, &walk, nbytes); + u128_to_be128(&ctrblks[i], iv); + u128_inc(iv); } - return err; + serpent_enc_blk_xway_xor(ctx, (u8 *)dst, (u8 *)ctrblks); } -static unsigned int __cbc_decrypt(struct blkcipher_desc *desc, - struct blkcipher_walk *walk) -{ - struct serpent_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - const unsigned int bsize = SERPENT_BLOCK_SIZE; - unsigned int nbytes = walk->nbytes; - u128 *src = (u128 *)walk->src.virt.addr; - u128 *dst = (u128 *)walk->dst.virt.addr; - u128 ivs[SERPENT_PARALLEL_BLOCKS - 1]; - u128 last_iv; - int i; - - /* Start of the last block. */ - src += nbytes / bsize - 1; - dst += nbytes / bsize - 1; - - last_iv = *src; - - /* Process multi-block batch */ - if (nbytes >= bsize * SERPENT_PARALLEL_BLOCKS) { - do { - nbytes -= bsize * (SERPENT_PARALLEL_BLOCKS - 1); - src -= SERPENT_PARALLEL_BLOCKS - 1; - dst -= SERPENT_PARALLEL_BLOCKS - 1; - - for (i = 0; i < SERPENT_PARALLEL_BLOCKS - 1; i++) - ivs[i] = src[i]; - - serpent_dec_blk_xway(ctx, (u8 *)dst, (u8 *)src); - - for (i = 0; i < SERPENT_PARALLEL_BLOCKS - 1; i++) - u128_xor(dst + (i + 1), dst + (i + 1), ivs + i); - - nbytes -= bsize; - if (nbytes < bsize) - goto done; - - u128_xor(dst, dst, src - 1); - src -= 1; - dst -= 1; - } while (nbytes >= bsize * SERPENT_PARALLEL_BLOCKS); - - if (nbytes < bsize) - goto done; - } - - /* Handle leftovers */ - for (;;) { - __serpent_decrypt(ctx, (u8 *)dst, (u8 *)src); - - nbytes -= bsize; - if (nbytes < bsize) - break; +static const struct common_glue_ctx serpent_enc = { + .num_funcs = 2, + .fpu_blocks_limit = SERPENT_PARALLEL_BLOCKS, + + .funcs = { { + .num_blocks = SERPENT_PARALLEL_BLOCKS, + .fn_u = { .ecb = GLUE_FUNC_CAST(serpent_enc_blk_xway) } + }, { + .num_blocks = 1, + .fn_u = { .ecb = GLUE_FUNC_CAST(__serpent_encrypt) } + } } +}; - u128_xor(dst, dst, src - 1); - src -= 1; - dst -= 1; - } +static const struct common_glue_ctx serpent_ctr = { + .num_funcs = 2, + .fpu_blocks_limit = SERPENT_PARALLEL_BLOCKS, + + .funcs = { { + .num_blocks = SERPENT_PARALLEL_BLOCKS, + .fn_u = { .ctr = GLUE_CTR_FUNC_CAST(serpent_crypt_ctr_xway) } + }, { + .num_blocks = 1, + .fn_u = { .ctr = GLUE_CTR_FUNC_CAST(serpent_crypt_ctr) } + } } +}; -done: - u128_xor(dst, dst, (u128 *)walk->iv); - *(u128 *)walk->iv = last_iv; +static const struct common_glue_ctx serpent_dec = { + .num_funcs = 2, + .fpu_blocks_limit = SERPENT_PARALLEL_BLOCKS, + + .funcs = { { + .num_blocks = SERPENT_PARALLEL_BLOCKS, + .fn_u = { .ecb = GLUE_FUNC_CAST(serpent_dec_blk_xway) } + }, { + .num_blocks = 1, + .fn_u = { .ecb = GLUE_FUNC_CAST(__serpent_decrypt) } + } } +}; - return nbytes; -} +static const struct common_glue_ctx serpent_dec_cbc = { + .num_funcs = 2, + .fpu_blocks_limit = SERPENT_PARALLEL_BLOCKS, + + .funcs = { { + .num_blocks = SERPENT_PARALLEL_BLOCKS, + .fn_u = { .cbc = GLUE_CBC_FUNC_CAST(serpent_decrypt_cbc_xway) } + }, { + .num_blocks = 1, + .fn_u = { .cbc = GLUE_CBC_FUNC_CAST(__serpent_decrypt) } + } } +}; -static int cbc_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, +static int ecb_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, struct scatterlist *src, unsigned int nbytes) { - bool fpu_enabled = false; - struct blkcipher_walk walk; - int err; - - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt(desc, &walk); - desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; - - while ((nbytes = walk.nbytes)) { - fpu_enabled = serpent_fpu_begin(fpu_enabled, nbytes); - nbytes = __cbc_decrypt(desc, &walk); - err = blkcipher_walk_done(desc, &walk, nbytes); - } - - serpent_fpu_end(fpu_enabled); - return err; + return glue_ecb_crypt_128bit(&serpent_enc, desc, dst, src, nbytes); } -static inline void u128_to_be128(be128 *dst, const u128 *src) +static int ecb_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) { - dst->a = cpu_to_be64(src->a); - dst->b = cpu_to_be64(src->b); + return glue_ecb_crypt_128bit(&serpent_dec, desc, dst, src, nbytes); } -static inline void be128_to_u128(u128 *dst, const be128 *src) +static int cbc_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) { - dst->a = be64_to_cpu(src->a); - dst->b = be64_to_cpu(src->b); + return glue_cbc_encrypt_128bit(GLUE_FUNC_CAST(__serpent_encrypt), desc, + dst, src, nbytes); } -static inline void u128_inc(u128 *i) +static int cbc_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) { - i->b++; - if (!i->b) - i->a++; + return glue_cbc_decrypt_128bit(&serpent_dec_cbc, desc, dst, src, + nbytes); } -static void ctr_crypt_final(struct blkcipher_desc *desc, - struct blkcipher_walk *walk) +static int ctr_crypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) { - struct serpent_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - u8 *ctrblk = walk->iv; - u8 keystream[SERPENT_BLOCK_SIZE]; - u8 *src = walk->src.virt.addr; - u8 *dst = walk->dst.virt.addr; - unsigned int nbytes = walk->nbytes; - - __serpent_encrypt(ctx, keystream, ctrblk); - crypto_xor(keystream, src, nbytes); - memcpy(dst, keystream, nbytes); - - crypto_inc(ctrblk, SERPENT_BLOCK_SIZE); + return glue_ctr_crypt_128bit(&serpent_ctr, desc, dst, src, nbytes); } -static unsigned int __ctr_crypt(struct blkcipher_desc *desc, - struct blkcipher_walk *walk) +static inline bool serpent_fpu_begin(bool fpu_enabled, unsigned int nbytes) { - struct serpent_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - const unsigned int bsize = SERPENT_BLOCK_SIZE; - unsigned int nbytes = walk->nbytes; - u128 *src = (u128 *)walk->src.virt.addr; - u128 *dst = (u128 *)walk->dst.virt.addr; - u128 ctrblk; - be128 ctrblocks[SERPENT_PARALLEL_BLOCKS]; - int i; - - be128_to_u128(&ctrblk, (be128 *)walk->iv); - - /* Process multi-block batch */ - if (nbytes >= bsize * SERPENT_PARALLEL_BLOCKS) { - do { - /* create ctrblks for parallel encrypt */ - for (i = 0; i < SERPENT_PARALLEL_BLOCKS; i++) { - if (dst != src) - dst[i] = src[i]; - - u128_to_be128(&ctrblocks[i], &ctrblk); - u128_inc(&ctrblk); - } - - serpent_enc_blk_xway_xor(ctx, (u8 *)dst, - (u8 *)ctrblocks); - - src += SERPENT_PARALLEL_BLOCKS; - dst += SERPENT_PARALLEL_BLOCKS; - nbytes -= bsize * SERPENT_PARALLEL_BLOCKS; - } while (nbytes >= bsize * SERPENT_PARALLEL_BLOCKS); - - if (nbytes < bsize) - goto done; - } - - /* Handle leftovers */ - do { - if (dst != src) - *dst = *src; - - u128_to_be128(&ctrblocks[0], &ctrblk); - u128_inc(&ctrblk); - - __serpent_encrypt(ctx, (u8 *)ctrblocks, (u8 *)ctrblocks); - u128_xor(dst, dst, (u128 *)ctrblocks); - - src += 1; - dst += 1; - nbytes -= bsize; - } while (nbytes >= bsize); - -done: - u128_to_be128((be128 *)walk->iv, &ctrblk); - return nbytes; + return glue_fpu_begin(SERPENT_BLOCK_SIZE, SERPENT_PARALLEL_BLOCKS, + NULL, fpu_enabled, nbytes); } -static int ctr_crypt(struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) +static inline void serpent_fpu_end(bool fpu_enabled) { - bool fpu_enabled = false; - struct blkcipher_walk walk; - int err; - - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt_block(desc, &walk, SERPENT_BLOCK_SIZE); - desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; - - while ((nbytes = walk.nbytes) >= SERPENT_BLOCK_SIZE) { - fpu_enabled = serpent_fpu_begin(fpu_enabled, nbytes); - nbytes = __ctr_crypt(desc, &walk); - err = blkcipher_walk_done(desc, &walk, nbytes); - } - - serpent_fpu_end(fpu_enabled); - - if (walk.nbytes) { - ctr_crypt_final(desc, &walk); - err = blkcipher_walk_done(desc, &walk, 0); - } - - return err; + glue_fpu_end(fpu_enabled); } struct crypt_priv { diff --git a/crypto/Kconfig b/crypto/Kconfig index 92b46970c85..a86c2fb92ea 100644 --- a/crypto/Kconfig +++ b/crypto/Kconfig @@ -842,6 +842,7 @@ config CRYPTO_SERPENT_AVX_X86_64 select CRYPTO_ALGAPI select CRYPTO_CRYPTD select CRYPTO_ABLK_HELPER_X86 + select CRYPTO_GLUE_HELPER_X86 select CRYPTO_SERPENT select CRYPTO_LRW select CRYPTO_XTS -- cgit v1.2.3 From 964263afdcbf9d1e85c021acfff0cc68dd168475 Mon Sep 17 00:00:00 2001 From: Jussi Kivilinna Date: Mon, 18 Jun 2012 14:07:29 +0300 Subject: crypto: camellia-x86_64 - remove duplicated glue code and use shared glue code from glue_helper Now that shared glue code is available, convert camellia-x86_64 to use it. Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu --- arch/x86/crypto/camellia_glue.c | 355 ++++++++++------------------------------ crypto/Kconfig | 1 + 2 files changed, 87 insertions(+), 269 deletions(-) (limited to 'crypto') diff --git a/arch/x86/crypto/camellia_glue.c b/arch/x86/crypto/camellia_glue.c index 3306dc0b139..eeb2b3b743e 100644 --- a/arch/x86/crypto/camellia_glue.c +++ b/arch/x86/crypto/camellia_glue.c @@ -5,10 +5,6 @@ * * Camellia parts based on code by: * Copyright (C) 2006 NTT (Nippon Telegraph and Telephone Corporation) - * CBC & ECB parts based on code (crypto/cbc.c,ecb.c) by: - * Copyright (c) 2006 Herbert Xu - * CTR part based on code (crypto/ctr.c) by: - * (C) Copyright IBM Corp. 2007 - Joy Latten * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -34,9 +30,9 @@ #include #include #include -#include #include #include +#include #define CAMELLIA_MIN_KEY_SIZE 16 #define CAMELLIA_MAX_KEY_SIZE 32 @@ -1312,307 +1308,128 @@ static int camellia_setkey(struct crypto_tfm *tfm, const u8 *in_key, &tfm->crt_flags); } -static int ecb_crypt(struct blkcipher_desc *desc, struct blkcipher_walk *walk, - void (*fn)(struct camellia_ctx *, u8 *, const u8 *), - void (*fn_2way)(struct camellia_ctx *, u8 *, const u8 *)) +static void camellia_decrypt_cbc_2way(void *ctx, u128 *dst, const u128 *src) { - struct camellia_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - unsigned int bsize = CAMELLIA_BLOCK_SIZE; - unsigned int nbytes; - int err; - - err = blkcipher_walk_virt(desc, walk); - - while ((nbytes = walk->nbytes)) { - u8 *wsrc = walk->src.virt.addr; - u8 *wdst = walk->dst.virt.addr; - - /* Process two block batch */ - if (nbytes >= bsize * 2) { - do { - fn_2way(ctx, wdst, wsrc); - - wsrc += bsize * 2; - wdst += bsize * 2; - nbytes -= bsize * 2; - } while (nbytes >= bsize * 2); - - if (nbytes < bsize) - goto done; - } - - /* Handle leftovers */ - do { - fn(ctx, wdst, wsrc); - - wsrc += bsize; - wdst += bsize; - nbytes -= bsize; - } while (nbytes >= bsize); - -done: - err = blkcipher_walk_done(desc, walk, nbytes); - } - - return err; -} - -static int ecb_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) -{ - struct blkcipher_walk walk; - - blkcipher_walk_init(&walk, dst, src, nbytes); - return ecb_crypt(desc, &walk, camellia_enc_blk, camellia_enc_blk_2way); -} + u128 iv = *src; -static int ecb_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) -{ - struct blkcipher_walk walk; - - blkcipher_walk_init(&walk, dst, src, nbytes); - return ecb_crypt(desc, &walk, camellia_dec_blk, camellia_dec_blk_2way); -} + camellia_dec_blk_2way(ctx, (u8 *)dst, (u8 *)src); -static unsigned int __cbc_encrypt(struct blkcipher_desc *desc, - struct blkcipher_walk *walk) -{ - struct camellia_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - unsigned int bsize = CAMELLIA_BLOCK_SIZE; - unsigned int nbytes = walk->nbytes; - u128 *src = (u128 *)walk->src.virt.addr; - u128 *dst = (u128 *)walk->dst.virt.addr; - u128 *iv = (u128 *)walk->iv; - - do { - u128_xor(dst, src, iv); - camellia_enc_blk(ctx, (u8 *)dst, (u8 *)dst); - iv = dst; - - src += 1; - dst += 1; - nbytes -= bsize; - } while (nbytes >= bsize); - - u128_xor((u128 *)walk->iv, (u128 *)walk->iv, iv); - return nbytes; + u128_xor(&dst[1], &dst[1], &iv); } -static int cbc_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) +static void camellia_crypt_ctr(void *ctx, u128 *dst, const u128 *src, u128 *iv) { - struct blkcipher_walk walk; - int err; + be128 ctrblk; - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt(desc, &walk); + if (dst != src) + *dst = *src; - while ((nbytes = walk.nbytes)) { - nbytes = __cbc_encrypt(desc, &walk); - err = blkcipher_walk_done(desc, &walk, nbytes); - } + u128_to_be128(&ctrblk, iv); + u128_inc(iv); - return err; + camellia_enc_blk_xor(ctx, (u8 *)dst, (u8 *)&ctrblk); } -static unsigned int __cbc_decrypt(struct blkcipher_desc *desc, - struct blkcipher_walk *walk) +static void camellia_crypt_ctr_2way(void *ctx, u128 *dst, const u128 *src, + u128 *iv) { - struct camellia_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - unsigned int bsize = CAMELLIA_BLOCK_SIZE; - unsigned int nbytes = walk->nbytes; - u128 *src = (u128 *)walk->src.virt.addr; - u128 *dst = (u128 *)walk->dst.virt.addr; - u128 ivs[2 - 1]; - u128 last_iv; + be128 ctrblks[2]; - /* Start of the last block. */ - src += nbytes / bsize - 1; - dst += nbytes / bsize - 1; - - last_iv = *src; - - /* Process two block batch */ - if (nbytes >= bsize * 2) { - do { - nbytes -= bsize * (2 - 1); - src -= 2 - 1; - dst -= 2 - 1; - - ivs[0] = src[0]; - - camellia_dec_blk_2way(ctx, (u8 *)dst, (u8 *)src); - - u128_xor(dst + 1, dst + 1, ivs + 0); - - nbytes -= bsize; - if (nbytes < bsize) - goto done; - - u128_xor(dst, dst, src - 1); - src -= 1; - dst -= 1; - } while (nbytes >= bsize * 2); - - if (nbytes < bsize) - goto done; + if (dst != src) { + dst[0] = src[0]; + dst[1] = src[1]; } - /* Handle leftovers */ - for (;;) { - camellia_dec_blk(ctx, (u8 *)dst, (u8 *)src); - - nbytes -= bsize; - if (nbytes < bsize) - break; + u128_to_be128(&ctrblks[0], iv); + u128_inc(iv); + u128_to_be128(&ctrblks[1], iv); + u128_inc(iv); - u128_xor(dst, dst, src - 1); - src -= 1; - dst -= 1; - } - -done: - u128_xor(dst, dst, (u128 *)walk->iv); - *(u128 *)walk->iv = last_iv; - - return nbytes; + camellia_enc_blk_xor_2way(ctx, (u8 *)dst, (u8 *)ctrblks); } -static int cbc_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) -{ - struct blkcipher_walk walk; - int err; - - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt(desc, &walk); +static const struct common_glue_ctx camellia_enc = { + .num_funcs = 2, + .fpu_blocks_limit = -1, + + .funcs = { { + .num_blocks = 2, + .fn_u = { .ecb = GLUE_FUNC_CAST(camellia_enc_blk_2way) } + }, { + .num_blocks = 1, + .fn_u = { .ecb = GLUE_FUNC_CAST(camellia_enc_blk) } + } } +}; - while ((nbytes = walk.nbytes)) { - nbytes = __cbc_decrypt(desc, &walk); - err = blkcipher_walk_done(desc, &walk, nbytes); - } +static const struct common_glue_ctx camellia_ctr = { + .num_funcs = 2, + .fpu_blocks_limit = -1, + + .funcs = { { + .num_blocks = 2, + .fn_u = { .ctr = GLUE_CTR_FUNC_CAST(camellia_crypt_ctr_2way) } + }, { + .num_blocks = 1, + .fn_u = { .ctr = GLUE_CTR_FUNC_CAST(camellia_crypt_ctr) } + } } +}; - return err; -} +static const struct common_glue_ctx camellia_dec = { + .num_funcs = 2, + .fpu_blocks_limit = -1, + + .funcs = { { + .num_blocks = 2, + .fn_u = { .ecb = GLUE_FUNC_CAST(camellia_dec_blk_2way) } + }, { + .num_blocks = 1, + .fn_u = { .ecb = GLUE_FUNC_CAST(camellia_dec_blk) } + } } +}; -static inline void u128_to_be128(be128 *dst, const u128 *src) -{ - dst->a = cpu_to_be64(src->a); - dst->b = cpu_to_be64(src->b); -} +static const struct common_glue_ctx camellia_dec_cbc = { + .num_funcs = 2, + .fpu_blocks_limit = -1, + + .funcs = { { + .num_blocks = 2, + .fn_u = { .cbc = GLUE_CBC_FUNC_CAST(camellia_decrypt_cbc_2way) } + }, { + .num_blocks = 1, + .fn_u = { .cbc = GLUE_CBC_FUNC_CAST(camellia_dec_blk) } + } } +}; -static inline void be128_to_u128(u128 *dst, const be128 *src) +static int ecb_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) { - dst->a = be64_to_cpu(src->a); - dst->b = be64_to_cpu(src->b); + return glue_ecb_crypt_128bit(&camellia_enc, desc, dst, src, nbytes); } -static inline void u128_inc(u128 *i) +static int ecb_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) { - i->b++; - if (!i->b) - i->a++; + return glue_ecb_crypt_128bit(&camellia_dec, desc, dst, src, nbytes); } -static void ctr_crypt_final(struct blkcipher_desc *desc, - struct blkcipher_walk *walk) +static int cbc_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) { - struct camellia_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - u8 keystream[CAMELLIA_BLOCK_SIZE]; - u8 *src = walk->src.virt.addr; - u8 *dst = walk->dst.virt.addr; - unsigned int nbytes = walk->nbytes; - u128 ctrblk; - - memcpy(keystream, src, nbytes); - camellia_enc_blk_xor(ctx, keystream, walk->iv); - memcpy(dst, keystream, nbytes); - - be128_to_u128(&ctrblk, (be128 *)walk->iv); - u128_inc(&ctrblk); - u128_to_be128((be128 *)walk->iv, &ctrblk); + return glue_cbc_encrypt_128bit(GLUE_FUNC_CAST(camellia_enc_blk), desc, + dst, src, nbytes); } -static unsigned int __ctr_crypt(struct blkcipher_desc *desc, - struct blkcipher_walk *walk) +static int cbc_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) { - struct camellia_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - unsigned int bsize = CAMELLIA_BLOCK_SIZE; - unsigned int nbytes = walk->nbytes; - u128 *src = (u128 *)walk->src.virt.addr; - u128 *dst = (u128 *)walk->dst.virt.addr; - u128 ctrblk; - be128 ctrblocks[2]; - - be128_to_u128(&ctrblk, (be128 *)walk->iv); - - /* Process two block batch */ - if (nbytes >= bsize * 2) { - do { - if (dst != src) { - dst[0] = src[0]; - dst[1] = src[1]; - } - - /* create ctrblks for parallel encrypt */ - u128_to_be128(&ctrblocks[0], &ctrblk); - u128_inc(&ctrblk); - u128_to_be128(&ctrblocks[1], &ctrblk); - u128_inc(&ctrblk); - - camellia_enc_blk_xor_2way(ctx, (u8 *)dst, - (u8 *)ctrblocks); - - src += 2; - dst += 2; - nbytes -= bsize * 2; - } while (nbytes >= bsize * 2); - - if (nbytes < bsize) - goto done; - } - - /* Handle leftovers */ - do { - if (dst != src) - *dst = *src; - - u128_to_be128(&ctrblocks[0], &ctrblk); - u128_inc(&ctrblk); - - camellia_enc_blk_xor(ctx, (u8 *)dst, (u8 *)ctrblocks); - - src += 1; - dst += 1; - nbytes -= bsize; - } while (nbytes >= bsize); - -done: - u128_to_be128((be128 *)walk->iv, &ctrblk); - return nbytes; + return glue_cbc_decrypt_128bit(&camellia_dec_cbc, desc, dst, src, + nbytes); } static int ctr_crypt(struct blkcipher_desc *desc, struct scatterlist *dst, struct scatterlist *src, unsigned int nbytes) { - struct blkcipher_walk walk; - int err; - - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt_block(desc, &walk, CAMELLIA_BLOCK_SIZE); - - while ((nbytes = walk.nbytes) >= CAMELLIA_BLOCK_SIZE) { - nbytes = __ctr_crypt(desc, &walk); - err = blkcipher_walk_done(desc, &walk, nbytes); - } - - if (walk.nbytes) { - ctr_crypt_final(desc, &walk); - err = blkcipher_walk_done(desc, &walk, 0); - } - - return err; + return glue_ctr_crypt_128bit(&camellia_ctr, desc, dst, src, nbytes); } static void encrypt_callback(void *priv, u8 *srcdst, unsigned int nbytes) diff --git a/crypto/Kconfig b/crypto/Kconfig index a86c2fb92ea..72828fafe09 100644 --- a/crypto/Kconfig +++ b/crypto/Kconfig @@ -671,6 +671,7 @@ config CRYPTO_CAMELLIA_X86_64 depends on X86 && 64BIT depends on CRYPTO select CRYPTO_ALGAPI + select CRYPTO_GLUE_HELPER_X86 select CRYPTO_LRW select CRYPTO_XTS help -- cgit v1.2.3 From 414cb5e7cc6e258fe36e2c3cc3ef1ff2e246c0e3 Mon Sep 17 00:00:00 2001 From: Jussi Kivilinna Date: Mon, 18 Jun 2012 14:07:34 +0300 Subject: crypto: twofish-x86_64-3way - remove duplicated glue code and use shared glue code from glue_helper Now that shared glue code is available, convert twofish-x86_64-3way to use it. Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu --- arch/x86/crypto/twofish_glue_3way.c | 365 +++++++++--------------------------- crypto/Kconfig | 1 + 2 files changed, 94 insertions(+), 272 deletions(-) (limited to 'crypto') diff --git a/arch/x86/crypto/twofish_glue_3way.c b/arch/x86/crypto/twofish_glue_3way.c index 77e4e55a266..25bf5e9b006 100644 --- a/arch/x86/crypto/twofish_glue_3way.c +++ b/arch/x86/crypto/twofish_glue_3way.c @@ -3,11 +3,6 @@ * * Copyright (c) 2011 Jussi Kivilinna * - * CBC & ECB parts based on code (crypto/cbc.c,ecb.c) by: - * Copyright (c) 2006 Herbert Xu - * CTR part based on code (crypto/ctr.c) by: - * (C) Copyright IBM Corp. 2007 - Joy Latten - * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by * the Free Software Foundation; either version 2 of the License, or @@ -33,6 +28,7 @@ #include #include #include +#include #include #include @@ -62,311 +58,136 @@ static inline void twofish_enc_blk_xor_3way(struct twofish_ctx *ctx, u8 *dst, __twofish_enc_blk_3way(ctx, dst, src, true); } -static int ecb_crypt(struct blkcipher_desc *desc, struct blkcipher_walk *walk, - void (*fn)(struct twofish_ctx *, u8 *, const u8 *), - void (*fn_3way)(struct twofish_ctx *, u8 *, const u8 *)) +static void twofish_dec_blk_cbc_3way(void *ctx, u128 *dst, const u128 *src) { - struct twofish_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - unsigned int bsize = TF_BLOCK_SIZE; - unsigned int nbytes; - int err; - - err = blkcipher_walk_virt(desc, walk); - - while ((nbytes = walk->nbytes)) { - u8 *wsrc = walk->src.virt.addr; - u8 *wdst = walk->dst.virt.addr; - - /* Process three block batch */ - if (nbytes >= bsize * 3) { - do { - fn_3way(ctx, wdst, wsrc); + u128 ivs[2]; - wsrc += bsize * 3; - wdst += bsize * 3; - nbytes -= bsize * 3; - } while (nbytes >= bsize * 3); + ivs[0] = src[0]; + ivs[1] = src[1]; - if (nbytes < bsize) - goto done; - } - - /* Handle leftovers */ - do { - fn(ctx, wdst, wsrc); - - wsrc += bsize; - wdst += bsize; - nbytes -= bsize; - } while (nbytes >= bsize); - -done: - err = blkcipher_walk_done(desc, walk, nbytes); - } + twofish_dec_blk_3way(ctx, (u8 *)dst, (u8 *)src); - return err; + u128_xor(&dst[1], &dst[1], &ivs[0]); + u128_xor(&dst[2], &dst[2], &ivs[1]); } -static int ecb_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) +static void twofish_enc_blk_ctr(void *ctx, u128 *dst, const u128 *src, u128 *iv) { - struct blkcipher_walk walk; + be128 ctrblk; - blkcipher_walk_init(&walk, dst, src, nbytes); - return ecb_crypt(desc, &walk, twofish_enc_blk, twofish_enc_blk_3way); -} + if (dst != src) + *dst = *src; -static int ecb_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) -{ - struct blkcipher_walk walk; + u128_to_be128(&ctrblk, iv); + u128_inc(iv); - blkcipher_walk_init(&walk, dst, src, nbytes); - return ecb_crypt(desc, &walk, twofish_dec_blk, twofish_dec_blk_3way); + twofish_enc_blk(ctx, (u8 *)&ctrblk, (u8 *)&ctrblk); + u128_xor(dst, dst, (u128 *)&ctrblk); } -static unsigned int __cbc_encrypt(struct blkcipher_desc *desc, - struct blkcipher_walk *walk) +static void twofish_enc_blk_ctr_3way(void *ctx, u128 *dst, const u128 *src, + u128 *iv) { - struct twofish_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - unsigned int bsize = TF_BLOCK_SIZE; - unsigned int nbytes = walk->nbytes; - u128 *src = (u128 *)walk->src.virt.addr; - u128 *dst = (u128 *)walk->dst.virt.addr; - u128 *iv = (u128 *)walk->iv; - - do { - u128_xor(dst, src, iv); - twofish_enc_blk(ctx, (u8 *)dst, (u8 *)dst); - iv = dst; - - src += 1; - dst += 1; - nbytes -= bsize; - } while (nbytes >= bsize); - - u128_xor((u128 *)walk->iv, (u128 *)walk->iv, iv); - return nbytes; -} + be128 ctrblks[3]; -static int cbc_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) -{ - struct blkcipher_walk walk; - int err; - - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt(desc, &walk); - - while ((nbytes = walk.nbytes)) { - nbytes = __cbc_encrypt(desc, &walk); - err = blkcipher_walk_done(desc, &walk, nbytes); - } - - return err; -} - -static unsigned int __cbc_decrypt(struct blkcipher_desc *desc, - struct blkcipher_walk *walk) -{ - struct twofish_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - unsigned int bsize = TF_BLOCK_SIZE; - unsigned int nbytes = walk->nbytes; - u128 *src = (u128 *)walk->src.virt.addr; - u128 *dst = (u128 *)walk->dst.virt.addr; - u128 ivs[3 - 1]; - u128 last_iv; - - /* Start of the last block. */ - src += nbytes / bsize - 1; - dst += nbytes / bsize - 1; - - last_iv = *src; - - /* Process three block batch */ - if (nbytes >= bsize * 3) { - do { - nbytes -= bsize * (3 - 1); - src -= 3 - 1; - dst -= 3 - 1; - - ivs[0] = src[0]; - ivs[1] = src[1]; - - twofish_dec_blk_3way(ctx, (u8 *)dst, (u8 *)src); - - u128_xor(dst + 1, dst + 1, ivs + 0); - u128_xor(dst + 2, dst + 2, ivs + 1); - - nbytes -= bsize; - if (nbytes < bsize) - goto done; - - u128_xor(dst, dst, src - 1); - src -= 1; - dst -= 1; - } while (nbytes >= bsize * 3); - - if (nbytes < bsize) - goto done; - } - - /* Handle leftovers */ - for (;;) { - twofish_dec_blk(ctx, (u8 *)dst, (u8 *)src); - - nbytes -= bsize; - if (nbytes < bsize) - break; - - u128_xor(dst, dst, src - 1); - src -= 1; - dst -= 1; + if (dst != src) { + dst[0] = src[0]; + dst[1] = src[1]; + dst[2] = src[2]; } -done: - u128_xor(dst, dst, (u128 *)walk->iv); - *(u128 *)walk->iv = last_iv; + u128_to_be128(&ctrblks[0], iv); + u128_inc(iv); + u128_to_be128(&ctrblks[1], iv); + u128_inc(iv); + u128_to_be128(&ctrblks[2], iv); + u128_inc(iv); - return nbytes; + twofish_enc_blk_xor_3way(ctx, (u8 *)dst, (u8 *)ctrblks); } -static int cbc_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) -{ - struct blkcipher_walk walk; - int err; - - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt(desc, &walk); +static const struct common_glue_ctx twofish_enc = { + .num_funcs = 2, + .fpu_blocks_limit = -1, + + .funcs = { { + .num_blocks = 3, + .fn_u = { .ecb = GLUE_FUNC_CAST(twofish_enc_blk_3way) } + }, { + .num_blocks = 1, + .fn_u = { .ecb = GLUE_FUNC_CAST(twofish_enc_blk) } + } } +}; - while ((nbytes = walk.nbytes)) { - nbytes = __cbc_decrypt(desc, &walk); - err = blkcipher_walk_done(desc, &walk, nbytes); - } +static const struct common_glue_ctx twofish_ctr = { + .num_funcs = 2, + .fpu_blocks_limit = -1, + + .funcs = { { + .num_blocks = 3, + .fn_u = { .ecb = GLUE_FUNC_CAST(twofish_enc_blk_ctr_3way) } + }, { + .num_blocks = 1, + .fn_u = { .ecb = GLUE_FUNC_CAST(twofish_enc_blk_ctr) } + } } +}; - return err; -} +static const struct common_glue_ctx twofish_dec = { + .num_funcs = 2, + .fpu_blocks_limit = -1, + + .funcs = { { + .num_blocks = 3, + .fn_u = { .ecb = GLUE_FUNC_CAST(twofish_dec_blk_3way) } + }, { + .num_blocks = 1, + .fn_u = { .ecb = GLUE_FUNC_CAST(twofish_dec_blk) } + } } +}; -static inline void u128_to_be128(be128 *dst, const u128 *src) -{ - dst->a = cpu_to_be64(src->a); - dst->b = cpu_to_be64(src->b); -} +static const struct common_glue_ctx twofish_dec_cbc = { + .num_funcs = 2, + .fpu_blocks_limit = -1, + + .funcs = { { + .num_blocks = 3, + .fn_u = { .cbc = GLUE_CBC_FUNC_CAST(twofish_dec_blk_cbc_3way) } + }, { + .num_blocks = 1, + .fn_u = { .cbc = GLUE_CBC_FUNC_CAST(twofish_dec_blk) } + } } +}; -static inline void be128_to_u128(u128 *dst, const be128 *src) +static int ecb_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) { - dst->a = be64_to_cpu(src->a); - dst->b = be64_to_cpu(src->b); + return glue_ecb_crypt_128bit(&twofish_enc, desc, dst, src, nbytes); } -static inline void u128_inc(u128 *i) +static int ecb_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) { - i->b++; - if (!i->b) - i->a++; + return glue_ecb_crypt_128bit(&twofish_dec, desc, dst, src, nbytes); } -static void ctr_crypt_final(struct blkcipher_desc *desc, - struct blkcipher_walk *walk) +static int cbc_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) { - struct twofish_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - u8 *ctrblk = walk->iv; - u8 keystream[TF_BLOCK_SIZE]; - u8 *src = walk->src.virt.addr; - u8 *dst = walk->dst.virt.addr; - unsigned int nbytes = walk->nbytes; - - twofish_enc_blk(ctx, keystream, ctrblk); - crypto_xor(keystream, src, nbytes); - memcpy(dst, keystream, nbytes); - - crypto_inc(ctrblk, TF_BLOCK_SIZE); + return glue_cbc_encrypt_128bit(GLUE_FUNC_CAST(twofish_enc_blk), desc, + dst, src, nbytes); } -static unsigned int __ctr_crypt(struct blkcipher_desc *desc, - struct blkcipher_walk *walk) +static int cbc_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) { - struct twofish_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - unsigned int bsize = TF_BLOCK_SIZE; - unsigned int nbytes = walk->nbytes; - u128 *src = (u128 *)walk->src.virt.addr; - u128 *dst = (u128 *)walk->dst.virt.addr; - u128 ctrblk; - be128 ctrblocks[3]; - - be128_to_u128(&ctrblk, (be128 *)walk->iv); - - /* Process three block batch */ - if (nbytes >= bsize * 3) { - do { - if (dst != src) { - dst[0] = src[0]; - dst[1] = src[1]; - dst[2] = src[2]; - } - - /* create ctrblks for parallel encrypt */ - u128_to_be128(&ctrblocks[0], &ctrblk); - u128_inc(&ctrblk); - u128_to_be128(&ctrblocks[1], &ctrblk); - u128_inc(&ctrblk); - u128_to_be128(&ctrblocks[2], &ctrblk); - u128_inc(&ctrblk); - - twofish_enc_blk_xor_3way(ctx, (u8 *)dst, - (u8 *)ctrblocks); - - src += 3; - dst += 3; - nbytes -= bsize * 3; - } while (nbytes >= bsize * 3); - - if (nbytes < bsize) - goto done; - } - - /* Handle leftovers */ - do { - if (dst != src) - *dst = *src; - - u128_to_be128(&ctrblocks[0], &ctrblk); - u128_inc(&ctrblk); - - twofish_enc_blk(ctx, (u8 *)ctrblocks, (u8 *)ctrblocks); - u128_xor(dst, dst, (u128 *)ctrblocks); - - src += 1; - dst += 1; - nbytes -= bsize; - } while (nbytes >= bsize); - -done: - u128_to_be128((be128 *)walk->iv, &ctrblk); - return nbytes; + return glue_cbc_decrypt_128bit(&twofish_dec_cbc, desc, dst, src, + nbytes); } static int ctr_crypt(struct blkcipher_desc *desc, struct scatterlist *dst, struct scatterlist *src, unsigned int nbytes) { - struct blkcipher_walk walk; - int err; - - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt_block(desc, &walk, TF_BLOCK_SIZE); - - while ((nbytes = walk.nbytes) >= TF_BLOCK_SIZE) { - nbytes = __ctr_crypt(desc, &walk); - err = blkcipher_walk_done(desc, &walk, nbytes); - } - - if (walk.nbytes) { - ctr_crypt_final(desc, &walk); - err = blkcipher_walk_done(desc, &walk, 0); - } - - return err; + return glue_ctr_crypt_128bit(&twofish_ctr, desc, dst, src, nbytes); } static void encrypt_callback(void *priv, u8 *srcdst, unsigned int nbytes) diff --git a/crypto/Kconfig b/crypto/Kconfig index 72828fafe09..fc559caaa9a 100644 --- a/crypto/Kconfig +++ b/crypto/Kconfig @@ -935,6 +935,7 @@ config CRYPTO_TWOFISH_X86_64_3WAY select CRYPTO_ALGAPI select CRYPTO_TWOFISH_COMMON select CRYPTO_TWOFISH_X86_64 + select CRYPTO_GLUE_HELPER_X86 select CRYPTO_LRW select CRYPTO_XTS help -- cgit v1.2.3 From a7378d4e552ac139ae1cbbdfebfeaa9b18c948d0 Mon Sep 17 00:00:00 2001 From: Jussi Kivilinna Date: Mon, 18 Jun 2012 14:07:39 +0300 Subject: crypto: twofish-avx - remove duplicated glue code and use shared glue code from glue_helper Now that shared glue code is available, convert twofish-avx to use it. Cc: Johannes Goetzfried Signed-off-by: Jussi Kivilinna Signed-off-by: Herbert Xu --- arch/x86/crypto/twofish_avx_glue.c | 556 ++++++---------------------------- arch/x86/crypto/twofish_glue_3way.c | 46 +-- arch/x86/include/asm/crypto/twofish.h | 46 +++ crypto/Kconfig | 1 + 4 files changed, 162 insertions(+), 487 deletions(-) create mode 100644 arch/x86/include/asm/crypto/twofish.h (limited to 'crypto') diff --git a/arch/x86/crypto/twofish_avx_glue.c b/arch/x86/crypto/twofish_avx_glue.c index cabe058eba1..782b67ddaf6 100644 --- a/arch/x86/crypto/twofish_avx_glue.c +++ b/arch/x86/crypto/twofish_avx_glue.c @@ -4,9 +4,6 @@ * Copyright (C) 2012 Johannes Goetzfried * * - * Glue code based on serpent_sse2_glue.c by: - * Copyright (C) 2011 Jussi Kivilinna - * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by * the Free Software Foundation; either version 2 of the License, or @@ -39,38 +36,21 @@ #include #include #include +#include #include +#include #include #include #include - #define TWOFISH_PARALLEL_BLOCKS 8 -/* regular block cipher functions from twofish_x86_64 module */ -asmlinkage void twofish_enc_blk(struct twofish_ctx *ctx, u8 *dst, - const u8 *src); -asmlinkage void twofish_dec_blk(struct twofish_ctx *ctx, u8 *dst, - const u8 *src); - -/* 3-way parallel cipher functions from twofish_x86_64-3way module */ -asmlinkage void __twofish_enc_blk_3way(struct twofish_ctx *ctx, u8 *dst, - const u8 *src, bool xor); -asmlinkage void twofish_dec_blk_3way(struct twofish_ctx *ctx, u8 *dst, - const u8 *src); - static inline void twofish_enc_blk_3way(struct twofish_ctx *ctx, u8 *dst, const u8 *src) { __twofish_enc_blk_3way(ctx, dst, src, false); } -static inline void twofish_enc_blk_3way_xor(struct twofish_ctx *ctx, u8 *dst, - const u8 *src) -{ - __twofish_enc_blk_3way(ctx, dst, src, true); -} - /* 8-way parallel cipher functions */ asmlinkage void __twofish_enc_blk_8way(struct twofish_ctx *ctx, u8 *dst, const u8 *src, bool xor); @@ -95,423 +75,142 @@ static inline void twofish_dec_blk_xway(struct twofish_ctx *ctx, u8 *dst, twofish_dec_blk_8way(ctx, dst, src); } - -static inline bool twofish_fpu_begin(bool fpu_enabled, unsigned int nbytes) +static void twofish_dec_blk_cbc_xway(void *ctx, u128 *dst, const u128 *src) { - if (fpu_enabled) - return true; + u128 ivs[TWOFISH_PARALLEL_BLOCKS - 1]; + unsigned int j; - /* AVX is only used when chunk to be processed is large enough, so - * do not enable FPU until it is necessary. - */ - if (nbytes < TF_BLOCK_SIZE * TWOFISH_PARALLEL_BLOCKS) - return false; + for (j = 0; j < TWOFISH_PARALLEL_BLOCKS - 1; j++) + ivs[j] = src[j]; - kernel_fpu_begin(); - return true; -} + twofish_dec_blk_xway(ctx, (u8 *)dst, (u8 *)src); -static inline void twofish_fpu_end(bool fpu_enabled) -{ - if (fpu_enabled) - kernel_fpu_end(); + for (j = 0; j < TWOFISH_PARALLEL_BLOCKS - 1; j++) + u128_xor(dst + (j + 1), dst + (j + 1), ivs + j); } -static int ecb_crypt(struct blkcipher_desc *desc, struct blkcipher_walk *walk, - bool enc) +static void twofish_enc_blk_ctr_xway(void *ctx, u128 *dst, const u128 *src, + u128 *iv) { - bool fpu_enabled = false; - struct twofish_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - const unsigned int bsize = TF_BLOCK_SIZE; - unsigned int nbytes; - int err; + be128 ctrblks[TWOFISH_PARALLEL_BLOCKS]; + unsigned int i; - err = blkcipher_walk_virt(desc, walk); - desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + for (i = 0; i < TWOFISH_PARALLEL_BLOCKS; i++) { + if (dst != src) + dst[i] = src[i]; - while ((nbytes = walk->nbytes)) { - u8 *wsrc = walk->src.virt.addr; - u8 *wdst = walk->dst.virt.addr; - - fpu_enabled = twofish_fpu_begin(fpu_enabled, nbytes); - - /* Process