diff options
author | shadchin <shadchin@yandex-team.ru> | 2022-02-10 16:44:30 +0300 |
---|---|---|
committer | Daniil Cherednik <dcherednik@yandex-team.ru> | 2022-02-10 16:44:30 +0300 |
commit | 2598ef1d0aee359b4b6d5fdd1758916d5907d04f (patch) | |
tree | 012bb94d777798f1f56ac1cec429509766d05181 /contrib/libs/lzmasdk | |
parent | 6751af0b0c1b952fede40b19b71da8025b5d8bcf (diff) | |
download | ydb-2598ef1d0aee359b4b6d5fdd1758916d5907d04f.tar.gz |
Restoring authorship annotation for <shadchin@yandex-team.ru>. Commit 1 of 2.
Diffstat (limited to 'contrib/libs/lzmasdk')
-rw-r--r-- | contrib/libs/lzmasdk/7zVersion.h | 54 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/Aes.c | 612 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/Aes.h | 76 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/AesOpt.c | 376 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/Bra.c | 460 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/Bra.h | 128 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/Bra86.c | 164 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/BraIA64.c | 106 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/CpuArch.c | 436 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/CpuArch.h | 672 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/Lzma2Dec.c | 976 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/Lzma2Dec.h | 240 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/Lzma2Enc.c | 1606 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/Lzma2Enc.h | 110 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/LzmaEnc.c | 5968 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/LzmaEnc.h | 166 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/MtCoder.h | 282 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/MtDec.h | 402 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/RotateDefs.h | 60 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/Sha256.c | 496 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/Sha256.h | 52 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/Threads.h | 136 | ||||
-rw-r--r-- | contrib/libs/lzmasdk/ya.make | 20 |
23 files changed, 6799 insertions, 6799 deletions
diff --git a/contrib/libs/lzmasdk/7zVersion.h b/contrib/libs/lzmasdk/7zVersion.h index 0074c64be9..fcc1ec405e 100644 --- a/contrib/libs/lzmasdk/7zVersion.h +++ b/contrib/libs/lzmasdk/7zVersion.h @@ -1,27 +1,27 @@ -#define MY_VER_MAJOR 19
-#define MY_VER_MINOR 00
-#define MY_VER_BUILD 0
-#define MY_VERSION_NUMBERS "19.00"
-#define MY_VERSION MY_VERSION_NUMBERS
-
-#ifdef MY_CPU_NAME
- #define MY_VERSION_CPU MY_VERSION " (" MY_CPU_NAME ")"
-#else
- #define MY_VERSION_CPU MY_VERSION
-#endif
-
-#define MY_DATE "2019-02-21"
-#undef MY_COPYRIGHT
-#undef MY_VERSION_COPYRIGHT_DATE
-#define MY_AUTHOR_NAME "Igor Pavlov"
-#define MY_COPYRIGHT_PD "Igor Pavlov : Public domain"
-#define MY_COPYRIGHT_CR "Copyright (c) 1999-2018 Igor Pavlov"
-
-#ifdef USE_COPYRIGHT_CR
- #define MY_COPYRIGHT MY_COPYRIGHT_CR
-#else
- #define MY_COPYRIGHT MY_COPYRIGHT_PD
-#endif
-
-#define MY_COPYRIGHT_DATE MY_COPYRIGHT " : " MY_DATE
-#define MY_VERSION_COPYRIGHT_DATE MY_VERSION_CPU " : " MY_COPYRIGHT " : " MY_DATE
