@pinkparrot/qsafe-mayo-wasm 0.0.3

package/mayo-c/src/common/aes_ctr.h

@@ -0,0 +1,32 @@
+// SPDX-License-Identifier: Apache-2.0
+
+#ifndef AESCTR_H
+#define AESCTR_H
+
+#include <stddef.h>
+#include <stdint.h>
+
+void AES_256_ECB(const uint8_t *input, const uint8_t *key, uint8_t *output);
+#define AES_ECB_encrypt AES_256_ECB
+
+#ifdef ENABLE_AESNI
+int AES_128_CTR_NI(unsigned char *output, size_t outputByteLen,
+                   const unsigned char *input, size_t inputByteLen);
+int AES_128_CTR_4R_NI(unsigned char *output, size_t outputByteLen,
+                      const unsigned char *input, size_t inputByteLen);
+#define AES_128_CTR AES_128_CTR_NI
+#else
+#ifdef ENABLE_AESNEON
+int AES_128_CTR_NEON(unsigned char *output, size_t outputByteLen,
+                   const unsigned char *input, size_t inputByteLen);
+int AES_128_CTR_4R_NI(unsigned char *output, size_t outputByteLen,
+                      const unsigned char *input, size_t inputByteLen);
+#define AES_128_CTR AES_128_CTR_NEON
+#else
+int AES_128_CTR(unsigned char *output, size_t outputByteLen,
+                const unsigned char *input, size_t inputByteLen);
+#endif
+#endif
+
+#endif
+

package/mayo-c/src/common/aes_neon.c

@@ -0,0 +1,201 @@
+// SPDX-License-Identifier: MIT and Public Domain
+
+#ifdef ENABLE_AESNEON
+
+// Code taken from https://github.com/ChristerKnorborg/post-quantum-signature-schemes/blob/main/src/genkat/aes_arm.c
+
+/* ARMv8 AES Implementation adapted
+  from liboqs/src/common/aes which
+  in turn takes it from:
+  crypto_core/aes128ncrypt/dolbeau/aesenc-int
+ (https://bench.cr.yp.to/supercop.html) */
+
+#include "mem.h"
+#include <string.h>
+#include <stdlib.h>
+
+#if defined(__arm__) || defined(__aarch32__) || defined(__arm64__) || defined(__aarch64__) || defined(_M_ARM) || defined(_M_ARM64)
+# if defined(__GNUC__)
+#  include <stdint.h>
+# endif
+# if defined(__ARM_NEON) || defined(_MSC_VER)
+#  include <arm_neon.h>
+# endif
+/* GCC and LLVM Clang, but not Apple Clang */
+# if defined(__GNUC__) && !defined(__apple_build_version__)
+#  if defined(__ARM_ACLE) || defined(__ARM_FEATURE_CRYPTO)
+#   include <arm_acle.h>
+#  endif
+# endif
+#endif  /* ARM Headers */
+
+// aes s-box
+static const uint8_t sbox[256] = {
+  //0     1    2      3     4    5     6     7      8    9     A      B    C     D     E     F