multi-block batch */ - if (nbytes >= bsize * TWOFISH_PARALLEL_BLOCKS) { - do { - if (enc) - twofish_enc_blk_xway(ctx, wdst, wsrc); - else - twofish_dec_blk_xway(ctx, wdst, wsrc); - - wsrc += bsize * TWOFISH_PARALLEL_BLOCKS; - wdst += bsize * TWOFISH_PARALLEL_BLOCKS; - nbytes -= bsize * TWOFISH_PARALLEL_BLOCKS; - } while (nbytes >= bsize * TWOFISH_PARALLEL_BLOCKS); - - if (nbytes < bsize) - goto done; - } - - /* Process three block batch */ - if (nbytes >= bsize * 3) { - do { - if (enc) - twofish_enc_blk_3way(ctx, wdst, wsrc); - else - twofish_dec_blk_3way(ctx, wdst, wsrc); - - wsrc += bsize * 3; - wdst += bsize * 3; - nbytes -= bsize * 3; - } while (nbytes >= bsize * 3); - - if (nbytes < bsize) - goto done; - } - - /* Handle leftovers */ - do { - if (enc) - twofish_enc_blk(ctx, wdst, wsrc); - else - twofish_dec_blk(ctx, wdst, wsrc); - - wsrc += bsize; - wdst += bsize; - nbytes -= bsize; - } while (nbytes >= bsize); - -done: - err = blkcipher_walk_done(desc, walk, nbytes); + u128_to_be128(&ctrblks[i], iv); + u128_inc(iv); } - twofish_fpu_end(fpu_enabled); - return err; + twofish_enc_blk_xway_xor(ctx, (u8 *)dst, (u8 *)ctrblks); } +static const struct common_glue_ctx twofish_enc = { + .num_funcs = 3, + .fpu_blocks_limit = TWOFISH_PARALLEL_BLOCKS, + + .funcs = { { + .num_blocks = TWOFISH_PARALLEL_BLOCKS, + .fn_u = { .ecb = GLUE_FUNC_CAST(twofish_enc_blk_xway) } + }, { + .num_blocks = 3, + .fn_u = { .ecb = GLUE_FUNC_CAST(twofish_enc_blk_3way) } + }, { + .num_blocks = 1, + .fn_u = { .ecb = GLUE_FUNC_CAST(twofish_enc_blk) } + } } +}; + +static const struct common_glue_ctx twofish_ctr = { + .num_funcs = 3, + .fpu_blocks_limit = TWOFISH_PARALLEL_BLOCKS, + + .funcs = { { + .num_blocks = TWOFISH_PARALLEL_BLOCKS, + .fn_u = { .ctr = GLUE_CTR_FUNC_CAST(twofish_enc_blk_ctr_xway) } + }, { + .num_blocks = 3, + .fn_u = { .ctr = GLUE_CTR_FUNC_CAST(twofish_enc_blk_ctr_3way) } + }, { + .num_blocks = 1, + .fn_u = { .ctr = GLUE_CTR_FUNC_CAST(twofish_enc_blk_ctr) } + } } +}; + +static const struct common_glue_ctx twofish_dec = { + .num_funcs = 3, + .fpu_blocks_limit = TWOFISH_PARALLEL_BLOCKS, + + .funcs = { { + .num_blocks = TWOFISH_PARALLEL_BLOCKS, + .fn_u = { .ecb = GLUE_FUNC_CAST(twofish_dec_blk_xway) } + }, { + .num_blocks = 3, + .fn_u = { .ecb = GLUE_FUNC_CAST(twofish_dec_blk_3way) } + }, { + .num_blocks = 1, + .fn_u = { .ecb = GLUE_FUNC_CAST(twofish_dec_blk) } + } } +}; + +static const struct common_glue_ctx twofish_dec_cbc = { + .num_funcs = 3, + .fpu_blocks_limit = TWOFISH_PARALLEL_BLOCKS, + + .funcs = { { + .num_blocks = TWOFISH_PARALLEL_BLOCKS, + .fn_u = { .cbc = GLUE_CBC_FUNC_CAST(twofish_dec_blk_cbc_xway) } + }, { + .num_blocks = 3, + .fn_u = { .cbc = GLUE_CBC_FUNC_CAST(twofish_dec_blk_cbc_3way) } + }, { + .num_blocks = 1, + .fn_u = { .cbc = GLUE_CBC_FUNC_CAST(twofish_dec_blk) } + } } +}; + static int ecb_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, struct scatterlist *src, unsigned int nbytes) { - struct blkcipher_walk walk; - - blkcipher_walk_init(&walk, dst, src, nbytes); - return ecb_crypt(desc, &walk, true); + return glue_ecb_crypt_128bit(&twofish_enc, desc, dst, src, nbytes); } static int ecb_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, struct scatterlist *src, unsigned int nbytes) { - struct blkcipher_walk walk; - - blkcipher_walk_init(&walk, dst, src, nbytes); - return ecb_crypt(desc, &walk, false); -} - -static unsigned int __cbc_encrypt(struct blkcipher_desc *desc, - struct blkcipher_walk *walk) -{ - struct twofish_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - const unsigned int bsize = TF_BLOCK_SIZE; - unsigned int nbytes = walk->nbytes; - u128 *src = (u128 *)walk->src.virt.addr; - u128 *dst = (u128 *)walk->dst.virt.addr; - u128 *iv = (u128 *)walk->iv; - - do { - u128_xor(dst, src, iv); - twofish_enc_blk(ctx, (u8 *)dst, (u8 *)dst); - iv = dst; - - src += 1; - dst += 1; - nbytes -= bsize; - } while (nbytes >= bsize); - - u128_xor((u128 *)walk->iv, (u128 *)walk->iv, iv); - return nbytes; + return glue_ecb_crypt_128bit(&twofish_dec, desc, dst, src, nbytes); } static int cbc_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, struct scatterlist *src, unsigned int nbytes) { - struct blkcipher_walk walk; - int err; - - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt(desc, &walk); - - while ((nbytes = walk.nbytes)) { - nbytes = __cbc_encrypt(desc, &walk); - err = blkcipher_walk_done(desc, &walk, nbytes); - } - - return err; -} - -static unsigned int __cbc_decrypt(struct blkcipher_desc *desc, - struct blkcipher_walk *walk) -{ - struct twofish_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - const unsigned int bsize = TF_BLOCK_SIZE; - unsigned int nbytes = walk->nbytes; - u128 *src = (u128 *)walk->src.virt.addr; - u128 *dst = (u128 *)walk->dst.virt.addr; - u128 ivs[TWOFISH_PARALLEL_BLOCKS - 1]; - u128 last_iv; - int i; - - /* Start of the last block. */ - src += nbytes / bsize - 1; - dst += nbytes / bsize - 1; - - last_iv = *src; - - /* Process multi-block batch */ - if (nbytes >= bsize * TWOFISH_PARALLEL_BLOCKS) { - do { - nbytes -= bsize * (TWOFISH_PARALLEL_BLOCKS - 1); - src -= TWOFISH_PARALLEL_BLOCKS - 1; - dst -= TWOFISH_PARALLEL_BLOCKS - 1; - - for (i = 0; i < TWOFISH_PARALLEL_BLOCKS - 1; i++) - ivs[i] = src[i]; - - twofish_dec_blk_xway(ctx, (u8 *)dst, (u8 *)src); - - for (i = 0; i < TWOFISH_PARALLEL_BLOCKS - 1; i++) - u128_xor(dst + (i + 1), dst + (i + 1), ivs + i); - - nbytes -= bsize; - if (nbytes < bsize) - goto done; - - u128_xor(dst, dst, src - 1); - src -= 1; - dst -= 1; - } while (nbytes >= bsize * TWOFISH_PARALLEL_BLOCKS); - - if (nbytes < bsize) - goto done; - } - - /* Process three block batch */ - if (nbytes >= bsize * 3) { - do { - nbytes -= bsize * (3 - 1); - src -= 3 - 1; - dst -= 3 - 1; - - ivs[0] = src[0]; - ivs[1] = src[1]; - - twofish_dec_blk_3way(ctx, (u8 *)dst, (u8 *)src); - - u128_xor(dst + 1, dst + 1, ivs + 0); - u128_xor(dst + 2, dst + 2, ivs + 1); - - nbytes -= bsize; - if (nbytes < bsize) - goto done; - - u128_xor(dst, dst, src - 1); - src -= 1; - dst -= 1; - } while (nbytes >= bsize * 3); - - if (nbytes < bsize) - goto done; - } - - /* Handle leftovers */ - for (;;) { - twofish_dec_blk(ctx, (u8 *)dst, (u8 *)src); - - nbytes -= bsize; - if (nbytes < bsize) - break; - - u128_xor(dst, dst, src - 1); - src -= 1; - dst -= 1; - } - -done: - u128_xor(dst, dst, (u128 *)walk->iv); - *(u128 *)walk->iv = last_iv; - - return nbytes; + return glue_cbc_encrypt_128bit(GLUE_FUNC_CAST(twofish_enc_blk), desc, + dst, src, nbytes); } static int cbc_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, struct scatterlist *src, unsigned int nbytes) { - bool fpu_enabled = false; - struct blkcipher_walk walk; - int err; - - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt(desc, &walk); - desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; - - while ((nbytes = walk.nbytes)) { - fpu_enabled = twofish_fpu_begin(fpu_enabled, nbytes); - nbytes = __cbc_decrypt(desc, &walk); - err = blkcipher_walk_done(desc, &walk, nbytes); - } - - twofish_fpu_end(fpu_enabled); - return err; -} - -static inline void u128_to_be128(be128 *dst, const u128 *src) -{ - dst->a = cpu_to_be64(src->a); - dst->b = cpu_to_be64(src->b); + return glue_cbc_decrypt_128bit(&twofish_dec_cbc, desc, dst, src, + nbytes); } -static inline void be128_to_u128(u128 *dst, const be128 *src) -{ - dst->a = be64_to_cpu(src->a); - dst->b = be64_to_cpu(src->b); -} - -static inline void u128_inc(u128 *i) +static int ctr_crypt(struct blkcipher_desc *desc, struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) { - i->b++; - if (!i->b) - i->a++; + return glue_ctr_crypt_128bit(&twofish_ctr, desc, dst, src, nbytes); } -static void ctr_crypt_final(struct blkcipher_desc *desc, - struct blkcipher_walk *walk) +static inline bool twofish_fpu_begin(bool fpu_enabled, unsigned int nbytes) { - struct twofish_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - u8 *ctrblk = walk->iv; - u8 keystream[TF_BLOCK_SIZE]; - u8 *src = walk->src.virt.addr; - u8 *dst = walk->dst.virt.addr; - unsigned int nbytes = walk->nbytes; - - twofish_enc_blk(ctx, keystream, ctrblk); - crypto_xor(keystream, src, nbytes); - memcpy(dst, keystream, nbytes); - - crypto_inc(ctrblk, TF_BLOCK_SIZE); + return glue_fpu_begin(TF_BLOCK_SIZE, TWOFISH_PARALLEL_BLOCKS, NULL, + fpu_enabled, nbytes); } -static unsigned int __ctr_crypt(struct blkcipher_desc *desc, - struct blkcipher_walk *walk) -{ - struct twofish_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - const unsigned int bsize = TF_BLOCK_SIZE; - unsigned int nbytes = walk->nbytes; - u128 *src = (u128 *)walk->src.virt.addr; - u128 *dst = (u128 *)walk->dst.virt.addr; - u128 ctrblk; - be128 ctrblocks[TWOFISH_PARALLEL_BLOCKS]; - int i; - - be128_to_u128(&ctrblk, (be128 *)walk->iv); - - /* Process multi-block batch */ - if (nbytes >= bsize * TWOFISH_PARALLEL_BLOCKS) { - do { - /* create ctrblks for parallel encrypt */ - for (i = 0; i < TWOFISH_PARALLEL_BLOCKS; i++) { - if (dst != src) - dst[i] = src[i]; - - u128_to_be128(&ctrblocks[i], &ctrblk); - u128_inc(&ctrblk); - } - - twofish_enc_blk_xway_xor(ctx, (u8 *)dst, - (u8 *)ctrblocks); - - src += TWOFISH_PARALLEL_BLOCKS; - dst += TWOFISH_PARALLEL_BLOCKS; - nbytes -= bsize * TWOFISH_PARALLEL_BLOCKS; - } while (nbytes >= bsize * TWOFISH_PARALLEL_BLOCKS); - - if (nbytes < bsize) - goto done; - } - - /* Process three block batch */ - if (nbytes >= bsize * 3) { - do { - if (dst != src) { - dst[0] = src[0]; - dst[1] = src[1]; - dst[2] = src[2]; - } - - /* create ctrblks for parallel encrypt */ - u128_to_be128(&ctrblocks[0], &ctrblk); - u128_inc(&ctrblk); - u128_to_be128(&ctrblocks[1], &ctrblk); - u128_inc(&ctrblk); - u128_to_be128(&ctrblocks[2], &ctrblk); - u128_inc(&ctrblk); - - twofish_enc_blk_3way_xor(ctx, (u8 *)dst, - (u8 *)ctrblocks); - - src += 3; - dst += 3; - nbytes -= bsize * 3; - } while (nbytes >= bsize * 3); - - if (nbytes < bsize) - goto done; - } - - /* Handle leftovers */ - do { - if (dst != src) - *dst = *src; - - u128_to_be128(&ctrblocks[0], &ctrblk); - u128_inc(&ctrblk); - - twofish_enc_blk(ctx, (u8 *)ctrblocks, (u8 *)ctrblocks); - u128_xor(dst, dst, (u128 *)ctrblocks); - - src += 1; - dst += 1; - nbytes -= bsize; - } while (nbytes >= bsize); - -done: - u128_to_be128((be128 *)walk->iv, &ctrblk); - return nbytes; -} - -static int ctr_crypt(struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) +static inline void twofish_fpu_end(bool fpu_enabled) { - bool fpu_enabled = false; - struct blkcipher_walk walk; - int err; - - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt_block(desc, &walk, TF_BLOCK_SIZE); - desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; - - while ((nbytes = walk.nbytes) >= TF_BLOCK_SIZE) { - fpu_enabled = twofish_fpu_begin(fpu_enabled, nbytes); - nbytes = __ctr_crypt(desc, &walk); - err = blkcipher_walk_done(desc, &walk, nbytes); - } - - twofish_fpu_end(fpu_enabled); - - if (walk.nbytes) { - ctr_crypt_final(desc, &walk); - err = blkcipher_walk_done(desc, &walk, 0); - } - - return err; + glue_fpu_end(fpu_enabled); } struct crypt_priv { @@ -563,26 +262,6 @@ static void decrypt_callback(void *priv, u8 *srcdst, unsigned int nbytes) twofish_dec_blk(ctx->ctx, srcdst, srcdst); } -struct twofish_lrw_ctx { - struct lrw_table_ctx lrw_table; - struct