wolfSSL/wolfssl
1
0
Fork 1
mirror of https://github.com/wolfSSL/wolfssl.git synced 2025-07-30 18:57:27 +02:00
Code Issues Packages Projects Releases Wiki Activity

Fixes for Windows AESNI with clang. Improve 32-bit support.

Browse Source
This commit is contained in:
David Garske
2022-03-17 14:05:24 -07:00
parent aa8e5a29d4
commit f954aef973
1 changed files with 35 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

46
wolfcrypt/src/aes.c
View File

@ -4773,11 +4773,22 @@ void AES_GCM_decrypt_avx2(const unsigned char *in, unsigned char *out,
#else /* _MSC_VER */ #else /* _MSC_VER */
#define S(w,z) ((char)((unsigned long long)(w) >> (8*(7-(z))) & 0xFF)) /* AESNI with Microsoft */
#define M128_INIT(x,y) { S((x),7), S((x),6), S((x),5), S((x),4), \ #ifdef __clang__
S((x),3), S((x),2), S((x),1), S((x),0), \ /* With Clang the __m128i in emmintrin.h is union using:
S((y),7), S((y),6), S((y),5), S((y),4), \ * "unsigned __int64 m128i_u64[2];"
S((y),3), S((y),2), S((y),1), S((y),0) } * Notes: Must add "-maes -msse4.1 -mpclmul" to compiler flags.
* Must mark "aes_asm.asm" as included/compiled C file.
*/
#define M128_INIT(x,y) { (long long)x, (long long)y }
#else
/* Typically this is array of 16 int8's */
#define S(w,z) ((char)((unsigned long long)(w) >> (8*(7-(z))) & 0xFF))
#define M128_INIT(x,y) { S((x),7), S((x),6), S((x),5), S((x),4), \
S((x),3), S((x),2), S((x),1), S((x),0), \
S((y),7), S((y),6), S((y),5), S((y),4), \
S((y),3), S((y),2), S((y),1), S((y),0) }
#endif
static const __m128i MOD2_128 = static const __m128i MOD2_128 =
M128_INIT(0x1, (long long int)0xc200000000000000UL); M128_INIT(0x1, (long long int)0xc200000000000000UL);
@ -4862,6 +4873,21 @@ do \
} \ } \
while (0) while (0)
#ifdef _M_X64
/* 64-bit */
#define AES_GCM_INSERT_EPI(tmp1, a, b) \
tmp1 = _mm_insert_epi64(tmp1, ((word64)(a))*8, 0); \
tmp1 = _mm_insert_epi64(tmp1, ((word64)(b))*8, 1);
#else
/* 32-bit */
#define AES_GCM_INSERT_EPI(tmp1, a, b) \
tmp1 = _mm_insert_epi32(tmp1, ((int)(a))*8, 0); \
tmp1 = _mm_insert_epi32(tmp1, 0, 1); \
tmp1 = _mm_insert_epi32(tmp1, ((int)(b))*8, 2); \
tmp1 = _mm_insert_epi32(tmp1, 0, 3);
#endif
#define aes_gcm_calc_iv(KEY, ivec, ibytes, nr, H, Y, T) \ #define aes_gcm_calc_iv(KEY, ivec, ibytes, nr, H, Y, T) \
do \ do \
{ \ { \
@ -4906,8 +4932,7 @@ do \
Y = _mm_xor_si128(Y, tmp1); \ Y = _mm_xor_si128(Y, tmp1); \
Y = gfmul_sw(Y, H); \ Y = gfmul_sw(Y, H); \
} \ } \
tmp1 = _mm_insert_epi64(tmp1, ibytes*8, 0); \ AES_GCM_INSERT_EPI(tmp1, ibytes, 0); \
tmp1 = _mm_insert_epi64(tmp1, 0, 1); \
Y = _mm_xor_si128(Y, tmp1); \ Y = _mm_xor_si128(Y, tmp1); \
Y = gfmul_sw(Y, H); \ Y = gfmul_sw(Y, H); \
Y = _mm_shuffle_epi8(Y, BSWAP_MASK); /* Compute E(K, Y0) */ \ Y = _mm_shuffle_epi8(Y, BSWAP_MASK); /* Compute E(K, Y0) */ \
@ -5539,8 +5564,7 @@ static WARN_UNUSED_RESULT int AES_GCM_encrypt(
X =_mm_xor_si128(X, tmp1); X =_mm_xor_si128(X, tmp1);
X = gfmul_shifted(X, H); X = gfmul_shifted(X, H);
} }
tmp1 = _mm_insert_epi64(tmp1, ((word64)nbytes)*8, 0); AES_GCM_INSERT_EPI(tmp1, nbytes, abytes);
tmp1 = _mm_insert_epi64(tmp1, ((word64)abytes)*8, 1);
X = _mm_xor_si128(X, tmp1); X = _mm_xor_si128(X, tmp1);
X = gfmul_shifted(X, H); X = gfmul_shifted(X, H);
X = _mm_shuffle_epi8(X, BSWAP_MASK); X = _mm_shuffle_epi8(X, BSWAP_MASK);
@ -5866,8 +5890,8 @@ static WARN_UNUSED_RESULT int AES_GCM_decrypt(
X = gfmul_shifted(XV, H); X = gfmul_shifted(XV, H);
} }
tmp1 = _mm_insert_epi64(tmp1, ((word64)nbytes)*8, 0); AES_GCM_INSERT_EPI(tmp1, nbytes, abytes);
tmp1 = _mm_insert_epi64(tmp1, ((word64)abytes)*8, 1);
/* 128 x 128 Carryless Multiply */ /* 128 x 128 Carryless Multiply */
X = _mm_xor_si128(X, tmp1); X = _mm_xor_si128(X, tmp1);
X = gfmul_shifted(X, H); X = gfmul_shifted(X, H);