diff options
author | Jonathon Murphy <jpmurphy@google.com> | 2023-04-21 16:22:51 +0000 |
---|---|---|
committer | Jonathon Murphy <jpmurphy@google.com> | 2023-04-21 16:24:57 +0000 |
commit | eaf2376ff817d2c0019a6064f9d77187b4f70507 (patch) | |
tree | 34a8bc11188cfcef6de74d26f849c32e8e784c96 | |
parent | 33d87e53009a84379747ae31efac39e067e23484 (diff) | |
download | chrome-ec-eaf2376ff817d2c0019a6064f9d77187b4f70507.tar.gz |
Revert "Reland "tree: Remove the stale boringssl AES code""
This reverts commit 2e03009c0dccc4ae9683d3b7d761e900b3cf2c46.
Reason for revert: Breaks postsubmit b/279174496
Original change's description:
> Reland "tree: Remove the stale boringssl AES code"
>
> This is a reland of commit 1eb725490d0e7ff726b9edcf11579080a412a02f
>
> Original change's description:
> > tree: Remove the stale boringssl AES code
> >
> > BUG=b:248508087, b:273639386
> > TEST=make runhosttests -j
> > TEST=make buildall -j
> > BRANCH=none
> >
> > Change-Id: If4d7ee558871debdca5998c51d75c1062dce2a50
> > Signed-off-by: Yi Chou <yich@google.com>
> > Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/ec/+/4388275
> > Reviewed-by: Tom Hughes <tomhughes@chromium.org>
>
> BUG=b:248508087, b:273639386
> TEST=make runhosttests -j
> TEST=make buildall -j
> BRANCH=none
>
> Signed-off-by: Yi Chou <yich@google.com>
> Change-Id: Ifca00792d0568ed767a1c9aea4953eb7a93c1c92
> Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/ec/+/4439570
> Reviewed-by: Tom Hughes <tomhughes@chromium.org>
Bug: b:248508087, b:273639386
Change-Id: I4c44cdd75a9cc38e3e0e688d91f2a64b908ae61d
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/ec/+/4459598
Tested-by: Jonathon Murphy <jpmurphy@google.com>
Reviewed-by: Tom Hughes <tomhughes@chromium.org>
Auto-Submit: Jonathon Murphy <jpmurphy@google.com>
Owners-Override: Jonathon Murphy <jpmurphy@google.com>
l--------- | common/aes-gcm.c | 1 | ||||
l--------- | common/aes.c | 1 | ||||
-rw-r--r-- | common/build.mk | 4 | ||||
l--------- | core/cortex-m/aes.S | 1 | ||||
-rw-r--r-- | core/cortex-m/build.mk | 2 | ||||
l--------- | core/cortex-m/ghash.S | 1 | ||||
l--------- | include/aes-gcm.h | 1 | ||||
l--------- | include/aes.h | 1 | ||||
-rw-r--r-- | include/config.h | 6 | ||||
-rw-r--r-- | third_party/boringssl/common/aes-gcm.c | 882 | ||||
-rw-r--r-- | third_party/boringssl/common/aes.c | 806 | ||||
-rw-r--r-- | third_party/boringssl/core/cortex-m/aes.S | 1189 | ||||
-rw-r--r-- | third_party/boringssl/core/cortex-m/ghash.S | 575 | ||||
-rw-r--r-- | third_party/boringssl/include/aes-gcm.h | 146 | ||||
-rw-r--r-- | third_party/boringssl/include/aes.h | 129 | ||||
-rw-r--r-- | util/config_allowed.txt | 2 |
16 files changed, 3747 insertions, 0 deletions
diff --git a/common/aes-gcm.c b/common/aes-gcm.c new file mode 120000 index 0000000000..3176d85ff8 --- /dev/null +++ b/common/aes-gcm.c @@ -0,0 +1 @@ +../third_party/boringssl/common/aes-gcm.c
\ No newline at end of file diff --git a/common/aes.c b/common/aes.c new file mode 120000 index 0000000000..ed10836943 --- /dev/null +++ b/common/aes.c @@ -0,0 +1 @@ +../third_party/boringssl/common/aes.c
\ No newline at end of file diff --git a/common/build.mk b/common/build.mk index 0c5250cf18..2da6a89b38 100644 --- a/common/build.mk +++ b/common/build.mk @@ -33,6 +33,10 @@ common-$(CONFIG_ACCEL_KX022)+=math_util.o common-$(CONFIG_BODY_DETECTION)+=math_util.o common-$(CONFIG_TEMP_SENSOR_TMP112)+=math_util.o common-$(CONFIG_TEMP_SENSOR_PCT2075)+=math_util.o +ifneq ($(CORE),cortex-m) +common-$(CONFIG_AES)+=aes.o +endif +common-$(CONFIG_AES_GCM)+=aes-gcm.o common-$(CONFIG_CMD_ADC)+=adc.o common-$(HAS_TASK_ALS)+=als.o common-$(CONFIG_AP_HANG_DETECT)+=ap_hang_detect.o diff --git a/core/cortex-m/aes.S b/core/cortex-m/aes.S new file mode 120000 index 0000000000..39d1286943 --- /dev/null +++ b/core/cortex-m/aes.S @@ -0,0 +1 @@ +../../third_party/boringssl/core/cortex-m/aes.S
\ No newline at end of file diff --git a/core/cortex-m/build.mk b/core/cortex-m/build.mk index cb5b12b380..1e2c47297b 100644 --- a/core/cortex-m/build.mk +++ b/core/cortex-m/build.mk @@ -44,6 +44,8 @@ core-y=cpu.o debug.o init.o vecttable.o ifneq ($(cc-name),clang) core-y+=ldivmod.o llsr.o uldivmod.o endif +core-$(CONFIG_AES)+=aes.o +core-$(CONFIG_AES_GCM)+=ghash.o core-$(CONFIG_ARMV7M_CACHE)+=cache.o core-$(CONFIG_COMMON_PANIC_OUTPUT)+=panic.o core-$(CONFIG_COMMON_RUNTIME)+=switch.o task.o diff --git a/core/cortex-m/ghash.S b/core/cortex-m/ghash.S new file mode 120000 index 0000000000..e9acbf4b25 --- /dev/null +++ b/core/cortex-m/ghash.S @@ -0,0 +1 @@ +../../third_party/boringssl/core/cortex-m/ghash.S
\ No newline at end of file diff --git a/include/aes-gcm.h b/include/aes-gcm.h new file mode 120000 index 0000000000..ba62939792 --- /dev/null +++ b/include/aes-gcm.h @@ -0,0 +1 @@ +../third_party/boringssl/include/aes-gcm.h
\ No newline at end of file diff --git a/include/aes.h b/include/aes.h new file mode 120000 index 0000000000..b30c680a6a --- /dev/null +++ b/include/aes.h @@ -0,0 +1 @@ +../third_party/boringssl/include/aes.h
\ No newline at end of file diff --git a/include/config.h b/include/config.h index 1b914bae1e..ee212a8fdc 100644 --- a/include/config.h +++ b/include/config.h @@ -327,6 +327,12 @@ #define CONFIG_ADC_PROFILE_SINGLE #undef CONFIG_ADC_PROFILE_FAST_CONTINUOUS +/* Support AES symmetric-key algorithm */ +#undef CONFIG_AES + +/* Support AES-GCM */ +#undef CONFIG_AES_GCM + /* * Some ALS modules may be connected to the EC. We need the command, and * specific drivers for each module. diff --git a/third_party/boringssl/common/aes-gcm.c b/third_party/boringssl/common/aes-gcm.c new file mode 100644 index 0000000000..8fe5f75307 --- /dev/null +++ b/third_party/boringssl/common/aes-gcm.c @@ -0,0 +1,882 @@ +/* ==================================================================== + * Copyright (c) 2008 The OpenSSL Project. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * + * 3. All advertising materials mentioning features or use of this + * software must display the following acknowledgment: + * "This product includes software developed by the OpenSSL Project + * for use in the OpenSSL Toolkit. (http://www.openssl.org/)" + * + * 4. The names "OpenSSL Toolkit" and "OpenSSL Project" must not be used to + * endorse or promote products derived from this software without + * prior written permission. For written permission, please contact + * openssl-core@openssl.org. + * + * 5. Products derived from this software may not be called "OpenSSL" + * nor may "OpenSSL" appear in their names without prior written + * permission of the OpenSSL Project. + * + * 6. Redistributions of any form whatsoever must retain the following + * acknowledgment: + * "This product includes software developed by the OpenSSL Project + * for use in the OpenSSL Toolkit (http://www.openssl.org/)" + * + * THIS SOFTWARE IS PROVIDED BY THE OpenSSL PROJECT ``AS IS'' AND ANY + * EXPRESSED OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE OpenSSL PROJECT OR + * ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT + * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + * ==================================================================== */ + +#include "aes-gcm.h" +#include "common.h" +#include "util.h" + +#define STRICT_ALIGNMENT 1 + +#define OPENSSL_memcpy memcpy +#define OPENSSL_memset memset +#define CRYPTO_memcmp safe_memcmp + +#ifdef CORE_CORTEX_M +#define GHASH_ASM +#define OPENSSL_ARM +#define __ARM_ARCH__ 7 +#endif + +static inline uint32_t CRYPTO_bswap4(uint32_t x) { + return __builtin_bswap32(x); +} + +static inline uint64_t CRYPTO_bswap8(uint64_t x) { + return __builtin_bswap64(x); +} + +static inline size_t load_word_le(const void *in) { + size_t v; + OPENSSL_memcpy(&v, in, sizeof(v)); + return v; +} + +static inline void store_word_le(void *out, size_t v) { + OPENSSL_memcpy(out, &v, sizeof(v)); +} + +#define PACK(s) ((size_t)(s) << (sizeof(size_t) * 8 - 16)) +#define REDUCE1BIT(V) \ + do { \ + if (sizeof(size_t) == 8) { \ + uint64_t T = UINT64_C(0xe100000000000000) & (0 - ((V).lo & 1)); \ + (V).lo = ((V).hi << 63) | ((V).lo >> 1); \ + (V).hi = ((V).hi >> 1) ^ T; \ + } else { \ + uint32_t T = 0xe1000000U & (0 - (uint32_t)((V).lo & 1)); \ + (V).lo = ((V).hi << 63) | ((V).lo >> 1); \ + (V).hi = ((V).hi >> 1) ^ ((uint64_t)T << 32); \ + } \ + } while (0) + +static void gcm_init_4bit(u128 Htable[16], uint64_t H[2]) { + u128 V; + + Htable[0].hi = 0; + Htable[0].lo = 0; + V.hi = H[0]; + V.lo = H[1]; + + Htable[8] = V; + REDUCE1BIT(V); + Htable[4] = V; + REDUCE1BIT(V); + Htable[2] = V; + REDUCE1BIT(V); + Htable[1] = V; + Htable[3].hi = V.hi ^ Htable[2].hi, Htable[3].lo = V.lo ^ Htable[2].lo; + V = Htable[4]; + Htable[5].hi = V.hi ^ Htable[1].hi, Htable[5].lo = V.lo ^ Htable[1].lo; + Htable[6].hi = V.hi ^ Htable[2].hi, Htable[6].lo = V.lo ^ Htable[2].lo; + Htable[7].hi = V.hi ^ Htable[3].hi, Htable[7].lo = V.lo ^ Htable[3].lo; + V = Htable[8]; + Htable[9].hi = V.hi ^ Htable[1].hi, Htable[9].lo = V.lo ^ Htable[1].lo; + Htable[10].hi = V.hi ^ Htable[2].hi, Htable[10].lo = V.lo ^ Htable[2].lo; + Htable[11].hi = V.hi ^ Htable[3].hi, Htable[11].lo = V.lo ^ Htable[3].lo; + Htable[12].hi = V.hi ^ Htable[4].hi, Htable[12].lo = V.lo ^ Htable[4].lo; + Htable[13].hi = V.hi ^ Htable[5].hi, Htable[13].lo = V.lo ^ Htable[5].lo; + Htable[14].hi = V.hi ^ Htable[6].hi, Htable[14].lo = V.lo ^ Htable[6].lo; + Htable[15].hi = V.hi ^ Htable[7].hi, Htable[15].lo = V.lo ^ Htable[7].lo; + +#if defined(GHASH_ASM) && defined(OPENSSL_ARM) + for (int j = 0; j < 16; ++j) { + V = Htable[j]; + Htable[j].hi = V.lo; + Htable[j].lo = V.hi; + } +#endif +} + +#if !defined(GHASH_ASM) || defined(OPENSSL_AARCH64) || defined(OPENSSL_PPC64LE) +static const size_t rem_4bit[16] = { + PACK(0x0000), PACK(0x1C20), PACK(0x3840), PACK(0x2460), + PACK(0x7080), PACK(0x6CA0), PACK(0x48C0), PACK(0x54E0), + PACK(0xE100), PACK(0xFD20), PACK(0xD940), PACK(0xC560), + PACK(0x9180), PACK(0x8DA0), PACK(0xA9C0), PACK(0xB5E0)}; + +static void gcm_gmult_4bit(uint64_t Xi[2], const u128 Htable[16]) { + u128 Z; + int cnt = 15; + size_t rem, nlo, nhi; + + nlo = ((const uint8_t *)Xi)[15]; + nhi = nlo >> 4; + nlo &= 0xf; + + Z.hi = Htable[nlo].hi; + Z.lo = Htable[nlo].lo; + + while (1) { + rem = (size_t)Z.lo & 0xf; + Z.lo = (Z.hi << 60) | (Z.lo >> 4); + Z.hi = (Z.hi >> 4); + if (sizeof(size_t) == 8) { + Z.hi ^= rem_4bit[rem]; + } else { + Z.hi ^= (uint64_t)rem_4bit[rem] << 32; + } + + Z.hi ^= Htable[nhi].hi; + Z.lo ^= Htable[nhi].lo; + + if (--cnt < 0) { + break; + } + + nlo = ((const uint8_t *)Xi)[cnt]; + nhi = nlo >> 4; + nlo &= 0xf; + + rem = (size_t)Z.lo & 0xf; + Z.lo = (Z.hi << 60) | (Z.lo >> 4); + Z.hi = (Z.hi >> 4); + if (sizeof(size_t) == 8) { + Z.hi ^= rem_4bit[rem]; + } else { + Z.hi ^= (uint64_t)rem_4bit[rem] << 32; + } + + Z.hi ^= Htable[nlo].hi; + Z.lo ^= Htable[nlo].lo; + } + + Xi[0] = CRYPTO_bswap8(Z.hi); + Xi[1] = CRYPTO_bswap8(Z.lo); +} + +// Streamed gcm_mult_4bit, see CRYPTO_gcm128_[en|de]crypt for +// details... Compiler-generated code doesn't seem to give any +// performance improvement, at least not on x86[_64]. It's here +// mostly as reference and a placeholder for possible future +// non-trivial optimization[s]... +static void gcm_ghash_4bit(uint64_t Xi[2], const u128 Htable[16], + const uint8_t *inp, size_t len) { + u128 Z; + int cnt; + size_t rem, nlo, nhi; + + do { + cnt = 15; + nlo = ((const uint8_t *)Xi)[15]; + nlo ^= inp[15]; + nhi = nlo >> 4; + nlo &= 0xf; + + Z.hi = Htable[nlo].hi; + Z.lo = Htable[nlo].lo; + + while (1) { + rem = (size_t)Z.lo & 0xf; + Z.lo = (Z.hi << 60) | (Z.lo >> 4); + Z.hi = (Z.hi >> 4); + if (sizeof(size_t) == 8) { + Z.hi ^= rem_4bit[rem]; + } else { + Z.hi ^= (uint64_t)rem_4bit[rem] << 32; + } + + Z.hi ^= Htable[nhi].hi; + Z.lo ^= Htable[nhi].lo; + + if (--cnt < 0) { + break; + } + + nlo = ((const uint8_t *)Xi)[cnt]; + nlo ^= inp[cnt]; + nhi = nlo >> 4; + nlo &= 0xf; + + rem = (size_t)Z.lo & 0xf; + Z.lo = (Z.hi << 60) | (Z.lo >> 4); + Z.hi = (Z.hi >> 4); + if (sizeof(size_t) == 8) { + Z.hi ^= rem_4bit[rem]; + } else { + Z.hi ^= (uint64_t)rem_4bit[rem] << 32; + } + + Z.hi ^= Htable[nlo].hi; + Z.lo ^= Htable[nlo].lo; + } + + Xi[0] = CRYPTO_bswap8(Z.hi); + Xi[1] = CRYPTO_bswap8(Z.lo); + } while (inp += 16, len -= 16); +} +#else // GHASH_ASM +void gcm_gmult_4bit(uint64_t Xi[2], const u128 Htable[16]); +void gcm_ghash_4bit(uint64_t Xi[2], const u128 Htable[16], const uint8_t *inp, + size_t len); +#endif + +#define GCM_MUL(ctx, Xi) gcm_gmult_4bit((ctx)->Xi.u, (ctx)->Htable) +#if defined(GHASH_ASM) +#define GHASH(ctx, in, len) gcm_ghash_4bit((ctx)->Xi.u, (ctx)->Htable, in, len) +// GHASH_CHUNK is "stride parameter" missioned to mitigate cache +// trashing effect. In other words idea is to hash data while it's +// still in L1 cache after encryption pass... +#define GHASH_CHUNK (3 * 1024) +#endif + + +#if defined(GHASH_ASM) + +#if defined(OPENSSL_X86) || defined(OPENSSL_X86_64) +#define GCM_FUNCREF_4BIT +void gcm_init_clmul(u128 Htable[16], const uint64_t Xi[2]); +void gcm_gmult_clmul(uint64_t Xi[2], const u128 Htable[16]); +void gcm_ghash_clmul(uint64_t Xi[2], const u128 Htable[16], const uint8_t *inp, + size_t len); + +#if defined(OPENSSL_X86_64) +#define GHASH_ASM_X86_64 +void gcm_init_avx(u128 Htable[16], const uint64_t Xi[2]); +void gcm_gmult_avx(uint64_t Xi[2], const u128 Htable[16]); +void gcm_ghash_avx(uint64_t Xi[2], const u128 Htable[16], const uint8_t *in, + size_t len); +#define AESNI_GCM +size_t aesni_gcm_encrypt(const uint8_t *in, uint8_t *out, size_t len, + const void *key, uint8_t ivec[16], uint64_t *Xi); +size_t aesni_gcm_decrypt(const uint8_t *in, uint8_t *out, size_t len, + const void *key, uint8_t ivec[16], uint64_t *Xi); +#endif + +#if defined(OPENSSL_X86) +#define GHASH_ASM_X86 +void gcm_gmult_4bit_mmx(uint64_t Xi[2], const u128 Htable[16]); +void gcm_ghash_4bit_mmx(uint64_t Xi[2], const u128 Htable[16], const uint8_t *inp, + size_t len); +#endif + +#elif defined(OPENSSL_ARM) || defined(OPENSSL_AARCH64) +#if __ARM_ARCH__ >= 7 +#define GHASH_ASM_ARM +#define GCM_FUNCREF_4BIT + +#if defined(OPENSSL_ARM_PMULL) +static int pmull_capable(void) { + return CRYPTO_is_ARMv8_PMULL_capable(); +} + +void gcm_init_v8(u128 Htable[16], const uint64_t Xi[2]); +void gcm_gmult_v8(uint64_t Xi[2], const u128 Htable[16]); +void gcm_ghash_v8(uint64_t Xi[2], const u128 Htable[16], const uint8_t *inp, + size_t len); +#else +static int pmull_capable(void) { + return 0; +} +static void gcm_init_v8(u128 Htable[16], const uint64_t Xi[2]) { + +} +static void gcm_gmult_v8(uint64_t Xi[2], const u128 Htable[16]) { + +} +static void gcm_ghash_v8(uint64_t Xi[2], const u128 Htable[16], + const uint8_t *inp, size_t len) { + +} +#endif + +#if defined(OPENSSL_ARM_NEON) +// 32-bit ARM also has support for doing GCM with NEON instructions. +static int neon_capable(void) { + return CRYPTO_is_NEON_capable(); +} + +void gcm_init_neon(u128 Htable[16], const uint64_t Xi[2]); +void gcm_gmult_neon(uint64_t Xi[2], const u128 Htable[16]); +void gcm_ghash_neon(uint64_t Xi[2], const u128 Htable[16], const uint8_t *inp, + size_t len); +#else +// AArch64 only has the ARMv8 versions of functions. +static int neon_capable(void) { + return 0; +} +static void