twofish_ctx twofish_ctx; -}; - -static int lrw_twofish_setkey(struct crypto_tfm *tfm, const u8 *key, - unsigned int keylen) -{ - struct twofish_lrw_ctx *ctx = crypto_tfm_ctx(tfm); - int err; - - err = __twofish_setkey(&ctx->twofish_ctx, key, - keylen - TF_BLOCK_SIZE, &tfm->crt_flags); - if (err) - return err; - - return lrw_init_table(&ctx->lrw_table, key + keylen - - TF_BLOCK_SIZE); -} - static int lrw_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, struct scatterlist *src, unsigned int nbytes) { @@ -635,43 +314,6 @@ static int lrw_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, return ret; } -static void lrw_exit_tfm(struct crypto_tfm *tfm) -{ - struct twofish_lrw_ctx *ctx = crypto_tfm_ctx(tfm); - - lrw_free_table(&ctx->lrw_table); -} - -struct twofish_xts_ctx { - struct twofish_ctx tweak_ctx; - struct twofish_ctx crypt_ctx; -}; - -static int xts_twofish_setkey(struct crypto_tfm *tfm, const u8 *key, - unsigned int keylen) -{ - struct twofish_xts_ctx *ctx = crypto_tfm_ctx(tfm); - u32 *flags = &tfm->crt_flags; - int err; - - /* key consists of keys of equal size concatenated, therefore - * the length must be even - */ - if (keylen % 2) { - *flags |= CRYPTO_TFM_RES_BAD_KEY_LEN; - return -EINVAL; - } - - /* first half of xts-key is for crypt */ - err = __twofish_setkey(&ctx->crypt_ctx, key, keylen / 2, flags); - if (err) - return err; - - /* second half of xts-key is for tweak */ - return __twofish_setkey(&ctx->tweak_ctx, - key + keylen / 2, keylen / 2, flags); -} - static int xts_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, struct scatterlist *src, unsigned int nbytes) { @@ -798,7 +440,7 @@ static struct crypto_alg twofish_algs[10] = { { .cra_type = &crypto_blkcipher_type, .cra_module = THIS_MODULE, .cra_list = LIST_HEAD_INIT(twofish_algs[3].cra_list), - .cra_exit = lrw_exit_tfm, + .cra_exit = lrw_twofish_exit_tfm, .cra_u = { .blkcipher = { .min_keysize = TF_MIN_KEY_SIZE + diff --git a/arch/x86/crypto/twofish_glue_3way.c b/arch/x86/crypto/twofish_glue_3way.c index 25bf5e9b006..15f9347316c 100644 --- a/arch/x86/crypto/twofish_glue_3way.c +++ b/arch/x86/crypto/twofish_glue_3way.c @@ -28,22 +28,12 @@ #include #include #include +#include #include #include #include -/* regular block cipher functions from twofish_x86_64 module */ -asmlinkage void twofish_enc_blk(struct twofish_ctx *ctx, u8 *dst, - const u8 *src); -asmlinkage void twofish_dec_blk(struct twofish_ctx *ctx, u8 *dst, - const u8 *src); - -/* 3-way parallel cipher functions */ -asmlinkage void __twofish_enc_blk_3way(struct twofish_ctx *ctx, u8 *dst, - const u8 *src, bool xor); EXPORT_SYMBOL_GPL(__twofish_enc_blk_3way); -asmlinkage void twofish_dec_blk_3way(struct twofish_ctx *ctx, u8 *dst, - const u8 *src); EXPORT_SYMBOL_GPL(twofish_dec_blk_3way); static inline void twofish_enc_blk_3way(struct twofish_ctx *ctx, u8 *dst, @@ -58,7 +48,7 @@ static inline void twofish_enc_blk_xor_3way(struct twofish_ctx *ctx, u8 *dst, __twofish_enc_blk_3way(ctx, dst, src, true); } -static void twofish_dec_blk_cbc_3way(void *ctx, u128 *dst, const u128 *src) +void twofish_dec_blk_cbc_3way(void *ctx, u128 *dst, const u128 *src) { u128 ivs[2]; @@ -70,8 +60,9 @@ static void twofish_dec_blk_cbc_3way(void *ctx, u128 *dst, const u128 *src) u128_xor(&dst[1], &dst[1], &ivs[0]); u128_xor(&dst[2], &dst[2], &ivs[1]); } +EXPORT_SYMBOL_GPL(twofish_dec_blk_cbc_3way); -static void twofish_enc_blk_ctr(void *ctx, u128 *dst, const u128 *src, u128 *iv) +void twofish_enc_blk_ctr(void *ctx, u128 *dst, const u128 *src, u128 *iv) { be128 ctrblk; @@ -84,8 +75,9 @@ static void twofish_enc_blk_ctr(void *ctx, u128 *dst, const u128 *src, u128 *iv) twofish_enc_blk(ctx, (u8 *)&ctrblk, (u8 *)&ctrblk); u128_xor(dst, dst, (u128 *)&ctrblk); } +EXPORT_SYMBOL_GPL(twofish_enc_blk_ctr); -static void twofish_enc_blk_ctr_3way(void *ctx, u128 *dst, const u128 *src, +void twofish_enc_blk_ctr_3way(void *ctx, u128 *dst, const u128 *src, u128 *iv) { be128 ctrblks[3]; @@ -105,6 +97,7 @@ static void twofish_enc_blk_ctr_3way(void *ctx, u128 *dst, const u128 *src, twofish_enc_blk_xor_3way(ctx, (u8 *)dst, (u8 *)ctrblks); } +EXPORT_SYMBOL_GPL(twofish_enc_blk_ctr_3way); static const struct common_glue_ctx twofish_enc = { .num_funcs = 2, @@ -220,13 +213,8 @@ static void decrypt_callback(void *priv, u8 *srcdst, unsigned int nbytes) twofish_dec_blk(ctx, srcdst, srcdst); } -struct twofish_lrw_ctx { - struct lrw_table_ctx lrw_table; - struct twofish_ctx twofish_ctx; -}; - -static int lrw_twofish_setkey(struct crypto_tfm *tfm, const u8 *key, - unsigned int keylen) +int lrw_twofish_setkey(struct crypto_tfm *tfm, const u8 *key, + unsigned int keylen) { struct twofish_lrw_ctx *ctx = crypto_tfm_ctx(tfm); int err; @@ -238,6 +226,7 @@ static int lrw_twofish_setkey(struct crypto_tfm *tfm, const u8 *key, return lrw_init_table(&ctx->lrw_table, key + keylen - TF_BLOCK_SIZE); } +EXPORT_SYMBOL_GPL(lrw_twofish_setkey); static int lrw_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, struct scatterlist *src, unsigned int nbytes) @@ -273,20 +262,16 @@ static int lrw_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, return lrw_crypt(desc, dst, src, nbytes, &req); } -static void lrw_exit_tfm(struct crypto_tfm *tfm) +void lrw_twofish_exit_tfm(struct crypto_tfm *tfm) { struct twofish_lrw_ctx *ctx = crypto_tfm_ctx(tfm); lrw_free_table(&ctx->lrw_table); } +EXPORT_SYMBOL_GPL(lrw_twofish_exit_tfm); -struct twofish_xts_ctx { - struct twofish_ctx tweak_ctx; - struct twofish_ctx crypt_ctx; -}; - -static int xts_twofish_setkey(struct crypto_tfm *tfm, const u8 *key, - unsigned int keylen) +int xts_twofish_setkey(struct crypto_tfm *tfm, const u8 *key, + unsigned int keylen) { struct twofish_xts_ctx *ctx = crypto_tfm_ctx(tfm); u32 *flags = &tfm->crt_flags; @@ -309,6 +294,7 @@ static int xts_twofish_setkey(struct crypto_tfm *tfm, const u8 *key, return __twofish_setkey(&ctx->tweak_ctx, key + keylen / 2, keylen / 2, flags); } +EXPORT_SYMBOL_GPL(xts_twofish_setkey); static int xts_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, struct scatterlist *src, unsigned int nbytes) @@ -419,7 +405,7 @@ static struct crypto_alg tf_algs[5] = { { .cra_type = &crypto_blkcipher_type, .cra_module = THIS_MODULE, .cra_list = LIST_HEAD_INIT(tf_algs[3].cra_list), - .cra_exit = lrw_exit_tfm, + .cra_exit = lrw_twofish_exit_tfm, .cra_u = { .blkcipher = { .min_keysize = TF_MIN_KEY_SIZE + TF_BLOCK_SIZE, diff --git a/arch/x86/include/asm/crypto/twofish.h b/arch/x86/include/asm/crypto/twofish.h new file mode 100644 index 00000000000..9d2c514bd5f --- /dev/null +++ b/arch/x86/include/asm/crypto/twofish.h @@ -0,0 +1,46 @@ +#ifndef ASM_X86_TWOFISH_H +#define ASM_X86_TWOFISH_H + +#include +#include +#include +#include + +struct twofish_lrw_ctx { + struct lrw_table_ctx lrw_table; + struct twofish_ctx twofish_ctx; +}; + +struct twofish_xts_ctx { + struct twofish_ctx tweak_ctx; + struct twofish_ctx crypt_ctx; +}; + +/* regular block cipher functions from twofish_x86_64 module */ +asmlinkage void twofish_enc_blk(struct twofish_ctx *ctx, u8 *dst, + const u8 *src); +asmlinkage void twofish_dec_blk(struct twofish_ctx *ctx, u8 *dst, + const u8 *src); + +/* 3-way parallel cipher functions */ +asmlinkage void __twofish_enc_blk_3way(struct twofish_ctx *ctx, u8 *dst, + const u8 *src, bool xor); +asmlinkage void twofish_dec_blk_3way(struct twofish_ctx *ctx, u8 *dst, + const u8 *src); + +/* helpers from twofish_x86_64-3way module */ +extern void twofish_dec_blk_cbc_3way(void *ctx, u128 *dst, const u128 *src); +extern void twofish_enc_blk_ctr(void *ctx, u128 *dst, const u128 *src, + u128 *iv); +extern void twofish_enc_blk_ctr_3way(void *ctx, u128 *dst, const u128 *src, + u128 *iv); + +extern int lrw_twofish_setkey(struct crypto_tfm *tfm, const u8 *key, + unsigned int keylen); + +extern void lrw_twofish_exit_tfm(struct crypto_tfm *tfm); + +extern int xts_twofish_setkey(struct crypto_tfm *tfm, const u8 *key, + unsigned int keylen); + +#endif /* ASM_X86_TWOFISH_H */ diff --git a/crypto/Kconfig b/crypto/Kconfig index fc559caaa9a..02e30377469 100644 --- a/crypto/Kconfig +++ b/crypto/Kconfig @@ -958,6 +958,7 @@ config CRYPTO_TWOFISH_AVX_X86_64 select CRYPTO_ALGAPI select CRYPTO_CRYPTD select CRYPTO_ABLK_HELPER_X86 + select CRYPTO_GLUE_HELPER_X86 select CRYPTO_TWOFISH_COMMON select CRYPTO_TWOFISH_X86_64 select CRYPTO_TWOFISH_X86_64_3WAY -- cgit v1.2.3 From b9b0f080fac541d88eee4d5175c3f6470a9e9189 Mon Sep 17 00:00:00 2001 From: Sebastian Andrzej Siewior Date: Tue, 26 Jun 2012 18:13:46 +0200 Subject: crypto: arc4 - now arc needs blockcipher support Since commit ce6dd368 ("crypto: arc4 - improve performance by adding ecb(arc4)) we need to pull in a blkcipher. |ERROR: "crypto_blkcipher_type" [crypto/arc4.ko] undefined! |ERROR: "blkcipher_walk_done" [crypto/arc4.ko] undefined! |ERROR: "blkcipher_walk_virt" [crypto/arc4.ko] undefined! Signed-off-by: Sebastian Andrzej Siewior Signed-off-by: Herbert Xu --- crypto/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'crypto') diff --git a/crypto/Kconfig b/crypto/Kconfig index 02e30377469..a3238051b03 100644 --- a/crypto/Kconfig +++ b/crypto/Kconfig @@ -604,7 +604,7 @@ config CRYPTO_ANUBIS config CRYPTO_ARC4 tristate "ARC4 cipher algorithm" - select CRYPTO_ALGAPI + select CRYPTO_BLKCIPHER help ARC4 cipher algorithm. -- cgit v1.2.3 From 26c8aaebc188b539a0a9077350009a059464097d Mon Sep 17 00:00:00 2001 From: Steffen Klassert Date: Wed, 27 Jun 2012 13:31:01 +0200 Subject: crypto: algapi - Fix hang on crypto allocation git commit 398710379 (crypto: algapi - Move larval completion into algboss) replaced accidentally a call to complete_all() by a call to complete(). This causes a hang on crypto allocation if we have more than one larval waiter. This pach restores the call to complete_all(). Signed-off-by: Steffen Klassert Signed-off-by: Herbert Xu --- crypto/algboss.