+  0x63, 0x7c, 0x77, 0x7b, 0xf2, 0x6b, 0x6f, 0xc5, 0x30, 0x01, 0x67, 0x2b, 0xfe, 0xd7, 0xab, 0x76,
+  0xca, 0x82, 0xc9, 0x7d, 0xfa, 0x59, 0x47, 0xf0, 0xad, 0xd4, 0xa2, 0xaf, 0x9c, 0xa4, 0x72, 0xc0,
+  0xb7, 0xfd, 0x93, 0x26, 0x36, 0x3f, 0xf7, 0xcc, 0x34, 0xa5, 0xe5, 0xf1, 0x71, 0xd8, 0x31, 0x15,
+  0x04, 0xc7, 0x23, 0xc3, 0x18, 0x96, 0x05, 0x9a, 0x07, 0x12, 0x80, 0xe2, 0xeb, 0x27, 0xb2, 0x75,
+  0x09, 0x83, 0x2c, 0x1a, 0x1b, 0x6e, 0x5a, 0xa0, 0x52, 0x3b, 0xd6, 0xb3, 0x29, 0xe3, 0x2f, 0x84,
+  0x53, 0xd1, 0x00, 0xed, 0x20, 0xfc, 0xb1, 0x5b, 0x6a, 0xcb, 0xbe, 0x39, 0x4a, 0x4c, 0x58, 0xcf,
+  0xd0, 0xef, 0xaa, 0xfb, 0x43, 0x4d, 0x33, 0x85, 0x45, 0xf9, 0x02, 0x7f, 0x50, 0x3c, 0x9f, 0xa8,
+  0x51, 0xa3, 0x40, 0x8f, 0x92, 0x9d, 0x38, 0xf5, 0xbc, 0xb6, 0xda, 0x21, 0x10, 0xff, 0xf3, 0xd2,
+  0xcd, 0x0c, 0x13, 0xec, 0x5f, 0x97, 0x44, 0x17, 0xc4, 0xa7, 0x7e, 0x3d, 0x64, 0x5d, 0x19, 0x73,
+  0x60, 0x81, 0x4f, 0xdc, 0x22, 0x2a, 0x90, 0x88, 0x46, 0xee, 0xb8, 0x14, 0xde, 0x5e, 0x0b, 0xdb,
+  0xe0, 0x32, 0x3a, 0x0a, 0x49, 0x06, 0x24, 0x5c, 0xc2, 0xd3, 0xac, 0x62, 0x91, 0x95, 0xe4, 0x79,
+  0xe7, 0xc8, 0x37, 0x6d, 0x8d, 0xd5, 0x4e, 0xa9, 0x6c, 0x56, 0xf4, 0xea, 0x65, 0x7a, 0xae, 0x08,
+  0xba, 0x78, 0x25, 0x2e, 0x1c, 0xa6, 0xb4, 0xc6, 0xe8, 0xdd, 0x74, 0x1f, 0x4b, 0xbd, 0x8b, 0x8a,
+  0x70, 0x3e, 0xb5, 0x66, 0x48, 0x03, 0xf6, 0x0e, 0x61, 0x35, 0x57, 0xb9, 0x86, 0xc1, 0x1d, 0x9e,
+  0xe1, 0xf8, 0x98, 0x11, 0x69, 0xd9, 0x8e, 0x94, 0x9b, 0x1e, 0x87, 0xe9, 0xce, 0x55, 0x28, 0xdf,
+  0x8c, 0xa1, 0x89, 0x0d, 0xbf, 0xe6, 0x42, 0x68, 0x41, 0x99, 0x2d, 0x0f, 0xb0, 0x54, 0xbb, 0x16 };
+
+
+
+
+// subword algorithm used in the aes key scheduling.
+uint32_t subword(uint32_t word) {
+    return (uint32_t)sbox[(word >> 24) & 0xFF] << 24 |
+           (uint32_t)sbox[(word >> 16) & 0xFF] << 16 |
+           (uint32_t)sbox[(word >> 8) & 0xFF] << 8 |
+           (uint32_t)sbox[word & 0xFF];
+}
+
+uint32x4_t aeskeygenassist(uint32x4_t a32, uint8_t rcon) {
+    // Extract words X1 and X3
+    uint32_t X1 = vgetq_lane_u32(a32, 1);
+    uint32_t X3 = vgetq_lane_u32(a32, 3);
+
+    // Apply SubWord (Manually implemented using AES S-box lookup table)
+    uint32_t subX1 = subword(X1);  // Implement this function based on AES S-box
+    uint32_t subX3 = subword(X3);  // Implement this function based on AES S-box
+
+    // RotWord