gcm_init_neon(u128 Htable[16], const uint64_t Xi[2]) { + +} +static void gcm_gmult_neon(uint64_t Xi[2], const u128 Htable[16]) { + +} +static void gcm_ghash_neon(uint64_t Xi[2], const u128 Htable[16], + const uint8_t *inp, size_t len) { + +} +#endif + +#endif +#elif defined(OPENSSL_PPC64LE) +#define GHASH_ASM_PPC64LE +#define GCM_FUNCREF_4BIT +void gcm_init_p8(u128 Htable[16], const uint64_t Xi[2]); +void gcm_gmult_p8(uint64_t Xi[2], const u128 Htable[16]); +void gcm_ghash_p8(uint64_t Xi[2], const u128 Htable[16], const uint8_t *inp, + size_t len); +#endif +#endif + +#ifdef GCM_FUNCREF_4BIT +#undef GCM_MUL +#define GCM_MUL(ctx, Xi) (*gcm_gmult_p)((ctx)->Xi.u, (ctx)->Htable) +#ifdef GHASH +#undef GHASH +#define GHASH(ctx, in, len) (*gcm_ghash_p)((ctx)->Xi.u, (ctx)->Htable, in, len) +#endif +#endif + +#ifdef GHASH +// kSizeTWithoutLower4Bits is a mask that can be used to zero the lower four +// bits of a |size_t|. +static const size_t kSizeTWithoutLower4Bits = (size_t) -16; +#endif + +static void CRYPTO_ghash_init(gmult_func *out_mult, ghash_func *out_hash, + u128 *out_key, u128 out_table[16], + const uint8_t *gcm_key) { + + union { + uint64_t u[2]; + uint8_t c[16]; + } H; + + OPENSSL_memcpy(H.c, gcm_key, 16); + + // H is stored in host byte order + H.u[0] = CRYPTO_bswap8(H.u[0]); + H.u[1] = CRYPTO_bswap8(H.u[1]); + + OPENSSL_memcpy(out_key, H.c, 16); + +#if defined(GHASH_ASM_X86_64) + if (crypto_gcm_clmul_enabled()) { + if (((OPENSSL_ia32cap_get()[1] >> 22) & 0x41) == 0x41) { // AVX+MOVBE + gcm_init_avx(out_table, H.u); + *out_mult = gcm_gmult_avx; + *out_hash = gcm_ghash_avx; + *out_is_avx = 1; + return; + } + gcm_init_clmul(out_table, H.u); + *out_mult = gcm_gmult_clmul; + *out_hash = gcm_ghash_clmul; + return; + } +#elif defined(GHASH_ASM_X86) + if (crypto_gcm_clmul_enabled()) { + gcm_init_clmul(out_table, H.u); + *out_mult = gcm_gmult_clmul; + *out_hash = gcm_ghash_clmul; + return; + } +#elif defined(GHASH_ASM_ARM) + if (pmull_capable()) { + gcm_init_v8(out_table, H.u); + *out_mult = gcm_gmult_v8; + *out_hash = gcm_ghash_v8; + return; + } + + if (neon_capable()) { + gcm_init_neon(out_table, H.u); + *out_mult = gcm_gmult_neon; + *out_hash = gcm_ghash_neon; + return; + } +#elif defined(GHASH_ASM_PPC64LE) + if (CRYPTO_is_PPC64LE_vcrypto_capable()) { + gcm_init_p8(out_table, H.u); + *out_mult = gcm_gmult_p8; + *out_hash = gcm_ghash_p8; + return; + } +#endif + + gcm_init_4bit(out_table, H.u); +#if defined(GHASH_ASM_X86) + *out_mult = gcm_gmult_4bit_mmx; + *out_hash = gcm_ghash_4bit_mmx; +#else + *out_mult = gcm_gmult_4bit; + *out_hash = gcm_ghash_4bit; +#endif +} + +void CRYPTO_gcm128_init(GCM128_CONTEXT *ctx, const void *aes_key, + block128_f block, int block_is_hwaes) { + OPENSSL_memset(ctx, 0, sizeof(*ctx)); + ctx->block = block; + + uint8_t gcm_key[16]; + OPENSSL_memset(gcm_key, 0, sizeof(gcm_key)); + (*block)(gcm_key, gcm_key, aes_key); + + CRYPTO_ghash_init(&ctx->gmult, &ctx->ghash, &ctx->H, ctx->Htable, + gcm_key); +} + +void CRYPTO_gcm128_setiv(GCM128_CONTEXT *ctx, const void *key, + const uint8_t *iv, size_t len) { + unsigned int ctr; +#ifdef GCM_FUNCREF_4BIT + void (*gcm_gmult_p)(uint64_t Xi[2], const u128 Htable[16]) = ctx->gmult; +#endif + + ctx->Yi.u[0] = 0; + ctx->Yi.u[1] = 0; + ctx->Xi.u[0] = 0; + ctx->Xi.u[1] = 0; + ctx->len.u[0] = 0; // AAD length + ctx->len.u[1] = 0; // message length + ctx->ares = 0; + ctx->mres = 0; + + if (len == 12) { + OPENSSL_memcpy(ctx->Yi.c, iv, 12); + ctx->Yi.c[15] = 1; + ctr = 1; + } else { + uint64_t len0 = len; + + while (len >= 16) { + for (size_t i = 0; i < 16; ++i) { + ctx->Yi.c[i] ^= iv[i]; + } + GCM_MUL(ctx, Yi); + iv += 16; + len -= 16; + } + if (len) { + for (size_t i = 0; i < len; ++i) { + ctx->Yi.c[i] ^= iv[i]; + } + GCM_MUL(ctx, Yi); + } + len0 <<= 3; + ctx->Yi.u[1] ^= CRYPTO_bswap8(len0); + + GCM_MUL(ctx, Yi); + ctr = CRYPTO_bswap4(ctx->Yi.d[3]); + } + + (*ctx->block)(ctx->Yi.c, ctx->EK0.c, key); + ++ctr; + ctx->Yi.d[3] = CRYPTO_bswap4(ctr); +} + +int CRYPTO_gcm128_aad(GCM128_CONTEXT *ctx, const uint8_t *aad, size_t len) { + unsigned int n; + uint64_t alen = ctx->len.u[0]; +#ifdef GCM_FUNCREF_4BIT + void (*gcm_gmult_p)(uint64_t Xi[2], const u128 Htable[16]) = ctx->gmult; +#ifdef GHASH + void (*gcm_ghash_p)(uint64_t Xi[2], const u128 Htable[16], const uint8_t *inp, + size_t len) = ctx->ghash; +#endif +#endif + + if (ctx->len.u[1]) { + return 0; + } + + alen += len; + if (alen > (UINT64_C(1) << 61) || (sizeof(len) == 8 && alen < len)) { + return 0; + } + ctx->len.u[0] = alen; + + n = ctx->ares; + if (n) { + while (n && len) { + ctx->Xi.c[n] ^= *(aad++); + --len; + n = (n + 1) % 16; + } + if (n == 0) { + GCM_MUL(ctx, Xi); + } else { + ctx->ares = n; + return 1; + } + } + + // Process a whole number of blocks. +#ifdef GHASH + size_t len_blocks = len & kSizeTWithoutLower4Bits; + if (len_blocks != 0) { + GHASH(ctx, aad, len_blocks); + aad += len_blocks; + len -= len_blocks; + } +#else + while (len >= 16) { + for (size_t i = 0; i < 16; ++i) { + ctx->Xi.c[i] ^= aad[i]; + } + GCM_MUL(ctx, Xi); + aad += 16; + len -= 16; + } +#endif + + // Process the remainder. + if (len != 0) { + n = (unsigned int)len; + for (size_t i = 0; i < len; ++i) { + ctx->Xi.c[i] ^= aad[i]; + } + } + + ctx->ares = n; + return 1; +} + +int CRYPTO_gcm128_encrypt(GCM128_CONTEXT *ctx, const void *key, + const uint8_t *in, uint8_t *out, size_t len) { + unsigned int n, ctr; + uint64_t mlen = ctx->len.u[1]; + block128_f block = ctx->block; +#ifdef GCM_FUNCREF_4BIT + void (*gcm_gmult_p)(uint64_t Xi[2], const u128 Htable[16]) = ctx->gmult; +#ifdef GHASH + void (*gcm_ghash_p)(uint64_t Xi[2], const u128 Htable[16], const uint8_t *inp, + size_t len) = ctx->ghash; +#endif +#endif + + mlen += len; + if (mlen > ((UINT64_C(1) << 36) - 32) || + (sizeof(len) == 8 && mlen < len)) { + return 0; + } + ctx->len.u[1] = mlen; + + if (ctx->ares) { + // First call to encrypt finalizes GHASH(AAD) + GCM_MUL(ctx, Xi); + ctx->ares = 0; + } + + ctr = CRYPTO_bswap4(ctx->Yi.d[3]); + + n = ctx->mres; + if (n) { + while (n && len) { + ctx->Xi.c[n] ^= *(out++) = *(in++) ^ ctx->EKi.c[n]; + --len; + n = (n + 1) % 16; + } + if (n == 0) { + GCM_MUL(ctx, Xi); + } else { + ctx->mres = n; + return 1; + } + } + if (STRICT_ALIGNMENT && + ((uintptr_t)in | (uintptr_t)out) % sizeof(size_t) != 0) { + for (size_t i = 0; i < len; ++i) { + if (n == 0) { + (*block)(ctx->Yi.c, ctx->EKi.c, key); + ++ctr; + ctx->Yi.d[3] = CRYPTO_bswap4(ctr); + } + ctx->Xi.c[n] ^= out[i] = in[i] ^ ctx->EKi.c[n]; + n = (n + 1) % 16; + if (n == 0) { + GCM_MUL(ctx, Xi); + } + } + + ctx->mres = n; + return 1; + } +#if defined(GHASH) && defined(GHASH_CHUNK) + while (len >= GHASH_CHUNK) { + size_t j = GHASH_CHUNK; + + while (j) { + (*block)(ctx->Yi.c, ctx->EKi.c, key); + ++ctr; + ctx->Yi.d[3] = CRYPTO_bswap4(ctr); + for (size_t i = 0; i < 16; i += sizeof(size_t)) { + store_word_le(out + i, + load_word_le(in + i) ^ ctx->EKi.t[i / sizeof(size_t)]); + } + out += 16; + in += 16; + j -= 16; + } + GHASH(ctx, out - GHASH_CHUNK, GHASH_CHUNK); + len -= GHASH_CHUNK; + } + size_t len_blocks = len & kSizeTWithoutLower4Bits; + if (len_blocks != 0) { + while (len >= 16) { + (*block)(ctx->Yi.c, ctx->EKi.c, key); + ++ctr; + ctx->Yi.d[3] = CRYPTO_bswap4(ctr); + for (size_t i = 0; i < 16; i += sizeof(size_t)) { + store_word_le(out + i, + load_word_le(in + i) ^ ctx->EKi.t[i / sizeof(size_t)]); + } + out += 16; + in += 16; + len -= 16; + } + GHASH(ctx, out - len_blocks, len_blocks); + } +#else + while (len >= 16) { + (*block)(ctx->Yi.c, ctx->EKi.c, key); + ++ctr; + ctx->Yi.d[3] = CRYPTO_bswap4(ctr); + for (size_t i = 0; i < 16; i += sizeof(size_t)) { + size_t tmp = load_word_le(in + i) ^ ctx->EKi.t[i / sizeof(size_t)]; + store_word_le(out + i, tmp); + ctx->Xi.t[i / sizeof(size_t)] ^= tmp; + } + GCM_MUL(ctx, Xi); + out += 16; + in += 16; + len -= 16; + } +#endif + if (len) { + (*block)(ctx->Yi.c, ctx->EKi.c, key); + ++ctr; + ctx->Yi.d[3] = CRYPTO_bswap4(ctr); + while (len--) { + ctx->Xi.c[n] ^= out[n] = in[n] ^ ctx->EKi.c[n]; + ++n; + } + } + + ctx->mres = n; + return 1; +} + +int CRYPTO_gcm128_decrypt(GCM128_CONTEXT *ctx, const void *key, + const unsigned char *in, unsigned char *out, + size_t len) { + unsigned int n, ctr; + uint64_t mlen = ctx->len.u[1]; + block128_f block = ctx->block; +#ifdef GCM_FUNCREF_4BIT + void (*gcm_gmult_p)(uint64_t Xi[2], const u128 Htable[16]) = ctx->gmult; +#ifdef GHASH + void (*gcm_ghash_p)(uint64_t Xi[2], const u128 Htable[16], const uint8_t *inp, + size_t len) = ctx->ghash; +#endif +#endif + + mlen += len; + if (mlen > ((UINT64_C(1) << 36) - 32) || + (sizeof(len) == 8 && mlen < len)) { + return 0; + } + ctx->len.u[1] = mlen; + + if (ctx->ares) { + // First call to decrypt finalizes GHASH(AAD) + GCM_MUL(ctx, Xi); + ctx->ares = 0; + } + + ctr = CRYPTO_bswap4(ctx->Yi.d[3]); + + n = ctx->mres; + if (n) { + while (n && len) { + uint8_t c = *(in++); + *(out++) = c ^ ctx->EKi.c[n]; + ctx->Xi.c[n] ^= c; + --len; + n = (n + 1) % 16; + } + if (n == 0) { + GCM_MUL(ctx, Xi); + } else { + ctx->mres = n; + return 1; + } + } + if (STRICT_ALIGNMENT && + ((uintptr_t)in | (uintptr_t)out) % sizeof(size_t) != 0) { + for (size_t i = 0; i < len; ++i) { + uint8_t c; + if (n == 0) { + (*block)(ctx->Yi.c, ctx->EKi.c, key); + ++ctr; + ctx->Yi.d[3] = CRYPTO_bswap4(ctr); + } + c = in[i]; + out[i] = c ^ ctx->EKi.c[n]; + ctx->Xi.c[n] ^= c; + n = (n + 1) % 16; + if (n == 0) { + GCM_MUL(ctx, Xi); + } + } + + ctx->mres = n; + return 1; + } +#if defined(GHASH) && defined(GHASH_CHUNK) + while (len >= GHASH_CHUNK) { + size_t j = GHASH_CHUNK; + + GHASH(ctx, in, GHASH_CHUNK); + while (j) { + (*block)(ctx->Yi.c, ctx->EKi.c, key); + ++ctr; + ctx->Yi.d[3] = CRYPTO_bswap4(ctr); + for (size_t i = 0; i < 16; i += sizeof(size_t)) { + store_word_le(out + i, + load_word_le(in + i) ^ ctx->EKi.t[i / sizeof(size_t)]); + } + out += 16; + in += 16; + j -= 16; + } + len -= GHASH_CHUNK; + } + size_t len_blocks = len & kSizeTWithoutLower4Bits; + if (len_blocks != 0) { + GHASH(ctx, in, len_blocks); + while (len >= 16) { + (*block)(ctx->Yi.c, ctx->EKi.c, key); + ++ctr; + ctx->Yi.d[3] = CRYPTO_bswap4(ctr); + for (size_t i = 0; i < 16; i += sizeof(size_t)) { + store_word_le(out + i, + load_word_le(in + i) ^ ctx->EKi.t[i / sizeof(size_t)]); + } + out += 16; + in += 16; + len -= 16; + } + } +#else + while (len >= 16) { + (*block)(ctx->Yi.c, ctx->EKi.c, key); + ++ctr; + ctx->Yi.d[3] = CRYPTO_bswap4(ctr); + for (size_t i = 0; i < 16; i += sizeof(size_t)) { + size_t c = load_word_le(in + i); + store_word_le(out + i, c ^ ctx->EKi.t[i / sizeof(size_t)]); + ctx->Xi.t[i / sizeof(size_t)] ^= c; + } + GCM_MUL(ctx, Xi); + out += 16; + in += 16; + len -= 16; + } +#endif + if (len) { + (*block)(ctx->Yi.c, ctx->EKi.c, key); + ++ctr; + ctx->Yi.d[3] = CRYPTO_bswap4(ctr); + while (len--) { + uint8_t c = in[n]; + ctx->Xi.c[n] ^= c; + out[n] = c ^ ctx->EKi.c[n]; + ++n; + } + } + + ctx->mres = n; + return 1; +} + +int CRYPTO_gcm128_finish(GCM128_CONTEXT *ctx, const uint8_t *tag, size_t len) { + uint64_t alen = ctx->len.u[0] << 3; + uint64_t clen = ctx->len.u[1] << 3; +#ifdef GCM_FUNCREF_4BIT + void (*gcm_gmult_p)(uint64_t Xi[2], const u128 Htable[16]) = ctx->gmult; +#endif + + if (ctx->mres || ctx->ares) { + GCM_MUL(ctx, Xi); + } + + alen = CRYPTO_bswap8(alen); + clen = CRYPTO_bswap8(clen); + + ctx->Xi.u[0] ^= alen; + ctx->Xi.u[1] ^= clen; + GCM_MUL(ctx, Xi); + + ctx->Xi.u[0] ^= ctx->EK0.u[0]; + ctx->Xi.u[1] ^= ctx->EK0.u[1]; + + if (tag && len <= sizeof(ctx->Xi)) { + return CRYPTO_memcmp(ctx->Xi.c, tag, len) == 0; + } else { + return 0; + } +} + +void CRYPTO_gcm128_tag(GCM128_CONTEXT *ctx, unsigned char *tag, size_t len) { + CRYPTO_gcm128_finish(ctx, NULL, 0); + OPENSSL_memcpy(tag, ctx->Xi.c, + len <= sizeof(ctx->Xi.c) ? len : sizeof(ctx->Xi.c)); +} + +#if defined(OPENSSL_X86) || defined(OPENSSL_X86_64) +int crypto_gcm_clmul_enabled(void) { +#ifdef GHASH_ASM + const uint32_t *ia32cap = OPENSSL_ia32cap_get(); + return (ia32cap[0] & (1 << 24)) && // check FXSR bit + (ia32cap[1] & (1 << 1)); // check PCLMULQDQ bit +#else + return 0; +#endif +} +#endif diff --git a/third_party/boringssl/common/aes.c b/third_party/boringssl/common/aes.c new file mode 100644 index 0000000000..aa213eb907 --- /dev/null +++ b/third_party/boringssl/common/aes.c @@ -0,0 +1,806 @@ +/* ==================================================================== + * Copyright (c) 2002-2006 The OpenSSL Project. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * + * 3. All advertising materials mentioning features or use of this + * software must display the following acknowledgment: + * "This product includes software developed by the OpenSSL Project + * for use in the OpenSSL Toolkit. (http://www.openssl.org/)" + * + * 4. The names "OpenSSL Toolkit" and "OpenSSL Project" must not be used to + * endorse or promote products derived from this software without + * prior written permission. For written permission, please contact + * openssl-core@openssl.org. + * + * 5. Products derived from this software may not be called "OpenSSL" + * nor may "OpenSSL" appear in their names without prior written + * permission of the OpenSSL Project. + * + * 6. Redistributions of any form whatsoever must retain the following + * acknowledgment: + * "This product includes software developed by the OpenSSL Project + * for use in the OpenSSL Toolkit (http://www.openssl.org/)" + * + * THIS SOFTWARE IS PROVIDED BY THE OpenSSL PROJECT ``AS IS'' AND ANY + * EXPRESSED OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE OpenSSL PROJECT OR + * ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT + * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + * ==================================================================== */ + +#include "aes.h" +#include "common.h" +#include "endian.h" + +static inline uint32_t GETU32(const void *in) { + return be32toh(*(uint32_t *)in); +} + +static inline void PUTU32(void *out, uint32_t v) { + *(uint32_t *)out = htobe32(v); +} + +// Te0[x] = S [x].[02, 01, 01, 03]; +// Te1[x] = S [x].[03, 02, 01, 01]; +// Te2[x] = S [x].[01, 03, 02, 01]; +// Te3[x] = S [x].[01, 01, 03, 02]; +// +// Td0[x] = Si[x].[0e, 09, 0d, 0b]; +// Td1[x] = Si[x].[0b, 0e, 09, 0d]; +// Td2[x] = Si[x].[0d, 0b, 0e, 09]; +// Td3[x] = Si[x].[09, 0d, 0b, 0e]; +// Td4[x] = Si[x].