+#define MY_VER_MAJOR 19 +#define MY_VER_MINOR 00 +#define MY_VER_BUILD 0 +#define MY_VERSION_NUMBERS "19.00" +#define MY_VERSION MY_VERSION_NUMBERS + +#ifdef MY_CPU_NAME + #define MY_VERSION_CPU MY_VERSION " (" MY_CPU_NAME ")" +#else + #define MY_VERSION_CPU MY_VERSION +#endif + +#define MY_DATE "2019-02-21" +#undef MY_COPYRIGHT +#undef MY_VERSION_COPYRIGHT_DATE +#define MY_AUTHOR_NAME "Igor Pavlov" +#define MY_COPYRIGHT_PD "Igor Pavlov : Public domain" +#define MY_COPYRIGHT_CR "Copyright (c) 1999-2018 Igor Pavlov" + +#ifdef USE_COPYRIGHT_CR + #define MY_COPYRIGHT MY_COPYRIGHT_CR +#else + #define MY_COPYRIGHT MY_COPYRIGHT_PD +#endif + +#define MY_COPYRIGHT_DATE MY_COPYRIGHT " : " MY_DATE +#define MY_VERSION_COPYRIGHT_DATE MY_VERSION_CPU " : " MY_COPYRIGHT " : " MY_DATE diff --git a/contrib/libs/lzmasdk/Aes.c b/contrib/libs/lzmasdk/Aes.c index 8f7d50ea24..dc85b78788 100644 --- a/contrib/libs/lzmasdk/Aes.c +++ b/contrib/libs/lzmasdk/Aes.c @@ -1,306 +1,306 @@ -/* Aes.c -- AES encryption / decryption
-2017-01-24 : Igor Pavlov : Public domain */
-
-#include "Precomp.h"
-
-#include "Aes.h"
-#include "CpuArch.h"
-
-static UInt32 T[256 * 4];
-static const Byte Sbox[256] = {
- 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};
-
-void MY_FAST_CALL AesCbc_Encode(UInt32 *ivAes, Byte *data, size_t numBlocks);
-void MY_FAST_CALL AesCbc_Decode(UInt32 *ivAes, Byte *data, size_t numBlocks);
-void MY_FAST_CALL AesCtr_Code(UInt32 *ivAes, Byte *data, size_t numBlocks);
-
-void MY_FAST_CALL AesCbc_Encode_Intel(UInt32 *ivAes, Byte *data, size_t numBlocks);
-void MY_FAST_CALL AesCbc_Decode_Intel(UInt32 *ivAes, Byte *data, size_t numBlocks);
-void MY_FAST_CALL AesCtr_Code_Intel(UInt32 *ivAes, Byte *data, size_t numBlocks);
-
-AES_CODE_FUNC g_AesCbc_Encode;
-AES_CODE_FUNC g_AesCbc_Decode;
-AES_CODE_FUNC g_AesCtr_Code;
-
-static UInt32 D[256 * 4];
-static Byte InvS[256];
-
-static const Byte Rcon[11] = { 0x00, 0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80, 0x1b, 0x36 };
-
-#define xtime(x) ((((x) << 1) ^ (((x) & 0x80) != 0 ? 0x1B : 0)) & 0xFF)
-
-#define Ui32(a0, a1, a2, a3) ((UInt32)(a0) | ((UInt32)(a1) << 8) | ((UInt32)(a2) << 16) | ((UInt32)(a3) << 24))
-
-#define gb0(x) ( (x) & 0xFF)
-#define gb1(x) (((x) >> ( 8)) & 0xFF)
-#define gb2(x) (((x) >> (16)) & 0xFF)
-#define gb3(x) (((x) >> (24)))
-
-#define gb(n, x) gb ## n(x)
-
-#define TT(x) (T + (x << 8))
-#define DD(x) (D + (x << 8))
-
-
-void AesGenTables(void)
-{
- unsigned i;
- for (i = 0; i < 256; i++)
- InvS[Sbox[i]] = (Byte)i;
-
- for (i = 0; i < 256; i++)
- {
- {
- UInt32 a1 = Sbox[i];
- UInt32 a2 = xtime(a1);
- UInt32 a3 = a2 ^ a1;
- TT(0)[i] = Ui32(a2, a1, a1, a3);
- TT(1)[i] = Ui32(a3, a2, a1, a1);
- TT(2)[i] = Ui32(a1, a3, a2, a1);
- TT(3)[i] = Ui32(a1, a1, a3, a2);
- }
- {
- UInt32 a1 = InvS[i];
- UInt32 a2 = xtime(a1);
- UInt32 a4 = xtime(a2);
- UInt32 a8 = xtime(a4);
- UInt32 a9 = a8 ^ a1;
- UInt32 aB = a8 ^ a2 ^ a1;
- UInt32 aD = a8 ^ a4 ^ a1;
- UInt32 aE = a8 ^ a4 ^ a2;
- DD(0)[i] = Ui32(aE, a9, aD, aB);
- DD(1)[i] = Ui32(aB, aE, a9, aD);
- DD(2)[i] = Ui32(aD, aB, aE, a9);
- DD(3)[i] = Ui32(a9, aD, aB, aE);
- }
- }
-
- g_AesCbc_Encode = AesCbc_Encode;
- g_AesCbc_Decode = AesCbc_Decode;
- g_AesCtr_Code = AesCtr_Code;
-
- #ifdef MY_CPU_X86_OR_AMD64
- if (CPU_Is_Aes_Supported())
- {
- g_AesCbc_Encode = AesCbc_Encode_Intel;
- g_AesCbc_Decode = AesCbc_Decode_Intel;