+    uint32_t rotX1 = (subX1 >> 8) | (subX1 << 24);
+    uint32_t rotX3 = (subX3 >> 8) | (subX3 << 24);
+
+    // Apply RCON
+    rotX1 ^= (uint32_t)rcon;
+    rotX3 ^= (uint32_t)rcon;
+
+    // Assemble the final vector
+    uint32x4_t result = a32;
+    result = vsetq_lane_u32(subX1, result, 0);
+    result = vsetq_lane_u32(rotX1, result, 1);
+    result = vsetq_lane_u32(subX3, result, 2);
+    result = vsetq_lane_u32(rotX3, result, 3);
+
+    return result;
+}
+
+static void aes_setkey_encrypt(const unsigned char *key, uint8x16_t rkeys[]) {
+    uint8x16_t key0 = vld1q_u8((const uint8_t *)(key));
+    uint32x4_t temp0, temp1, temp4;
+    int idx = 0;
+
+    temp0 = vreinterpretq_u32_u8(key0);
+    temp4 = vdupq_n_u32(0);
+
+    #define BLOCK1(IMM)                                                     \
+      temp1 = aeskeygenassist(temp0, IMM);                                  \
+      rkeys[idx++] = vreinterpretq_u8_u32(temp0);                           \
+      temp4 = vsetq_lane_u32(vgetq_lane_u32(temp0, 0), temp4, 1);           \
+      temp4 = vsetq_lane_u32(vgetq_lane_u32(temp0, 1), temp4, 2);           \
+      temp4 = vsetq_lane_u32(vgetq_lane_u32(temp0, 2), temp4, 3);           \
+      temp0 = veorq_u32(temp0, temp4);                                      \
+      temp0 = vreinterpretq_u32_u64((vsetq_lane_u64(((uint64_t) veor_u32(vget_high_u32(temp0), vget_low_u32(temp0))), vreinterpretq_u64_u32(temp0), 1)));      \
+      temp1 = vdupq_n_u32(vgetq_lane_u32(temp1, 3));                        \
+      temp0 = veorq_u32(temp0, temp1);                                      \
+
+    BLOCK1(0x01);
+    BLOCK1(0x02);
+    BLOCK1(0x04);
+    BLOCK1(0x08);
+    BLOCK1(0x10);
+    BLOCK1(0x20);
+    BLOCK1(0x40);
+    BLOCK1(0x80);
+    BLOCK1(0x1b);
+    BLOCK1(0x36);
+    rkeys[idx++] = vreinterpretq_u8_u32(temp0);
+}
+
+void arm_aes128_load_schedule(const uint8_t *key, void **_schedule) {
+    *_schedule = malloc(11 * sizeof(uint8x16_t));
+    // assert(*_schedule != NULL);
+    uint8x16_t *schedule = (uint8x16_t *)*_schedule;
+    aes_setkey_encrypt(key, schedule);
+}
+
+
+// AES encryption using NEON intrinsics. Round constants are 0 as the function mimics Intel AVX2 implementation,
+// which applies in order: ShiftRows, SubBytes, MixColumns, AddRoundKey.
+// vaeseq_u8 applies SubBytes, ShiftRows, AddRoundKey.