[01]; + +static const uint32_t Te0[256] = { + 0xc66363a5U, 0xf87c7c84U, 0xee777799U, 0xf67b7b8dU, 0xfff2f20dU, + 0xd66b6bbdU, 0xde6f6fb1U, 0x91c5c554U, 0x60303050U, 0x02010103U, + 0xce6767a9U, 0x562b2b7dU, 0xe7fefe19U, 0xb5d7d762U, 0x4dababe6U, + 0xec76769aU, 0x8fcaca45U, 0x1f82829dU, 0x89c9c940U, 0xfa7d7d87U, + 0xeffafa15U, 0xb25959ebU, 0x8e4747c9U, 0xfbf0f00bU, 0x41adadecU, + 0xb3d4d467U, 0x5fa2a2fdU, 0x45afafeaU, 0x239c9cbfU, 0x53a4a4f7U, + 0xe4727296U, 0x9bc0c05bU, 0x75b7b7c2U, 0xe1fdfd1cU, 0x3d9393aeU, + 0x4c26266aU, 0x6c36365aU, 0x7e3f3f41U, 0xf5f7f702U, 0x83cccc4fU, + 0x6834345cU, 0x51a5a5f4U, 0xd1e5e534U, 0xf9f1f108U, 0xe2717193U, + 0xabd8d873U, 0x62313153U, 0x2a15153fU, 0x0804040cU, 0x95c7c752U, + 0x46232365U, 0x9dc3c35eU, 0x30181828U, 0x379696a1U, 0x0a05050fU, + 0x2f9a9ab5U, 0x0e070709U, 0x24121236U, 0x1b80809bU, 0xdfe2e23dU, + 0xcdebeb26U, 0x4e272769U, 0x7fb2b2cdU, 0xea75759fU, 0x1209091bU, + 0x1d83839eU, 0x582c2c74U, 0x341a1a2eU, 0x361b1b2dU, 0xdc6e6eb2U, + 0xb45a5aeeU, 0x5ba0a0fbU, 0xa45252f6U, 0x763b3b4dU, 0xb7d6d661U, + 0x7db3b3ceU, 0x5229297bU, 0xdde3e33eU, 0x5e2f2f71U, 0x13848497U, + 0xa65353f5U, 0xb9d1d168U, 0x00000000U, 0xc1eded2cU, 0x40202060U, + 0xe3fcfc1fU, 0x79b1b1c8U, 0xb65b5bedU, 0xd46a6abeU, 0x8dcbcb46U, + 0x67bebed9U, 0x7239394bU, 0x944a4adeU, 0x984c4cd4U, 0xb05858e8U, + 0x85cfcf4aU, 0xbbd0d06bU, 0xc5efef2aU, 0x4faaaae5U, 0xedfbfb16U, + 0x864343c5U, 0x9a4d4dd7U, 0x66333355U, 0x11858594U, 0x8a4545cfU, + 0xe9f9f910U, 0x04020206U, 0xfe7f7f81U, 0xa05050f0U, 0x783c3c44U, + 0x259f9fbaU, 0x4ba8a8e3U, 0xa25151f3U, 0x5da3a3feU, 0x804040c0U, + 0x058f8f8aU, 0x3f9292adU, 0x219d9dbcU, 0x70383848U, 0xf1f5f504U, + 0x63bcbcdfU, 0x77b6b6c1U, 0xafdada75U, 0x42212163U, 0x20101030U, + 0xe5ffff1aU, 0xfdf3f30eU, 0xbfd2d26dU, 0x81cdcd4cU, 0x180c0c14U, + 0x26131335U, 0xc3ecec2fU, 0xbe5f5fe1U, 0x359797a2U, 0x884444ccU, + 0x2e171739U, 0x93c4c457U, 0x55a7a7f2U, 0xfc7e7e82U, 0x7a3d3d47U, + 0xc86464acU, 0xba5d5de7U, 0x3219192bU, 0xe6737395U, 0xc06060a0U, + 0x19818198U, 0x9e4f4fd1U, 0xa3dcdc7fU, 0x44222266U, 0x542a2a7eU, + 0x3b9090abU, 0x0b888883U, 0x8c4646caU, 0xc7eeee29U, 0x6bb8b8d3U, + 0x2814143cU, 0xa7dede79U, 0xbc5e5ee2U, 0x160b0b1dU, 0xaddbdb76U, + 0xdbe0e03bU, 0x64323256U, 0x743a3a4eU, 0x140a0a1eU, 0x924949dbU, + 0x0c06060aU, 0x4824246cU, 0xb85c5ce4U, 0x9fc2c25dU, 0xbdd3d36eU, + 0x43acacefU, 0xc46262a6U, 0x399191a8U, 0x319595a4U, 0xd3e4e437U, + 0xf279798bU, 0xd5e7e732U, 0x8bc8c843U, 0x6e373759U, 0xda6d6db7U, + 0x018d8d8cU, 0xb1d5d564U, 0x9c4e4ed2U, 0x49a9a9e0U, 0xd86c6cb4U, + 0xac5656faU, 0xf3f4f407U, 0xcfeaea25U, 0xca6565afU, 0xf47a7a8eU, + 0x47aeaee9U, 0x10080818U, 0x6fbabad5U, 0xf0787888U, 0x4a25256fU, + 0x5c2e2e72U, 0x381c1c24U, 0x57a6a6f1U, 0x73b4b4c7U, 0x97c6c651U, + 0xcbe8e823U, 0xa1dddd7cU, 0xe874749cU, 0x3e1f1f21U, 0x964b4bddU, + 0x61bdbddcU, 0x0d8b8b86U, 0x0f8a8a85U, 0xe0707090U, 0x7c3e3e42U, + 0x71b5b5c4U, 0xcc6666aaU, 0x904848d8U, 0x06030305U, 0xf7f6f601U, + 0x1c0e0e12U, 0xc26161a3U, 0x6a35355fU, 0xae5757f9U, 0x69b9b9d0U, + 0x17868691U, 0x99c1c158U, 0x3a1d1d27U, 0x279e9eb9U, 0xd9e1e138U, + 0xebf8f813U, 0x2b9898b3U, 0x22111133U, 0xd26969bbU, 0xa9d9d970U, + 0x078e8e89U, 0x339494a7U, 0x2d9b9bb6U, 0x3c1e1e22U, 0x15878792U, + 0xc9e9e920U, 0x87cece49U, 0xaa5555ffU, 0x50282878U, 0xa5dfdf7aU, + 0x038c8c8fU, 0x59a1a1f8U, 0x09898980U, 0x1a0d0d17U, 0x65bfbfdaU, + 0xd7e6e631U, 0x844242c6U, 0xd06868b8U, 0x824141c3U, 0x299999b0U, + 0x5a2d2d77U, 0x1e0f0f11U, 0x7bb0b0cbU, 0xa85454fcU, 0x6dbbbbd6U, + 0x2c16163aU, }; + +static const uint32_t Te1[256] = { + 0xa5c66363U, 0x84f87c7cU, 0x99ee7777U, 0x8df67b7bU, 0x0dfff2f2U, + 0xbdd66b6bU, 0xb1de6f6fU, 0x5491c5c5U, 0x50603030U, 0x03020101U, + 0xa9ce6767U, 0x7d562b2bU, 0x19e7fefeU, 0x62b5d7d7U, 0xe64dababU, + 0x9aec7676U, 0x458fcacaU, 0x9d1f8282U, 0x4089c9c9U, 0x87fa7d7dU, + 0x15effafaU, 0xebb25959U, 0xc98e4747U, 0x0bfbf0f0U, 0xec41adadU, + 0x67b3d4d4U, 0xfd5fa2a2U, 0xea45afafU, 0xbf239c9cU, 0xf753a4a4U, + 0x96e47272U, 0x5b9bc0c0U, 0xc275b7b7U, 0x1ce1fdfdU, 0xae3d9393U, + 0x6a4c2626U, 0x5a6c3636U, 0x417e3f3fU, 0x02f5f7f7U, 0x4f83ccccU, + 0x5c683434U, 0xf451a5a5U, 0x34d1e5e5U, 0x08f9f1f1U, 0x93e27171U, + 0x73abd8d8U, 0x53623131U, 0x3f2a1515U, 0x0c080404U, 0x5295c7c7U, + 0x65462323U, 0x5e9dc3c3U, 0x28301818U, 0xa1379696U, 0x0f0a0505U, + 0xb52f9a9aU, 0x090e0707U, 0x36241212U, 0x9b1b8080U, 0x3ddfe2e2U, + 0x26cdebebU, 0x694e2727U, 0xcd7fb2b2U, 0x9fea7575U, 0x1b120909U, + 0x9e1d8383U, 0x74582c2cU, 0x2e341a1aU, 0x2d361b1bU, 0xb2dc6e6eU, + 0xeeb45a5aU, 0xfb5ba0a0U, 0xf6a45252U, 0x4d763b3bU, 0x61b7d6d6U, + 0xce7db3b3U, 0x7b522929U, 0x3edde3e3U, 0x715e2f2fU, 0x97138484U, + 0xf5a65353U, 0x68b9d1d1U, 0x00000000U, 0x2cc1ededU, 0x60402020U, + 0x1fe3fcfcU, 0xc879b1b1U, 0xedb65b5bU, 0xbed46a6aU, 0x468dcbcbU, + 0xd967bebeU, 0x4b723939U, 0xde944a4aU, 0xd4984c4cU, 0xe8b05858U, + 0x4a85cfcfU, 0x6bbbd0d0U, 0x2ac5efefU, 0xe54faaaaU, 0x16edfbfbU, + 0xc5864343U, 0xd79a4d4dU, 0x55663333U, 0x94118585U, 0xcf8a4545U, + 0x10e9f9f9U, 0x06040202U, 0x81fe7f7fU, 0xf0a05050U, 0x44783c3cU, + 0xba259f9fU, 0xe34ba8a8U, 0xf3a25151U, 0xfe5da3a3U, 0xc0804040U, + 0x8a058f8fU, 0xad3f9292U, 0xbc219d9dU, 0x48703838U, 0x04f1f5f5U, + 0xdf63bcbcU, 0xc177b6b6U, 0x75afdadaU, 0x63422121U, 0x30201010U, + 0x1ae5ffffU, 0x0efdf3f3U, 0x6dbfd2d2U, 0x4c81cdcdU, 0x14180c0cU, + 0x35261313U, 0x2fc3ececU, 0xe1be5f5fU, 0xa2359797U, 0xcc884444U, + 0x392e1717U, 0x5793c4c4U, 0xf255a7a7U, 0x82fc7e7eU, 0x477a3d3dU, + 0xacc86464U, 0xe7ba5d5dU, 0x2b321919U, 0x95e67373U, 0xa0c06060U, + 0x98198181U, 0xd19e4f4fU, 0x7fa3dcdcU, 0x66442222U, 0x7e542a2aU, + 0xab3b9090U, 0x830b8888U, 0xca8c4646U, 0x29c7eeeeU, 0xd36bb8b8U, + 0x3c281414U, 0x79a7dedeU, 0xe2bc5e5eU, 0x1d160b0bU, 0x76addbdbU, + 0x3bdbe0e0U, 0x56643232U, 0x4e743a3aU, 0x1e140a0aU, 0xdb924949U, + 0x0a0c0606U, 0x6c482424U, 0xe4b85c5cU, 0x5d9fc2c2U, 0x6ebdd3d3U, + 0xef43acacU, 0xa6c46262U, 0xa8399191U, 0xa4319595U, 0x37d3e4e4U, + 0x8bf27979U, 0x32d5e7e7U, 0x438bc8c8U, 0x596e3737U, 0xb7da6d6dU, + 0x8c018d8dU, 0x64b1d5d5U, 0xd29c4e4eU, 0xe049a9a9U, 0xb4d86c6cU, + 0xfaac5656U, 0x07f3f4f4U, 0x25cfeaeaU, 0xafca6565U, 0x8ef47a7aU, + 0xe947aeaeU, 0x18100808U, 0xd56fbabaU, 0x88f07878U, 0x6f4a2525U, + 0x725c2e2eU, 0x24381c1cU, 0xf157a6a6U, 0xc773b4b4U, 0x5197c6c6U, + 0x23cbe8e8U, 0x7ca1ddddU, 0x9ce87474U, 0x213e1f1fU, 0xdd964b4bU, + 0xdc61bdbdU, 0x860d8b8bU, 0x850f8a8aU, 0x90e07070U, 0x427c3e3eU, + 0xc471b5b5U, 0xaacc6666U, 0xd8904848U, 0x05060303U, 0x01f7f6f6U, + 0x121c0e0eU, 0xa3c26161U, 0x5f6a3535U, 0xf9ae5757U, 0xd069b9b9U, + 0x91178686U, 0x5899c1c1U, 0x273a1d1dU, 0xb9279e9eU, 0x38d9e1e1U, + 0x13ebf8f8U, 0xb32b9898U, 0x33221111U, 0xbbd26969U, 0x70a9d9d9U, + 0x89078e8eU, 0xa7339494U, 0xb62d9b9bU, 0x223c1e1eU, 0x92158787U, + 0x20c9e9e9U, 0x4987ceceU, 0xffaa5555U, 0x78502828U, 0x7aa5dfdfU, + 0x8f038c8cU, 0xf859a1a1U, 0x80098989U, 0x171a0d0dU, 0xda65bfbfU, + 0x31d7e6e6U, 0xc6844242U, 0xb8d06868U, 0xc3824141U, 0xb0299999U, + 0x775a2d2dU, 0x111e0f0fU, 0xcb7bb0b0U, 0xfca85454U, 0xd66dbbbbU, + 0x3a2c1616U, }; + +static const uint32_t Te2[256] = { + 0x63a5c663U, 0x7c84f87cU, 0x7799ee77U, 0x7b8df67bU, 0xf20dfff2U, + 0x6bbdd66bU, 0x6fb1de6fU, 0xc55491c5U, 0x30506030U, 0x01030201U, + 0x67a9ce67U, 0x2b7d562bU, 0xfe19e7feU, 0xd762b5d7U, 0xabe64dabU, + 0x769aec76U, 0xca458fcaU, 0x829d1f82U, 0xc94089c9U, 0x7d87fa7dU, + 0xfa15effaU, 0x59ebb259U, 0x47c98e47U, 0xf00bfbf0U, 0xadec41adU, + 0xd467b3d4U, 0xa2fd5fa2U, 0xafea45afU, 0x9cbf239cU, 0xa4f753a4U, + 0x7296e472U, 0xc05b9bc0U, 0xb7c275b7U, 0xfd1ce1fdU, 0x93ae3d93U, + 0x266a4c26U, 0x365a6c36U, 0x3f417e3fU, 0xf702f5f7U, 0xcc4f83ccU, + 0x345c6834U, 0xa5f451a5U, 0xe534d1e5U, 0xf108f9f1U, 0x7193e271U, + 0xd873abd8U, 0x31536231U, 0x153f2a15U, 0x040c0804U, 0xc75295c7U, + 0x23654623U, 0xc35e9dc3U, 0x18283018U, 0x96a13796U, 0x050f0a05U, + 0x9ab52f9aU, 0x07090e07U, 0x12362412U, 0x809b1b80U, 0xe23ddfe2U, + 0xeb26cdebU, 0x27694e27U, 0xb2cd7fb2U, 0x759fea75U, 0x091b1209U, + 0x839e1d83U, 0x2c74582cU, 0x1a2e341aU, 0x1b2d361bU, 0x6eb2dc6eU, + 0x5aeeb45aU, 0xa0fb5ba0U, 0x52f6a452U, 0x3b4d763bU, 0xd661b7d6U, + 0xb3ce7db3U, 0x297b5229U, 0xe33edde3U, 0x2f715e2fU, 0x84971384U, + 0x53f5a653U, 0xd168b9d1U, 0x00000000U, 0xed2cc1edU, 0x20604020U, + 0xfc1fe3fcU, 0xb1c879b1U, 0x5bedb65bU, 0x6abed46aU, 0xcb468dcbU, + 0xbed967beU, 0x394b7239U, 0x4ade944aU, 0x4cd4984cU, 0x58e8b058U, + 0xcf4a85cfU, 0xd06bbbd0U, 0xef2ac5efU, 0xaae54faaU, 0xfb16edfbU, + 0x43c58643U, 0x4dd79a4dU, 0x33556633U, 0x85941185U, 0x45cf8a45U, + 0xf910e9f9U, 0x02060402U, 0x7f81fe7fU, 0x50f0a050U, 0x3c44783cU, + 0x9fba259fU, 0xa8e34ba8U, 0x51f3a251U, 0xa3fe5da3U, 0x40c08040U, + 0x8f8a058fU, 0x92ad3f92U, 0x9dbc219dU, 0x38487038U, 0xf504f1f5U, + 0xbcdf63bcU, 0xb6c177b6U, 0xda75afdaU, 0x21634221U, 0x10302010U, + 0xff1ae5ffU, 0xf30efdf3U, 0xd26dbfd2U, 0xcd4c81cdU, 0x0c14180cU, + 0x13352613U, 0xec2fc3ecU, 0x5fe1be5fU, 0x97a23597U, 0x44cc8844U, + 0x17392e17U, 0xc45793c4U, 0xa7f255a7U, 0x7e82fc7eU, 0x3d477a3dU, + 0x64acc864U, 0x5de7ba5dU, 0x192b3219U, 0x7395e673U, 0x60a0c060U, + 0x81981981U, 0x4fd19e4fU, 0xdc7fa3dcU, 0x22664422U, 0x2a7e542aU, + 0x90ab3b90U, 0x88830b88U, 0x46ca8c46U, 0xee29c7eeU, 0xb8d36bb8U, + 0x143c2814U, 0xde79a7deU, 0x5ee2bc5eU, 0x0b1d160bU, 0xdb76addbU, + 0xe03bdbe0U, 0x32566432U, 0x3a4e743aU, 0x0a1e140aU, 0x49db9249U, + 0x060a0c06U, 0x246c4824U, 0x5ce4b85cU, 0xc25d9fc2U, 0xd36ebdd3U, + 0xacef43acU, 0x62a6c462U, 0x91a83991U, 0x95a43195U, 0xe437d3e4U, + 0x798bf279U, 0xe732d5e7U, 0xc8438bc8U, 0x37596e37U, 0x6db7da6dU, + 0x8d8c018dU, 0xd564b1d5U, 0x4ed29c4eU, 0xa9e049a9U, 0x6cb4d86cU, + 0x56faac56U, 0xf407f3f4U, 0xea25cfeaU, 0x65afca65U, 0x7a8ef47aU, + 0xaee947aeU, 0x08181008U, 0xbad56fbaU, 0x7888f078U, 0x256f4a25U, + 0x2e725c2eU, 0x1c24381cU, 0xa6f157a6U, 0xb4c773b4U, 0xc65197c6U, + 0xe823cbe8U, 0xdd7ca1ddU, 0x749ce874U, 0x1f213e1fU, 0x4bdd964bU, + 0xbddc61bdU, 0x8b860d8bU, 0x8a850f8aU, 0x7090e070U, 0x3e427c3eU, + 0xb5c471b5U, 0x66aacc66U, 0x48d89048U, 0x03050603U, 0xf601f7f6U, + 0x0e121c0eU, 0x61a3c261U, 0x355f6a35U, 0x57f9ae57U, 0xb9d069b9U, + 0x86911786U, 0xc15899c1U, 0x1d273a1dU, 0x9eb9279eU, 0xe138d9e1U, + 0xf813ebf8U, 0x98b32b98U, 0x11332211U, 0x69bbd269U, 0xd970a9d9U, + 0x8e89078eU, 0x94a73394U, 0x9bb62d9bU, 0x1e223c1eU, 0x87921587U, + 0xe920c9e9U, 0xce4987ceU, 0x55ffaa55U, 0x28785028U, 0xdf7aa5dfU, + 0x8c8f038cU, 0xa1f859a1U, 0x89800989U, 0x0d171a0dU, 0xbfda65bfU, + 0xe631d7e6U, 0x42c68442U, 0x68b8d068U, 0x41c38241U, 0x99b02999U, + 0x2d775a2dU, 0x0f111e0fU, 0xb0cb7bb0U, 0x54fca854U, 0xbbd66dbbU, + 0x163a2c16U, }; + +static const uint32_t Te3[256] = { + 0x6363a5c6U, 0x7c7c84f8U, 0x777799eeU, 0x7b7b8df6U, 0xf2f20dffU, + 0x6b6bbdd6U, 0x6f6fb1deU, 0xc5c55491U, 0x30305060U, 0x01010302U, + 0x6767a9ceU, 0x2b2b7d56U, 0xfefe19e7U, 0xd7d762b5U, 0xababe64dU, + 0x76769aecU, 0xcaca458fU, 0x82829d1fU, 0xc9c94089U, 0x7d7d87faU, + 0xfafa15efU, 0x5959ebb2U, 0x4747c98eU, 0xf0f00bfbU, 0xadadec41U, + 0xd4d467b3U, 0xa2a2fd5fU, 0xafafea45U, 0x9c9cbf23U, 0xa4a4f753U, + 0x727296e4U, 0xc0c05b9bU, 0xb7b7c275U, 0xfdfd1ce1U, 0x9393ae3dU, + 0x26266a4cU, 0x36365a6cU, 0x3f3f417eU, 0xf7f702f5U, 0xcccc4f83U, + 0x34345c68U, 0xa5a5f451U, 0xe5e534d1U, 0xf1f108f9U, 0x717193e2U, + 0xd8d873abU, 0x31315362U, 0x15153f2aU, 0x04040c08U, 0xc7c75295U, + 0x23236546U, 0xc3c35e9dU, 0x18182830U, 0x9696a137U, 0x05050f0aU, + 0x9a9ab52fU, 0x0707090eU, 0x12123624U, 0x80809b1bU, 0xe2e23ddfU, + 0xebeb26cdU, 0x2727694eU, 0xb2b2cd7fU, 0x75759feaU, 0x09091b12U, + 0x83839e1dU, 0x2c2c7458U, 0x1a1a2e34U, 0x1b1b2d36U, 0x6e6eb2dcU, + 0x5a5aeeb4U, 0xa0a0fb5bU, 0x5252f6a4U, 0x3b3b4d76U, 0xd6d661b7U, + 0xb3b3ce7dU, 0x29297b52U, 0xe3e33eddU, 0x2f2f715eU, 0x84849713U, + 0x5353f5a6U, 0xd1d168b9U, 0x00000000U, 0xeded2cc1U, 0x20206040U, + 0xfcfc1fe3U, 0xb1b1c879U, 0x5b5bedb6U, 0x6a6abed4U, 0xcbcb468dU, + 0xbebed967U, 0x39394b72U, 0x4a4ade94U, 0x4c4cd498U, 0x5858e8b0U, + 0xcfcf4a85U, 0xd0d06bbbU, 0xefef2ac5U, 0xaaaae54fU, 0xfbfb16edU, + 0x4343c586U, 0x4d4dd79aU, 0x33335566U, 0x85859411U, 0x4545cf8aU, + 0xf9f910e9U, 0x02020604U, 0x7f7f81feU, 0x5050f0a0U, 0x3c3c4478U, + 0x9f9fba25U, 0xa8a8e34bU, 0x5151f3a2U, 0xa3a3fe5dU, 0x4040c080U, + 0x8f8f8a05U, 0x9292ad3fU, 0x9d9dbc21U, 0x38384870U, 0xf5f504f1U, + 0xbcbcdf63U, 0xb6b6c177U, 0xdada75afU, 0x21216342U, 0x10103020U, + 0xffff1ae5U, 0xf3f30efdU, 0xd2d26dbfU, 0xcdcd4c81U, 0x0c0c1418U, + 0x13133526U, 0xecec2fc3U, 0x5f5fe1beU, 0x9797a235U, 0x4444cc88U, + 0x1717392eU, 0xc4c45793U, 0xa7a7f255U, 0x7e7e82fcU, 0x3d3d477aU, + 0x6464acc8U, 0x5d5de7baU, 0x19192b32U, 0x737395e6U, 0x6060a0c0U, + 0x81819819U, 0x4f4fd19eU, 0xdcdc7fa3U, 0x22226644U, 0x2a2a7e54U, + 0x9090ab3bU, 0x8888830bU, 0x4646ca8cU, 0xeeee29c7U, 0xb8b8d36bU, + 0x14143c28U, 0xdede79a7U, 0x5e5ee2bcU, 0x0b0b1d16U, 0xdbdb76adU, + 0xe0e03bdbU, 0x32325664U, 0x3a3a4e74U, 0x0a0a1e14U, 0x4949db92U, + 0x06060a0cU, 0x24246c48U, 0x5c5ce4b8U, 0xc2c25d9fU, 0xd3d36ebdU, + 0xacacef43U, 0x6262a6c4U, 0x9191a839U, 0x9595a431U, 0xe4e437d3U, + 0x79798bf2U, 0xe7e732d5U, 0xc8c8438bU, 0x3737596eU, 0x6d6db7daU, + 0x8d8d8c01U, 0xd5d564b1U, 0x4e4ed29cU, 0xa9a9e049U, 0x6c6cb4d8U, + 0x5656faacU, 0xf4f407f3U, 0xeaea25cfU, 0x6565afcaU, 0x7a7a8ef4U, + 0xaeaee947U, 0x08081810U, 0xbabad56fU, 0x787888f0U, 0x25256f4aU, + 0x2e2e725cU, 0x1c1c2438U, 0xa6a6f157U, 0xb4b4c773U, 0xc6c65197U, + 0xe8e823cbU, 0xdddd7ca1U, 0x74749ce8U, 0x1f1f213eU, 0x4b4bdd96U, + 0xbdbddc61U, 0x8b8b860dU, 0x8a8a850fU, 0x707090e0U, 0x3e3e427cU, + 0xb5b5c471U, 0x6666aaccU, 0x4848d890U, 0x03030506U, 0xf6f601f7U, + 0x0e0e121cU, 0x6161a3c2U, 0x35355f6aU, 0x5757f9aeU, 0xb9b9d069U, + 0x86869117U, 0xc1c15899U, 0x1d1d273aU, 0x9e9eb927U, 0xe1e138d9U, + 0xf8f813ebU, 0x9898b32bU, 0x11113322U, 0x6969bbd2U, 0xd9d970a9U, + 0x8e8e8907U, 0x9494a733U, 0x9b9bb62dU, 0x1e1e223cU, 0x87879215U, + 0xe9e920c9U, 0xcece4987U, 0x5555ffaaU, 0x28287850U, 0xdfdf7aa5U, + 0x8c8c8f03U, 0xa1a1f859U, 0x89898009U, 0x0d0d171aU, 0xbfbfda65U, + 0xe6e631d7U, 0x4242c684U, 0x6868b8d0U, 0x4141c382U, 0x9999b029U, + 0x2d2d775aU, 0x0f0f111eU, 0xb0b0cb7bU, 0x5454fca8U, 0xbbbbd66dU, + 0x16163a2cU, }; + +static const uint32_t Td0[256] = { + 0x51f4a750U, 0x7e416553U, 0x1a17a4c3U, 0x3a275e96U, 0x3bab6bcbU, + 0x1f9d45f1U, 0xacfa58abU, 0x4be30393U, 0x2030fa55U, 0xad766df6U, + 0x88cc7691U, 0xf5024c25U, 0x4fe5d7fcU, 0xc52acbd7U, 0x26354480U, + 0xb562a38fU, 0xdeb15a49U, 0x25ba1b67U, 0x45ea0e98U, 0x5dfec0e1U, + 0xc32f7502U, 0x814cf012U, 0x8d4697a3U, 0x6bd3f9c6U, 0x038f5fe7U, + 0x15929c95U, 0xbf6d7aebU, 0x955259daU, 0xd4be832dU, 0x587421d3U, + 0x49e06929U, 0x8ec9c844U, 0x75c2896aU, 0xf48e7978U, 0x99583e6bU, + 0x27b971ddU, 0xbee14fb6U, 0xf088ad17U, 0xc920ac66U, 0x7dce3ab4U, + 0x63df4a18U, 0xe51a3182U, 0x97513360U, 0x62537f45U, 0xb16477e0U, + 0xbb6bae84U, 0xfe81a01cU, 0xf9082b94U, 0x70486858U, 0x8f45fd19U, + 0x94de6c87U, 0x527bf8b7U, 0xab73d323U, 0x724b02e2U, 0xe31f8f57U, + 0x6655ab2aU, 0xb2eb2807U, 0x2fb5c203U, 0x86c57b9aU, 0xd33708a5U, + 0x302887f2U, 0x23bfa5b2U, 0x02036abaU, 0xed16825cU, 0x8acf1c2bU, + 0xa779b492U, 0xf307f2f0U, 0x4e69e2a1U, 0x65daf4cdU, 0x0605bed5U, + 0xd134621fU, 0xc4a6fe8aU, 0x342e539dU, 0xa2f355a0U, 0x058ae132U, + 0xa4f6eb75U, 0x0b83ec39U, 0x4060efaaU, 0x5e719f06U, 0xbd6e1051U, + 0x3e218af9U, 0x96dd063dU, 0xdd3e05aeU, 0x4de6bd46U, 0x91548db5U, + 0x71c45d05U, 0x0406d46fU, 0x605015ffU, 0x1998fb24U, 0xd6bde997U, + 0x894043ccU, 0x67d99e77U, 0xb0e842bdU, 0x07898b88U, 0xe7195b38U, + 0x79c8eedbU, 0xa17c0a47U, 0x7c420fe9U, 0xf8841ec9U, 0x00000000U, + 0x09808683U, 0x322bed48U, 0x1e1170acU, 0x6c5a724eU, 0xfd0efffbU, + 0x0f853856U, 0x3daed51eU, 0x362d3927U, 0x0a0fd964U, 0x685ca621U, + 0x9b5b54d1U, 0x24362e3aU, 0x0c0a67b1U, 0x9357e70fU, 0xb4ee96d2U, + 0x1b9b919eU, 0x80c0c54fU, 0x61dc20a2U, 0x5a774b69U, 0x1c121a16U, + 0xe293ba0aU, 0xc0a02ae5U, 0x3c22e043U, 0x121b171dU, 0x0e090d0bU, + 0xf28bc7adU, 0x2db6a8b9U, 0x141ea9c8U, 0x57f11985U, 0xaf75074cU, + 0xee99ddbbU, 0xa37f60fdU, 0xf701269fU, 0x5c72f5bcU, 0x44663bc5U, + 0x5bfb7e34U, 0x8b432976U, 0xcb23c6dcU, 0xb6edfc68U, 0xb8e4f163U, + 0xd731dccaU, 0x42638510U, 0x13972240U, 0x84c61120U, 0x854a247dU, + 0xd2bb3df8U, 0xaef93211U, 0xc729a16dU, 0x1d9e2f4bU, 0xdcb230f3U, + 0x0d8652ecU, 0x77c1e3d0U, 0x2bb3166cU, 0xa970b999U, 0x119448faU, + 0x47e96422U, 0xa8fc8cc4U, 0xa0f03f1aU, 0x567d2cd8U, 0x223390efU, + 0x87494ec7U, 0xd938d1c1U, 