- g_AesCtr_Code = AesCtr_Code_Intel;
- }
- #endif
-}
-
-
-#define HT(i, x, s) TT(x)[gb(x, s[(i + x) & 3])]
-
-#define HT4(m, i, s, p) m[i] = \
- HT(i, 0, s) ^ \
- HT(i, 1, s) ^ \
- HT(i, 2, s) ^ \
- HT(i, 3, s) ^ w[p + i]
-
-#define HT16(m, s, p) \
- HT4(m, 0, s, p); \
- HT4(m, 1, s, p); \
- HT4(m, 2, s, p); \
- HT4(m, 3, s, p); \
-
-#define FT(i, x) Sbox[gb(x, m[(i + x) & 3])]
-#define FT4(i) dest[i] = Ui32(FT(i, 0), FT(i, 1), FT(i, 2), FT(i, 3)) ^ w[i];
-
-
-#define HD(i, x, s) DD(x)[gb(x, s[(i - x) & 3])]
-
-#define HD4(m, i, s, p) m[i] = \
- HD(i, 0, s) ^ \
- HD(i, 1, s) ^ \
- HD(i, 2, s) ^ \
- HD(i, 3, s) ^ w[p + i];
-
-#define HD16(m, s, p) \
- HD4(m, 0, s, p); \
- HD4(m, 1, s, p); \
- HD4(m, 2, s, p); \
- HD4(m, 3, s, p); \
-
-#define FD(i, x) InvS[gb(x, m[(i - x) & 3])]
-#define FD4(i) dest[i] = Ui32(FD(i, 0), FD(i, 1), FD(i, 2), FD(i, 3)) ^ w[i];
-
-void MY_FAST_CALL Aes_SetKey_Enc(UInt32 *w, const Byte *key, unsigned keySize)
-{
- unsigned i, wSize;
- wSize = keySize + 28;
- keySize /= 4;
- w[0] = ((UInt32)keySize / 2) + 3;
- w += 4;
-
- for (i = 0; i < keySize; i++, key += 4)
- w[i] = GetUi32(key);
-
- for (; i < wSize; i++)
- {
- UInt32 t = w[(size_t)i - 1];
- unsigned rem = i % keySize;
- if (rem == 0)
- t = Ui32(Sbox[gb1(t)] ^ Rcon[i / keySize], Sbox[gb2(t)], Sbox[gb3(t)], Sbox[gb0(t)]);
- else if (keySize > 6 && rem == 4)
- t = Ui32(Sbox[gb0(t)], Sbox[gb1(t)], Sbox[gb2(t)], Sbox[gb3(t)]);
- w[i] = w[i - keySize] ^ t;
- }
-}
-
-void MY_FAST_CALL Aes_SetKey_Dec(UInt32 *w, const Byte *key, unsigned keySize)
-{
- unsigned i, num;
- Aes_SetKey_Enc(w, key, keySize);
- num = keySize + 20;
- w += 8;
- for (i = 0; i < num; i++)
- {
- UInt32 r = w[i];
- w[i] =
- DD(0)[Sbox[gb0(r)]] ^
- DD(1)[Sbox[gb1(r)]] ^
- DD(2)[Sbox[gb2(r)]] ^
- DD(3)[Sbox[gb3(r)]];
- }
-}
-
-/* Aes_Encode and Aes_Decode functions work with little-endian words.
- src and dest are pointers to 4 UInt32 words.
- src and dest can point to same block */
-
-static void Aes_Encode(const UInt32 *w, UInt32 *dest, const UInt32 *src)
-{
- UInt32 s[4];
- UInt32 m[4];
- UInt32 numRounds2 = w[0];
- w += 4;
- s[0] = src[0] ^ w[0];
- s[1] = src[1] ^ w[1];
- s[2] = src[2] ^ w[2];
- s[3] = src[3] ^ w[3];
- w += 4;
- for (;;)
- {
- HT16(m, s, 0);
- if (--numRounds2 == 0)
- break;
- HT16(s, m, 4);
- w += 8;
- }
- w += 4;
- FT4(0); FT4(1); FT4(2); FT4(3);
-}
-
-static void Aes_Decode(const UInt32 *w, UInt32 *dest, const UInt32 *src)
-{
- UInt32 s[4];
- UInt32 m[4];
- UInt32 numRounds2 = w[0];
- w += 4 + numRounds2 * 8;
- s[0] = src[0] ^ w[0];
- s[1] = src[1] ^ w[1];
- s[2] = src[2] ^ w[2];
- s[3] = src[3] ^ w[3];
- for (;;)
- {
- w -= 8;
- HD16(m, s, 4);
- if (--numRounds2 == 0)
- break;
- HD16(s, m, 0);
- }
- FD4(0); FD4(1); FD4(2); FD4(3);
-}
-
-void AesCbc_Init(UInt32 *p, const Byte *iv)
-{
- unsigned i;
- for (i = 0; i < 4; i++)
- p[i] = GetUi32(iv + i * 4);
-}
-
-void MY_FAST_CALL AesCbc_Encode(UInt32 *p, Byte *data, size_t numBlocks)
-{
- for (; numBlocks != 0; numBlocks--, data += AES_BLOCK_SIZE)
- {
- p[0] ^= GetUi32(data);
- p[1] ^= GetUi32(data + 4);
- p[2] ^= GetUi32(data + 8);
- p[3] ^= GetUi32(data + 12);
-
- Aes_Encode(p + 4, p, p);
-
- SetUi32(data, p[0]);
- SetUi32(data + 4, p[1]);
- SetUi32(data + 8, p[2]);
- SetUi32(data + 12, p[3]);
- }
-}
-