+static void arm_aes128_encrypt(const uint8x16_t rkeys[11], uint8x16_t nv, unsigned char *out) {
+
+    uint8x16_t temp = vaeseq_u8(nv, rkeys[0]);
+
+    temp = vaesmcq_u8(temp);
+    temp = vaeseq_u8(temp, rkeys[1]);
+
+    temp = vaesmcq_u8(temp);
+    temp = vaeseq_u8(temp, rkeys[2]);
+
+    temp = vaesmcq_u8(temp);
+    temp = vaeseq_u8(temp, rkeys[3]);
+
+    temp = vaesmcq_u8(temp);
+    temp = vaeseq_u8(temp, rkeys[4]);
+
+    temp = vaesmcq_u8(temp);
+    temp = vaeseq_u8(temp, rkeys[5]);
+
+    temp = vaesmcq_u8(temp);
+    temp = vaeseq_u8(temp, rkeys[6]);
+
+    temp = vaesmcq_u8(temp);
+    temp = vaeseq_u8(temp, rkeys[7]);
+
+    temp = vaesmcq_u8(temp);
+    temp = vaeseq_u8(temp, rkeys[8]);
+
+    temp = vaesmcq_u8(temp);
+    temp = vaeseq_u8(temp, rkeys[9]);
+
+    temp = veorq_u8(temp, rkeys[10]);
+    vst1q_u8(out, temp);
+}
+
+void arm_aes128_free_schedule(void *schedule) {
+    if (schedule != NULL) {
+        mayo_secure_free(schedule, 11 * sizeof(uint16x8_t));
+    }
+}
+
+
+static void arm_aes128_ctr_enc_sch(const void *schedule, uint8_t *out,
+                                        size_t out_len) {
+    uint8x16_t mask = {0, 1, 2, 3, 4, 5, 6, 7, 15, 14, 13, 12, 11, 10, 9, 8};
+    uint8x16_t block = vdupq_n_u8(0); // Initialize block to zero
+    while (out_len >= 16) {
+        arm_aes128_encrypt(schedule, block, out);
+        out += 16;
+        out_len -= 16;
+        block = vqtbl1q_u8(vreinterpretq_u8_u64(vaddq_u64(vreinterpretq_u64_u8(vqtbl1q_u8(block, mask)), (uint64x2_t) {0,1}) ), mask);
+    }
+    if (out_len > 0) {
+        uint8_t tmp[16];
+        arm_aes128_encrypt(schedule, block, tmp);
+        memcpy(out, tmp, out_len);
+    }
+}
+
+int AES_128_CTR_NEON(unsigned char *output, size_t outputByteLen,
+                   const unsigned char *input) {
+    void *schedule = NULL;
+    arm_aes128_load_schedule(input, &schedule);
+    arm_aes128_ctr_enc_sch(schedule, output, outputByteLen);
+    arm_aes128_free_schedule(schedule);
+    return (int)outputByteLen;
+}
+
+#endif

package/mayo-c/src/common/debug_bench_tools.h

@@ -0,0 +1,69 @@
+// SPDX-License-Identifier: Apache-2.0
+
+#ifndef DEBUG_BENCH_TOOLS_H
+#define DEBUG_BENCH_TOOLS_H
+
+#include <stdio.h>
+#include <sys/time.h>
+
+static inline int64_t cpucycles(void) {
+#if (defined(TARGET_AMD64) || defined(TARGET_X86))
+    unsigned int hi, lo;
+
+    asm volatile ("rdtsc" : "=a" (lo), "=d"(hi));
+    return ((int64_t) lo) | (((int64_t) hi) << 32);
+#elif (defined(TARGET_S390X))
+    uint64_t tod;
+    asm volatile("stckf %0\n" : "=Q" (tod) : : "cc");
+    return (tod * 1000 / 4096);
+#else
+    struct timespec time;
+    clock_gettime(CLOCK_REALTIME, &time);
+    return (int64_t)(time.tv_sec * 1e9 + time.tv_nsec);
+#endif
+}
+
+#ifdef TICTOC
+#define TIC printf("\n"); \
+        int64_t tic_toc_cycles = cpucycles();
+
+#define TOC(name) printf(" %-30s cycles: %lu \n", name, cpucycles() - tic_toc_cycles); \
+        tic_toc_cycles = cpucycles();
+#else
+#define TIC
+#define TOC(name)
+#endif
+
+#ifdef MAYO_AVX
+
+#include <immintrin.h>
+
+static inline void print_avx2(__m256i a){
+    unsigned char *temp = (unsigned char*) &a;
+    for (size_t i = 0; i < 32; i++)
+    {
+        printf("%X", temp[i] & 0xf);
+        printf("%X", temp[i] >> 4);
+        if(i%4 == 3){
+            printf(" ");
+        }
+    }
+    printf("\n");
+}
+
+static inline void print_avx2_(__m256i a){
+    unsigned char *temp = (unsigned char*) &a;
+    for (size_t i = 0; i < 32; i++)
+    {
+        printf("%X", temp[i] & 0xf);
+        printf("%X", temp[i] >> 4);
+        if(i%4 == 3){
+            printf(" ");
+        }
+    }
+}
+
+#endif
+
+#endif
+