0x8ccaa2feU, 0x98d40b36U, 0xa6f581cfU, + 0xa57ade28U, 0xdab78e26U, 0x3fadbfa4U, 0x2c3a9de4U, 0x5078920dU, + 0x6a5fcc9bU, 0x547e4662U, 0xf68d13c2U, 0x90d8b8e8U, 0x2e39f75eU, + 0x82c3aff5U, 0x9f5d80beU, 0x69d0937cU, 0x6fd52da9U, 0xcf2512b3U, + 0xc8ac993bU, 0x10187da7U, 0xe89c636eU, 0xdb3bbb7bU, 0xcd267809U, + 0x6e5918f4U, 0xec9ab701U, 0x834f9aa8U, 0xe6956e65U, 0xaaffe67eU, + 0x21bccf08U, 0xef15e8e6U, 0xbae79bd9U, 0x4a6f36ceU, 0xea9f09d4U, + 0x29b07cd6U, 0x31a4b2afU, 0x2a3f2331U, 0xc6a59430U, 0x35a266c0U, + 0x744ebc37U, 0xfc82caa6U, 0xe090d0b0U, 0x33a7d815U, 0xf104984aU, + 0x41ecdaf7U, 0x7fcd500eU, 0x1791f62fU, 0x764dd68dU, 0x43efb04dU, + 0xccaa4d54U, 0xe49604dfU, 0x9ed1b5e3U, 0x4c6a881bU, 0xc12c1fb8U, + 0x4665517fU, 0x9d5eea04U, 0x018c355dU, 0xfa877473U, 0xfb0b412eU, + 0xb3671d5aU, 0x92dbd252U, 0xe9105633U, 0x6dd64713U, 0x9ad7618cU, + 0x37a10c7aU, 0x59f8148eU, 0xeb133c89U, 0xcea927eeU, 0xb761c935U, + 0xe11ce5edU, 0x7a47b13cU, 0x9cd2df59U, 0x55f2733fU, 0x1814ce79U, + 0x73c737bfU, 0x53f7cdeaU, 0x5ffdaa5bU, 0xdf3d6f14U, 0x7844db86U, + 0xcaaff381U, 0xb968c43eU, 0x3824342cU, 0xc2a3405fU, 0x161dc372U, + 0xbce2250cU, 0x283c498bU, 0xff0d9541U, 0x39a80171U, 0x080cb3deU, + 0xd8b4e49cU, 0x6456c190U, 0x7bcb8461U, 0xd532b670U, 0x486c5c74U, + 0xd0b85742U, }; + +static const uint32_t Td1[256] = { + 0x5051f4a7U, 0x537e4165U, 0xc31a17a4U, 0x963a275eU, 0xcb3bab6bU, + 0xf11f9d45U, 0xabacfa58U, 0x934be303U, 0x552030faU, 0xf6ad766dU, + 0x9188cc76U, 0x25f5024cU, 0xfc4fe5d7U, 0xd7c52acbU, 0x80263544U, + 0x8fb562a3U, 0x49deb15aU, 0x6725ba1bU, 0x9845ea0eU, 0xe15dfec0U, + 0x02c32f75U, 0x12814cf0U, 0xa38d4697U, 0xc66bd3f9U, 0xe7038f5fU, + 0x9515929cU, 0xebbf6d7aU, 0xda955259U, 0x2dd4be83U, 0xd3587421U, + 0x2949e069U, 0x448ec9c8U, 0x6a75c289U, 0x78f48e79U, 0x6b99583eU, + 0xdd27b971U, 0xb6bee14fU, 0x17f088adU, 0x66c920acU, 0xb47dce3aU, + 0x1863df4aU, 0x82e51a31U, 0x60975133U, 0x4562537fU, 0xe0b16477U, + 0x84bb6baeU, 0x1cfe81a0U, 0x94f9082bU, 0x58704868U, 0x198f45fdU, + 0x8794de6cU, 0xb7527bf8U, 0x23ab73d3U, 0xe2724b02U, 0x57e31f8fU, + 0x2a6655abU, 0x07b2eb28U, 0x032fb5c2U, 0x9a86c57bU, 0xa5d33708U, + 0xf2302887U, 0xb223bfa5U, 0xba02036aU, 0x5ced1682U, 0x2b8acf1cU, + 0x92a779b4U, 0xf0f307f2U, 0xa14e69e2U, 0xcd65daf4U, 0xd50605beU, + 0x1fd13462U, 0x8ac4a6feU, 0x9d342e53U, 0xa0a2f355U, 0x32058ae1U, + 0x75a4f6ebU, 0x390b83ecU, 0xaa4060efU, 0x065e719fU, 0x51bd6e10U, + 0xf93e218aU, 0x3d96dd06U, 0xaedd3e05U, 0x464de6bdU, 0xb591548dU, + 0x0571c45dU, 0x6f0406d4U, 0xff605015U, 0x241998fbU, 0x97d6bde9U, + 0xcc894043U, 0x7767d99eU, 0xbdb0e842U, 0x8807898bU, 0x38e7195bU, + 0xdb79c8eeU, 0x47a17c0aU, 0xe97c420fU, 0xc9f8841eU, 0x00000000U, + 0x83098086U, 0x48322bedU, 0xac1e1170U, 0x4e6c5a72U, 0xfbfd0effU, + 0x560f8538U, 0x1e3daed5U, 0x27362d39U, 0x640a0fd9U, 0x21685ca6U, + 0xd19b5b54U, 0x3a24362eU, 0xb10c0a67U, 0x0f9357e7U, 0xd2b4ee96U, + 0x9e1b9b91U, 0x4f80c0c5U, 0xa261dc20U, 0x695a774bU, 0x161c121aU, + 0x0ae293baU, 0xe5c0a02aU, 0x433c22e0U, 0x1d121b17U, 0x0b0e090dU, + 0xadf28bc7U, 0xb92db6a8U, 0xc8141ea9U, 0x8557f119U, 0x4caf7507U, + 0xbbee99ddU, 0xfda37f60U, 0x9ff70126U, 0xbc5c72f5U, 0xc544663bU, + 0x345bfb7eU, 0x768b4329U, 0xdccb23c6U, 0x68b6edfcU, 0x63b8e4f1U, + 0xcad731dcU, 0x10426385U, 0x40139722U, 0x2084c611U, 0x7d854a24U, + 0xf8d2bb3dU, 0x11aef932U, 0x6dc729a1U, 0x4b1d9e2fU, 0xf3dcb230U, + 0xec0d8652U, 0xd077c1e3U, 0x6c2bb316U, 0x99a970b9U, 0xfa119448U, + 0x2247e964U, 0xc4a8fc8cU, 0x1aa0f03fU, 0xd8567d2cU, 0xef223390U, + 0xc787494eU, 0xc1d938d1U, 0xfe8ccaa2U, 0x3698d40bU, 0xcfa6f581U, + 0x28a57adeU, 0x26dab78eU, 0xa43fadbfU, 0xe42c3a9dU, 0x0d507892U, + 0x9b6a5fccU, 0x62547e46U, 0xc2f68d13U, 0xe890d8b8U, 0x5e2e39f7U, + 0xf582c3afU, 0xbe9f5d80U, 0x7c69d093U, 0xa96fd52dU, 0xb3cf2512U, + 0x3bc8ac99U, 0xa710187dU, 0x6ee89c63U, 0x7bdb3bbbU, 0x09cd2678U, + 0xf46e5918U, 0x01ec9ab7U, 0xa8834f9aU, 0x65e6956eU, 0x7eaaffe6U, + 0x0821bccfU, 0xe6ef15e8U, 0xd9bae79bU, 0xce4a6f36U, 0xd4ea9f09U, + 0xd629b07cU, 0xaf31a4b2U, 0x312a3f23U, 0x30c6a594U, 0xc035a266U, + 0x37744ebcU, 0xa6fc82caU, 0xb0e090d0U, 0x1533a7d8U, 0x4af10498U, + 0xf741ecdaU, 0x0e7fcd50U, 0x2f1791f6U, 0x8d764dd6U, 0x4d43efb0U, + 0x54ccaa4dU, 0xdfe49604U, 0xe39ed1b5U, 0x1b4c6a88U, 0xb8c12c1fU, + 0x7f466551U, 0x049d5eeaU, 0x5d018c35U, 0x73fa8774U, 0x2efb0b41U, + 0x5ab3671dU, 0x5292dbd2U, 0x33e91056U, 0x136dd647U, 0x8c9ad761U, + 0x7a37a10cU, 0x8e59f814U, 0x89eb133cU, 0xeecea927U, 0x35b761c9U, + 0xede11ce5U, 0x3c7a47b1U, 0x599cd2dfU, 0x3f55f273U, 0x791814ceU, + 0xbf73c737U, 0xea53f7cdU, 0x5b5ffdaaU, 0x14df3d6fU, 0x867844dbU, + 0x81caaff3U, 0x3eb968c4U, 0x2c382434U, 0x5fc2a340U, 0x72161dc3U, + 0x0cbce225U, 0x8b283c49U, 0x41ff0d95U, 0x7139a801U, 0xde080cb3U, + 0x9cd8b4e4U, 0x906456c1U, 0x617bcb84U, 0x70d532b6U, 0x74486c5cU, + 0x42d0b857U, }; + +static const uint32_t Td2[256] = { + 0xa75051f4U, 0x65537e41U, 0xa4c31a17U, 0x5e963a27U, 0x6bcb3babU, + 0x45f11f9dU, 0x58abacfaU, 0x03934be3U, 0xfa552030U, 0x6df6ad76U, + 0x769188ccU, 0x4c25f502U, 0xd7fc4fe5U, 0xcbd7c52aU, 0x44802635U, + 0xa38fb562U, 0x5a49deb1U, 0x1b6725baU, 0x0e9845eaU, 0xc0e15dfeU, + 0x7502c32fU, 0xf012814cU, 0x97a38d46U, 0xf9c66bd3U, 0x5fe7038fU, + 0x9c951592U, 0x7aebbf6dU, 0x59da9552U, 0x832dd4beU, 0x21d35874U, + 0x692949e0U, 0xc8448ec9U, 0x896a75c2U, 0x7978f48eU, 0x3e6b9958U, + 0x71dd27b9U, 0x4fb6bee1U, 0xad17f088U, 0xac66c920U, 0x3ab47dceU, + 0x4a1863dfU, 0x3182e51aU, 0x33609751U, 0x7f456253U, 0x77e0b164U, + 0xae84bb6bU, 0xa01cfe81U, 0x2b94f908U, 0x68587048U, 0xfd198f45U, + 0x6c8794deU, 0xf8b7527bU, 0xd323ab73U, 0x02e2724bU, 0x8f57e31fU, + 0xab2a6655U, 0x2807b2ebU, 0xc2032fb5U, 0x7b9a86c5U, 0x08a5d337U, + 0x87f23028U, 0xa5b223bfU, 0x6aba0203U, 0x825ced16U, 0x1c2b8acfU, + 0xb492a779U, 0xf2f0f307U, 0xe2a14e69U, 0xf4cd65daU, 0xbed50605U, + 0x621fd134U, 0xfe8ac4a6U, 0x539d342eU, 0x55a0a2f3U, 0xe132058aU, + 0xeb75a4f6U, 0xec390b83U, 0xefaa4060U, 0x9f065e71U, 0x1051bd6eU, + 0x8af93e21U, 0x063d96ddU, 0x05aedd3eU, 0xbd464de6U, 0x8db59154U, + 0x5d0571c4U, 0xd46f0406U, 0x15ff6050U, 0xfb241998U, 0xe997d6bdU, + 0x43cc8940U, 0x9e7767d9U, 0x42bdb0e8U, 0x8b880789U, 0x5b38e719U, + 0xeedb79c8U, 0x0a47a17cU, 0x0fe97c42U, 0x1ec9f884U, 0x00000000U, + 0x86830980U, 0xed48322bU, 0x70ac1e11U, 0x724e6c5aU, 0xfffbfd0eU, + 0x38560f85U, 0xd51e3daeU, 0x3927362dU, 0xd9640a0fU, 0xa621685cU, + 0x54d19b5bU, 0x2e3a2436U, 0x67b10c0aU, 0xe70f9357U, 0x96d2b4eeU, + 0x919e1b9bU, 0xc54f80c0U, 0x20a261dcU, 0x4b695a77U, 0x1a161c12U, + 0xba0ae293U, 0x2ae5c0a0U, 0xe0433c22U, 0x171d121bU, 0x0d0b0e09U, + 0xc7adf28bU, 0xa8b92db6U, 0xa9c8141eU, 0x198557f1U, 0x074caf75U, + 0xddbbee99U, 0x60fda37fU, 0x269ff701U, 0xf5bc5c72U, 0x3bc54466U, + 0x7e345bfbU, 0x29768b43U, 0xc6dccb23U, 0xfc68b6edU, 0xf163b8e4U, + 0xdccad731U, 0x85104263U, 0x22401397U, 0x112084c6U, 0x247d854aU, + 0x3df8d2bbU, 0x3211aef9U, 0xa16dc729U, 0x2f4b1d9eU, 0x30f3dcb2U, + 0x52ec0d86U, 0xe3d077c1U, 0x166c2bb3U, 0xb999a970U, 0x48fa1194U, + 0x642247e9U, 0x8cc4a8fcU, 0x3f1aa0f0U, 0x2cd8567dU, 0x90ef2233U, + 0x4ec78749U, 0xd1c1d938U, 0xa2fe8ccaU, 0x0b3698d4U, 0x81cfa6f5U, + 0xde28a57aU, 0x8e26dab7U, 0xbfa43fadU, 0x9de42c3aU, 0x920d5078U, + 0xcc9b6a5fU, 0x4662547eU, 0x13c2f68dU, 0xb8e890d8U, 0xf75e2e39U, + 0xaff582c3U, 0x80be9f5dU, 0x937c69d0U, 0x2da96fd5U, 0x12b3cf25U, + 0x993bc8acU, 0x7da71018U, 0x636ee89cU, 0xbb7bdb3bU, 0x7809cd26U, + 0x18f46e59U, 0xb701ec9aU, 0x9aa8834fU, 0x6e65e695U, 0xe67eaaffU, + 0xcf0821bcU, 0xe8e6ef15U, 0x9bd9bae7U, 0x36ce4a6fU, 0x09d4ea9fU, + 0x7cd629b0U, 0xb2af31a4U, 0x23312a3fU, 0x9430c6a5U, 0x66c035a2U, + 0xbc37744eU, 0xcaa6fc82U, 0xd0b0e090U, 0xd81533a7U, 0x984af104U, + 0xdaf741ecU, 0x500e7fcdU, 0xf62f1791U, 0xd68d764dU, 0xb04d43efU, + 0x4d54ccaaU, 0x04dfe496U, 0xb5e39ed1U, 0x881b4c6aU, 0x1fb8c12cU, + 0x517f4665U, 0xea049d5eU, 0x355d018cU, 0x7473fa87U, 0x412efb0bU, + 0x1d5ab367U, 0xd25292dbU, 0x5633e910U, 0x47136dd6U, 0x618c9ad7U, + 0x0c7a37a1U, 0x148e59f8U, 0x3c89eb13U, 0x27eecea9U, 0xc935b761U, + 0xe5ede11cU, 0xb13c7a47U, 0xdf599cd2U, 0x733f55f2U, 0xce791814U, + 0x37bf73c7U, 0xcdea53f7U, 0xaa5b5ffdU, 0x6f14df3dU, 0xdb867844U, + 0xf381caafU, 0xc43eb968U, 0x342c3824U, 0x405fc2a3U, 0xc372161dU, + 0x250cbce2U, 0x498b283cU, 0x9541ff0dU, 0x017139a8U, 0xb3de080cU, + 0xe49cd8b4U, 0xc1906456U, 0x84617bcbU, 0xb670d532U, 0x5c74486cU, + 0x5742d0b8U, }; + +static const uint32_t Td3[256] = { + 0xf4a75051U, 0x4165537eU, 0x17a4c31aU, 0x275e963aU, 0xab6bcb3bU, + 0x9d45f11fU, 0xfa58abacU, 0xe303934bU, 0x30fa5520U, 0x766df6adU, + 0xcc769188U, 0x024c25f5U, 0xe5d7fc4fU, 0x2acbd7c5U, 0x35448026U, + 0x62a38fb5U, 0xb15a49deU, 0xba1b6725U, 0xea0e9845U, 0xfec0e15dU, + 0x2f7502c3U, 0x4cf01281U, 0x4697a38dU, 0xd3f9c66bU, 0x8f5fe703U, + 0x929c9515U, 0x6d7aebbfU, 0x5259da95U, 0xbe832dd4U, 0x7421d358U, + 0xe0692949U, 0xc9c8448eU, 0xc2896a75U, 0x8e7978f4U, 0x583e6b99U, + 0xb971dd27U, 0xe14fb6beU, 0x88ad17f0U, 0x20ac66c9U, 0xce3ab47dU, + 0xdf4a1863U, 0x1a3182e5U, 0x51336097U, 0x537f4562U, 0x6477e0b1U, + 0x6bae84bbU, 0x81a01cfeU, 0x082b94f9U, 0x48685870U, 0x45fd198fU, + 0xde6c8794U, 0x7bf8b752U, 0x73d323abU, 0x4b02e272U, 0x1f8f57e3U, + 0x55ab2a66U, 0xeb2807b2U, 0xb5c2032fU, 0xc57b9a86U, 0x3708a5d3U, + 0x2887f230U, 0xbfa5b223U, 0x036aba02U, 0x16825cedU, 0xcf1c2b8aU, + 0x79b492a7U, 0x07f2f0f3U, 0x69e2a14eU, 0xdaf4cd65U, 0x05bed506U, + 0x34621fd1U, 0xa6fe8ac4U, 0x2e539d34U, 0xf355a0a2U, 0x8ae13205U, + 0xf6eb75a4U, 0x83ec390bU, 0x60efaa40U, 0x719f065eU, 0x6e1051bdU, + 0x218af93eU, 0xdd063d96U, 0x3e05aeddU, 0xe6bd464dU, 0x548db591U, + 0xc45d0571U, 0x06d46f04U, 0x5015ff60U, 0x98fb2419U, 0xbde997d6U, + 0x4043cc89U, 0xd99e7767U, 0xe842bdb0U, 0x898b8807U, 0x195b38e7U, + 0xc8eedb79U, 0x7c0a47a1U, 0x420fe97cU, 0x841ec9f8U, 0x00000000U, + 0x80868309U, 0x2bed4832U, 0x1170ac1eU, 0x5a724e6cU, 0x0efffbfdU, + 0x8538560fU, 0xaed51e3dU, 0x2d392736U, 0x0fd9640aU, 0x5ca62168U, + 0x5b54d19bU, 0x362e3a24U, 0x0a67b10cU, 0x57e70f93U, 0xee96d2b4U, + 0x9b919e1bU, 0xc0c54f80U, 0xdc20a261U, 0x774b695aU, 0x121a161cU, + 0x93ba0ae2U, 0xa02ae5c0U, 0x22e0433cU, 0x1b171d12U, 0x090d0b0eU, + 0x8bc7adf2U, 0xb6a8b92dU, 0x1ea9c814U, 0xf1198557U, 0x75074cafU, + 0x99ddbbeeU, 0x7f60fda3U, 0x01269ff7U, 0x72f5bc5cU, 0x663bc544U, + 0xfb7e345bU, 0x4329768bU, 0x23c6dccbU, 0xedfc68b6U, 0xe4f163b8U, + 0x31dccad7U, 0x63851042U, 0x97224013U, 0xc6112084U, 0x4a247d85U, + 0xbb3df8d2U, 0xf93211aeU, 0x29a16dc7U, 0x9e2f4b1dU, 0xb230f3dcU, + 0x8652ec0dU, 0xc1e3d077U, 0xb3166c2bU, 0x70b999a9U, 0x9448fa11U, + 0xe9642247U, 0xfc8cc4a8U, 0xf03f1aa0U, 0x7d2cd856U, 0x3390ef22U, + 0x494ec787U, 0x38d1c1d9U, 0xcaa2fe8cU, 0xd40b3698U, 0xf581cfa6U, + 0x7ade28a5U, 0xb78e26daU, 0xadbfa43fU, 0x3a9de42cU, 0x78920d50U, + 0x5fcc9b6aU, 0x7e466254U, 0x8d13c2f6U, 0xd8b8e890U, 0x39f75e2eU, + 0xc3aff582U, 0x5d80be9fU, 0xd0937c69U, 0xd52da96fU, 0x2512b3cfU, + 0xac993bc8U, 0x187da710U, 0x9c636ee8U, 0x3bbb7bdbU, 0x267809cdU, + 0x5918f46eU, 0x9ab701ecU, 0x4f9aa883U, 0x956e65e6U, 0xffe67eaaU, + 0xbccf0821U, 0x15e8e6efU, 0xe79bd9baU, 0x6f36ce4aU, 0x9f09d4eaU, + 0xb07cd629U, 0xa4b2af31U, 0x3f23312aU, 0xa59430c6U, 0xa266c035U, + 0x4ebc3774U, 0x82caa6fcU, 0x90d0b0e0U, 0xa7d81533U, 0x04984af1U, + 0xecdaf741U, 0xcd500e7fU, 0x91f62f17U, 0x4dd68d76U, 0xefb04d43U, + 0xaa4d54ccU, 0x9604dfe4U, 0xd1b5e39eU, 0x6a881b4cU, 0x2c1fb8c1U, + 0x65517f46U, 0x5eea049dU, 0x8c355d01U, 0x877473faU, 0x0b412efbU, + 0x671d5ab3U, 0xdbd25292U, 0x105633e9U, 0xd647136dU, 0xd7618c9aU, + 0xa10c7a37U, 0xf8148e59U, 0x133c89ebU, 0xa927eeceU, 0x61c935b7U, + 0x1ce5ede1U, 0x47b13c7aU, 0xd2df599cU, 0xf2733f55U, 0x14ce7918U, + 0xc737bf73U, 0xf7cdea53U, 0xfdaa5b5fU, 0x3d6f14dfU, 0x44db8678U, + 0xaff381caU, 0x68c43eb9U, 0x24342c38U, 0xa3405fc2U, 0x1dc37216U, + 0xe2250cbcU, 0x3c498b28U, 0x0d9541ffU, 0xa8017139U, 0x0cb3de08U, + 0xb4e49cd8U, 0x56c19064U, 0xcb84617bU, 0x32b670d5U, 0x6c5c7448U, + 0xb85742d0U, }; + +static const uint8_t Td4[256] = { + 0x52U, 0x09U, 0x6aU, 0xd5U, 0x30U, 0x36U, 0xa5U, 0x38U, 0xbfU, 0x40U, 0xa3U, + 0x9eU, 0x81U, 0xf3U, 0xd7U, 0xfbU, 0x7cU, 0xe3U, 0x39U, 0x82U, 0x9bU, 0x2fU, + 0xffU, 0x87U, 0x34U, 0x8eU, 0x43U, 0x44U, 0xc4U, 0xdeU, 0xe9U, 0xcbU, 0x54U, + 0x7bU, 0x94U, 0x32U, 0xa6U, 0xc2U, 0x23U, 0x3dU, 0xeeU, 0x4cU, 0x95U, 0x0bU, + 0x42U, 0xfaU, 0xc3U, 0x4eU, 0x08U, 0x2eU, 0xa1U, 0x66U, 0x28U, 0xd9U, 0x24U, + 0xb2U, 0x76U, 0x5bU, 0xa2U, 0x49U, 0x6dU, 0x8bU, 0xd1U, 0x25U, 0x72U, 0xf8U, + 0xf6U, 0x64U, 0x86U, 0x68U, 0x98U, 0x16U, 0xd4U, 0xa4U, 0x5cU, 0xccU, 0x5dU, + 0x65U, 0xb6U, 0x92U, 0x6cU, 0x70U, 0x48U, 0x50U, 0xfdU, 0xedU, 0xb9U, 0xdaU, + 0x5eU, 0x15U, 0x46U, 0x57U, 0xa7U, 0x8dU, 0x9dU, 0x84U, 0x90U, 0xd8U, 0xabU, + 0x00U, 0x8cU, 0xbcU, 0xd3U, 0x0aU, 0xf7U, 0xe4U, 0x58U, 0x05U, 0xb8U, 0xb3U, + 0x45U, 0x06U, 0xd0U, 0x2cU, 0x1eU, 0x8fU, 0xcaU, 0x3fU, 0x0fU, 0x02U, 0xc1U, + 0xafU, 0xbdU, 0x03U, 0x01U, 0x13U, 0x8aU, 0x6bU, 0x3aU, 0x91U, 0x11U, 0x41U, + 0x4fU, 0x67U, 0xdcU, 0xeaU, 0x97U, 0xf2U, 0xcfU, 0xceU, 0xf0U, 0xb4U, 0xe6U, + 0x73U, 0x96U, 0xacU, 0x74U, 0x22U, 0xe7U, 0xadU, 0x35U, 0x85U, 0xe2U, 0xf9U, + 0x37U, 0xe8U, 0x1cU, 0x75U, 0xdfU, 0x6eU, 0x47U, 0xf1U, 0x1aU, 0x71U, 0x1dU, + 0x29U, 0xc5U, 0x89U, 0x6fU, 0xb7U, 0x62U, 0x0eU, 0xaaU, 0x18U, 0xbeU, 0x1bU, + 0xfcU, 0x56U, 0x3eU, 0x4bU, 0xc6U, 0xd2U, 0x79U, 0x20U, 0x9aU, 0xdbU, 0xc0U, + 0xfeU, 0x78U, 0xcdU, 0x5aU, 0xf4U, 0x1fU, 0xddU, 0xa8U, 0x33U, 0x88U, 0x07U, + 0xc7U, 0x31U, 0xb1U, 0x12U, 0x10U, 0x59U, 0x27U, 0x80U, 0xecU, 0x5fU, 0x60U, + 0x51U, 0x7fU, 0xa9U, 0x19U, 0xb5U, 0x4aU, 0x0dU, 0x2dU, 0xe5U, 0x7aU, 0x9fU, + 0x93U, 0xc9U, 0x9cU, 0xefU, 0xa0U, 0xe0U, 0x3bU, 0x4dU, 0xaeU, 0x2aU, 0xf5U, + 0xb0U, 0xc8U, 0xebU, 0xbbU, 0x3cU, 0x83U, 0x53U, 0x99U, 0x61U, 0x17U, 0x2bU, + 0x04U, 0x7eU, 0xbaU, 0x77U, 0xd6U, 0x26U, 0xe1U, 0x69U, 0x14U, 0x63U, 0x55U, + 0x21U, 0x0cU, 0x7dU, }; + +static const uint32_t rcon[] = { + 0x01000000, 0x02000000, 0x04000000, 0x08000000, 0x10000000, + 0x20000000, 0x40000000, 0x80000000, 0x1B000000, 0x36000000, + // for 128-bit blocks, Rijndael never uses more than 10 rcon values +}; + +int aes_nohw_set_encrypt_key(const uint8_t *key, unsigned bits, + AES_KEY *aeskey) { + uint32_t *rk; + int i = 0; + uint32_t temp; + + if (!key || !aeskey) { + return -1; + } + + switch (bits) { + case 128: + aeskey->rounds = 10; + break; + case 192: + aeskey->rounds = 12; + break; + case 256: + aeskey->rounds = 14; + break; + default: + return -2; + } + + rk = aeskey->rd_key; + + rk[0] = GETU32(key); + rk[1] = GETU32(key + 4); + rk[2] = GETU32(key + 8); + rk[3] = GETU32(key + 12); + if (bits == 128) { + while (1) { + temp = rk[3]; + rk[4] = rk[0] ^ (Te2[(temp >> 16) & 0xff] & 0xff000000) ^ + (Te3[(temp >> 8) & 0xff] & 0x00ff0000) ^ + (Te0[(temp) & 0xff] & 0x0000ff00) ^ + (Te1[(temp >> 24)] & 0x000000ff) ^ rcon[i]; + rk[5] = rk[1] ^ rk[4]; + rk[6] = rk[2] ^ rk[5]; + rk[7] = rk[3] ^ rk[6]; + if (++i == 10) { + return 0; + } + rk += 4; + } + } + rk[4] = GETU32(key + 16); + rk[5] = GETU32(key + 20); + if (bits == 192) { + while (1) { + temp = rk[5]; + rk[6] = rk[0] ^ (Te2[(temp >> 16) & 0xff] & 0xff000000) ^ + (Te3[(temp >> 8) & 0xff] & 0x00ff0000) ^ + (Te0[(temp) & 0xff] & 0x0000ff00) ^ + (Te1[(temp >> 24)] & 0x000000ff) ^ rcon[i]; + rk[7] = rk[1] ^ rk[6]; + rk[8] = rk[2] ^ rk[7]; + rk[9] = rk[3] ^ rk[8]; + if (++i == 8) { + return 0; + } + rk[10] = rk[4] ^ rk[9]; + rk[11] = rk[5] ^ rk[10]; + rk += 6; + } + } + rk[6] = GETU32(key + 24); + rk[7] = GETU32(key + 28); + if (bits == 256) { + while (1) { + temp = rk[7]; + rk[8] = rk[0] ^ (Te2[(temp >> 16) & 0xff] & 0xff000000) ^ + (Te3[(temp >> 8) & 0xff] & 0x00ff0000) ^ + (Te0[(temp) & 0xff] & 0x0000ff00) ^ + (Te1[(temp >> 24)] & 0x000000ff) ^ rcon[i]; + rk[9] = rk[1] ^ rk[8]; + rk[10] = rk[2] ^ rk[9]; + rk[11] = rk[3] ^ rk[10]; + if (++i == 7) { + return 0; + } + temp = rk[11]; + rk[12] = rk[4] ^ (Te2[(temp >> 24)] & 0xff000000) ^ + (Te3[(temp >> 16) & 0xff] & 0x00ff0000) ^ + (Te0[(temp >> 8) & 0xff] & 0x0000ff00) ^ + (Te1[(temp) & 0xff] & 0x000000ff); + rk[13] = rk[5] ^ rk[12]; + rk[14] = rk[6] ^ rk[13]; + rk[15] = rk[7] ^ rk[14]; + + rk += 8; + } + } + return 0; +} + +int aes_nohw_set_decrypt_key(const uint8_t *key, unsigned bits, + AES_KEY *aeskey) { + uint32_t *rk; + int i, j, status; + uint32_t temp; + + // first, start with an encryption schedule + status = AES_set_encrypt_key(key, bits, aeskey); + if (status < 0) { + return status; + } + + rk = aeskey->rd_key; + + // invert the order of the round keys: + for (i = 0, j = 4 * aeskey->rounds; i < j; i += 4, j -= 4) { + temp = rk[i]; + rk[i] = rk[j]; + rk[j] = temp; + temp = rk[i + 1]; + rk[i + 1] = rk[j + 1]; + rk[j + 1] = temp; + temp = rk[i + 2]; + rk[i + 2] = rk[j + 2]; + rk[j + 2] = temp; + temp = rk[i + 3]; + rk[i + 3] = rk[j + 3]; + rk[j + 3] = temp; + } + // apply the inverse MixColumn transform to all round keys but the first and + // the last: + for (i = 1; i < (int)aeskey->rounds; i++) { + rk += 4; + rk[0] = + Td0[Te1[(rk[0] >> 24)] & 0xff] ^ Td1[Te1[(rk[0] >> 16) & 0xff] & 0xff] ^ + Td2[Te1[(rk[0] >> 8) & 0xff] & 0xff] ^ Td3[Te1[(rk[0]) & 0xff] & 0xff]; + rk[1] = + Td0[Te1[(rk[1] >> 24)] & 0xff] ^ Td1[Te1[(rk[1] >> 16) & 0xff] & 0xff] ^ + Td2[Te1[(rk[1] >> 8) & 0xff] & 0xff] ^ Td3[Te1[(rk[1]) & 0xff] & 0xff]; + rk[2] = + Td0[Te1[(rk[2] >> 24)] & 0xff] ^ Td1[Te1[(rk[2] >> 16) & 0xff] & 0xff] ^ + Td2[Te1[(rk[2] >> 8) & 0xff] & 0xff] ^ Td3[Te1[(rk[2]) & 0xff] & 0xff]; + rk[3] = + Td0[Te1[(rk[3] >> 24)] & 0xff] ^ Td1[Te1[(rk[3] >> 16) & 0xff] & 0xff] ^ + Td2[Te1[(rk[3] >> 8) & 0xff] & 0xff] ^ Td3[Te1[(rk[3]) & 0xff] & 0xff]; + } + return 0; +} + +void aes_nohw_encrypt(const uint8_t *in, uint8_t *out, + const AES_KEY *key) { + const uint32_t *rk; + uint32_t s0, s1, s2, s3, t0, t1, t2, t3; + int r; + + rk = key->rd_key; + + // map byte array block to cipher state + // and add initial round key: + s0 = GETU32(in) ^ rk[0]; + s1 = GETU32(in + 4) ^ rk[1]; + s2 = GETU32(in + 8) ^ rk[2]; + s3 = GETU32(in + 12) ^ rk[3]; + + // Nr - 1 full rounds: + r = key->rounds >> 1; + for (;;) { + t0 = Te0[(s0 >> 24)] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ + Te3[(s3) & 0xff] ^ rk[4]; + t1 = Te0[(s1 >> 24)] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ + Te3[(s0) & 0xff] ^ rk[5]; + t2 = Te0[(s2 >> 24)] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ + Te3[(s1) & 0xff] ^ rk[6]; + t3 = Te0[(s3 >> 24)] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ + Te3[(s2) & 0xff] ^ rk[7]; + + rk += 8; + if (--r == 0) { + break; + } + + s0 = Te0[(t0 >> 24)] ^ Te1[(t1 >> 16) & 0xff] ^ Te2[(t2 >> 8) & 0xff] ^ + Te3[(t3) & 0xff] ^ rk[0]; + s1 = Te0[(t1 >> 24)] ^ Te1[(t2 >> 16) & 0xff] ^ Te2[(t3 >> 8) & 0xff] ^ + Te3[(t0) & 0xff] ^ rk[1]; + s2 = Te0[(t2 >> 24)] ^ Te1[(t3 >> 16) & 0xff] ^ Te2[(t0 >> 8) & 0xff] ^ + Te3[(t1) & 0xff] ^ rk[2]; + s3 = Te0[(t3 >> 24)] ^ Te1[(t0 >> 16) & 0xff] ^ Te2[(t1 >> 8) & 0xff] ^ + Te3[(t2) & 0xff] ^ rk[3]; + } + + // apply last round and map cipher state to byte array block: + s0 = (Te2[(t0 >> 24)] & 0xff000000) ^ (Te3[(t1 >> 16) & 0xff] & 0x00ff0000) ^ + (Te0[(t2 >> 8) & 0xff] & 0x0000ff00) ^ (Te1[(t3) & 0xff] & 0x000000ff) ^ + rk[0]; + PUTU32(out, s0); + s1 = (Te2[(t1 >> 24)] & 0xff000000) ^ (Te3[(t2 >> 16) & 0xff] & 0x00ff0000) ^ + (Te0[(t3 >> 8) & 0xff] & 0x0000ff00) ^ (Te1[(t0) & 0xff] & 0x000000ff) ^ + rk[1]; + PUTU32(out + 4, s1); + s2 = (Te2[(t2 >> 24)] & 0xff000000) ^ (Te3[(t3 >> 16) & 0xff] & 0x00ff0000) ^ + (Te0[(t0 >> 8) & 0xff] & 0x0000ff00) ^ (Te1[(t1) & 0xff] & 0x000000ff) ^ + rk[2]; + PUTU32(out + 8, s2); + s3 = (Te2[(t3 >> 24)] & 0xff000000) ^ (Te3[(t0 >> 16) & 0xff] & 0x00ff0000) ^ + (Te0[(t1 >> 8) & 0xff] & 0x0000ff00) ^ (Te1[(t2) & 0xff] & 0x000000ff) ^ + rk[3]; + PUTU32(out + 12, s3); +} + +void aes_nohw_decrypt(const uint8_t *in, uint8_t *out, + const AES_KEY *key) { + const uint32_t *rk; + uint32_t s0, s1, s2, s3, t0, t1, t2, t3; + int r; + + rk = key->rd_key; + + // map byte array block to cipher state + // and add initial round key: + s0 = GETU32(in) ^ rk[0]; + s1 = GETU32(in + 4) ^ rk[1]; + s2 = GETU32(in + 8) ^ rk[2]; + s3 = GETU32(in + 12) ^ rk[3]; + + // Nr - 1 full rounds: + r = key->rounds >> 1; + for (;;) { + t0 = Td0[(s0 >> 24)] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ + Td3[(s1) & 0xff] ^ rk[4]; + t1 = Td0[(s1 >> 24)] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ + Td3[(s2) & 0xff] ^ rk[5]; + t2 = Td0[(s2 >> 24)] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ + Td3[(s3) & 0xff] ^ rk[6]; + t3 = Td0[(s3 >> 24)] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ + Td3[(s0) & 0xff] ^ rk[7]; + + rk += 8; + if (--r == 0) { + break; + } + + s0 = Td0[(t0 >> 24)] ^ Td1[(t3 >> 16) & 0xff] ^ Td2[(t2 >> 8) & 0xff] ^ + Td3[(t1) & 0xff] ^ rk[0]; + s1 = Td0[(t1 >> 24)] ^ Td1[(t0 >> 16) & 0xff] ^ Td2[(t3 >> 8) & 0xff] ^ + Td3[(t2) & 0xff] ^ rk[1]; + s2 = Td0[(t2 >> 24)] ^ Td1[(t1 >> 16) & 0xff] ^ Td2[(t0 >> 8) & 0xff] ^ + Td3[(t3) & 0xff] ^ rk[2]; + s3 = Td0[(t3 >> 24)] ^ Td1[(t2 >> 16) & 0xff] ^ Td2[(t1 >> 8) & 0xff] ^ + Td3[(t0) & 0xff] ^ rk[3]; + } + + // apply last round and + // map cipher state to byte array block: + s0 = ((uint32_t)Td4[(t0 >> 24)] << 24) ^ + ((uint32_t)Td4[(t3 >> 16) & 0xff] << 16) ^ + ((uint32_t)Td4[(t2 >> 8) & 0xff] << 8) ^ + ((uint32_t)Td4[(t1) & 0xff]) ^ rk[0]; + PUTU32(out, s0); + s1 = ((uint32_t)Td4[(t1 >> 24)] << 24) ^ + ((uint32_t)Td4[(t0 >> 16) & 0xff] << 16) ^ + ((uint32_t)Td4[(t3 >> 8) & 0xff] << 8) ^ + ((uint32_t)Td4[(t2) & 0xff]) ^ rk[1]; + PUTU32(out + 4, s1); + s2 = ((uint32_t)Td4[(t2 >> 24)] << 24) ^ + ((uint32_t)Td4[(t1 >> 16) & 0xff] << 16) ^ + ((uint32_t)Td4[(t0 >> 8) & 0xff] << 8) ^ + ((uint32_t)Td4[(t3) & 0xff]) ^ rk[2]; + PUTU32(out + 8, s2); + s3 = ((uint32_t)Td4[(t3 >> 24)] << 24) ^ + ((uint32_t)Td4[(t2 >> 16) & 0xff] << 16) ^ + ((uint32_t)Td4[(t1 >> 8) & 0xff] << 8) ^ + ((uint32_t)Td4[(t0) & 0xff]) ^ rk[3]; + PUTU32(out + 12, s3); +} diff --git a/third_party/boringssl/core/cortex-m/aes.S b/third_party/boringssl/core/cortex-m/aes.S new file mode 100644 index 0000000000..10d3cdabba --- /dev/null +++ b/third_party/boringssl/core/cortex-m/aes.S @@ -0,0 +1,1189 @@ +@ Copyright 2007-2016 The OpenSSL Project Authors. All Rights Reserved. +@ +@ Licensed under the OpenSSL license (the "License"). You may not use +@ this file except in compliance with the License. You can obtain a copy +@ in the file LICENSE in the source distribution or at +@ https://www.openssl.org/source/license.html + + +@ ==================================================================== +@ Written by Andy Polyakov <appro@openssl.org> for the OpenSSL +@ project. The module is, however, dual licensed under OpenSSL and +@ CRYPTOGAMS licenses depending on where you obtain it. For further +@ details see http://www.openssl.org/~appro/cryptogams/. +@ ==================================================================== + +@ AES for ARMv4 + +@ January 2007. +@ +@ Code uses single 1K S-box and is >2 times faster than code generated +@ by gcc-3.4.1. This is thanks to unique feature of ARMv4 ISA, which +@ allows to merge logical or arithmetic operation with shift or rotate +@ in one instruction and emit combined result every cycle. The module +@ is endian-neutral. The performance is ~42 cycles/byte for 128-bit +@ key [on single-issue Xscale PXA250 core]. + +@ May 2007. +@ +@ AES_set_[en|de]crypt_key is added. + +@ July 2010. +@ +@ Rescheduling for dual-issue pipeline resulted in 12% improvement on +@ Cortex A8 core and ~25 cycles per byte processed with 128-bit key. + +@ February 2011. +@ +@ Profiler-assisted and platform-specific optimization resulted in 16% +@ improvement on Cortex A8 core and ~21.5 cycles per byte. + +#define __ARM_ARCH__ 7 + +.text +#if defined(__thumb2__) && !defined(__APPLE__) +.syntax unified +.thumb +#else +.code 32 +#undef __thumb2__ +#endif + +.type AES_Te,%object +.align 5 +AES_Te: +.word 0xc66363a5, 0xf87c7c84, 0xee777799, 0xf67b7b8d +.word 0xfff2f20d, 0xd66b6bbd, 0xde6f6fb1, 0x91c5c554 +.word 0x60303050, 0x02010103, 0xce6767a9, 0x562b2b7d +.word 0xe7fefe19, 0xb5d7d762, 0x4dababe6, 0xec76769a +.word 0x8fcaca45, 0x1f82829d, 0x89c9c940, 0xfa7d7d87 +.word 0xeffafa15, 0xb25959eb, 0x8e4747c9, 0xfbf0f00b +.word 0x41adadec, 0xb3d4d467, 0x5fa2a2fd, 0x45afafea +.word 0x239c9cbf, 0x53a4a4f7, 0xe4727296, 0x9bc0c05b +.word 0x75b7b7c2, 0xe1fdfd1c, 0x3d9393ae, 0x4c26266a +.word 0x6c36365a, 0x7e3f3f41, 0xf5f7f702, 0x83cccc4f +.word 0x6834345c, 0x51a5a5f4, 0xd1e5e534, 0xf9f1f108 +.word 0xe2717193, 0xabd8d873, 0x62313153, 0x2a15153f +.word 0x0804040c, 0x95c7c752, 0x46232365, 0x9dc3c35e +.word 0x30181828, 0x379696a1, 0x0a05050f, 0x2f9a9ab5 +.word 0x0e070709, 0x24121236, 0x1b80809b, 0xdfe2e23d +.word 0xcdebeb26, 0x4e272769, 0x7fb2b2cd, 0xea75759f +.word 0x1209091b, 0x1d83839e, 0x582c2c74, 0x341a1a2e +.word 0x361b1b2d, 0xdc6e6eb2, 0xb45a5aee, 0x5ba0a0fb +.word 0xa45252f6, 0x763b3b4d, 0xb7d6d661, 0x7db3b3ce +.word 0x5229297b, 0xdde3e33e, 0x5e2f2f71, 0x13848497 +.word 0xa65353f5, 0xb9d1d168, 0x00000000, 0xc1eded2c +.word 0x40202060, 0xe3fcfc1f, 0x79b1b1c8, 0xb65b5bed +.word 0xd46a6abe, 0x8dcbcb46, 0x67bebed9, 0x7239394b +.word 0x944a4ade, 0x984c4cd4, 0xb05858e8, 0x85cfcf4a +.word 0xbbd0d06b, 0xc5efef2a, 0x4faaaae5, 0xedfbfb16 +.word 0x864343c5, 0x9a4d4dd7, 0x66333355, 0x11858594 +.word 0x8a4545cf, 0xe9f9f910, 0x04020206, 0xfe7f7f81 +.word 0xa05050f0, 0x783c3c44, 0x259f9fba, 0x4ba8a8e3 +.word 0xa25151f3, 0x5da3a3fe, 0x804040c0, 0x058f8f8a +.word 0x3f9292ad, 0x219d9dbc, 0x70383848, 0xf1f5f504 +.word 0x63bcbcdf, 0x77b6b6c1, 0xafdada75, 0x42212163 +.word 0x20101030, 0xe5ffff1a, 0xfdf3f30e, 0xbfd2d26d +.word 0x81cdcd4c, 0x180c0c14, 0x26131335, 0xc3ecec2f +.word 0xbe5f5fe1, 0x359797a2, 0x884444cc, 0x2e171739 +.word 0x93c4c457, 0x55a7a7f2, 0xfc7e7e82, 0x7a3d3d47 +.word 0xc86464ac, 0xba5d5de7, 0x3219192b, 0xe6737395 +.word 0xc06060a0, 0x19818198, 0x9e4f4fd1, 0xa3dcdc7f +.word 0x44222266, 0x542a2a7e, 0x3b9090ab, 0x0b888883 +.word 0x8c4646ca, 0xc7eeee29, 0x6bb8b8d3, 0x2814143c +.word 0xa7dede79, 0xbc5e5ee2, 0x160b0b1d, 0xaddbdb76 +.word 0xdbe0e03b, 0x64323256, 0x743a3a4e, 0x140a0a1e +.word 0x924949db, 0x0c06060a, 0x4824246c, 0xb85c5ce4 +.word 0x9fc2c25d, 0xbdd3d36e, 0x43acacef, 0xc46262a6 +.word 0x399191a8, 0x319595a4, 0xd3e4e437, 0xf279798b +.word 0xd5e7e732, 0x8bc8c843, 0x6e373759, 0xda6d6db7 +.word 0x018d8d8c, 0xb1d5d564, 0x9c4e4ed2, 0x49a9a9e0 +.word 0xd86c6cb4, 0xac5656fa, 0xf3f4f407, 0xcfeaea25 +.word 0xca6565af, 0xf47a7a8e, 0x47aeaee9, 0x10080818 +.word 0x6fbabad5, 0xf0787888, 0x4a25256f, 0x5c2e2e72 +.word 0x381c1c24, 0x57a6a6f1, 0x73b4b4c7, 0x97c6c651 +.word 0xcbe8e823, 0xa1dddd7c, 0xe874749c, 0x3e1f1f21 +.word 0x964b4bdd, 0x61bdbddc, 0x0d8b8b86, 0x0f8a8a85 +.word 0xe0707090, 0x7c3e3e42, 0x71b5b5c4, 0xcc6666aa +.word 0x904848d8, 0x06030305, 0xf7f6f601, 0x1c0e0e12 +.word 0xc26161a3, 0x6a35355f, 0xae5757f9, 0x69b9b9d0 +.word 0x17868691, 0x99c1c158, 0x3a1d1d27, 0x279e9eb9 +.word 0xd9e1e138, 0xebf8f813, 0x2b9898b3, 0x22111133 +.word 0xd26969bb, 0xa9d9d970, 0x078e8e89, 0x339494a7 +.word 0x2d9b9bb6, 0x3c1e1e22, 0x15878792, 0xc9e9e920 +.word 0x87cece49, 0xaa5555ff, 0x50282878, 0xa5dfdf7a +.word 0x038c8c8f, 0x59a1a1f8, 0x09898980, 0x1a0d0d17 +.word 0x65bfbfda, 0xd7e6e631, 0x844242c6, 0xd06868b8 +.word 0x824141c3, 0x299999b0, 0x5a2d2d77, 0x1e0f0f11 +.word 0x7bb0b0cb, 0xa85454fc, 0x6dbbbbd6, 0x2c16163a +@ Te4[256] +.byte 0x63, 0x7c, 0x77, 0x7b, 0xf2, 0x6b, 0x6f, 0xc5 +.byte 0x30, 0x01, 0x67, 0x2b, 0xfe, 0xd7, 0xab, 0x76 +.byte 0xca, 0x82, 0xc9, 0x7d, 0xfa, 0x59, 0x47, 0xf0 +.byte 0xad, 0xd4, 0xa2, 0xaf, 0x9c, 0xa4, 0x72, 0xc0 +.byte 0xb7, 0xfd, 0x93, 0x26, 0x36, 0x3f, 0xf7, 0xcc +.byte 0x34, 0xa5, 0xe5, 0xf1, 0x71, 0xd8, 0x31, 0x15 +.byte 0x04, 0xc7, 0x23, 0xc3, 0x18, 0x96, 0x05, 0x9a +.byte 0x07, 0x12, 0x80, 0xe2, 0xeb, 0x27, 0xb2, 0x75 +.byte 0x09, 0x83, 0x2c, 0x1a, 0x1b, 0x6e, 0x5a, 0xa0 +.byte 0x52, 0x3b, 0xd6, 0xb3, 0x29, 0xe3, 0x2f, 0x84 +.byte 0x53, 0xd1, 0x00, 0xed, 0x20, 0xfc, 0xb1, 0x5b +.byte 0x6a, 0xcb, 0xbe, 0x39, 0x4a, 0x4c, 0x58, 0xcf +.byte 0xd0, 0xef, 0xaa, 0xfb, 0x43, 0x4d, 0x33, 0x85 +.byte 0x45, 0xf9, 0x02, 0x7f, 0x50, 0x3c, 0x9f, 0xa8 +.byte 0x51, 0xa3, 0x40, 0x8f, 0x92, 0x9d, 0x38, 0xf5 +.byte 0xbc, 0xb6, 0xda, 0x21, 0x10, 0xff, 0xf3, 0xd2 +.byte 0xcd, 0x0c, 0x13, 0xec, 0x5f, 0x97, 0x44, 0x17 +.byte 0xc4, 0xa7, 0x7e, 0x3d, 0x64, 0x5d, 0x19, 0x73 +.byte 0x60, 0x81, 0x4f, 0xdc, 0x22, 0x2a, 0x90, 0x88 +.byte 0x46, 0xee, 0xb8, 0x14, 0xde, 0x5e, 0x0b, 0xdb +.byte 0xe0, 0x32, 0x3a, 0x0a, 0x49, 0x06, 0x24, 0x5c +.byte 0xc2, 0xd3, 0xac, 0x62, 0x91, 0x95, 0xe4, 0x79 +.byte 0xe7, 0xc8, 0x37, 0x6d, 0x8d, 0xd5, 0x4e, 0xa9 +.byte 0x6c, 0x56, 0xf4, 0xea, 0x65, 0x7a, 0xae, 0x08 +.byte 0xba, 0x78, 0x25, 0x2e, 0x1c, 0xa6, 0xb4, 0xc6 +.byte 0xe8, 0xdd, 0x74, 0x1f, 0x4b, 0xbd, 0x8b, 0x8a +.byte 0x70, 0x3e, 0xb5, 0x66, 0x48, 0x03, 0xf6, 0x0e +.byte 0x61, 0x35, 0x57, 0xb9, 0x86, 0xc1, 0x1d, 0x9e +.byte 0xe1, 0xf8, 0x98, 0x11, 0x69, 0xd9, 0x8e, 0x94 +.byte 0x9b, 0x1e, 0x87, 0xe9, 0xce, 0x55, 0x28, 0xdf +.byte 0x8c, 0xa1, 0x89, 0x0d, 0xbf, 0xe6, 0x42, 0x68 +.byte 0x41, 0x99, 0x2d, 0x0f, 0xb0, 0x54, 0xbb, 0x16 +@ rcon[] +.word 0x01000000, 0x02000000, 0x04000000, 0x08000000 +.word 0x10000000, 0x20000000, 0x40000000, 0x80000000 +.word 0x1B000000, 0x36000000, 0, 0, 0, 0, 0, 0 +.size AES_Te,.-AES_Te + +@ void aes_nohw_encrypt(const unsigned char *in, unsigned char *out, +@ const AES_KEY *key) { +.global aes_nohw_encrypt +.type aes_nohw_encrypt,%function +.align 5 +aes_nohw_encrypt: +#ifndef __thumb2__ + sub r3,pc,#8 @ aes_nohw_encrypt +#else + adr r3,. +#endif + stmdb sp!,{r1,r4-r12,lr} +#if defined(__thumb2__) || defined(__APPLE__) + adr r10,AES_Te +#else + sub r10,r3,#aes_nohw_encrypt-AES_Te @ Te +#endif + mov r12,r0 @ inp + mov r11,r2 +#if __ARM_ARCH__<7 + ldrb r0,[r12,#3] @ load input data in endian-neutral + ldrb r4,[r12,#2] @ manner... + ldrb r5,[r12,#1] + ldrb r6,[r12,#0] + orr r0,r0,r4,lsl#8 + ldrb r1,[r12,#7] + orr r0,r0,r5,lsl#16 + ldrb r4,[r12,#6] + orr r0,r0,r6,lsl#24 + ldrb r5,[r12,#5] + ldrb r6,[r12,#4] + orr r1,r1,r4,lsl#8 + ldrb r2,[r12,#11] + orr r1,r1,r5,lsl#16 + ldrb r4,[r12,#10] + orr r1,r1,r6,lsl#24 + ldrb r5,[r12,#9] + ldrb r6,[r12,#8] + orr r2,r2,r4,lsl#8 + ldrb r3,[r12,#15] + orr r2,r2,r5,lsl#16 + ldrb r4,[r12,#14] + orr r2,r2,r6,lsl#24 + ldrb r5,[r12,#13] + ldrb r6,[r12,#12] + orr r3,r3,r4,lsl#8 + orr r3,r3,r5,lsl#16 + orr r3,r3,r6,lsl#24 +#else + ldr r0,[r12,#0] + ldr r1,[r12,#4] + ldr r2,[r12,#8] + ldr r3,[r12,#12] +#ifdef __ARMEL__ + rev r0,r0 + rev r1,r1 + rev r2,r2 + rev r3,r3 +#endif +#endif + bl _armv4_AES_encrypt + + ldr r12,[sp],#4 @ pop out +#if __ARM_ARCH__>=7 +#ifdef __ARMEL__ + rev r0,r0 + rev r1,r1 + rev r2,r2 + rev r3,r3 +#endif + str r0,[r12,#0] + str r1,[r12,#4] + str r2,[r12,#8] + str r3,[r12,#12] +#else + mov r4,r0,lsr#24 @ write output in endian-neutral + mov r5,r0,lsr#16 @ manner... + mov r6,r0,lsr#8 + strb r4,[r12,#0] + strb r5,[r12,#1] + mov r4,r1,lsr#24 + strb r6,[r12,#2] + mov r5,r1,lsr#16 + strb r0,[r12,#3] + mov r6,r1,lsr#8 + strb r4,[r12,#4] + strb r5,[r12,#5] + mov r4,r2,lsr#24 + strb r6,[r12,#6] + mov r5,r2,lsr#16 + strb r1,[r12,#7] + mov r6,r2,lsr#8 + strb r4,[r12,#8] + strb r5,[r12,#9] + mov r4,r3,lsr#24 + strb r6,[r12,#10] + mov r5,r3,lsr#16 + strb r2,[r12,#11] + mov r6,r3,lsr#8 + strb r4,[r12,#12] + strb r5,[r12,#13] + strb r6,[r12,#14] + strb r3,[r12,#15] +#endif +#if __ARM_ARCH__>=5 + ldmia sp!,{r4-r12,pc} +#else + ldmia sp!,{r4-r12,lr} + tst lr,#1 + moveq pc,lr @ be binary compatible with V4, yet + .word 0xe12fff1e @ interoperable with Thumb ISA:-) +#endif +.size aes_nohw_encrypt,.