-void MY_FAST_CALL AesCbc_Decode(UInt32 *p, Byte *data, size_t numBlocks)
-{
- UInt32 in[4], out[4];
- for (; numBlocks != 0; numBlocks--, data += AES_BLOCK_SIZE)
- {
- in[0] = GetUi32(data);
- in[1] = GetUi32(data + 4);
- in[2] = GetUi32(data + 8);
- in[3] = GetUi32(data + 12);
-
- Aes_Decode(p + 4, out, in);
-
- SetUi32(data, p[0] ^ out[0]);
- SetUi32(data + 4, p[1] ^ out[1]);
- SetUi32(data + 8, p[2] ^ out[2]);
- SetUi32(data + 12, p[3] ^ out[3]);
-
- p[0] = in[0];
- p[1] = in[1];
- p[2] = in[2];
- p[3] = in[3];
- }
-}
-
-void MY_FAST_CALL AesCtr_Code(UInt32 *p, Byte *data, size_t numBlocks)
-{
- for (; numBlocks != 0; numBlocks--)
- {
- UInt32 temp[4];
- unsigned i;
-
- if (++p[0] == 0)
- p[1]++;
-
- Aes_Encode(p + 4, temp, p);
-
- for (i = 0; i < 4; i++, data += 4)
- {
- UInt32 t = temp[i];
-
- #ifdef MY_CPU_LE_UNALIGN
- *((UInt32 *)data) ^= t;
- #else
- data[0] ^= (t & 0xFF);
- data[1] ^= ((t >> 8) & 0xFF);
- data[2] ^= ((t >> 16) & 0xFF);
- data[3] ^= ((t >> 24));
- #endif
- }
- }
-}
+/* Aes.c -- AES encryption / decryption +2017-01-24 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include "Aes.h" +#include "CpuArch.h" + +static UInt32 T[256 * 4]; +static const Byte Sbox[256] = { + 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}; + +void MY_FAST_CALL AesCbc_Encode(UInt32 *ivAes, Byte *data, size_t numBlocks); +void MY_FAST_CALL AesCbc_Decode(UInt32 *ivAes, Byte *data, size_t numBlocks); +void MY_FAST_CALL AesCtr_Code(UInt32 *ivAes, Byte *data, size_t numBlocks); + +void MY_FAST_CALL AesCbc_Encode_Intel(UInt32 *ivAes, Byte *data, size_t numBlocks); +void MY_FAST_CALL AesCbc_Decode_Intel(UInt32 *ivAes, Byte *data, size_t numBlocks); +void MY_FAST_CALL AesCtr_Code_Intel(UInt32 *ivAes, Byte *data, size_t numBlocks); + +AES_CODE_FUNC g_AesCbc_Encode; +AES_CODE_FUNC g_AesCbc_Decode; +AES_CODE_FUNC g_AesCtr_Code; + +static UInt32 D[256 * 4]; +static Byte InvS[256]; + +static const Byte Rcon[11] = { 0x00, 0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80, 0x1b, 0x36 }; + +#define xtime(x) ((((x) << 1) ^ (((x) & 0x80) != 0 ? 0x1B : 0)) & 0xFF) + +#define Ui32(a0, a1, a2, a3) ((UInt32)(a0) | ((UInt32)(a1) << 8) | ((UInt32)(a2) << 16) | ((UInt32)(a3) << 24)) + +#define gb0(x) ( (x) & 0xFF) +#define gb1(x) (((x) >> ( 8)) & 0xFF) +#define gb2(x) (((x) >> (16)) & 0xFF) +#define gb3(x) (((x) >> (24))) + +#define gb(n, x) gb ## n(x) + +#define TT(x) (T + (x << 8)) +#define DD(x) (D + (x << 8)) + + +void AesGenTables(void) +{ + unsigned i; + for (i = 0; i < 256; i++) + InvS[Sbox[i]] = (Byte)i; + + for (i = 0; i < 256; i++) + { + { + UInt32 a1 = Sbox[i]; + UInt32 a2 = xtime(a1); + UInt32 a3 = a2 ^ a1; + TT(0)[i] = Ui32(a2, a1, a1, a3); + TT(1)[i] = Ui32(a3, a2, a1, a1); + TT(2)[i] = Ui32(a1, a3, a2, a1); + TT(3)[i] = Ui32(a1, a1, a3, a2); + } + { + UInt32 a1 = InvS[i]; + UInt32 a2 = xtime(a1); + UInt32 a4 = xtime(a2); + UInt32 a8 = xtime(a4); + UInt32 a9 = a8 ^ a1; + UInt32 aB = a8 ^ a2 ^ a1; + UInt32 aD = a8 ^ a4 ^ a1; + UInt32 aE = a8 ^ a4 ^ a2; + DD(0)[i] = Ui32(aE, a9, aD, aB); + DD(1)[i] = Ui32(aB, aE, a9, aD); + DD(2)[i] = Ui32(aD, aB, aE, a9); + DD(3)[i] = Ui32(a9, aD, aB, aE); + } + } + + g_AesCbc_Encode = AesCbc_Encode; + g_AesCbc_Decode = AesCbc_Decode; + g_AesCtr_Code = AesCtr_Code; + + #ifdef MY_CPU_X86_OR_AMD64 + if (CPU_Is_Aes_Supported()) + { + g_AesCbc_Encode = AesCbc_Encode_Intel; + g_AesCbc_Decode = AesCbc_Decode_Intel; + g_AesCtr_Code = AesCtr_Code_Intel; + } + #endif +} + + +#define HT(i, x, s) TT(x)[gb(x, s[(i + x) & 3])] + +#define HT4(m, i, s, p) m[i] = \ + HT(i, 0, s) ^ \ + HT(i, 1, s) ^ \ + HT(i, 2, s) ^ \ + HT(i, 3, s) ^ w[p + i] + +#define HT16(m, s, p) \ + HT4(m, 0, s, p); \ + HT4(m, 1, s, p); \ + HT4(m, 2, s, p); \ + HT4(m, 3, s, p); \ + +#define FT(i, x) Sbox[gb(x, m[(i + x) & 3])] +#define FT4(i) dest[i] = Ui32(FT(i, 0), FT(i, 1), FT(i, 2), FT(i, 3)) ^ w[i]; + + +#define HD(i, x, s) DD(x)[gb(x, s[(i - x) & 3])] + +#define HD4(m, i, s, p) m[i] = \ + HD(i, 0, s) ^ \ + HD(i, 1, s) ^ \ + HD(i, 2, s) ^ \ + HD(i, 3, s) ^ w[p + i]; + +#define HD16(m, s, p) \ + HD4(m, 0, s, p); \ + HD4(m, 1, s, p); \ + HD4(m, 2, s, p); \ + HD4(m, 3, s, p); \ + +#define FD(i, x) InvS[gb(x, m[(i - x) & 3])] +#define FD4(i) dest[i] = Ui32(FD(i, 0), FD(i, 1), FD(i, 2), FD(i, 3)) ^ w[i]; + +void MY_FAST_CALL Aes_SetKey_Enc(UInt32 *w, const Byte *key, unsigned keySize) +{ + unsigned i, wSize; + wSize = keySize + 28; + keySize /= 4; + w[0] = ((UInt32)keySize / 2) + 3; + w += 4; + + for (i = 0; i < keySize; i++, key += 4) + w[i] = GetUi32(key); + + for (; i < wSize; i++) + { + UInt32 t = w[(size_t)i - 1]; + unsigned rem = i % keySize; + if (rem == 0) + t = Ui32(Sbox[gb1(t)] ^ Rcon[i / keySize], Sbox[gb2(t)], Sbox[gb3(t)], Sbox[gb0(t)]); + else if (keySize > 6 && rem == 4) + t = Ui32(Sbox[gb0(t)], Sbox[gb1(t)], Sbox[gb2(t)], Sbox[gb3(t)]); + w[i] = w[i - keySize] ^ t; + } +} + +void MY_FAST_CALL Aes_SetKey_Dec(UInt32 *w, const Byte *key, unsigned keySize) +{ + unsigned i, num; + Aes_SetKey_Enc(w, key, keySize); + num = keySize + 20; + w += 8; + for (i = 0; i < num; i++) + { + UInt32 r = w[i]; + w[i] = + DD(0)[Sbox[gb0(r)]] ^ + DD(1)[Sbox[gb1(r)]] ^ + DD(2)[Sbox[gb2(r)]] ^ + DD(3)[Sbox[gb3(r)]]; + } +} + +/* Aes_Encode and Aes_Decode functions work with little-endian words. + src and dest are pointers to 4 UInt32 words. + src and dest can point to same block */ + +static void Aes_Encode(const UInt32 *w, UInt32 *dest, const UInt32 *src) +{ + UInt32 s[4]; + UInt32 m[4]; + UInt32 numRounds2 = w[0]; + w += 4; + s[0] = src[0] ^ w[0]; + s[1] = src[1] ^ w[1]; + s[2] = src[2] ^ w[2]; + s[3] = src[3] ^ w[3]; + w += 4; + for (;;) + { + HT16(m, s, 0); + if (--numRounds2 == 0) + break; + HT16(s, m, 4); + w += 8; + } + w += 4; + FT4(0); FT4(1); FT4(2); FT4(3); +} + +static void Aes_Decode(const UInt32 *w, UInt32 *dest, const UInt32 *src) +{ + UInt32 s[4]; + UInt32 m[4]; + UInt32 numRounds2 = w[0]; + w += 4 + numRounds2 * 8; + s[0] = src[0] ^ w[0]; + s[1] = src[1] ^ w[1]; + s[2] = src[2] ^ w[2]; + s[3] = src[3] ^ w[3]; + for (;;) + { + w -= 8; + HD16(m, s, 4); + if (--numRounds2 == 0) + break; + HD16(s, m, 0); + } + FD4(0); FD4(1); FD4(2); FD4(3); +} + +void AesCbc_Init(UInt32 *p, const Byte *iv) +{ + unsigned i; + for (i = 0; i < 4; i++) + p[i] = GetUi32(iv + i * 4); +} + +void MY_FAST_CALL AesCbc_Encode(UInt32 *p, Byte *data, size_t numBlocks) +{ + for (; numBlocks != 0; numBlocks--, data += AES_BLOCK_SIZE) + { + p[0] ^= GetUi32(data); + p[1] ^= GetUi32(data + 4); + p[2] ^= GetUi32(data + 8); + p[3] ^= GetUi32(data + 12); + + Aes_Encode(p + 4, p, p); + + SetUi32(data, p[0]); + SetUi32(data + 4, p[1]); + SetUi32(data + 8, p[2]); + SetUi32(data + 12, p[3]); + } +} + +void MY_FAST_CALL AesCbc_Decode(UInt32 *p, Byte *data, size_t numBlocks) +{ + UInt32 in[4], out[4]; + for (; numBlocks != 0; numBlocks--, data += AES_BLOCK_SIZE) + { + in[0] = GetUi32(data); + in[1] = GetUi32(data + 4); + in[2] = GetUi32(data + 8); + in[3] = GetUi32(data + 12); + + Aes_Decode(p + 4, out, in); + + SetUi32(data, p[0] ^ out[0]); + SetUi32(data + 4, p[1] ^ out[1]); + SetUi32(data + 8, p[2] ^ out[2]); + SetUi32(data + 12, p[3] ^ out[3]); + + p[0] = in[0]; + p[1] = in[1]; + p[2] = in[2]; + p[3] = in[3]; + } +} + +void MY_FAST_CALL AesCtr_Code(UInt32 *p, Byte *data, size_t numBlocks) +{ + for (; numBlocks != 0; numBlocks--) + { + UInt32 temp[4]; + unsigned i; + + if (++p[0] == 0) + p[1]++; + + Aes_Encode(p + 4, temp, p); + + for (i = 0; i < 4; i++, data += 4) + { + UInt32 t = temp[i]; + + #ifdef MY_CPU_LE_UNALIGN + *((UInt32 *)data) ^= t; + #else + data[0] ^= (t & 0xFF); + data[1] ^= ((t >> 8) & 0xFF); + data[2] ^= ((t >> 16) & 0xFF); + data[3] ^= ((t >> 24)); + #endif + } + } +} diff --git a/contrib/libs/lzmasdk/Aes.h b/contrib/libs/lzmasdk/Aes.h index 381e979d1b..167865600c 100644 --- a/contrib/libs/lzmasdk/Aes.h +++ b/contrib/libs/lzmasdk/Aes.h @@ -1,38 +1,38 @@ -/* Aes.h -- AES encryption / decryption
-2013-01-18 : Igor Pavlov : Public domain */
-
-#ifndef __AES_H
-#define __AES_H
-
-#include "7zTypes.h"
-
-EXTERN_C_BEGIN
-
-#define AES_BLOCK_SIZE 16
-
-/* Call AesGenTables one time before other AES functions */
-void AesGenTables(void);
-
-/* UInt32 pointers must be 16-byte aligned */
-
-/* 16-byte (4 * 32-bit words) blocks: 1 (IV) + 1 (keyMode) + 15 (AES-256 roundKeys) */
-#define AES_NUM_IVMRK_WORDS ((1 + 1 + 15) * 4)
-
-/* aes - 16-byte aligned pointer to keyMode+roundKeys sequence */
-/* keySize = 16 or 24 or 32 (bytes) */
-typedef void (MY_FAST_CALL *AES_SET_KEY_FUNC)(UInt32 *aes, const Byte *key, unsigned keySize);
-void MY_FAST_CALL Aes_SetKey_Enc(UInt32 *aes, const Byte *key, unsigned keySize);
-void MY_FAST_CALL Aes_SetKey_Dec(UInt32 *aes, const Byte *key, unsigned keySize);
-
-/* ivAes - 16-byte aligned pointer to iv+keyMode+roundKeys sequence: UInt32[AES_NUM_IVMRK_WORDS] */
-void AesCbc_Init(UInt32 *ivAes, const Byte *iv); /* iv size is AES_BLOCK_SIZE */
-/* data - 16-byte aligned pointer to data */
-/* numBlocks - the number of 16-byte blocks in data array */
-typedef void (MY_FAST_CALL *AES_CODE_FUNC)(UInt32 *ivAes, Byte *data, size_t numBlocks);
-extern AES_CODE_FUNC g_AesCbc_Encode;
-extern AES_CODE_FUNC g_AesCbc_Decode;
-extern AES_CODE_FUNC g_AesCtr_Code;
-
-EXTERN_C_END
-
-#endif
+/* Aes.h -- AES encryption / decryption +2013-01-18 : Igor Pavlov : Public domain */ + +#ifndef __AES_H +#define __AES_H + +#include "7zTypes.h" + +EXTERN_C_BEGIN + +#define AES_BLOCK_SIZE 16 + +/* Call AesGenTables one time before other AES functions */ +void AesGenTables(void); + +/* UInt32 pointers must be 16-byte aligned */ + +/* 16-byte (4 * 32-bit words) blocks: 1 (IV) + 1 (keyMode) + 15 (AES-256 roundKeys) */ +#define AES_NUM_IVMRK_WORDS ((1 + 1 + 15) * 4) + +/* aes - 16-byte aligned pointer to keyMode+roundKeys sequence */ +/* keySize = 16 or 24 or 32 (bytes) */ +typedef void (MY_FAST_CALL *AES_SET_KEY_FUNC)(UInt32 *aes, const