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'crypto') diff --git a/crypto/algboss.c b/crypto/algboss.c index f97027e7d99..769219b2930 100644 --- a/crypto/algboss.c +++ b/crypto/algboss.c @@ -87,7 +87,7 @@ static int cryptomgr_probe(void *data) crypto_tmpl_put(tmpl); out: - complete(param->completion); + complete_all(param->completion); kfree(param); module_put_and_exit(0); } -- cgit v1.2.3 From 6c79294f44fd7d1122cbaabff3b9815b074c0dd0 Mon Sep 17 00:00:00 2001 From: Milan Broz Date: Fri, 29 Jun 2012 22:08:09 +0200 Subject: crypto: testmgr - allow aesni-intel and ghash_clmulni-intel in fips mode Patch 863b557a88f8c033f7419fabafef4712a5055f85 added NULL entries for intel accelerated drivers but did not marked these fips allowed. This cause panic if running tests with fips=1. For ghash, fips_allowed flag was added in patch 18c0ebd2d8194cce4b3f67e2903fa01bea892cbc. Without patch, "modprobe tcrypt" fails with alg: skcipher: Failed to load transform for cbc-aes-aesni: -2 cbc-aes-aesni: cbc(aes) alg self test failed in fips mode! (panic) Also add missing cryptd(__driver-cbc-aes-aesni) and cryptd(__driver-gcm-aes-aesni) test to complement null tests above, otherwise system complains with alg: No test for __cbc-aes-aesni (cryptd(__driver-cbc-aes-aesni)) alg: No test for __gcm-aes-aesni (cryptd(__driver-gcm-aes-aesni)) Signed-off-by: Milan Broz Signed-off-by: Paul Wouters Signed-off-by: Herbert Xu --- crypto/testmgr.c | 38 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 38 insertions(+) (limited to 'crypto') diff --git a/crypto/testmgr.c b/crypto/testmgr.c index 36748a5996e..4308a11f129 100644 --- a/crypto/testmgr.c +++ b/crypto/testmgr.c @@ -1581,6 +1581,7 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "__driver-cbc-aes-aesni", .test = alg_test_null, + .fips_allowed = 1, .suite = { .cipher = { .enc = { @@ -1641,6 +1642,7 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "__driver-ecb-aes-aesni", .test = alg_test_null, + .fips_allowed = 1, .suite = { .cipher = { .enc = { @@ -1701,6 +1703,7 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "__ghash-pclmulqdqni", .test = alg_test_null, + .fips_allowed = 1, .suite = { .hash = { .vecs = NULL, @@ -1865,9 +1868,26 @@ static const struct alg_test_desc alg_test_descs[] = { .count = CRC32C_TEST_VECTORS } } + }, { + .alg = "cryptd(__driver-cbc-aes-aesni)", + .test = alg_test_null, + .fips_allowed = 1, + .suite = { + .cipher = { + .enc = { + .vecs = NULL, + .count = 0 + }, + .dec = { + .vecs = NULL, + .count = 0 + } + } + } }, { .alg = "cryptd(__driver-ecb-aes-aesni)", .test = alg_test_null, + .fips_allowed = 1, .suite = { .cipher = { .enc = { @@ -1925,9 +1945,26 @@ static const struct alg_test_desc alg_test_descs[] = { } } } + }, { + .alg = "cryptd(__driver-gcm-aes-aesni)", + .test = alg_test_null, + .fips_allowed = 1, + .suite = { + .cipher = { + .enc = { + .vecs = NULL, + .count = 0 + }, + .dec = { + .vecs = NULL, + .count = 0 + } + } + } }, { .alg = "cryptd(__ghash-pclmulqdqni)", .test = alg_test_null, + .fips_allowed = 1, .suite = { .hash = { .vecs = NULL, @@ -2043,6 +2080,7 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "ecb(__aes-aesni)", .test = alg_test_null, + .fips_allowed = 1, .suite = { .cipher = { .enc = { -- cgit v1.2.3 From de197533485c09598215802b0e401a688e172573 Mon Sep 17 00:00:00 2001 From: Nicolas Royer Date: Sun, 1 Jul 2012 19:19:47 +0200 Subject: crypto: atmel - add new tests to tcrypt MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit - set sg buffers size equals to message size - add cfb & ofb tests for AES, DES & TDES Signed-off-by: Nicolas Royer Acked-by: Nicolas Ferre Acked-by: Eric Bénard Tested-by: Eric Bénard Signed-off-by: Herbert Xu --- crypto/tcrypt.c | 50 +++++++++++++++++++++++++++++++++++++++++++++----- 1 file changed, 45 insertions(+), 5 deletions(-) (limited to 'crypto') diff --git a/crypto/tcrypt.c b/crypto/tcrypt.c index 58ad4fe5d20..5cf2ccb1540 100644 --- a/crypto/tcrypt.c +++ b/crypto/tcrypt.c @@ -809,7 +809,7 @@ static void test_acipher_speed(const char *algo, int enc, unsigned int sec, struct cipher_speed_template *template, unsigned int tcount, u8 *keysize) { - unsigned int ret, i, j, iv_len; + unsigned int ret, i, j, k, iv_len; struct tcrypt_result tresult; const char *key; char iv[128]; @@ -883,11 +883,23 @@ static void test_acipher_speed(const char *algo, int enc, unsigned int sec, } sg_init_table(sg, TVMEMSIZE); - sg_set_buf(sg, tvmem[0] + *keysize, + + k = *keysize + *b_size; + if (k > PAGE_SIZE) { + sg_set_buf(sg, tvmem[0] + *keysize, PAGE_SIZE - *keysize); - for (j = 1; j < TVMEMSIZE; j++) { - sg_set_buf(sg + j, tvmem[j], PAGE_SIZE); - memset(tvmem[j], 0xff, PAGE_SIZE); + k -= PAGE_SIZE; + j = 1; + while (k > PAGE_SIZE) { + sg_set_buf(sg + j, tvmem[j], PAGE_SIZE); + memset(tvmem[j], 0xff, PAGE_SIZE); + j++; + k -= PAGE_SIZE; + } + sg_set_buf(sg + j, tvmem[j], k); + memset(tvmem[j], 0xff, k); + } else { + sg_set_buf(sg, tvmem[0] + *keysize, *b_size); } iv_len = crypto_ablkcipher_ivsize(tfm); @@ -1520,6 +1532,14 @@ static int do_test(int m) speed_template_16_24_32); test_acipher_speed("ctr(aes)", DECRYPT, sec, NULL, 0, speed_template_16_24_32); + test_acipher_speed("cfb(aes)", ENCRYPT, sec, NULL, 0, + speed_template_16_24_32); + test_acipher_speed("cfb(aes)", DECRYPT, sec, NULL, 0, + speed_template_16_24_32); + test_acipher_speed("ofb(aes)", ENCRYPT, sec, NULL, 0, + speed_template_16_24_32); + test_acipher_speed("ofb(aes)", DECRYPT, sec, NULL, 0, + speed_template_16_24_32); break; case 501: @@ -1535,6 +1555,18 @@ static int do_test(int m) test_acipher_speed("cbc(des3_ede)", DECRYPT, sec, des3_speed_template, DES3_SPEED_VECTORS, speed_template_24); + test_acipher_speed("cfb(des3_ede)", ENCRYPT, sec, + des3_speed_template, DES3_SPEED_VECTORS, + speed_template_24); + test_acipher_speed("cfb(des3_ede)", DECRYPT, sec, + des3_speed_template, DES3_SPEED_VECTORS, + speed_template_24); + test_acipher_speed("ofb(des3_ede)", ENCRYPT, sec, + des3_speed_template, DES3_SPEED_VECTORS, + speed_template_24); + test_acipher_speed("ofb(des3_ede)", DECRYPT, sec, + des3_speed_template, DES3_SPEED_VECTORS, + speed_template_24); break; case 502: @@ -1546,6 +1578,14 @@ static int do_test(int m) speed_template_8); test_acipher_speed("cbc(des)", DECRYPT, sec, NULL, 0, speed_template_8); + test_acipher_speed("cfb(des)", ENCRYPT, sec, NULL, 0, + speed_template_8); + test_acipher_speed("cfb(des)", DECRYPT, sec, NULL, 0, + speed_template_8); + test_acipher_speed("ofb(des)", ENCRYPT, sec, NULL, 0, + speed_template_8); + test_acipher_speed("ofb(des)", DECRYPT, sec, NULL, 0, + speed_template_8); break; case 503: -- cgit v1.2.3 From e46e9a46386bca8e80a6467b5c643dc494861896 Mon Sep 17 00:00:00 2001 From: Horia Geanta Date: Tue, 3 Jul 2012 19:16:54 +0300 Subject: crypto: testmgr - add aead cbc aes hmac sha1,256,512 test vectors Test vectors were generated starting from existing CBC(AES) test vectors (RFC3602, NIST SP800-38A) and adding HMAC(SHA*) computed with Crypto++ and double-checked with HashCalc. Signed-off-by: Horia Geanta Signed-off-by: Herbert Xu --- crypto/testmgr.c | 36 +++ crypto/testmgr.h | 834 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 870 insertions(+) (limited to 'crypto') diff --git a/crypto/testmgr.c b/crypto/testmgr.c index 4308a11f129..a2ca7431760 100644 --- a/crypto/testmgr.c +++ b/crypto/testmgr.c @@ -1720,6 +1720,42 @@ static const struct alg_test_desc alg_test_descs[] = { .count = ANSI_CPRNG_AES_TEST_VECTORS } } + }, { + .alg = "authenc(hmac(sha1),cbc(aes))", + .test = alg_test_aead, + .fips_allowed = 1, + .suite = { + .aead = { + .enc = { + .vecs = hmac_sha1_aes_cbc_enc_tv_template, + .count = HMAC_SHA1_AES_CBC_ENC_TEST_VECTORS + } + } + } + }, { + .alg = "authenc(hmac(sha256),cbc(aes))", + .test = alg_test_aead, + .fips_allowed = 1, + .suite = { + .aead = { + .enc = { + .vecs = hmac_sha256_aes_cbc_enc_tv_template, + .count = HMAC_SHA256_AES_CBC_ENC_TEST_VECTORS + } + } + } + }, { + .alg = "authenc(hmac(sha512),cbc(aes))", + .test = alg_test_aead, + .fips_allowed = 1, + .suite = { + .aead = { + .enc = { + .vecs = hmac_sha512_aes_cbc_enc_tv_template, + .count = HMAC_SHA512_AES_CBC_ENC_TEST_VECTORS + } + } + } }, { .alg = "cbc(aes)", .test = alg_test_skcipher, diff --git a/crypto/testmgr.h b/crypto/testmgr.h index 4d84fe48df0..f8179e0344e 100644 --- a/crypto/testmgr.h +++ b/crypto/testmgr.h @@ -6975,6 +6975,9 @@ static struct cipher_testvec cast6_dec_tv_template[] = { #define AES_DEC_TEST_VECTORS 3 #define AES_CBC_ENC_TEST_VECTORS 4 #define AES_CBC_DEC_TEST_VECTORS 4 +#define HMAC_SHA1_AES_CBC_ENC_TEST_VECTORS 7 +#define HMAC_SHA256_AES_CBC_ENC_TEST_VECTORS 7 +#define HMAC_SHA512_AES_CBC_ENC_TEST_VECTORS 7 #define AES_LRW_ENC_TEST_VECTORS 8 #define AES_LRW_DEC_TEST_VECTORS 8 #define AES_XTS_ENC_TEST_VECTORS 5 @@ -7232,6 +7235,837 @@ static struct cipher_testvec aes_cbc_dec_tv_template[] = { }, }; +static struct aead_testvec hmac_sha1_aes_cbc_enc_tv_template[] = { + { /* RFC 3602 Case 1 */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x10" /* enc key length */ + "\x00\x00\x00\x00\x00\x00\x00\x00" + "\x00\x00\x00\x00\x00\x00\x00\x00" + "\x00\x00\x00\x00" + "\x06\xa9\x21\x40\x36\xb8\xa1\x5b" + "\x51\x2e\x03\xd5\x34\x12\x00\x06", + .klen = 8 + 20 + 16, + .iv = "\x3d\xaf\xba\x42\x9d\x9e\xb4\x30" + "\xb4\x22\xda\x80\x2c\x9f\xac\x41", + .input = "Single block msg", + .ilen = 16, + .result = "\xe3\x53\x77\x9c\x10\x79\xae\xb8" + "\x27\x08\x94\x2d\xbe\x77\x18\x1a" + "\x1b\x13\xcb\xaf\x89\x5e\xe1\x2c" + "\x13\xc5\x2e\xa3\xcc\xed\xdc\xb5" + "\x03\x71\xa2\x06", + .rlen = 16 + 20, + }, { /* RFC 3602 Case 2 */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x10" /* enc key length */ + "\x20\x21\x22\x23\x24\x25\x26\x27" + "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f" + "\x30\x31\x32\x33" + "\xc2\x86\x69\x6d\x88\x7c\x9a\xa0" + "\x61\x1b\xbb\x3e\x20\x25\xa4\x5a", + .klen = 8 + 20 + 16, + .iv = "\x56\x2e\x17\x99\x6d\x09\x3d\x28" + "\xdd\xb3\xba\x69\x5a\x2e\x6f\x58", + .input = "\x00\x01\x02\x03\x04\x05\x06\x07" + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f" + "\x10\x11\x12\x13\x14\x15\x16\x17" + "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f", + .ilen = 32, + .result = "\xd2\x96\xcd\x94\xc2\xcc\xcf\x8a" + "\x3a\x86\x30\x28\xb5\xe1\xdc\x0a" + "\x75\x86\x60\x2d\x25\x3c\xff\xf9" + "\x1b\x82\x66\xbe\xa6\xd6\x1a\xb1" + "\xad\x9b\x4c\x5c\x85\xe1\xda\xae" + "\xee\x81\x4e\xd7\xdb\x74\xcf\x58" + "\x65\x39\xf8\xde", + .rlen = 32 + 20, + }, { /* RFC 3602 Case 3 */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x10" /* enc key length */ + "\x11\x22\x33\x44\x55\x66\x77\x88" + "\x99\xaa\xbb\xcc\xdd\xee\xff\x11" + "\x22\x33\x44\x55" + "\x6c\x3e\xa0\x47\x76\x30\xce\x21" + "\xa2\xce\x33\x4a\xa7\x46\xc2\xcd", + .klen = 8 + 20 + 16, + .iv = "\xc7\x82\xdc\x4c\x09\x8c\x66\xcb" + "\xd9\xcd\x27\xd8\x25\x68\x2c\x81", + .input = "This is a 48-byte message (exactly 3 AES blocks)", + .ilen = 48, + .result = "\xd0\xa0\x2b\x38\x36\x45\x17\x53" + "\xd4\x93\x66\x5d\x33\xf0\xe8\x86" + "\x2d\xea\x54\xcd\xb2\x93\xab\xc7" + "\x50\x69\x39\x27\x67\x72\xf8\xd5" + "\x02\x1c\x19\x21\x6b\xad\x52\x5c" + "\x85\x79\x69\x5d\x83\xba\x26\x84" + "\xc2\xec\x0c\xf8\x7f\x05\xba\xca" + "\xff\xee\x4c\xd0\x93\xe6\x36\x7f" + "\x8d\x62\xf2\x1e", + .rlen = 48 + 20, + }, { /* RFC 3602 Case 4 */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x10" /* enc key length */ + "\x11\x22\x33\x44\x55\x66\x77\x88" + "\x99\xaa\xbb\xcc\xdd\xee\xff\x11" + "\x22\x33\x44\x55" + "\x56\xe4\x7a\x38\xc5\x59\x89\x74" + "\xbc\x46\x90\x3d\xba\x29\x03\x49", + .klen = 8 + 20 + 16, + .iv = "\x8c\xe8\x2e\xef\xbe\xa0\xda\x3c" + "\x44\x69\x9e\xd7\xdb\x51\xb7\xd9", + .input = "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7" + "\xa8\xa9\xaa\xab\xac\xad\xae\xaf" + "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7" + "\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf" + "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7" + "\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf" + "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7" + "\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf", + .ilen = 64, + .result = "\xc3\x0e\x32\xff\xed\xc0\x77\x4e" + "\x6a\xff\x6a\xf0\x86\x9f\x71\xaa" + "\x0f\x3a\xf0\x7a\x9a\x31\xa9\xc6" + "\x84\xdb\x20\x7e\xb0\xef\x8e\x4e" + "\x35\x90\x7a\xa6\x32\xc3\xff\xdf" + "\x86\x8b\xb7\xb2\x9d\x3d\x46\xad" + "\x83\xce\x9f\x9a\x10\x2e\xe9\x9d" + "\x49\xa5\x3e\x87\xf4\xc3\xda\x55" + "\x1c\x45\x57\xa9\x56\xcb\xa9\x2d" + "\x18\xac\xf1\xc7\x5d\xd1\xcd\x0d" + "\x1d\xbe\xc6\xe9", + .rlen = 64 + 20, + }, { /* RFC 3602 Case 5 */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x10" /* enc key length */ + "\x11\x22\x33\x44\x55\x66\x77\x88" + "\x99\xaa\xbb\xcc\xdd\xee\xff\x11" + "\x22\x33\x44\x55" + "\x90\xd3\x82\xb4\x10\xee\xba\x7a" + "\xd9\x38\xc4\x6c\xec\x1a\x82\xbf", + .klen = 8 + 20 + 16, + .iv = "\xe9\x6e\x8c\x08\xab\x46\x57\x63" + "\xfd\x09\x8d\x45\xdd\x3f\xf8\x93", + .assoc = "\x00\x00\x43\x21\x00\x00\x00\x01", + .alen = 8, + .input = "\x08\x00\x0e\xbd\xa7\x0a\x00\x00" + "\x8e\x9c\x08\x3d\xb9\x5b\x07\x00" + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f" + "\x10\x11\x12\x13\x14\x15\x16\x17" + "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f" + "\x20\x21\x22\x23\x24\x25\x26\x27" + "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f" + "\x30\x31\x32\x33\x34\x35\x36\x37" + "\x01\x02\x03\x04\x05\x06\x07\x08" + "\x09\x0a\x0b\x0c\x0d\x0e\x0e\x01", + .ilen = 80, + .result = "\xf6\x63\xc2\x5d\x32\x5c\x18\xc6" + "\xa9\x45\x3e\x19\x4e\x12\x08\x49" + "\xa4\x87\x0b\x66\xcc\x6b\x99\x65" + "\x33\x00\x13\xb4\x89\x8d\xc8\x56" + "\xa4\x69\x9e\x52\x3a\x55\xdb\x08" + "\x0b\x59\xec\x3a\x8e\x4b\x7e\x52" + "\x77\x5b\x07\xd1\xdb\x34\xed\x9c" + "\x53\x8a\xb5\x0c\x55\x1b\x87\x4a" + "\xa2\x69\xad\xd0\x47\xad\x2d\x59" + "\x13\xac\x19\xb7\xcf\xba\xd4\xa6" + "\x58\xc6\x84\x75\xe4\xe9\x6b\x0c" + "\xe1\xc5\x0b\x73\x4d\x82\x55\xa8" + "\x85\xe1\x59\xf7", + .rlen = 80 + 20, + }, { /* NIST SP800-38A F.2.3 CBC-AES192.Encrypt */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x18" /* enc key length */ + "\x11\x22\x33\x44\x55\x66\x77\x88" + "\x99\xaa\xbb\xcc\xdd\xee\xff\x11" + "\x22\x33\x44\x55" + "\x8e\x73\xb0\xf7\xda\x0e\x64\x52" + "\xc8\x10\xf3\x2b\x80\x90\x79\xe5" + "\x62\xf8\xea\xd2\x52\x2c\x6b\x7b", + .klen = 8 + 20 + 24, + .iv = "\x00\x01\x02\x03\x04\x05\x06\x07" + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f", + .input = "\x6b\xc1\xbe\xe2\x2e\x40\x9f\x96" + "\xe9\x3d\x7e\x11\x73\x93\x17\x2a" + "\xae\x2d\x8a\x57\x1e\x03\xac\x9c" + "\x9e\xb7\x6f\xac\x45\xaf\x8e\x51" + "\x30\xc8\x1c\x46\xa3\x5c\xe4\x11" + "\xe5\xfb\xc1\x19\x1a\x0a\x52\xef" + "\xf6\x9f\x24\x45\xdf\x4f\x9b\x17" + "\xad\x2b\x41\x7b\xe6\x6c\x37\x10", + .ilen = 64, + .result = "\x4f\x02\x1d\xb2\x43\xbc\x63\x3d" + "\x71\x78\x18\x3a\x9f\xa0\x71\xe8" + "\xb4\xd9\xad\xa9\xad\x7d\xed\xf4" + "\xe5\xe7\x38\x76\x3f\x69\x14\x5a" + "\x57\x1b\x24\x20\x12\xfb\x7a\xe0" + "\x7f\xa9\xba\xac\x3d\xf1\x02\xe0" + "\x08\xb0\xe2\x79\x88\x59\x88\x81" + "\xd9\x20\xa9\xe6\x4f\x56\x15\xcd" + "\x73\xe3\x19\x3f\x8b\xc9\xc6\xf4" + "\x5a\xf1\x5b\xa8\x98\x07\xc5\x36" + "\x47\x4c\xfc\x36", + .rlen = 64 + 20, + }, { /* NIST SP800-38A F.2.5 CBC-AES256.Encrypt */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x20" /* enc key length */ + "\x11\x22\x33\x44\x55\x66\x77\x88" + "\x99\xaa\xbb\xcc\xdd\xee\xff\x11" + "\x22\x33\x44\x55" + "\x60\x3d\xeb\x10\x15\xca\x71\xbe" + "\x2b\x73\xae\xf0\x85\x7d\x77\x81" + "\x1f\x35\x2c\x07\x3b\x61\x08\xd7" + "\x2d\x98\x10\xa3\x09\x14\xdf\xf4", + .klen = 8 + 20 + 32, + .iv = "\x00\x01\x02\x03\x04\x05\x06\x07" + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f", + .input = "\x6b\xc1\xbe\xe2\x2e\x40\x9f\x96" + "\xe9\x3d\x7e\x11\x73\x93\x17\x2a" + "\xae\x2d\x8a\x57\x1e\x03\xac\x9c" + "\x9e\xb7\x6f\xac\x45\xaf\x8e\x51" + "\x30\xc8\x1c\x46\xa3\x5c\xe4\x11" + "\xe5\xfb\xc1\x19\x1a\x0a\x52\xef" + "\xf6\x9f\x24\x45\xdf\x4f\x9b\x17" + "\xad\x2b\x41\x7b\xe6\x6c\x37\x10", + .ilen = 64, + .result = "\xf5\x8c\x4c\x04\xd6\xe5\xf1\xba" + "\x77\x9e\xab\xfb\x5f\x7b\xfb\xd6" + "\x9c\xfc\x4e\x96\x7e\xdb\x80\x8d" + "\x67\x9f\x77\x7b\xc6\x70\x2c\x7d" + "\x39\xf2\x33\x69\xa9\xd9\xba\xcf" + "\xa5\x30\xe2\x63\x04\x23\x14\x61" + "\xb2\xeb\x05\xe2\xc3\x9b\xe9\xfc" + "\xda\x6c\x19\x07\x8c\x6a\x9d\x1b" + "\xa3\xe8\x9b\x17\xe3\xf4\x7f\xde" + "\x1b\x9f\xc6\x81\x26\x43\x4a\x87" + "\x51\xee\xd6\x4e", + .rlen = 64 + 20, + }, +}; + +static struct aead_testvec hmac_sha256_aes_cbc_enc_tv_template[] = { + { /* RFC 3602 Case 1 */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x10" /* enc key length */ + "\x00\x00\x00\x00\x00\x00\x00\x00" + "\x00\x00\x00\x00\x00\x00\x00\x00" + "\x00\x00\x00\x00\x00\x00\x00\x00" + "\x00\x00\x00\x00\x00\x00\x00\x00" + "\x06\xa9\x21\x40\x36\xb8\xa1\x5b" + "\x51\x2e\x03\xd5\x34\x12\x00\x06", + .klen = 8 + 32 + 16, + .iv = "\x3d\xaf\xba\x42\x9d\x9e\xb4\x30" + "\xb4\x22\xda\x80\x2c\x9f\xac\x41", + .input = "Single block msg", + .ilen = 16, + .result = "\xe3\x53\x77\x9c\x10\x79\xae\xb8" + "\x27\x08\x94\x2d\xbe\x77\x18\x1a" + "\xcc\xde\x2d\x6a\xae\xf1\x0b\xcc" + "\x38\x06\x38\x51\xb4\xb8\xf3\x5b" + "\x5c\x34\xa6\xa3\x6e\x0b\x05\xe5" + "\x6a\x6d\x44\xaa\x26\xa8\x44\xa5", + .rlen = 16 + 32, + }, { /* RFC 3602 Case 2 */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x10" /* enc key length */ + "\x20\x21\x22\x23\x24\x25\x26\x27" + "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f" + "\x30\x31\x32\x33\x34\x35\x36\x37" + "\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f" + "\xc2\x86\x69\x6d\x88\x7c\x9a\xa0" + "\x61\x1b\xbb\x3e\x20\x25\xa4\x5a", + .klen = 8 + 32 + 16, + .iv = "\x56\x2e\x17\x99\x6d\x09\x3d\x28" + "\xdd\xb3\xba\x69\x5a\x2e\x6f\x58", + .input = "\x00\x01\x02\x03\x04\x05\x06\x07" + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f" + "\x10\x11\x12\x13\x14\x15\x16\x17" + "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f", + .ilen = 32, + .result = "\xd2\x96\xcd\x94\xc2\xcc\xcf\x8a" + "\x3a\x86\x30\x28\xb5\xe1\xdc\x0a" + "\x75\x86\x60\x2d\x25\x3c\xff\xf9" + "\x1b\x82\x66\xbe\xa6\xd6\x1a\xb1" + "\xf5\x33\x53\xf3\x68\x85\x2a\x99" + "\x0e\x06\x58\x8f\xba\xf6\x06\xda" + "\x49\x69\x0d\x5b\xd4\x36\x06\x62" + "\x35\x5e\x54\x58\x53\x4d\xdf\xbf", + .rlen = 32 + 32, + }, { /* RFC 3602 Case 3 */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x10" /* enc key length */ + "\x11\x22\x33\x44\x55\x66\x77\x88" + "\x99\xaa\xbb\xcc\xdd\xee\xff\x11" + "\x22\x33\x44\x55\x66\x77\x88\x99" + "\xaa\xbb\xcc\xdd\xee\xff\x11\x22" + "\x6c\x3e\xa0\x47\x76\x30\xce\x21" + "\xa2\xce\x33\x4a\xa7\x46\xc2\xcd", + .klen = 8 + 32 + 16, + .iv = "\xc7\x82\xdc\x4c\x09\x8c\x66\xcb" + "\xd9\xcd\x27\xd8\x25\x68\x2c\x81", + .input = "This is a 48-byte message (exactly 3 AES blocks)", + .ilen = 48, + .result = "\xd0\xa0\x2b\x38\x36\x45\x17\x53" + "\xd4\x93\x66\x5d\x33\xf0\xe8\x86" + "\x2d\xea\x54\xcd\xb2\x93\xab\xc7" + "\x50\x69\x39\x27\x67\x72\xf8\xd5" + "\x02\x1c\x19\x21\x6b\xad\x52\x5c" + "\x85\x79\x69\x5d\x83\xba\x26\x84" + "\x68\xb9\x3e\x90\x38\xa0\x88\x01" + "\xe7\xc6\xce\x10\x31\x2f\x9b\x1d" + "\x24\x78\xfb\xbe\x02\xe0\x4f\x40" + "\x10\xbd\xaa\xc6\xa7\x79\xe0\x1a", + .rlen = 48 + 32, + }, { /* RFC 3602 Case 4 */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x10" /* enc key length */ + "\x11\x22\x33\x44\x55\x66\x77\x88" + "\x99\xaa\xbb\xcc\xdd\xee\xff\x11" + "\x22\x33\x44\x55\x66\x77\x88\x99" + "\xaa\xbb\xcc\xdd\xee\xff\x11\x22" + "\x56\xe4\x7a\x38\xc5\x59\x89\x74" + "\xbc\x46\x90\x3d\xba\x29\x03\x49", + .klen = 8 + 32 + 16, + .iv = "\x8c\xe8\x2e\xef\xbe\xa0\xda\x3c" + "\x44\x69\x9e\xd7\xdb\x51\xb7\xd9", + .input = "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7" + "\xa8\xa9\xaa\xab\xac\xad\xae\xaf" + "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7" + "\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf" + "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7" + "\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf" + "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7" + "\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf", + .ilen = 64, + .result = "\xc3\x0e\x32\xff\xed\xc0\x77\x4e" + "\x6a\xff\x6a\xf0\x86\x9f\x71\xaa" + "\x0f\x3a\xf0\x7a\x9a\x31\xa9\xc6" + "\x84\xdb\x20\x7e\xb0\xef\x8e\x4e" + "\x35\x90\x7a\xa6\x32\xc3\xff\xdf" + "\x86\x8b\xb7\xb2\x9d\x3d\x46\xad" + "\x83\xce\x9f\x9a\x10\x2e\xe9\x9d" + "\x49\xa5\x3e\x87\xf4\xc3\xda\x55" + "\x7a\x1b\xd4\x3c\xdb\x17\x95\xe2" + "\xe0\x93\xec\xc9\x9f\xf7\xce\xd8" + "\x3f\x54\xe2\x49\x39\xe3\x71\x25" + "\x2b\x6c\xe9\x5d\xec\xec\x2b\x64", + .rlen = 64 + 32, + }, { /* RFC 3602 Case 5 */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x10" /* enc key length */ + "\x11\x22\x33\x44\x55\x66\x77\x88" + "\x99\xaa\xbb\xcc\xdd\xee\xff\x11" + "\x22\x33\x44\x55\x66\x77\x88\x99" + "\xaa\xbb\xcc\xdd\xee\xff\x11\x22" + "\x90\xd3\x82\xb4\x10\xee\xba\x7a" + "\xd9\x38\xc4\x6c\xec\x1a\x82\xbf", + .klen = 8 + 32 + 16, + .iv = "\xe9\x6e\x8c\x08\xab\x46\x57\x63" + "\xfd\x09\x8d\x45\xdd\x3f\xf8\x93", + .assoc = "\x00\x00\x43\x21\x00\x00\x00\x01", + .alen = 8, + .input = "\x08\x00\x0e\xbd\xa7\x0a\x00\x00" + "\x8e\x9c\x08\x3d\xb9\x5b\x07\x00" + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f" + "\x10\x11\x12\x13\x14\x15\x16\x17" + "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f" + "\x20\x21\x22\x23\x24\x25\x26\x27" + "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f" + "\x30\x31\x32\x33\x34\x35\x36\x37" + "\x01\x02\x03\x04\x05\x06\x07\x08" + "\x09\x0a\x0b\x0c\x0d\x0e\x0e\x01", + .ilen = 80, + .result = "\xf6\x63\xc2\x5d\x32\x5c\x18\xc6" + "\xa9\x45\x3e\x19\x4e\x12\x08\x49" + "\xa4\x87\x0b\x66\xcc\x6b\x99\x65" + "\x33\x00\x13\xb4\x89\x8d\xc8\x56" + "\xa4\x69\x9e\x52\x3a\x55\xdb\x08" + "\x0b\x59\xec\x3a\x8e\x4b\x7e\x52" + "\x77\x5b\x07\xd1\xdb\x34\xed\x9c" + "\x53\x8a\xb5\x0c\x55\x1b\x87\x4a" + "\xa2\x69\xad\xd0\x47\xad\x2d\x59" + "\x13\xac\x19\xb7\xcf\xba\xd4\xa6" + "\xbb\xd4\x0f\xbe\xa3\x3b\x4c\xb8" + "\x3a\xd2\xe1\x03\x86\xa5\x59\xb7" + "\x73\xc3\x46\x20\x2c\xb1\xef\x68" + "\xbb\x8a\x32\x7e\x12\x8c\x69\xcf", + .rlen = 80 + 32, + }, { /* NIST SP800-38A F.2.3 CBC-AES192.Encrypt */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x18" /* enc key length */ + "\x11\x22\x33\x44\x55\x66\x77\x88" + "\x99\xaa\xbb\xcc\xdd\xee\xff\x11" + "\x22\x33\x44\x55\x66\x77\x88\x99" + "\xaa\xbb\xcc\xdd\xee\xff\x11\x22" + "\x8e\x73\xb0\xf7\xda\x0e\x64\x52" + "\xc8\x10\xf3\x2b\x80\x90\x79\xe5" + "\x62\xf8\xea\xd2\x52\x2c\x6b\x7b", + .klen = 8 + 32 + 24, + .iv = "\x00\x01\x02\x03\x04\x05\x06\x07" + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f", + .input = "\x6b\xc1\xbe\xe2\x2e\x40\x9f\x96" + "\xe9\x3d\x7e\x11\x73\x93\x17\x2a" + "\xae\x2d\x8a\x57\x1e\x03\xac\x9c" + "\x9e\xb7\x6f\xac\x45\xaf\x8e\x51" + "\x30\xc8\x1c\x46\xa3\x5c\xe4\x11" + "\xe5\xfb\xc1\x19\x1a\x0a\x52\xef" + "\xf6\x9f\x24\x45\xdf\x4f\x9b\x17" + "\xad\x2b\x41\x7b\xe6\x6c\x37\x10", + .ilen = 64, + .result = "\x4f\x02\x1d\xb2\x43\xbc\x63\x3d" + "\x71\x78\x18\x3a\x9f\xa0\x71\xe8" + "\xb4\xd9\xad\xa9\xad\x7d\xed\xf4" + "\xe5\xe7\x38\x76\x3f\x69\x14\x5a" + "\x57\x1b\x24\x20\x12\xfb\x7a\xe0" + "\x7f\xa9\xba\xac\x3d\xf1\x02\xe0" + "\x08\xb0\xe2\x79\x88\x59\x88\x81" + "\xd9\x20\xa9\xe6\x4f\x56\x15\xcd" + "\x2f\xee\x5f\xdb\x66\xfe\x79\x09" + "\x61\x81\x31\xea\x5b\x3d\x8e\xfb" + "\xca\x71\x85\x93\xf7\x85\x55\x8b" + "\x7a\xe4\x94\xca\x8b\xba\x19\x33", + .rlen = 64 + 32, + }, { /* NIST SP800-38A F.2.5 CBC-AES256.Encrypt */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x20" /* enc key length */ + "\x11\x22\x33\x44\x55\x66\x77\x88" + "\x99\xaa\xbb\xcc\xdd\xee\xff\x11" + "\x22\x33\x44\x55\x66\x77\x88\x99" + "\xaa\xbb\xcc\xdd\xee\xff\x11\x22" + "\x60\x3d\xeb\x10\x15\xca\x71\xbe" + "\x2b\x73\xae\xf0\x85\x7d\x77\x81" + "\x1f\x35\x2c\x07\x3b\x61\x08\xd7" + "\x2d\x98\x10\xa3\x09\x14\xdf\xf4", + .klen = 8 + 32 + 32, + .iv = "\x00\x01\x02\x03\x04\x05\x06\x07" + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f", + .input = "\x6b\xc1\xbe\xe2\x2e\x40\x9f\x96" + "\xe9\x3d\x7e\x11\x73\x93\x17\x2a" + "\xae\x2d\x8a\x57\x1e\x03\xac\x9c" + "\x9e\xb7\x6f\xac\x45\xaf\x8e\x51" + "\x30\xc8\x1c\x46\xa3\x5c\xe4\x11" + "\xe5\xfb\xc1\x19\x1a\x0a\x52\xef" + "\xf6\x9f\x24\x45\xdf\x4f\x9b\x17" + "\xad\x2b\x41\x7b\xe6\x6c\x37\x10", + .ilen = 64, + .result = "\xf5\x8c\x4c\x04\xd6\xe5\xf1\xba" + "\x77\x9e\xab\xfb\x5f\x7b\xfb\xd6" + "\x9c\xfc\x4e\x96\x7e\xdb\x80\x8d" + "\x67\x9f\x77\x7b\xc6\x70\x2c\x7d" + "\x39\xf2\x33\x69\xa9\xd9\xba\xcf" + "\xa5\x30\xe2\x63\x04\x23\x14\x61" + "\xb2\xeb\x05\xe2\xc3\x9b\xe9\xfc" + "\xda\x6c\x19\x07\x8c\x6a\x9d\x1b" + "\x24\x29\xed\xc2\x31\x49\xdb\xb1" + "\x8f\x74\xbd\x17\x92\x03\xbe\x8f" + "\xf3\x61\xde\x1c\xe9\xdb\xcd\xd0" + "\xcc\xce\xe9\x85\x57\xcf\x6f\x5f", + .rlen = 64 + 32, + }, +}; + +static struct aead_testvec hmac_sha512_aes_cbc_enc_tv_template[] = { + { /* RFC 3602 Case 1 */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x10" /* enc key length */ + "\x00\x00\x00\x00\x00\x00\x00\x00" + "\x00\x00\x00\x00\x00\x00\x00\x00" + "\x00\x00\x00\x00\x00\x00\x00\x00" + "\x00\x00\x00\x00\x00\x00\x00\x00" + "\x00\x00\x00\x00\x00\x00\x00\x00" + "\x00\x00\x00\x00\x00\x00\x00\x00" + "\x00\x00\x00\x00\x00\x00\x00\x00" + "\x00\x00\x00\x00\x00\x00\x00\x00" + "\x06\xa9\x21\x40\x36\xb8\xa1\x5b" + "\x51\x2e\x03\xd5\x34\x12\x00\x06", + .klen = 8 + 64 + 16, + .iv = "\x3d\xaf\xba\x42\x9d\x9e\xb4\x30" + "\xb4\x22\xda\x80\x2c\x9f\xac\x41", + .input = "Single block msg", + .ilen = 16, + .result = "\xe3\x53\x77\x9c\x10\x79\xae\xb8" + "\x27\x08\x94\x2d\xbe\x77\x18\x1a" + "\x3f\xdc\xad\x90\x03\x63\x5e\x68" + "\xc3\x13\xdd\xa4\x5c\x4d\x54\xa7" + "\x19\x6e\x03\x75\x2b\xa1\x62\xce" + "\xe0\xc6\x96\x75\xb2\x14\xca\x96" + "\xec\xbd\x50\x08\x07\x64\x1a\x49" + "\xe8\x9a\x7c\x06\x3d\xcb\xff\xb2" + "\xfa\x20\x89\xdd\x9c\xac\x9e\x16" + "\x18\x8a\xa0\x6d\x01\x6c\xa3\x3a", + .rlen = 16 + 64, + }, { /* RFC 3602 Case 2 */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x10" /* enc key length */ + "\x20\x21\x22\x23\x24\x25\x26\x27" + "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f" + "\x30\x31\x32\x33\x34\x35\x36\x37" + "\x38\x39\x3a\x3b\x3c\x3d\x3e\x3f" + "\x40\x41\x42\x43\x44\x45\x46\x47" + "\x48\x49\x4a\x4b\x4c\x4d\x4e\x4f" + "\x50\x51\x52\x53\x54\x55\x56\x57" + "\x58\x59\x5a\x5b\x5c\x5d\x5e\x5f" + "\xc2\x86\x69\x6d\x88\x7c\x9a\xa0" + "\x61\x1b\xbb\x3e\x20\x25\xa4\x5a", + .klen = 8 + 64 + 16, + .iv = "\x56\x2e\x17\x99\x6d\x09\x3d\x28" + "\xdd\xb3\xba\x69\x5a\x2e\x6f\x58", + .input = "\x00\x01\x02\x03\x04\x05\x06\x07" + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f" + "\x10\x11\x12\x13\x14\x15\x16\x17" + "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f", + .ilen = 32, + .result = "\xd2\x96\xcd\x94\xc2\xcc\xcf\x8a" + "\x3a\x86\x30\x28\xb5\xe1\xdc\x0a" + "\x75\x86\x60\x2d\x25\x3c\xff\xf9" + "\x1b\x82\x66\xbe\xa6\xd6\x1a\xb1" + "\xda\xb2\x0c\xb2\x26\xc4\xd5\xef" + "\x60\x38\xa4\x5e\x9a\x8c\x1b\x41" + "\x03\x9f\xc4\x64\x7f\x01\x42\x9b" + "\x0e\x1b\xea\xef\xbc\x88\x19\x5e" + "\x31\x7e\xc2\x95\xfc\x09\x32\x0a" + "\x46\x32\x7c\x41\x9c\x59\x3e\xe9" + "\x8f\x9f\xd4\x31\xd6\x22\xbd\xf8" + "\xf7\x0a\x94\xe5\xa9\xc3\xf6\x9d", + .rlen = 32 + 64, + }, { /* RFC 3602 Case 3 */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x10" /* enc key length */ + "\x11\x22\x33\x44\x55\x66\x77\x88" + "\x99\xaa\xbb\xcc\xdd\xee\xff\x11" + "\x22\x33\x44\x55\x66\x77\x88\x99" + "\xaa\xbb\xcc\xdd\xee\xff\x11\x22" + "\x33\x44\x55\x66\x77\x88\x99\xaa" + "\xbb\xcc\xdd\xee\xff\x11\x22\x33" + "\x44\x55\x66\x77\x88\x99\xaa\xbb" + "\xcc\xdd\xee\xff\x11\x22\x33\x44" + "\x6c\x3e\xa0\x47\x76\x30\xce\x21" + "\xa2\xce\x33\x4a\xa7\x46\xc2\xcd", + .klen = 8 + 64 + 16, + .iv = "\xc7\x82\xdc\x4c\x09\x8c\x66\xcb" + "\xd9\xcd\x27\xd8\x25\x68\x2c\x81", + .input = "This is a 48-byte message (exactly 3 AES blocks)", + .ilen = 48, + .result = "\xd0\xa0\x2b\x38\x36\x45\x17\x53" + "\xd4\x93\x66\x5d\x33\xf0\xe8\x86" + "\x2d\xea\x54\xcd\xb2\x93\xab\xc7" + "\x50\x69\x39\x27\x67\x72\xf8\xd5" + "\x02\x1c\x19\x21\x6b\xad\x52\x5c" + "\x85\x79\x69\x5d\x83\xba\x26\x84" + "\x64\x19\x17\x5b\x57\xe0\x21\x0f" + "\xca\xdb\xa1\x26\x38\x14\xa2\x69" + "\xdb\x54\x67\x80\xc0\x54\xe0\xfd" + "\x3e\x91\xe7\x91\x7f\x13\x38\x44" + "\xb7\xb1\xd6\xc8\x7d\x48\x8d\x41" + "\x08\xea\x29\x6c\x74\x67\x3f\xb0" + "\xac\x7f\x5c\x1d\xf5\xee\x22\x66" + "\x27\xa6\xb6\x13\xba\xba\xf0\xc2", + .rlen = 48 + 64, + }, { /* RFC 3602 Case 4 */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x10" /* enc key length */ + "\x11\x22\x33\x44\x55\x66\x77\x88" + "\x99\xaa\xbb\xcc\xdd\xee\xff\x11" + "\x22\x33\x44\x55\x66\x77\x88\x99" + "\xaa\xbb\xcc\xdd\xee\xff\x11\x22" + "\x33\x44\x55\x66\x77\x88\x99\xaa" + "\xbb\xcc\xdd\xee\xff\x11\x22\x33" + "\x44\x55\x66\x77\x88\x99\xaa\xbb" + "\xcc\xdd\xee\xff\x11\x22\x33\x44" + "\x56\xe4\x7a\x38\xc5\x59\x89\x74" + "\xbc\x46\x90\x3d\xba\x29\x03\x49", + .klen = 8 + 64 + 16, + .iv = "\x8c\xe8\x2e\xef\xbe\xa0\xda\x3c" + "\x44\x69\x9e\xd7\xdb\x51\xb7\xd9", + .input = "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7" + "\xa8\xa9\xaa\xab\xac\xad\xae\xaf" + "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7" + "\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf" + "\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7" + "\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf" + "\xd0\xd1\xd2\xd3\xd4\xd5\xd6\xd7" + "\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf", + .ilen = 64, + .result = "\xc3\x0e\x32\xff\xed\xc0\x77\x4e" + "\x6a\xff\x6a\xf0\x86\x9f\x71\xaa" + "\x0f\x3a\xf0\x7a\x9a\x31\xa9\xc6" + "\x84\xdb\x20\x7e\xb0\xef\x8e\x4e" + "\x35\x90\x7a\xa6\x32\xc3\xff\xdf" + "\x86\x8b\xb7\xb2\x9d\x3d\x46\xad" + "\x83\xce\x9f\x9a\x10\x2e\xe9\x9d" + "\x49\xa5\x3e\x87\xf4\xc3\xda\x55" + "\x82\xcd\x42\x28\x21\x20\x15\xcc" + "\xb7\xb2\x48\x40\xc7\x64\x41\x3a" + "\x61\x32\x82\x85\xcf\x27\xed\xb4" + "\xe4\x68\xa2\xf5\x79\x26\x27\xb2" + "\x51\x67\x6a\xc4\xf0\x66\x55\x50" + "\xbc\x6f\xed\xd5\x8d\xde\x23\x7c" + "\x62\x98\x14\xd7\x2f\x37\x8d\xdf" + "\xf4\x33\x80\xeb\x8e\xb4\xa4\xda", + .rlen = 64 + 64, + }, { /* RFC 3602 Case 5 */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x10" /* enc key length */ + "\x11\x22\x33\x44\x55\x66\x77\x88" + "\x99\xaa\xbb\xcc\xdd\xee\xff\x11" + "\x22\x33\x44\x55\x66\x77\x88\x99" + "\xaa\xbb\xcc\xdd\xee\xff\x11\x22" + "\x33\x44\x55\x66\x77\x88\x99\xaa" + "\xbb\xcc\xdd\xee\xff\x11\x22\x33" + "\x44\x55\x66\x77\x88\x99\xaa\xbb" + "\xcc\xdd\xee\xff\x11\x22\x33\x44" + "\x90\xd3\x82\xb4\x10\xee\xba\x7a" + "\xd9\x38\xc4\x6c\xec\x1a\x82\xbf", + .klen = 8 + 64 + 16, + .iv = "\xe9\x6e\x8c\x08\xab\x46\x57\x63" + "\xfd\x09\x8d\x45\xdd\x3f\xf8\x93", + .assoc = "\x00\x00\x43\x21\x00\x00\x00\x01", + .alen = 8, + .input = "\x08\x00\x0e\xbd\xa7\x0a\x00\x00" + "\x8e\x9c\x08\x3d\xb9\x5b\x07\x00" + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f" + "\x10\x11\x12\x13\x14\x15\x16\x17" + "\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f" + "\x20\x21\x22\x23\x24\x25\x26\x27" + "\x28\x29\x2a\x2b\x2c\x2d\x2e\x2f" + "\x30\x31\x32\x33\x34\x35\x36\x37" + "\x01\x02\x03\x04\x05\x06\x07\x08" + "\x09\x0a\x0b\x0c\x0d\x0e\x0e\x01", + .ilen = 80, + .result = "\xf6\x63\xc2\x5d\x32\x5c\x18\xc6" + "\xa9\x45\x3e\x19\x4e\x12\x08\x49" + "\xa4\x87\x0b\x66\xcc\x6b\x99\x65" + "\x33\x00\x13\xb4\x89\x8d\xc8\x56" + "\xa4\x69\x9e\x52\x3a\x55\xdb\x08" + "\x0b\x59\xec\x3a\x8e\x4b\x7e\x52" + "\x77\x5b\x07\xd1\xdb\x34\xed\x9c" + "\x53\x8a\xb5\x0c\x55\x1b\x87\x4a" + "\xa2\x69\xad\xd0\x47\xad\x2d\x59" + "\x13\xac\x19\xb7\xcf\xba\xd4\xa6" + "\x74\x84\x94\xe2\xd7\x7a\xf9\xbf" + "\x00\x8a\xa2\xd5\xb7\xf3\x60\xcf" + "\xa0\x47\xdf\x4e\x09\xf4\xb1\x7f" + "\x14\xd9\x3d\x53\x8e\x12\xb3\x00" + "\x4c\x0a\x4e\x32\x40\x43\x88\xce" + "\x92\x26\xc1\x76\x20\x11\xeb\xba" + "\x62\x4f\x9a\x62\x25\xc3\x75\x80" + "\xb7\x0a\x17\xf5\xd7\x94\xb4\x14", + .rlen = 80 + 64, + }, { /* NIST SP800-38A F.2.3 CBC-AES192.Encrypt */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x18" /* enc key length */ + "\x11\x22\x33\x44\x55\x66\x77\x88" + "\x99\xaa\xbb\xcc\xdd\xee\xff\x11" + "\x22\x33\x44\x55\x66\x77\x88\x99" + "\xaa\xbb\xcc\xdd\xee\xff\x11\x22" + "\x33\x44\x55\x66\x77\x88\x99\xaa" + "\xbb\xcc\xdd\xee\xff\x11\x22\x33" + "\x44\x55\x66\x77\x88\x99\xaa\xbb" + "\xcc\xdd\xee\xff\x11\x22\x33\x44" + "\x8e\x73\xb0\xf7\xda\x0e\x64\x52" + "\xc8\x10\xf3\x2b\x80\x90\x79\xe5" + "\x62\xf8\xea\xd2\x52\x2c\x6b\x7b", + .klen = 8 + 64 + 24, + .iv = "\x00\x01\x02\x03\x04\x05\x06\x07" + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f", + .input = "\x6b\xc1\xbe\xe2\x2e\x40\x9f\x96" + "\xe9\x3d\x7e\x11\x73\x93\x17\x2a" + "\xae\x2d\x8a\x57\x1e\x03\xac\x9c" + "\x9e\xb7\x6f\xac\x45\xaf\x8e\x51" + "\x30\xc8\x1c\x46\xa3\x5c\xe4\x11" + "\xe5\xfb\xc1\x19\x1a\x0a\x52\xef" + "\xf6\x9f\x24\x45\xdf\x4f\x9b\x17" + "\xad\x2b\x41\x7b\xe6\x6c\x37\x10", + .ilen = 64, + .result = "\x4f\x02\x1d\xb2\x43\xbc\x63\x3d" + "\x71\x78\x18\x3a\x9f\xa0\x71\xe8" + "\xb4\xd9\xad\xa9\xad\x7d\xed\xf4" + "\xe5\xe7\x38\x76\x3f\x69\x14\x5a" + "\x57\x1b\x24\x20\x12\xfb\x7a\xe0" + "\x7f\xa9\xba\xac\x3d\xf1\x02\xe0" + "\x08\xb0\xe2\x79\x88\x59\x88\x81" + "\xd9\x20\xa9\xe6\x4f\x56\x15\xcd" + "\x77\x4b\x69\x9d\x3a\x0d\xb4\x99" + "\x8f\xc6\x8e\x0e\x72\x58\xe3\x56" + "\xbb\x21\xd2\x7d\x93\x11\x17\x91" + "\xc4\x83\xfd\x0a\xea\x71\xfe\x77" + "\xae\x6f\x0a\xa5\xf0\xcf\xe1\x35" + "\xba\x03\xd5\x32\xfa\x5f\x41\x58" + "\x8d\x43\x98\xa7\x94\x16\x07\x02" + "\x0f\xb6\x81\x50\x28\x95\x2e\x75", + .rlen = 64 + 64, + }, { /* NIST SP800-38A F.2.5 CBC-AES256.Encrypt */ +#ifdef __LITTLE_ENDIAN + .key = "\x08\x00" /* rta length */ + "\x01\x00" /* rta type */ +#else + .key = "\x00\x08" /* rta length */ + "\x00\x01" /* rta type */ +#endif + "\x00\x00\x00\x20" /* enc key length */ + "\x11\x22\x33\x44\x55\x66\x77\x88" + "\x99\xaa\xbb\xcc\xdd\xee\xff\x11" + "\x22\x33\x44\x55\x66\x77\x88\x99" + "\xaa\xbb\xcc\xdd\xee\xff\x11\x22" + "\x33\x44\x55\x66\x77\x88\x99\xaa" + "\xbb\xcc\xdd\xee\xff\x11\x22\x33" + "\x44\x55\x66\x77\x88\x99\xaa\xbb" + "\xcc\xdd\xee\xff\x11\x22\x33\x44" + "\x60\x3d\xeb\x10\x15\xca\x71\xbe" + "\x2b\x73\xae\xf0\x85\x7d\x77\x81" + "\x1f\x35\x2c\x07\x3b\x61\x08\xd7" + "\x2d\x98\x10\xa3\x09\x14\xdf\xf4", + .klen = 8 + 64 + 32, + .iv = "\x00\x01\x02\x03\x04\x05\x06\x07" + "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f", + .input = "\x6b\xc1\xbe\xe2\x2e\x40\x9f\x96" + "\xe9\x3d\x7e\x11\x73\x93\x17\x2a" + "\xae\x2d\x8a\x57\x1e\x03\xac\x9c" + "\x9e\xb7\x6f\xac\x45\xaf\x8e\x51" + "\x30\xc8\x1c\x46\xa3\x5c\xe4\x11" + "\xe5\xfb\xc1\x19\x1a\x0a\x52\xef" + "\xf6\x9f\x24\x45\xdf\x4f\x9b\x17" + "\xad\x2b\x41\x7b\xe6\x6c\x37\x10", + .ilen = 64, + .result = "\xf5\x8c\x4c\x04\xd6\xe5\xf1\xba" + "\x77\x9e\xab\xfb\x5f\x7b\xfb\xd6" + "\x9c\xfc\x4e\x96\x7e\xdb\x80\x8d" + "\x67\x9f\x77\x7b\xc6\x70\x2c\x7d" + "\x39\xf2\x33\x69\xa9\xd9\xba\xcf" + "\xa5\x30\xe2\x63\x04\x23\x14\x61" + "\xb2\xeb\x05\xe2\xc3\x9b\xe9\xfc" + "\xda\x6c\x19\x07\x8c\x6a\x9d\x1b" + "\xb2\x27\x69\x7f\x45\x64\x79\x2b" + "\xb7\xb8\x4c\xd4\x75\x94\x68\x40" + "\x2a\xea\x91\xc7\x3f\x7c\xed\x7b" + "\x95\x2c\x9b\xa8\xf5\xe5\x52\x8d" + "\x6b\xe1\xae\xf1\x74\xfa\x0d\x0c" + "\xe3\x8d\x64\xc3\x8d\xff\x7c\x8c" + "\xdb\xbf\xa0\xb4\x01\xa2\xa8\xa2" + "\x2c\xb1\x62\x2c\x10\xca\xf1\x21", + .rlen = 64 + 64, + }, +}; + static struct cipher_testvec aes_lrw_enc_tv_template[] = { /* from http://grouper.ieee.org/groups/1619/email/pdf00017.pdf */ { /* LRW-32-AES 1 */ -- cgit v1.2.3