-aes_nohw_encrypt + +.type _armv4_AES_encrypt,%function +.align 2 +_armv4_AES_encrypt: + str lr,[sp,#-4]! @ push lr + ldmia r11!,{r4-r7} + eor r0,r0,r4 + ldr r12,[r11,#240-16] + eor r1,r1,r5 + eor r2,r2,r6 + eor r3,r3,r7 + sub r12,r12,#1 + mov lr,#255 + + and r7,lr,r0 + and r8,lr,r0,lsr#8 + and r9,lr,r0,lsr#16 + mov r0,r0,lsr#24 +.Lenc_loop: + ldr r4,[r10,r7,lsl#2] @ Te3[s0>>0] + and r7,lr,r1,lsr#16 @ i0 + ldr r5,[r10,r8,lsl#2] @ Te2[s0>>8] + and r8,lr,r1 + ldr r6,[r10,r9,lsl#2] @ Te1[s0>>16] + and r9,lr,r1,lsr#8 + ldr r0,[r10,r0,lsl#2] @ Te0[s0>>24] + mov r1,r1,lsr#24 + + ldr r7,[r10,r7,lsl#2] @ Te1[s1>>16] + ldr r8,[r10,r8,lsl#2] @ Te3[s1>>0] + ldr r9,[r10,r9,lsl#2] @ Te2[s1>>8] + eor r0,r0,r7,ror#8 + ldr r1,[r10,r1,lsl#2] @ Te0[s1>>24] + and r7,lr,r2,lsr#8 @ i0 + eor r5,r5,r8,ror#8 + and r8,lr,r2,lsr#16 @ i1 + eor r6,r6,r9,ror#8 + and r9,lr,r2 + ldr r7,[r10,r7,lsl#2] @ Te2[s2>>8] + eor r1,r1,r4,ror#24 + ldr r8,[r10,r8,lsl#2] @ Te1[s2>>16] + mov r2,r2,lsr#24 + + ldr r9,[r10,r9,lsl#2] @ Te3[s2>>0] + eor r0,r0,r7,ror#16 + ldr r2,[r10,r2,lsl#2] @ Te0[s2>>24] + and r7,lr,r3 @ i0 + eor r1,r1,r8,ror#8 + and r8,lr,r3,lsr#8 @ i1 + eor r6,r6,r9,ror#16 + and r9,lr,r3,lsr#16 @ i2 + ldr r7,[r10,r7,lsl#2] @ Te3[s3>>0] + eor r2,r2,r5,ror#16 + ldr r8,[r10,r8,lsl#2] @ Te2[s3>>8] + mov r3,r3,lsr#24 + + ldr r9,[r10,r9,lsl#2] @ Te1[s3>>16] + eor r0,r0,r7,ror#24 + ldr r7,[r11],#16 + eor r1,r1,r8,ror#16 + ldr r3,[r10,r3,lsl#2] @ Te0[s3>>24] + eor r2,r2,r9,ror#8 + ldr r4,[r11,#-12] + eor r3,r3,r6,ror#8 + + ldr r5,[r11,#-8] + eor r0,r0,r7 + ldr r6,[r11,#-4] + and r7,lr,r0 + eor r1,r1,r4 + and r8,lr,r0,lsr#8 + eor r2,r2,r5 + and r9,lr,r0,lsr#16 + eor r3,r3,r6 + mov r0,r0,lsr#24 + + subs r12,r12,#1 + bne .Lenc_loop + + add r10,r10,#2 + + ldrb r4,[r10,r7,lsl#2] @ Te4[s0>>0] + and r7,lr,r1,lsr#16 @ i0 + ldrb r5,[r10,r8,lsl#2] @ Te4[s0>>8] + and r8,lr,r1 + ldrb r6,[r10,r9,lsl#2] @ Te4[s0>>16] + and r9,lr,r1,lsr#8 + ldrb r0,[r10,r0,lsl#2] @ Te4[s0>>24] + mov r1,r1,lsr#24 + + ldrb r7,[r10,r7,lsl#2] @ Te4[s1>>16] + ldrb r8,[r10,r8,lsl#2] @ Te4[s1>>0] + ldrb r9,[r10,r9,lsl#2] @ Te4[s1>>8] + eor r0,r7,r0,lsl#8 + ldrb r1,[r10,r1,lsl#2] @ Te4[s1>>24] + and r7,lr,r2,lsr#8 @ i0 + eor r5,r8,r5,lsl#8 + and r8,lr,r2,lsr#16 @ i1 + eor r6,r9,r6,lsl#8 + and r9,lr,r2 + ldrb r7,[r10,r7,lsl#2] @ Te4[s2>>8] + eor r1,r4,r1,lsl#24 + ldrb r8,[r10,r8,lsl#2] @ Te4[s2>>16] + mov r2,r2,lsr#24 + + ldrb r9,[r10,r9,lsl#2] @ Te4[s2>>0] + eor r0,r7,r0,lsl#8 + ldrb r2,[r10,r2,lsl#2] @ Te4[s2>>24] + and r7,lr,r3 @ i0 + eor r1,r1,r8,lsl#16 + and r8,lr,r3,lsr#8 @ i1 + eor r6,r9,r6,lsl#8 + and r9,lr,r3,lsr#16 @ i2 + ldrb r7,[r10,r7,lsl#2] @ Te4[s3>>0] + eor r2,r5,r2,lsl#24 + ldrb r8,[r10,r8,lsl#2] @ Te4[s3>>8] + mov r3,r3,lsr#24 + + ldrb r9,[r10,r9,lsl#2] @ Te4[s3>>16] + eor r0,r7,r0,lsl#8 + ldr r7,[r11,#0] + ldrb r3,[r10,r3,lsl#2] @ Te4[s3>>24] + eor r1,r1,r8,lsl#8 + ldr r4,[r11,#4] + eor r2,r2,r9,lsl#16 + ldr r5,[r11,#8] + eor r3,r6,r3,lsl#24 + ldr r6,[r11,#12] + + eor r0,r0,r7 + eor r1,r1,r4 + eor r2,r2,r5 + eor r3,r3,r6 + + sub r10,r10,#2 + ldr pc,[sp],#4 @ pop and return +.size _armv4_AES_encrypt,.-_armv4_AES_encrypt + +.global aes_nohw_set_encrypt_key +.type aes_nohw_set_encrypt_key,%function +.align 5 +aes_nohw_set_encrypt_key: +_armv4_AES_set_encrypt_key: +#ifndef __thumb2__ + sub r3,pc,#8 @ aes_nohw_set_encrypt_key +#else + adr r3,. +#endif + teq r0,#0 +#ifdef __thumb2__ + itt eq @ Thumb2 thing, validity check in ARM +#endif + moveq r0,#-1 + beq .Labrt + teq r2,#0 +#ifdef __thumb2__ + itt eq @ Thumb2 thing, validity check in ARM +#endif + moveq r0,#-1 + beq .Labrt + + teq r1,#128 + beq .Lok + teq r1,#192 + beq .Lok + teq r1,#256 +#ifdef __thumb2__ + itt ne @ Thumb2 thing, validity check in ARM +#endif + movne r0,#-1 + bne .Labrt + +.Lok: stmdb sp!,{r4-r12,lr} + mov r12,r0 @ inp + mov lr,r1 @ bits + mov r11,r2 @ key + +#if defined(__thumb2__) || defined(__APPLE__) + adr r10,AES_Te+1024 @ Te4 +#else + sub r10,r3,#_armv4_AES_set_encrypt_key-AES_Te-1024 @ Te4 +#endif + +#if __ARM_ARCH__<7 + ldrb r0,[r12,#3] @ load input data in endian-neutral + ldrb r4,[r12,#2] @ manner... + ldrb r5,[r12,#1] + ldrb r6,[r12,#0] + orr r0,r0,r4,lsl#8 + ldrb r1,[r12,#7] + orr r0,r0,r5,lsl#16 + ldrb r4,[r12,#6] + orr r0,r0,r6,lsl#24 + ldrb r5,[r12,#5] + ldrb r6,[r12,#4] + orr r1,r1,r4,lsl#8 + ldrb r2,[r12,#11] + orr r1,r1,r5,lsl#16 + ldrb r4,[r12,#10] + orr r1,r1,r6,lsl#24 + ldrb r5,[r12,#9] + ldrb r6,[r12,#8] + orr r2,r2,r4,lsl#8 + ldrb r3,[r12,#15] + orr r2,r2,r5,lsl#16 + ldrb r4,[r12,#14] + orr r2,r2,r6,lsl#24 + ldrb r5,[r12,#13] + ldrb r6,[r12,#12] + orr r3,r3,r4,lsl#8 + str r0,[r11],#16 + orr r3,r3,r5,lsl#16 + str r1,[r11,#-12] + orr r3,r3,r6,lsl#24 + str r2,[r11,#-8] + str r3,[r11,#-4] +#else + ldr r0,[r12,#0] + ldr r1,[r12,#4] + ldr r2,[r12,#8] + ldr r3,[r12,#12] +#ifdef __ARMEL__ + rev r0,r0 + rev r1,r1 + rev r2,r2 + rev r3,r3 +#endif + str r0,[r11],#16 + str r1,[r11,#-12] + str r2,[r11,#-8] + str r3,[r11,#-4] +#endif + + teq lr,#128 + bne .Lnot128 + mov r12,#10 + str r12,[r11,#240-16] + add r6,r10,#256 @ rcon + mov lr,#255 + +.L128_loop: + and r5,lr,r3,lsr#24 + and r7,lr,r3,lsr#16 + ldrb r5,[r10,r5] + and r8,lr,r3,lsr#8 + ldrb r7,[r10,r7] + and r9,lr,r3 + ldrb r8,[r10,r8] + orr r5,r5,r7,lsl#24 + ldrb r9,[r10,r9] + orr r5,r5,r8,lsl#16 + ldr r4,[r6],#4 @ rcon[i++] + orr r5,r5,r9,lsl#8 + eor r5,r5,r4 + eor r0,r0,r5 @ rk[4]=rk[0]^... + eor r1,r1,r0 @ rk[5]=rk[1]^rk[4] + str r0,[r11],#16 + eor r2,r2,r1 @ rk[6]=rk[2]^rk[5] + str r1,[r11,#-12] + eor r3,r3,r2 @ rk[7]=rk[3]^rk[6] + str r2,[r11,#-8] + subs r12,r12,#1 + str r3,[r11,#-4] + bne .L128_loop + sub r2,r11,#176 + b .Ldone + +.Lnot128: +#if __ARM_ARCH__<7 + ldrb r8,[r12,#19] + ldrb r4,[r12,#18] + ldrb r5,[r12,#17] + ldrb r6,[r12,#16] + orr r8,r8,r4,lsl#8 + ldrb r9,[r12,#23] + orr r8,r8,r5,lsl#16 + ldrb r4,[r12,#22] + orr r8,r8,r6,lsl#24 + ldrb r5,[r12,#21] + ldrb r6,[r12,#20] + orr r9,r9,r4,lsl#8 + orr r9,r9,r5,lsl#16 + str r8,[r11],#8 + orr r9,r9,r6,lsl#24 + str r9,[r11,#-4] +#else + ldr r8,[r12,#16] + ldr r9,[r12,#20] +#ifdef __ARMEL__ + rev r8,r8 + rev r9,r9 +#endif + str r8,[r11],#8 + str r9,[r11,#-4] +#endif + + teq lr,#192 + bne .Lnot192 + mov r12,#12 + str r12,[r11,#240-24] + add r6,r10,#256 @ rcon + mov lr,#255 + mov r12,#8 + +.L192_loop: + and r5,lr,r9,lsr#24 + and r7,lr,r9,lsr#16 + ldrb r5,[r10,r5] + and r8,lr,r9,lsr#8 + ldrb r7,[r10,r7] + and r9,lr,r9 + ldrb r8,[r10,r8] + orr r5,r5,r7,lsl#24 + ldrb r9,[r10,r9] + orr r5,r5,r8,lsl#16 + ldr r4,[r6],#4 @ rcon[i++] + orr r5,r5,r9,lsl#8 + eor r9,r5,r4 + eor r0,r0,r9 @ rk[6]=rk[0]^... + eor r1,r1,r0 @ rk[7]=rk[1]^rk[6] + str r0,[r11],#24 + eor r2,r2,r1 @ rk[8]=rk[2]^rk[7] + str r1,[r11,#-20] + eor r3,r3,r2 @ rk[9]=rk[3]^rk[8] + str r2,[r11,#-16] + subs r12,r12,#1 + str r3,[r11,#-12] +#ifdef __thumb2__ + itt eq @ Thumb2 thing, validity check in ARM +#endif + subeq r2,r11,#216 + beq .Ldone + + ldr r7,[r11,#-32] + ldr r8,[r11,#-28] + eor r7,r7,r3 @ rk[10]=rk[4]^rk[9] + eor r9,r8,r7 @ rk[11]=rk[5]^rk[10] + str r7,[r11,#-8] + str r9,[r11,#-4] + b .L192_loop + +.Lnot192: +#if __ARM_ARCH__<7 + ldrb r8,[r12,#27] + ldrb r4,[r12,#26] + ldrb r5,[r12,#25] + ldrb r6,[r12,#24] + orr r8,r8,r4,lsl#8 + ldrb r9,[r12,#31] + orr r8,r8,r5,lsl#16 + ldrb r4,[r12,#30] + orr r8,r8,r6,lsl#24 + ldrb r5,[r12,#29] + ldrb r6,[r12,#28] + orr r9,r9,r4,lsl#8 + orr r9,r9,r5,lsl#16 + str r8,[r11],#8 + orr r9,r9,r6,lsl#24 + str r9,[r11,#-4] +#else + ldr r8,[r12,#24] + ldr r9,[r12,#28] +#ifdef __ARMEL__ + rev r8,r8 + rev r9,r9 +#endif + str r8,[r11],#8 + str r9,[r11,#-4] +#endif + + mov r12,#14 + str r12,[r11,#240-32] + add r6,r10,#256 @ rcon + mov lr,#255 + mov r12,#7 + +.L256_loop: + and r5,lr,r9,lsr#24 + and r7,lr,r9,lsr#16 + ldrb r5,[r10,r5] + and r8,lr,r9,lsr#8 + ldrb r7,[r10,r7] + and r9,lr,r9 + ldrb r8,[r10,r8] + orr r5,r5,r7,lsl#24 + ldrb r9,[r10,r9] + orr r5,r5,r8,lsl#16 + ldr r4,[r6],#4 @ rcon[i++] + orr r5,r5,r9,lsl#8 + eor r9,r5,r4 + eor r0,r0,r9 @ rk[8]=rk[0]^... + eor r1,r1,r0 @ rk[9]=rk[1]^rk[8] + str r0,[r11],#32 + eor r2,r2,r1 @ rk[10]=rk[2]^rk[9] + str r1,[r11,#-28] + eor r3,r3,r2 @ rk[11]=rk[3]^rk[10] + str r2,[r11,#-24] + subs r12,r12,#1 + str r3,[r11,#-20] +#ifdef __thumb2__ + itt eq @ Thumb2 thing, validity check in ARM +#endif + subeq r2,r11,#256 + beq .Ldone + + and r5,lr,r3 + and r7,lr,r3,lsr#8 + ldrb r5,[r10,r5] + and r8,lr,r3,lsr#16 + ldrb r7,[r10,r7] + and r9,lr,r3,lsr#24 + ldrb r8,[r10,r8] + orr r5,r5,r7,lsl#8 + ldrb r9,[r10,r9] + orr r5,r5,r8,lsl#16 + ldr r4,[r11,#-48] + orr r5,r5,r9,lsl#24 + + ldr r7,[r11,#-44] + ldr r8,[r11,#-40] + eor r4,r4,r5 @ rk[12]=rk[4]^... + ldr r9,[r11,#-36] + eor r7,r7,r4 @ rk[13]=rk[5]^rk[12] + str r4,[r11,#-16] + eor r8,r8,r7 @ rk[14]=rk[6]^rk[13] + str r7,[r11,#-12] + eor r9,r9,r8 @ rk[15]=rk[7]^rk[14] + str r8,[r11,#-8] + str r9,[r11,#-4] + b .L256_loop + +.align 2 +.Ldone: mov r0,#0 + ldmia sp!,{r4-r12,lr} +.Labrt: +#if __ARM_ARCH__>=5 + bx lr @ .word 0xe12fff1e +#else + tst lr,#1 + moveq pc,lr @ be binary compatible with V4, yet + .word 0xe12fff1e @ interoperable with Thumb ISA:-) +#endif +.size aes_nohw_set_encrypt_key,.-aes_nohw_set_encrypt_key + +.global aes_nohw_set_decrypt_key +.type aes_nohw_set_decrypt_key,%function +.align 5 +aes_nohw_set_decrypt_key: + str lr,[sp,#-4]! @ push lr + bl _armv4_AES_set_encrypt_key + teq r0,#0 + ldr lr,[sp],#4 @ pop lr + bne .Labrt + + mov r0,r2 @ aes_nohw_set_encrypt_key preserves r2, + mov r1,r2 @ which is AES_KEY *key + b _armv4_AES_set_enc2dec_key +.size aes_nohw_set_decrypt_key,.-aes_nohw_set_decrypt_key + +@ void AES_set_enc2dec_key(const AES_KEY *inp,AES_KEY *out) +.global AES_set_enc2dec_key +.type AES_set_enc2dec_key,%function +.align 5 +AES_set_enc2dec_key: +_armv4_AES_set_enc2dec_key: + stmdb sp!,{r4-r12,lr} + + ldr r12,[r0,#240] + mov r7,r0 @ input + add r8,r0,r12,lsl#4 + mov r11,r1 @ output + add r10,r1,r12,lsl#4 + str r12,[r1,#240] + +.Linv: ldr r0,[r7],#16 + ldr r1,[r7,#-12] + ldr r2,[r7,#-8] + ldr r3,[r7,#-4] + ldr r4,[r8],#-16 + ldr r5,[r8,#16+4] + ldr r6,[r8,#16+8] + ldr r9,[r8,#16+12] + str r0,[r10],#-16 + str r1,[r10,#16+4] + str r2,[r10,#16+8] + str r3,[r10,#16+12] + str r4,[r11],#16 + str r5,[r11,#-12] + str r6,[r11,#-8] + str r9,[r11,#-4] + teq r7,r8 + bne .Linv + + ldr r0,[r7] + ldr r1,[r7,#4] + ldr r2,[r7,#8] + ldr r3,[r7,#12] + str r0,[r11] + str r1,[r11,#4] + str r2,[r11,#8] + str r3,[r11,#12] + sub r11,r11,r12,lsl#3 + ldr r0,[r11,#16]! @ prefetch tp1 + mov r7,#0x80 + mov r8,#0x1b + orr r7,r7,#0x8000 + orr r8,r8,#0x1b00 + orr r7,r7,r7,lsl#16 + orr r8,r8,r8,lsl#16 + sub r12,r12,#1 + mvn r9,r7 + mov r12,r12,lsl#2 @ (rounds-1)*4 + +.Lmix: and r4,r0,r7 + and r1,r0,r9 + sub r4,r4,r4,lsr#7 + and r4,r4,r8 + eor r1,r4,r1,lsl#1 @ tp2 + + and r4,r1,r7 + and r2,r1,r9 + sub r4,r4,r4,lsr#7 + and r4,r4,r8 + eor r2,r4,r2,lsl#1 @ tp4 + + and r4,r2,r7 + and r3,r2,r9 + sub r4,r4,r4,lsr#7 + and r4,r4,r8 + eor r3,r4,r3,lsl#1 @ tp8 + + eor r4,r1,r2 + eor r5,r0,r3 @ tp9 + eor r4,r4,r3 @ tpe + eor r4,r4,r1,ror#24 + eor r4,r4,r5,ror#24 @ ^= ROTATE(tpb=tp9^tp2,8) + eor r4,r4,r2,ror#16 + eor r4,r4,r5,ror#16 @ ^= ROTATE(tpd=tp9^tp4,16) + eor r4,r4,r5,ror#8 @ ^= ROTATE(tp9,24) + + ldr r0,[r11,#4] @ prefetch tp1 + str r4,[r11],#4 + subs r12,r12,#1 + bne .Lmix + + mov r0,#0 +#if __ARM_ARCH__>=5 + ldmia sp!,{r4-r12,pc} +#else + ldmia sp!,{r4-r12,lr} + tst lr,#1 + moveq pc,lr @ be binary compatible with V4, yet + .word 0xe12fff1e @ interoperable with Thumb ISA:-) +#endif +.size AES_set_enc2dec_key,.-AES_set_enc2dec_key + +.type AES_Td,%object +.align 5 +AES_Td: +.word 0x51f4a750, 0x7e416553, 0x1a17a4c3, 0x3a275e96 +.word 0x3bab6bcb, 0x1f9d45f1, 0xacfa58ab, 0x4be30393 +.word 0x2030fa55, 0xad766df6, 0x88cc7691, 0xf5024c25 +.word 0x4fe5d7fc, 0xc52acbd7, 0x26354480, 0xb562a38f +.word 0xdeb15a49, 0x25ba1b67, 0x45ea0e98, 0x5dfec0e1 +.word 0xc32f7502, 0x814cf012, 0x8d4697a3, 0x6bd3f9c6 +.word 0x038f5fe7, 0x15929c95, 0xbf6d7aeb, 0x955259da +.word 0xd4be832d, 0x587421d3, 0x49e06929, 0x8ec9c844 +.word 0x75c2896a, 0xf48e7978, 0x99583e6b, 0x27b971dd +.word 0xbee14fb6, 0xf088ad17, 0xc920ac66, 0x7dce3ab4 +.word 0x63df4a18, 0xe51a3182, 0x97513360, 0x62537f45 +.word 0xb16477e0, 0xbb6bae84, 0xfe81a01c, 0xf9082b94 +.word 0x70486858, 0x8f45fd19, 0x94de6c87, 0x527bf8b7 +.word 0xab73d323, 0x724b02e2, 0xe31f8f57, 0x6655ab2a +.word 0xb2eb2807, 0x2fb5c203, 0x86c57b9a, 0xd33708a5 +.word 0x302887f2, 0x23bfa5b2, 0x02036aba, 0xed16825c +.word 0x8acf1c2b, 0xa779b492, 0xf307f2f0, 0x4e69e2a1 +.word 0x65daf4cd, 0x0605bed5, 0xd134621f, 0xc4a6fe8a +.word 0x342e539d, 0xa2f355a0, 0x058ae132, 0xa4f6eb75 +.word 0x0b83ec39, 0x4060efaa, 0x5e719f06, 0xbd6e1051 +.word 0x3e218af9, 0x96dd063d, 0xdd3e05ae, 0x4de6bd46 +.word 0x91548db5, 0x71c45d05, 0x0406d46f, 0x605015ff +.word 0x1998fb24, 0xd6bde997, 0x894043cc, 0x67d99e77 +.word 0xb0e842bd, 0x07898b88, 0xe7195b38, 0x79c8eedb +.word 0xa17c0a47, 0x7c420fe9, 0xf8841ec9, 0x00000000 +.word 0x09808683, 0x322bed48, 0x1e1170ac, 0x6c5a724e +.word 0xfd0efffb, 0x0f853856, 0x3daed51e, 0x362d3927 +.word 0x0a0fd964, 0x685ca621, 0x9b5b54d1, 0x24362e3a +.word 0x0c0a67b1, 0x9357e70f, 0xb4ee96d2, 0x1b9b919e +.word 0x80c0c54f, 0x61dc20a2, 0x5a774b69, 0x1c121a16 +.word 0xe293ba0a, 0xc0a02ae5, 0x3c22e043, 0x121b171d +.word 0x0e090d0b, 0xf28bc7ad, 0x2db6a8b9, 0x141ea9c8 +.word 0x57f11985, 0xaf75074c, 0xee99ddbb, 0xa37f60fd +.word 0xf701269f, 0x5c72f5bc, 0x44663bc5, 0x5bfb7e34 +.word 0x8b432976, 0xcb23c6dc, 0xb6edfc68, 0xb8e4f163 +.word 0xd731dcca, 0x42638510, 0x13972240, 0x84c61120 +.word 0x854a247d, 0xd2bb3df8, 0xaef93211, 0xc729a16d +.word 0x1d9e2f4b, 0xdcb230f3, 0x0d8652ec, 0x77c1e3d0 +.word 0x2bb3166c, 0xa970b999, 0x119448fa, 0x47e96422 +.word 0xa8fc8cc4, 0xa0f03f1a, 0x567d2cd8, 0x223390ef +.word 0x87494ec7, 0xd938d1c1, 0x8ccaa2fe, 0x98d40b36 +.word 0xa6f581cf, 0xa57ade28, 0xdab78e26, 0x3fadbfa4 +.word 0x2c3a9de4, 0x5078920d, 0x6a5fcc9b, 0x547e4662 +.word 0xf68d13c2, 0x90d8b8e8, 0x2e39f75e, 0x82c3aff5 +.word 0x9f5d80be, 0x69d0937c, 0x6fd52da9, 0xcf2512b3 +.word 0xc8ac993b, 0x10187da7, 0xe89c636e, 0xdb3bbb7b +.word 0xcd267809, 0x6e5918f4, 0xec9ab701, 0x834f9aa8 +.word 0xe6956e65, 0xaaffe67e, 0x21bccf08, 0xef15e8e6 +.word 0xbae79bd9, 0x4a6f36ce, 0xea9f09d4, 0x29b07cd6 +.word 0x31a4b2af, 0x2a3f2331, 0xc6a59430, 0x35a266c0 +.word 0x744ebc37, 0xfc82caa6, 0xe090d0b0, 0x33a7d815 +.word 0xf104984a, 0x41ecdaf7, 0x7fcd500e, 0x1791f62f +.word 0x764dd68d, 0x43efb04d, 0xccaa4d54, 