Byte *key, unsigned keySize); +void MY_FAST_CALL Aes_SetKey_Enc(UInt32 *aes, const Byte *key, unsigned keySize); +void MY_FAST_CALL Aes_SetKey_Dec(UInt32 *aes, const Byte *key, unsigned keySize); + +/* ivAes - 16-byte aligned pointer to iv+keyMode+roundKeys sequence: UInt32[AES_NUM_IVMRK_WORDS] */ +void AesCbc_Init(UInt32 *ivAes, const Byte *iv); /* iv size is AES_BLOCK_SIZE */ +/* data - 16-byte aligned pointer to data */ +/* numBlocks - the number of 16-byte blocks in data array */ +typedef void (MY_FAST_CALL *AES_CODE_FUNC)(UInt32 *ivAes, Byte *data, size_t numBlocks); +extern AES_CODE_FUNC g_AesCbc_Encode; +extern AES_CODE_FUNC g_AesCbc_Decode; +extern AES_CODE_FUNC g_AesCtr_Code; + +EXTERN_C_END + +#endif diff --git a/contrib/libs/lzmasdk/AesOpt.c b/contrib/libs/lzmasdk/AesOpt.c index 65936a414f..6ade8f42ff 100644 --- a/contrib/libs/lzmasdk/AesOpt.c +++ b/contrib/libs/lzmasdk/AesOpt.c @@ -1,188 +1,188 @@ -/* AesOpt.c -- Intel's AES
-2017-06-08 : Igor Pavlov : Public domain */
-
-#include "Precomp.h"
-
-#include "CpuArch.h"
-
-#ifdef MY_CPU_X86_OR_AMD64
-#if (_MSC_VER > 1500) || (_MSC_FULL_VER >= 150030729)
-#define USE_INTEL_AES
-#endif
-#endif
-
-#ifdef USE_INTEL_AES
-
-#if defined(__clang__)
-#define TARGET_AES __attribute__((__target__("aes")))
-#else
-#define TARGET_AES
-#endif
-
-#include <wmmintrin.h>
-
-void TARGET_AES MY_FAST_CALL AesCbc_Encode_Intel(__m128i *p, __m128i *data, size_t numBlocks)
-{
- __m128i m = *p;
- for (; numBlocks != 0; numBlocks--, data++)
- {
- UInt32 numRounds2 = *(const UInt32 *)(p + 1) - 1;
- const __m128i *w = p + 3;
- m = _mm_xor_si128(m, *data);
- m = _mm_xor_si128(m, p[2]);
- do
- {
- m = _mm_aesenc_si128(m, w[0]);
- m = _mm_aesenc_si128(m, w[1]);
- w += 2;
- }
- while (--numRounds2 != 0);
- m = _mm_aesenc_si128(m, w[0]);
- m = _mm_aesenclast_si128(m, w[1]);
- *data = m;
- }
- *p = m;
-}
-
-#define NUM_WAYS 3
-
-#define AES_OP_W(op, n) { \
- const __m128i t = w[n]; \
- m0 = op(m0, t); \
- m1 = op(m1, t); \
- m2 = op(m2, t); \
- }
-
-#define AES_DEC(n) AES_OP_W(_mm_aesdec_si128, n)
-#define AES_DEC_LAST(n) AES_OP_W(_mm_aesdeclast_si128, n)
-#define AES_ENC(n) AES_OP_W(_mm_aesenc_si128, n)
-#define AES_ENC_LAST(n) AES_OP_W(_mm_aesenclast_si128, n)
-
-void TARGET_AES MY_FAST_CALL AesCbc_Decode_Intel(__m128i *p, __m128i *data, size_t numBlocks)
-{
- __m128i iv = *p;
- for (; numBlocks >= NUM_WAYS; numBlocks -= NUM_WAYS, data += NUM_WAYS)
- {
- UInt32 numRounds2 = *(const UInt32 *)(p + 1);
- const __m128i *w = p + numRounds2 * 2;
- __m128i m0, m1, m2;
- {
- const __m128i t = w[2];
- m0 = _mm_xor_si128(t, data[0]);
- m1 = _mm_xor_si128(t, data[1]);
- m2 = _mm_xor_si128(t, data[2]);
- }
- numRounds2--;
- do
- {
- AES_DEC(1)
- AES_DEC(0)
- w -= 2;
- }
- while (--numRounds2 != 0);
- AES_DEC(1)
- AES_DEC_LAST(0)
-
- {
- __m128i t;
- t = _mm_xor_si128(m0, iv); iv = data[0]; data[0] = t;
- t = _mm_xor_si128(m1, iv); iv = data[1]; data[1] = t;
- t = _mm_xor_si128(m2, iv); iv = data[2]; data[2] = t;
- }
- }
- for (; numBlocks != 0; numBlocks--, data++)
- {
- UInt32 numRounds2 = *(const UInt32 *)(p + 1);
- const __m128i *w = p + numRounds2 * 2;
- __m128i m = _mm_xor_si128(w[2], *data);
- numRounds2--;
- do
- {
- m = _mm_aesdec_si128(m, w[1]);
- m = _mm_aesdec_si128(m, w[0]);
- w -= 2;
- }