0xe49604df +.word 0x9ed1b5e3, 0x4c6a881b, 0xc12c1fb8, 0x4665517f +.word 0x9d5eea04, 0x018c355d, 0xfa877473, 0xfb0b412e +.word 0xb3671d5a, 0x92dbd252, 0xe9105633, 0x6dd64713 +.word 0x9ad7618c, 0x37a10c7a, 0x59f8148e, 0xeb133c89 +.word 0xcea927ee, 0xb761c935, 0xe11ce5ed, 0x7a47b13c +.word 0x9cd2df59, 0x55f2733f, 0x1814ce79, 0x73c737bf +.word 0x53f7cdea, 0x5ffdaa5b, 0xdf3d6f14, 0x7844db86 +.word 0xcaaff381, 0xb968c43e, 0x3824342c, 0xc2a3405f +.word 0x161dc372, 0xbce2250c, 0x283c498b, 0xff0d9541 +.word 0x39a80171, 0x080cb3de, 0xd8b4e49c, 0x6456c190 +.word 0x7bcb8461, 0xd532b670, 0x486c5c74, 0xd0b85742 +@ Td4[256] +.byte 0x52, 0x09, 0x6a, 0xd5, 0x30, 0x36, 0xa5, 0x38 +.byte 0xbf, 0x40, 0xa3, 0x9e, 0x81, 0xf3, 0xd7, 0xfb +.byte 0x7c, 0xe3, 0x39, 0x82, 0x9b, 0x2f, 0xff, 0x87 +.byte 0x34, 0x8e, 0x43, 0x44, 0xc4, 0xde, 0xe9, 0xcb +.byte 0x54, 0x7b, 0x94, 0x32, 0xa6, 0xc2, 0x23, 0x3d +.byte 0xee, 0x4c, 0x95, 0x0b, 0x42, 0xfa, 0xc3, 0x4e +.byte 0x08, 0x2e, 0xa1, 0x66, 0x28, 0xd9, 0x24, 0xb2 +.byte 0x76, 0x5b, 0xa2, 0x49, 0x6d, 0x8b, 0xd1, 0x25 +.byte 0x72, 0xf8, 0xf6, 0x64, 0x86, 0x68, 0x98, 0x16 +.byte 0xd4, 0xa4, 0x5c, 0xcc, 0x5d, 0x65, 0xb6, 0x92 +.byte 0x6c, 0x70, 0x48, 0x50, 0xfd, 0xed, 0xb9, 0xda +.byte 0x5e, 0x15, 0x46, 0x57, 0xa7, 0x8d, 0x9d, 0x84 +.byte 0x90, 0xd8, 0xab, 0x00, 0x8c, 0xbc, 0xd3, 0x0a +.byte 0xf7, 0xe4, 0x58, 0x05, 0xb8, 0xb3, 0x45, 0x06 +.byte 0xd0, 0x2c, 0x1e, 0x8f, 0xca, 0x3f, 0x0f, 0x02 +.byte 0xc1, 0xaf, 0xbd, 0x03, 0x01, 0x13, 0x8a, 0x6b +.byte 0x3a, 0x91, 0x11, 0x41, 0x4f, 0x67, 0xdc, 0xea +.byte 0x97, 0xf2, 0xcf, 0xce, 0xf0, 0xb4, 0xe6, 0x73 +.byte 0x96, 0xac, 0x74, 0x22, 0xe7, 0xad, 0x35, 0x85 +.byte 0xe2, 0xf9, 0x37, 0xe8, 0x1c, 0x75, 0xdf, 0x6e +.byte 0x47, 0xf1, 0x1a, 0x71, 0x1d, 0x29, 0xc5, 0x89 +.byte 0x6f, 0xb7, 0x62, 0x0e, 0xaa, 0x18, 0xbe, 0x1b +.byte 0xfc, 0x56, 0x3e, 0x4b, 0xc6, 0xd2, 0x79, 0x20 +.byte 0x9a, 0xdb, 0xc0, 0xfe, 0x78, 0xcd, 0x5a, 0xf4 +.byte 0x1f, 0xdd, 0xa8, 0x33, 0x88, 0x07, 0xc7, 0x31 +.byte 0xb1, 0x12, 0x10, 0x59, 0x27, 0x80, 0xec, 0x5f +.byte 0x60, 0x51, 0x7f, 0xa9, 0x19, 0xb5, 0x4a, 0x0d +.byte 0x2d, 0xe5, 0x7a, 0x9f, 0x93, 0xc9, 0x9c, 0xef +.byte 0xa0, 0xe0, 0x3b, 0x4d, 0xae, 0x2a, 0xf5, 0xb0 +.byte 0xc8, 0xeb, 0xbb, 0x3c, 0x83, 0x53, 0x99, 0x61 +.byte 0x17, 0x2b, 0x04, 0x7e, 0xba, 0x77, 0xd6, 0x26 +.byte 0xe1, 0x69, 0x14, 0x63, 0x55, 0x21, 0x0c, 0x7d +.size AES_Td,.-AES_Td + +@ void aes_nohw_decrypt(const unsigned char *in, unsigned char *out, +@ const AES_KEY *key) { +.global aes_nohw_decrypt +.type aes_nohw_decrypt,%function +.align 5 +aes_nohw_decrypt: +#ifndef __thumb2__ + sub r3,pc,#8 @ aes_nohw_decrypt +#else + adr r3,. +#endif + stmdb sp!,{r1,r4-r12,lr} +#if defined(__thumb2__) || defined(__APPLE__) + adr r10,AES_Td +#else + sub r10,r3,#aes_nohw_decrypt-AES_Td @ Td +#endif + mov r12,r0 @ inp + mov r11,r2 +#if __ARM_ARCH__<7 + ldrb r0,[r12,#3] @ load input data in endian-neutral + ldrb r4,[r12,#2] @ manner... + ldrb r5,[r12,#1] + ldrb r6,[r12,#0] + orr r0,r0,r4,lsl#8 + ldrb r1,[r12,#7] + orr r0,r0,r5,lsl#16 + ldrb r4,[r12,#6] + orr r0,r0,r6,lsl#24 + ldrb r5,[r12,#5] + ldrb r6,[r12,#4] + orr r1,r1,r4,lsl#8 + ldrb r2,[r12,#11] + orr r1,r1,r5,lsl#16 + ldrb r4,[r12,#10] + orr r1,r1,r6,lsl#24 + ldrb r5,[r12,#9] + ldrb r6,[r12,#8] + orr r2,r2,r4,lsl#8 + ldrb r3,[r12,#15] + orr r2,r2,r5,lsl#16 + ldrb r4,[r12,#14] + orr r2,r2,r6,lsl#24 + ldrb r5,[r12,#13] + ldrb r6,[r12,#12] + orr r3,r3,r4,lsl#8 + orr r3,r3,r5,lsl#16 + orr r3,r3,r6,lsl#24 +#else + ldr r0,[r12,#0] + ldr r1,[r12,#4] + ldr r2,[r12,#8] + ldr r3,[r12,#12] +#ifdef __ARMEL__ + rev r0,r0 + rev r1,r1 + rev r2,r2 + rev r3,r3 +#endif +#endif + bl _armv4_AES_decrypt + + ldr r12,[sp],#4 @ pop out +#if __ARM_ARCH__>=7 +#ifdef __ARMEL__ + rev r0,r0 + rev r1,r1 + rev r2,r2 + rev r3,r3 +#endif + str r0,[r12,#0] + str r1,[r12,#4] + str r2,[r12,#8] + str r3,[r12,#12] +#else + mov r4,r0,lsr#24 @ write output in endian-neutral + mov r5,r0,lsr#16 @ manner... + mov r6,r0,lsr#8 + strb r4,[r12,#0] + strb r5,[r12,#1] + mov r4,r1,lsr#24 + strb r6,[r12,#2] + mov r5,r1,lsr#16 + strb r0,[r12,#3] + mov r6,r1,lsr#8 + strb r4,[r12,#4] + strb r5,[r12,#5] + mov r4,r2,lsr#24 + strb r6,[r12,#6] + mov r5,r2,lsr#16 + strb r1,[r12,#7] + mov r6,r2,lsr#8 + strb r4,[r12,#8] + strb r5,[r12,#9] + mov r4,r3,lsr#24 + strb r6,[r12,#10] + mov r5,r3,lsr#16 + strb r2,[r12,#11] + mov r6,r3,lsr#8 + strb r4,[r12,#12] + strb r5,[r12,#13] + strb r6,[r12,#14] + strb r3,[r12,#15] +#endif +#if __ARM_ARCH__>=5 + ldmia sp!,{r4-r12,pc} +#else + ldmia sp!,{r4-r12,lr} + tst lr,#1 + moveq pc,lr @ be binary compatible with V4, yet + .word 0xe12fff1e @ interoperable with Thumb ISA:-) +#endif +.size aes_nohw_decrypt,.-aes_nohw_decrypt + +.type _armv4_AES_decrypt,%function +.align 2 +_armv4_AES_decrypt: + str lr,[sp,#-4]! @ push lr + ldmia r11!,{r4-r7} + eor r0,r0,r4 + ldr r12,[r11,#240-16] + eor r1,r1,r5 + eor r2,r2,r6 + eor r3,r3,r7 + sub r12,r12,#1 + mov lr,#255 + + and r7,lr,r0,lsr#16 + and r8,lr,r0,lsr#8 + and r9,lr,r0 + mov r0,r0,lsr#24 +.Ldec_loop: + ldr r4,[r10,r7,lsl#2] @ Td1[s0>>16] + and r7,lr,r1 @ i0 + ldr r5,[r10,r8,lsl#2] @ Td2[s0>>8] + and r8,lr,r1,lsr#16 + ldr r6,[r10,r9,lsl#2] @ Td3[s0>>0] + and r9,lr,r1,lsr#8 + ldr r0,[r10,r0,lsl#2] @ Td0[s0>>24] + mov r1,r1,lsr#24 + + ldr r7,[r10,r7,lsl#2] @ Td3[s1>>0] + ldr r8,[r10,r8,lsl#2] @ Td1[s1>>16] + ldr r9,[r10,r9,lsl#2] @ Td2[s1>>8] + eor r0,r0,r7,ror#24 + ldr r1,[r10,r1,lsl#2] @ Td0[s1>>24] + and r7,lr,r2,lsr#8 @ i0 + eor r5,r8,r5,ror#8 + and r8,lr,r2 @ i1 + eor r6,r9,r6,ror#8 + and r9,lr,r2,lsr#16 + ldr r7,[r10,r7,lsl#2] @ Td2[s2>>8] + eor r1,r1,r4,ror#8 + ldr r8,[r10,r8,lsl#2] @ Td3[s2>>0] + mov r2,r2,lsr#24 + + ldr r9,[r10,r9,lsl#2] @ Td1[s2>>16] + eor r0,r0,r7,ror#16 + ldr r2,[r10,r2,lsl#2] @ Td0[s2>>24] + and r7,lr,r3,lsr#16 @ i0 + eor r1,r1,r8,ror#24 + and r8,lr,r3,lsr#8 @ i1 + eor r6,r9,r6,ror#8 + and r9,lr,r3 @ i2 + ldr r7,[r10,r7,lsl#2] @ Td1[s3>>16] + eor r2,r2,r5,ror#8 + ldr r8,[r10,r8,lsl#2] @ Td2[s3>>8] + mov r3,r3,lsr#24 + + ldr r9,[r10,r9,lsl#2] @ Td3[s3>>0] + eor r0,r0,r7,ror#8 + ldr r7,[r11],#16 + eor r1,r1,r8,ror#16 + ldr r3,[r10,r3,lsl#2] @ Td0[s3>>24] + eor r2,r2,r9,ror#24 + + ldr r4,[r11,#-12] + eor r0,r0,r7 + ldr r5,[r11,#-8] + eor r3,r3,r6,ror#8 + ldr r6,[r11,#-4] + and r7,lr,r0,lsr#16 + eor r1,r1,r4 + and r8,lr,r0,lsr#8 + eor r2,r2,r5 + and r9,lr,r0 + eor r3,r3,r6 + mov r0,r0,lsr#24 + + subs r12,r12,#1 + bne .Ldec_loop + + add r10,r10,#1024 + + ldr r5,[r10,#0] @ prefetch Td4 + ldr r6,[r10,#32] + ldr r4,[r10,#64] + ldr r5,[r10,#96] + ldr r6,[r10,#128] + ldr r4,[r10,#160] + ldr r5,[r10,#192] + ldr r6,[r10,#224] + + ldrb r0,[r10,r0] @ Td4[s0>>24] + ldrb r4,[r10,r7] @ Td4[s0>>16] + and r7,lr,r1 @ i0 + ldrb r5,[r10,r8] @ Td4[s0>>8] + and r8,lr,r1,lsr#16 + ldrb r6,[r10,r9] @ Td4[s0>>0] + and r9,lr,r1,lsr#8 + + add r1,r10,r1,lsr#24 + ldrb r7,[r10,r7] @ Td4[s1>>0] + ldrb r1,[r1] @ Td4[s1>>24] + ldrb r8,[r10,r8] @ Td4[s1>>16] + eor r0,r7,r0,lsl#24 + ldrb r9,[r10,r9] @ Td4[s1>>8] + eor r1,r4,r1,lsl#8 + and r7,lr,r2,lsr#8 @ i0 + eor r5,r5,r8,lsl#8 + and r8,lr,r2 @ i1 + ldrb r7,[r10,r7] @ Td4[s2>>8] + eor r6,r6,r9,lsl#8 + ldrb r8,[r10,r8] @ Td4[s2>>0] + and r9,lr,r2,lsr#16 + + add r2,r10,r2,lsr#24 + ldrb r2,[r2] @ Td4[s2>>24] + eor r0,r0,r7,lsl#8 + ldrb r9,[r10,r9] @ Td4[s2>>16] + eor r1,r8,r1,lsl#16 + and r7,lr,r3,lsr#16 @ i0 + eor r2,r5,r2,lsl#16 + and r8,lr,r3,lsr#8 @ i1 + ldrb r7,[r10,r7] @ Td4[s3>>16] + eor r6,r6,r9,lsl#16 + ldrb r8,[r10,r8] @ Td4[s3>>8] + and r9,lr,r3 @ i2 + + add r3,r10,r3,lsr#24 + ldrb r9,[r10,r9] @ Td4[s3>>0] + ldrb r3,[r3] @ Td4[s3>>24] + eor r0,r0,r7,lsl#16 + ldr r7,[r11,#0] + eor r1,r1,r8,lsl#8 + ldr r4,[r11,#4] + eor r2,r9,r2,lsl#8 + ldr r5,[r11,#8] + eor r3,r6,r3,lsl#24 + ldr r6,[r11,#12] + + eor r0,r0,r7 + eor r1,r1,r4 + eor r2,r2,r5 + eor r3,r3,r6 + + sub r10,r10,#1024 + ldr pc,[sp],#4 @ pop and return +.size _armv4_AES_decrypt,.-_armv4_AES_decrypt +.asciz "AES for ARMv4, CRYPTOGAMS by <appro@openssl.org>" +.align 2 diff --git a/third_party/boringssl/core/cortex-m/ghash.S b/third_party/boringssl/core/cortex-m/ghash.S new file mode 100644 index 0000000000..a1eb97b9c5 --- /dev/null +++ b/third_party/boringssl/core/cortex-m/ghash.S @@ -0,0 +1,575 @@ +@ Generated by crypto/fipsmodule/modes/asm/ghash-armv4.pl, which carries +@ this license: +@ +@ Copyright 2010-2016 The OpenSSL Project Authors. All Rights Reserved. +@ +@ Licensed under the OpenSSL license (the "License"). You may not use +@ this file except in compliance with the License. You can obtain a copy +@ in the file LICENSE in the source distribution or at +@ https://www.openssl.org/source/license.html + +#define __ARM_ARCH__ 7 + +.text +#if defined(__thumb2__) || defined(__clang__) +.syntax unified +#endif +#if defined(__thumb2__) +.thumb +#else +.code 32 +#endif + +#ifdef __clang__ +#define ldrplb ldrbpl +#define ldrneb ldrbne +#endif + +.type rem_4bit,%object +.align 5 +rem_4bit: +.short 0x0000,0x1C20,0x3840,0x2460 +.short 0x7080,0x6CA0,0x48C0,0x54E0 +.short 0xE100,0xFD20,0xD940,0xC560 +.short 0x9180,0x8DA0,0xA9C0,0xB5E0 +.size rem_4bit,.-rem_4bit + +.type rem_4bit_get,%function +rem_4bit_get: +#if defined(__thumb2__) + adr r2,rem_4bit +#else + sub r2,pc,#8+32 @ &rem_4bit +#endif + b .Lrem_4bit_got + nop + nop +.size rem_4bit_get,.-rem_4bit_get + +.global gcm_ghash_4bit +.type gcm_ghash_4bit,%function +.align 4 +gcm_ghash_4bit: +#if defined(__thumb2__) + adr r12,rem_4bit +#else + sub r12,pc,#8+48 @ &rem_4bit +#endif + add r3,r2,r3 @ r3 to point at the end + stmdb sp!,{r3-r11,lr} @ save r3/end too + + ldmia r12,{r4-r11} @ copy rem_4bit ... + stmdb sp!,{r4-r11} @ ... to stack + + ldrb r12,[r2,#15] + ldrb r14,[r0,#15] +.Louter: + eor r12,r12,r14 + and r14,r12,#0xf0 + and r12,r12,#0x0f + mov r3,#14 + + add r7,r1,r12,lsl#4 + ldmia r7,{r4-r7} @ load Htbl[nlo] + add r11,r1,r14 + ldrb r12,[r2,#14] + + and r14,r4,#0xf @ rem + ldmia r11,{r8-r11} @ load Htbl[nhi] + add r14,r14,r14 + eor r4,r8,r4,lsr#4 + ldrh r8,[sp,r14] @ rem_4bit[rem] + eor r4,r4,r5,lsl#28 + ldrb r14,[r0,#14] + eor r5,r9,r5,lsr#4 + eor r5,r5,r6,lsl#28 + eor r6,r10,r6,lsr#4 + eor r6,r6,r7,lsl#28 + eor r7,r11,r7,lsr#4 + eor r12,r12,r14 + and r14,r12,#0xf0 + and r12,r12,#0x0f + eor r7,r7,r8,lsl#16 + +.Linner: + add r11,r1,r12,lsl#4 + and r12,r4,#0xf @ rem + subs r3,r3,#1 + add r12,r12,r12 + ldmia r11,{r8-r11} @ load Htbl[nlo] + eor r4,r8,r4,lsr#4 + eor r4,r4,r5,lsl#28 + eor r5,r9,r5,lsr#4 + eor r5,r5,r6,lsl#28 + ldrh r8,[sp,r12] @ rem_4bit[rem] + eor r6,r10,r6,lsr#4 +#ifdef __thumb2__ + it pl +#endif + ldrbpl r12,[r2,r3] + eor r6,r6,r7,lsl#28 + eor r7,r11,r7,lsr#4 + + add r11,r1,r14 + and r14,r4,#0xf @ rem + eor r7,r7,r8,lsl#16 @ ^= rem_4bit[rem] + add r14,r14,r14 + ldmia r11,{r8-r11} @ load Htbl[nhi] + eor r4,r8,r4,lsr#4 +#ifdef __thumb2__ + it pl +#endif + ldrbpl r8,[r0,r3] + eor r4,r4,r5,lsl#28 + eor r5,r9,r5,lsr#4 + ldrh r9,[sp,r14] + eor r5,r5,r6,lsl#28 + eor r6,r10,r6,lsr#4 + eor r6,r6,r7,lsl#28 +#ifdef __thumb2__ + it pl +#endif + eorpl r12,r12,r8 + eor r7,r11,r7,lsr#4 +#ifdef __thumb2__ + itt pl +#endif + andpl r14,r12,#0xf0 + andpl r12,r12,#0x0f + eor r7,r7,r9,lsl#16 @ ^= rem_4bit[rem] + bpl .Linner + + ldr r3,[sp,#32] @ re-load r3/end + add r2,r2,#16 + mov r14,r4 +#if __ARM_ARCH__>=7 && defined(__ARMEL__) + rev r4,r4 + str r4,[r0,#12] +#elif defined(__ARMEB__) + str r4,[r0,#12] +#else + mov r9,r4,lsr#8 + strb r4,[r0,#12+3] + mov r10,r4,lsr#16 + strb r9,[r0,#12+2] + mov r11,r4,lsr#24 + strb r10,[r0,#12+1] + strb r11,[r0,#12] +#endif + cmp r2,r3 +#if __ARM_ARCH__>=7 && defined(__ARMEL__) + rev r5,r5 + str r5,[r0,#8] +#elif defined(__ARMEB__) + str r5,[r0,#8] +#else + mov r9,r5,lsr#8 + strb r5,[r0,#8+3] + mov r10,r5,lsr#16 + strb r9,[r0,#8+2] + mov r11,r5,lsr#24 + strb r10,[r0,#8+1] + strb r11,[r0,#8] +#endif + +#ifdef __thumb2__ + it ne +#endif + ldrbne r12,[r2,#15] +#if __ARM_ARCH__>=7 && defined(__ARMEL__) + rev r6,r6 + str r6,[r0,#4] +#elif defined(__ARMEB__) + str r6,[r0,#4] +#else + mov r9,r6,lsr#8 + strb r6,[r0,#4+3] + mov r10,r6,lsr#16 + strb r9,[r0,#4+2] + mov r11,r6,lsr#24 + strb r10,[r0,#4+1] + strb r11,[r0,#4] +#endif + +#if __ARM_ARCH__>=7 && defined(__ARMEL__) + rev r7,r7 + str r7,[r0,#0] +#elif defined(__ARMEB__) + str r7,[r0,#0] +#else + mov r9,r7,lsr#8 + strb r7,[r0,#0+3] + mov r10,r7,lsr#16 + strb r9,[r0,#0+2] + mov r11,r7,lsr#24 + strb r10,[r0,#0+1] + strb r11,[r0,#0] +#endif + + bne .Louter + + add sp,sp,#36 +#if __ARM_ARCH__>=5 + ldmia sp!,{r4-r11,pc} +#else + ldmia sp!,{r4-r11,lr} + tst lr,#1 + moveq pc,lr @ be binary compatible with V4, yet + .word 0xe12fff1e @ interoperable with Thumb ISA:-) +#endif +.size gcm_ghash_4bit,.-gcm_ghash_4bit + +.global gcm_gmult_4bit +.type gcm_gmult_4bit,%function +gcm_gmult_4bit: + stmdb sp!,{r4-r11,lr} + ldrb r12,[r0,#15] + b rem_4bit_get +.Lrem_4bit_got: + and r14,r12,#0xf0 + and r12,r12,#0x0f + mov r3,#14 + + add r7,r1,r12,lsl#4 + ldmia r7,{r4-r7} @ load Htbl[nlo] + ldrb r12,[r0,#14] + + add r11,r1,r14 + and r14,r4,#0xf @ rem + ldmia r11,{r8-r11} @ load Htbl[nhi] + add r14,r14,r14 + eor r4,r8,r4,lsr#4 + ldrh r8,[r2,r14] @ rem_4bit[rem] + eor r4,r4,r5,lsl#28 + eor r5,r9,r5,lsr#4 + eor r5,r5,r6,lsl#28 + eor r6,r10,r6,lsr#4 + eor r6,r6,r7,lsl#28 + eor r7,r11,r7,lsr#4 + and r14,r12,#0xf0 + eor r7,r7,r8,lsl#16 + and r12,r12,#0x0f + +.Loop: + add r11,r1,r12,lsl#4 + and r12,r4,#0xf @ rem + subs r3,r3,#1 + add r12,r12,r12 + ldmia r11,{r8-r11} @ load Htbl[nlo] + eor r4,r8,r4,lsr#4 + eor r4,r4,r5,lsl#28 + eor r5,r9,r5,lsr#4 + eor r5,r5,r6,lsl#28 + ldrh r8,[r2,r12] @ rem_4bit[rem] + eor r6,r10,r6,lsr#4 +#ifdef __thumb2__ + it pl +#endif + ldrbpl r12,[r0,r3] + eor r6,r6,r7,lsl#28 + eor r7,r11,r7,lsr#4 + + add r11,r1,r14 + and r14,r4,#0xf @ rem + eor r7,r7,r8,lsl#16 @ ^= rem_4bit[rem] + add r14,r14,r14 + ldmia r11,{r8-r11} @ load Htbl[nhi] + eor r4,r8,r4,lsr#4 + eor r4,r4,r5,lsl#28 + eor r5,r9,r5,lsr#4 + ldrh r8,[r2,r14] @ rem_4bit[rem] + eor r5,r5,r6,lsl#28 + eor r6,r10,r6,lsr#4 + eor r6,r6,r7,lsl#28 + eor r7,r11,r7,lsr#4 +#ifdef __thumb2__ + itt pl +#endif + andpl r14,r12,#0xf0 + andpl r12,r12,#0x0f + eor r7,r7,r8,lsl#16 @ ^= rem_4bit[rem] + bpl .Loop +#if __ARM_ARCH__>=7 && defined(__ARMEL__) + rev r4,r4 + str r4,[r0,#12] +#elif defined(__ARMEB__) + str r4,[r0,#12] +#else + mov r9,r4,lsr#8 + strb r4,[r0,#12+3] + mov r10,r4,lsr#16 + strb r9,[r0,#12+2] + mov r11,r4,lsr#24 + strb r10,[r0,#12+1] + strb r11,[r0,#12] +#endif + +#if __ARM_ARCH__>=7 && defined(__ARMEL__) + rev r5,r5 + str r5,[r0,#8] +#elif defined(__ARMEB__) + str r5,[r0,#8] +#else + mov r9,r5,lsr#8 + strb r5,[r0,#8+3] + mov r10,r5,lsr#16 + strb r9,[r0,#8+2] + mov r11,r5,lsr#24 + strb r10,[r0,#8+1] + strb r11,[r0,#8] +#endif + +#if __ARM_ARCH__>=7 && defined(__ARMEL__) + rev r6,r6 + str r6,[r0,#4] +#elif defined(__ARMEB__) + str r6,[r0,#4] +#else + mov r9,r6,lsr#8 + strb r6,[r0,#4+3] + mov r10,r6,lsr#16 + strb r9,[r0,#4+2] + mov r11,r6,lsr#24 + strb r10,[r0,#4+1] + strb r11,[r0,#4] +#endif + +#if __ARM_ARCH__>=7 && defined(__ARMEL__) + rev r7,r7 + str r7,[r0,#0] +#elif defined(__ARMEB__) + str r7,[r0,#0] +#else + mov r9,r7,lsr#8 + strb r7,[r0,#0+3] + mov r10,r7,lsr#16 + strb r9,[r0,#0+2] + mov r11,r7,lsr#24 + strb r10,[r0,#0+1] + strb r11,[r0,#0] +#endif + +#if __ARM_ARCH__>=5 + ldmia sp!,{r4-r11,pc} +#else + ldmia sp!,{r4-r11,lr} + tst lr,#1 + moveq pc,lr @ be binary compatible with V4, yet + .word 0xe12fff1e @ interoperable with Thumb ISA:-) +#endif +.size gcm_gmult_4bit,.-gcm_gmult_4bit +#ifdef __ARM_NEON__ +.arch armv7-a +.fpu neon + +.global gcm_init_neon +.type gcm_init_neon,%function +.align 4 +gcm_init_neon: + vld1.64 d7,[r1]! @ load H + vmov.i8 q8,#0xe1 + vld1.64 d6,[r1] + vshl.i64 d17,#57 + vshr.u64 d16,#63 @ t0=0xc2....01 + vdup.8 q9,d7[7] + vshr.u64 d26,d6,#63 + vshr.s8 q9,#7 @ broadcast carry bit + vshl.i64 q3,q3,#1 + vand q8,q8,q9 + vorr d7,d26 @ H<<<=1 + veor q3,q3,q8 @ twisted H + vstmia r0,{q3} + + bx lr @ bx lr +.size gcm_init_neon,.-gcm_init_neon + +.global gcm_gmult_neon +.type gcm_gmult_neon,%function +.align 4 +gcm_gmult_neon: + vld1.64 d7,[r0]! @ load Xi + vld1.64 d6,[r0]! + vmov.i64 d29,#0x0000ffffffffffff + vldmia r1,{d26-d27} @ load twisted H + vmov.i64 d30,#0x00000000ffffffff +#ifdef __ARMEL__ + vrev64.8 q3,q3 +#endif + vmov.i64 d31,#0x000000000000ffff + veor d28,d26,d27 @ Karatsuba pre-processing + mov r3,#16 + b .Lgmult_neon +.size gcm_gmult_neon,.-gcm_gmult_neon + +.global gcm_ghash_neon +.type gcm_ghash_neon,%function +.align 4 +gcm_ghash_neon: + vld1.64 d1,[r0]! @ load Xi + vld1.64 d0,[r0]! + vmov.i64 d29,#0x0000ffffffffffff + vldmia r1,{d26-d27} @ load twisted H + vmov.i64 d30,#0x00000000ffffffff +#ifdef __ARMEL__ + vrev64.8 q0,q0 +#endif + vmov.i64 d31,#0x000000000000ffff + veor d28,d26,d27 @ Karatsuba pre-processing + +.Loop_neon: + vld1.64 d7,[r2]! @ load inp + vld1.64 d6,[r2]! +#ifdef __ARMEL__ + vrev64.8 q3,q3 +#endif + veor q3,q0 @ inp^=Xi +.Lgmult_neon: + vext.8 d16, d26, d26, #1 @ A1 + vmull.p8 q8, d16, d6 @ F = A1*B + vext.8 d0, d6, d6, #1 @ B1 + vmull.p8 q0, d26, d0 @ E = A*B1 + vext.8 d18, d26, d26, #2 @ A2 + vmull.p8 q9, d18, d6 @ H = A2*B + vext.8 d22, d6, d6, #2 @ B2 + vmull.p8 q11, d26, d22 @ G = A*B2 + vext.8 d20, d26, d26, #3 @ A3 + veor q8, q8, q0 @ L = E + F + vmull.p8 q10, d20, d6 @ J = A3*B + vext.8 d0, d6, d6, #3 @ B3 + veor q9, q9, q11 @ M = G + H + vmull.p8 q0, d26, d0 @ I = A*B3 + veor d16, d16, d17 @ t0 = (L) (P0 + P1) << 8 + vand d17, d17, d29 + vext.8 d22, d6, d6, #4 @ B4 + veor d18, d18, d19 @ t1 = (M) (P2 + P3) << 16 + vand d19, d19, d30 + vmull.p8 q11, d26, d22 @ K = A*B4 + veor q10, q10, q0 @ N = I + J + veor d16, d16, d17 + veor d18, d18, d19 + veor d20, d20, d21 @ t2 = (N) (P4 + P5) << 24 + vand d21, d21, d31 + vext.8 q8, q8, q8, #15 + veor d22, d22, d23 @ t3 = (K) (P6 + P7) << 32 + vmov.i64 d23, #0 + vext.8 q9, q9, q9, #14 + veor d20, d20, d21 + vmull.p8 q0, d26, d6 @ D = A*B + vext.8 q11, q11, q11, #12 + vext.8 q10, q10, q10, #13 + veor q8, q8, q9 + veor q10, q10, q11 + veor q0, q0, q8 + veor q0, q0, q10 + veor d6,d6,d7 @ Karatsuba pre-processing + vext.8 d16, d28, d28, #1 @ A1 + vmull.p8 q8, d16, d6 @ F = A1*B + vext.8 d2, d6, d6, #1 @ B1 + vmull.p8 q1, d28, d2 @ E = A*B1 + vext.8 d18, d28, d28, #2 @ A2 + vmull.p8 q9, d18, d6 @ H = A2*B + vext.8 d22, d6, d6, #2 @ B2 + vmull.p8 q11, d28, d22 @ G = A*B2 + vext.8 d20, d28, d28, #3 @ A3 + veor q8, q8, q1 @ L = E + F + vmull.p8 q10, d20, d6 @ J = A3*B + vext.8 d2, d6, d6, #3 @ B3 + veor q9, q9, q11 @ M = G + H + vmull.p8 q1, d28, d2 @ I = A*B3 + veor d16, d16, d17 @ t0 = (L) (P0 + P1) << 8 + vand d17, d17, d29 + vext.8 d22, d6, d6, #4 @ B4 + veor d18, d18, d19 @ t1 = (M) (P2 + P3) << 16 + vand d19, d19, d30 + vmull.p8 q11, d28, d22 @ K = A*B4 + veor q10, q10, q1 @ N = I + J + veor d16, d16, d17 + veor d18, d18, d19 + veor d20, d20, d21 @ t2 = (N) (P4 + P5) << 24 + vand d21, d21, d31 + vext.8 q8, q8, q8, #15 + veor d22, d22, d23 @ t3 = (K) (P6 + P7) << 32 + vmov.i64 d23, #0 + vext.8 q9, q9, q9, #14 + veor d20, d20, d21 + vmull.p8 q1, d28, d6 @ D = A*B + vext.8 q11, q11, q11, #12 + vext.8 q10, q10, q10, #13 + veor q8, q8, q9 + veor q10, q10, q11 + veor q1, q1, q8 + veor q1, q1, q10 + vext.8 d16, d27, d27, #1 @ A1 + vmull.p8 q8, d16, d7 @ F = A1*B + vext.8 d4, d7, d7, #1 @ B1 + vmull.p8 q2, d27, d4 @ E = A*B1 + vext.8 d18, d27, d27, #2 @ A2 + vmull.p8 q9, d18, d7 @ H = A2*B + vext.8 d22, d7, d7, #2 @ B2 + vmull.p8 q11, d27, d22 @ G = A*B2 + vext.8 d20, d27, d27, #3 @ A3 + veor q8, q8, q2 @ L = E + F + vmull.p8 q10, d20, d7 @ J = A3*B + vext.8 d4, d7, d7, #3 @ B3 + veor q9, q9, q11 @ M = G + H + vmull.p8 q2, d27, d4 @ I = A*B3 + veor d16, d16, d17 @ t0 = (L) (P0 + P1) << 8 + vand d17, d17, d29 + vext.8 d22, d7, d7, #4 @ B4 + veor d18, d18, d19 @ t1 = (M) (P2 + P3) << 16 + vand d19, d19, d30 + vmull.p8 q11, d27, d22 @ K = A*B4 + veor q10, q10, q2 @ N = I + J + veor d16, d16, d17 + veor d18, d18, d19 + veor d20, d20, d21 @ t2 = (N) (P4 + P5) << 24 + vand d21, d21, d31 + vext.8 q8, q8, q8, #15 + veor d22, d22, d23 @ t3 = (K) (P6 + P7) << 32 + vmov.i64 d23, #0 + vext.8 q9, q9, q9, #14 + veor d20, d20, d21 + vmull.p8 q2, d27, d7 @ D = A*B + vext.8 q11, q11, q11, #12 + vext.8 q10, q10, q10, #13 + veor q8, q8, q9 + veor q10, q10, q11 + veor q2, q2, q8 + veor q2, q2, q10 + veor q1,q1,q0 @ Karatsuba post-processing + veor q1,q1,q2 + veor d1,d1,d2 + veor d4,d4,d3 @ Xh|Xl - 256-bit result + + @ equivalent of reduction_avx from ghash-x86_64.pl + vshl.i64 q9,q0,#57 @ 1st phase + vshl.i64 q10,q0,#62 + veor q10,q10,q9 @ + vshl.i64 q9,q0,#63 + veor q10, q10, q9 @ + veor d1,d1,d20 @ + veor d4,d4,d21 + + vshr.u64 q10,q0,#1 @ 2nd phase + veor q2,q2,q0 + veor q0,q0,q10 @ + vshr.u64 q10,q10,#6 + vshr.u64 q0,q0,#1 @ + veor q0,q0,q2 @ + veor q0,q0,q10 @ + + subs r3,#16 + bne .Loop_neon + +#ifdef __ARMEL__ + vrev64.8 q0,q0 +#endif + sub r0,#16 + vst1.64 d1,[r0]! @ write out Xi + vst1.64 d0,[r0] + + bx lr @ bx lr +.size gcm_ghash_neon,.-gcm_ghash_neon +#endif +.asciz "GHASH for ARMv4/NEON, CRYPTOGAMS by <appro@openssl.org>" +.align 2 diff --git a/third_party/boringssl/include/aes-gcm.h b/third_party/boringssl/include/aes-gcm.h new file mode 100644 index 0000000000..77ca52a3ed --- /dev/null +++ b/third_party/boringssl/include/aes-gcm.h @@ -0,0 +1,146 @@ +/* ==================================================================== + * Copyright (c) 2008 The OpenSSL Project. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * + * 3. All advertising materials mentioning features or use of this + * software must display the following acknowledgment: + * "This product includes software developed by the OpenSSL Project + * for use in the OpenSSL Toolkit. (http://www.openssl.org/)" + * + * 4. The names "OpenSSL Toolkit" and "OpenSSL Project" must not be used to + * endorse or promote products derived from this software without + * prior written permission. For written permission, please contact + * openssl-core@openssl.org. + * + * 5. Products derived from this software may not be called "OpenSSL" + * nor may "OpenSSL" appear in their names without prior written + * permission of the OpenSSL Project. + * + * 6. Redistributions of any form whatsoever must retain the following + * acknowledgment: + * "This product includes software developed by the OpenSSL Project + * for use in the OpenSSL Toolkit (http://www.openssl.org/)" + * + * THIS SOFTWARE IS PROVIDED BY THE OpenSSL PROJECT ``AS IS'' AND ANY + * EXPRESSED OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE OpenSSL PROJECT OR + * ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT + * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + * ==================================================================== */ + +#ifndef __CROS_EC_AES_GCM_H +#define __CROS_EC_AES_GCM_H + +#include "common.h" +#include "util.h" + +// block128_f is the type of a 128-bit, block cipher. +typedef void (*block128_f)(const uint8_t in[16], uint8_t out[16], + const void *key); + +// GCM definitions +typedef struct { uint64_t hi,lo; } u128; + +// gmult_func multiplies |Xi| by the GCM key and writes the result back to +// |Xi|. +typedef void (*gmult_func)(uint64_t Xi[2], const u128 Htable[16]); + +// ghash_func repeatedly multiplies |Xi| by the GCM key and adds in blocks from +// |inp|. The result is written back to |Xi| and the |len| argument must be a +// multiple of 16. +typedef void (*ghash_func)(uint64_t Xi[2], const u128 Htable[16], + const uint8_t *inp, size_t len); + +// This differs from upstream's |gcm128_context| in that it does not have the +// |key| pointer, in order to make it |memcpy|-friendly. Rather the key is +// passed into each call that needs it. +struct gcm128_context { + // Following 6 names follow names in GCM specification + union { + uint64_t u[2]; + uint32_t d[4]; + uint8_t c[16]; + size_t t[16 / sizeof(size_t)]; + } Yi, EKi, EK0, len, Xi; + + // Note that the order of |Xi|, |H| and |Htable| is fixed by the MOVBE-based, + // x86-64, GHASH assembly. + u128 H; + u128 Htable[16]; + gmult_func gmult; + ghash_func ghash; + + unsigned int mres, ares; + block128_f block; +}; + + +// GCM. +// +// This API differs from the upstream API slightly. The |GCM128_CONTEXT| does +// not have a |key| pointer that points to the key as upstream's version does. +// Instead, every function takes a |key| parameter. This way |GCM128_CONTEXT| +// can be safely copied. + +typedef struct gcm128_context GCM128_CONTEXT; + +// CRYPTO_gcm128_init initialises |ctx| to use |block| (typically AES) with +// the given key. |block_is_hwaes| is one if |block| is |aes_hw_encrypt|. +void CRYPTO_gcm128_init(GCM128_CONTEXT *ctx, const void *key, + block128_f block, int block_is_hwaes); + +// CRYPTO_gcm128_setiv sets the IV (nonce) for |ctx|. The |key| must be the +// same key that was passed to |CRYPTO_gcm128_init|. +void CRYPTO_gcm128_setiv(GCM128_CONTEXT *ctx, const void *key, + const uint8_t *iv, size_t iv_len); + +// CRYPTO_gcm128_aad sets the authenticated data for an instance of GCM. +// This must be called before and data is encrypted. It returns one on success +// and zero otherwise. +int CRYPTO_gcm128_aad(GCM128_CONTEXT *ctx, const uint8_t *aad, + size_t len); + +// CRYPTO_gcm128_encrypt encrypts |len| bytes from |in| to |out|. The |key| +// must be the same key that was passed to |CRYPTO_gcm128_init|. It returns one +// on success and zero otherwise. +int CRYPTO_gcm128_encrypt(GCM128_CONTEXT *ctx, const void *key, + const uint8_t *in, uint8_t *out, + size_t len); + +// CRYPTO_gcm128_decrypt decrypts |len| bytes from |in| to |out|. The |key| +// must be the same key that was passed to |CRYPTO_gcm128_init|. It returns one +// on success and zero otherwise. +int CRYPTO_gcm128_decrypt(GCM128_CONTEXT *ctx, const void *key, + const uint8_t *in, uint8_t *out, + size_t len); + +// CRYPTO_gcm128_finish calculates the authenticator and compares it against +// |len| bytes of |tag|. It returns one on success and zero otherwise. +int CRYPTO_gcm128_finish(GCM128_CONTEXT *ctx, const uint8_t *tag, + size_t len); + +// CRYPTO_gcm128_tag calculates the authenticator and copies it into |tag|. +// The minimum of |len| and 16 bytes are copied into |tag|. +void CRYPTO_gcm128_tag(GCM128_CONTEXT *ctx, uint8_t *tag, + size_t len); + + +#endif // __CROS_EC_AES_GCM_H diff --git a/third_party/boringssl/include/aes.h b/third_party/boringssl/include/aes.h new file mode 100644 index 0000000000..6418a350eb --- /dev/null +++ b/third_party/boringssl/include/aes.h @@ -0,0 +1,129 @@ +/* ==================================================================== + * Copyright (c) 2002-2006 The OpenSSL Project. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * + * 3. All advertising materials mentioning features or use of this + * software must display the following acknowledgment: + * "This product includes software developed by the OpenSSL Project + * for use in the OpenSSL Toolkit. (http://www.openssl.org/)" + * + * 4. The names "OpenSSL Toolkit" and "OpenSSL Project" must not be used to + * endorse or promote products derived from this software without + * prior written permission. For written permission, please contact + * openssl-core@openssl.org. + * + * 5. Products derived from this software may not be called "OpenSSL" + * nor may "OpenSSL" appear in their names without prior written + * permission of the OpenSSL Project. + * + * 6. Redistributions of any form whatsoever must retain the following + * acknowledgment: + * "This product includes software developed by the OpenSSL Project + * for use in the OpenSSL Toolkit (http://www.openssl.org/)" + * + * THIS SOFTWARE IS PROVIDED BY THE OpenSSL PROJECT ``AS IS'' AND ANY + * EXPRESSED OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE OpenSSL PROJECT OR + * ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT + * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + * OF THE POSSIBILITY OF SUCH DAMAGE. + * ==================================================================== */ + +#ifndef __CROS_EC_AES_H +#define __CROS_EC_AES_H + +#include <stdint.h> + +#define AES_ENCRYPT 1 +#define AES_DECRYPT 0 + +/* AES_MAXNR is the maximum number of AES rounds. */ +#define AES_MAXNR 14 + +#define AES_BLOCK_SIZE 16 + +/* + * aes_key_st should be an opaque type, but EVP requires that the size be + * known. + */ +struct aes_key_st { + uint32_t rd_key[4 * (AES_MAXNR + 1)]; + unsigned rounds; +}; +typedef struct aes_key_st AES_KEY; + +/* + * These functions are provided by either common/aes.c, or assembly code, + * and should not be called directly. + */ +void aes_nohw_encrypt(const uint8_t *in, uint8_t *out, const AES_KEY *key); +void aes_nohw_decrypt(const uint8_t *in, uint8_t *out, const AES_KEY *key); +int aes_nohw_set_encrypt_key(const uint8_t *key, unsigned bits, + AES_KEY *aeskey); +int aes_nohw_set_decrypt_key(const uint8_t *key, unsigned bits, + AES_KEY *aeskey); + +/** + * AES_set_encrypt_key configures |aeskey| to encrypt with the |bits|-bit key, + * |key|. + * + * WARNING: unlike other OpenSSL functions, this returns zero on success and a + * negative number on error. + */ +static inline int AES_set_encrypt_key(const uint8_t *key, unsigned int bits, + AES_KEY *aeskey) +{ + return aes_nohw_set_encrypt_key(key, bits, aeskey); +} + +/** + * AES_set_decrypt_key configures |aeskey| to decrypt with the |bits|-bit key, + * |key|. + * + * WARNING: unlike other OpenSSL functions, this returns zero on success and a + * negative number on error. + */ +static inline int AES_set_decrypt_key(const uint8_t *key, unsigned int bits, + AES_KEY *aeskey) +{ + return aes_nohw_set_decrypt_key(key, bits, aeskey); +} + +/** + * AES_encrypt encrypts a single block from |in| to |out| with |key|. The |in| + * and |out| pointers may overlap. + */ +static inline void AES_encrypt(const uint8_t *in, uint8_t *out, + const AES_KEY *key) +{ + aes_nohw_encrypt(in, out, key); +} + +/** + * AES_decrypt decrypts a single block from |in| to |out| with |key|. The |in| + * and |out| pointers may overlap. + */ +static inline void AES_decrypt(const uint8_t *in, uint8_t *out, + const AES_KEY *key) +{ + aes_nohw_decrypt(in, out, key); +} + +#endif /* __CROS_EC_AES_H */ diff --git a/util/config_allowed.txt b/util/config_allowed.txt index b2a32b11aa..9ca270e42a 100644 --- a/util/config_allowed.txt +++ b/util/config_allowed.txt @@ -37,6 +37,8 @@ CONFIG_ADC_PROFILE_SINGLE CONFIG_ADC_SAMPLE_TIME CONFIG_ADC_VOLTAGE_COMPARATOR CONFIG_ADC_WATCHDOG +CONFIG_AES +CONFIG_AES_GCM CONFIG_ALS_AL3010 CONFIG_ALS_BH1730 CONFIG_ALS_BH1730_LUXTH_PARAMS |