- while (--numRounds2 != 0);
- m = _mm_aesdec_si128(m, w[1]);
- m = _mm_aesdeclast_si128(m, w[0]);
-
- m = _mm_xor_si128(m, iv);
- iv = *data;
- *data = m;
- }
- *p = iv;
-}
-
-void TARGET_AES MY_FAST_CALL AesCtr_Code_Intel(__m128i *p, __m128i *data, size_t numBlocks)
-{
- __m128i ctr = *p;
- __m128i one = _mm_set_epi64x(1, 0);
- for (; numBlocks >= NUM_WAYS; numBlocks -= NUM_WAYS, data += NUM_WAYS)
- {
- UInt32 numRounds2 = *(const UInt32 *)(p + 1) - 1;
- const __m128i *w = p;
- __m128i m0, m1, m2;
- {
- const __m128i t = w[2];
- ctr = _mm_add_epi64(ctr, one); m0 = _mm_xor_si128(ctr, t);
- ctr = _mm_add_epi64(ctr, one); m1 = _mm_xor_si128(ctr, t);
- ctr = _mm_add_epi64(ctr, one); m2 = _mm_xor_si128(ctr, t);
- }
- w += 3;
- do
- {
- AES_ENC(0)
- AES_ENC(1)
- w += 2;
- }
- while (--numRounds2 != 0);
- AES_ENC(0)
- AES_ENC_LAST(1)
- data[0] = _mm_xor_si128(data[0], m0);
- data[1] = _mm_xor_si128(data[1], m1);
- data[2] = _mm_xor_si128(data[2], m2);
- }
- for (; numBlocks != 0; numBlocks--, data++)
- {
- UInt32 numRounds2 = *(const UInt32 *)(p + 1) - 1;
- const __m128i *w = p;
- __m128i m;
- ctr = _mm_add_epi64(ctr, one);
- m = _mm_xor_si128(ctr, p[2]);
- w += 3;
- do
- {
- m = _mm_aesenc_si128(m, w[0]);
- m = _mm_aesenc_si128(m, w[1]);
- w += 2;
- }
- while (--numRounds2 != 0);
- m = _mm_aesenc_si128(m, w[0]);
- m = _mm_aesenclast_si128(m, w[1]);
- *data = _mm_xor_si128(*data, m);
- }
- *p = ctr;
-}
-
-#else
-
-void MY_FAST_CALL AesCbc_Encode(UInt32 *ivAes, Byte *data, size_t numBlocks);
-void MY_FAST_CALL AesCbc_Decode(UInt32 *ivAes, Byte *data, size_t numBlocks);
-void MY_FAST_CALL AesCtr_Code(UInt32 *ivAes, Byte *data, size_t numBlocks);
-
-void MY_FAST_CALL AesCbc_Encode_Intel(UInt32 *p, Byte *data, size_t numBlocks)
-{
- AesCbc_Encode(p, data, numBlocks);
-}
-
-void MY_FAST_CALL AesCbc_Decode_Intel(UInt32 *p, Byte *data, size_t numBlocks)
-{
- AesCbc_Decode(p, data, numBlocks);
-}
-
-void MY_FAST_CALL AesCtr_Code_Intel(UInt32 *p, Byte *data, size_t numBlocks)
-{
- AesCtr_Code(p, data, numBlocks);
-}
-
-#endif
+/* AesOpt.c -- Intel's AES +2017-06-08 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include "CpuArch.h" + +#ifdef MY_CPU_X86_OR_AMD64 +#if (_MSC_VER > 1500) || (_MSC_FULL_VER >= 150030729) +#define USE_INTEL_AES +#endif +#endif + +#ifdef USE_INTEL_AES + +#if defined(__clang__) +#define TARGET_AES __attribute__((__target__("aes"))) +#else +#define TARGET_AES +#endif + +#include <wmmintrin.h> + +void TARGET_AES MY_FAST_CALL AesCbc_Encode_Intel(__m128i *p, __m128i *data, size_t numBlocks) +{ + __m128i m = *p; + for (; numBlocks != 0; numBlocks--, data++) + { + UInt32 numRounds2 = *(const UInt32 *)(p + 1) - 1; + const __m128i *w = p + 3; + m = _mm_xor_si128(m, *data); + m = _mm_xor_si128(m, p[2]); + do + { + m = _mm_aesenc_si128(m, w[0]); + m = _mm_aesenc_si128(m, w[1]); + w += 2; + } + while (--numRounds2 != 0); + m = _mm_aesenc_si128(m, w[0]); + m = _mm_aesenclast_si128(m, w[1]); + *data = m; + } + *p = m; +} + +#define NUM_WAYS 3 + +#define AES_OP_W(op, n) { \ + const __m128i t = w[n]; \ + m0 = op(m0, t); \ + m1 = op(m1, t); \ + m2 = op(m2, t); \ + } + +#define AES_DEC(n) AES_OP_W(_mm_aesdec_si128, n) +#define AES_DEC_LAST(n) AES_OP_W(_mm_aesdeclast_si128, n) +#define AES_ENC(n) AES_OP_W(_mm_aesenc_si128, n) +#define AES_ENC_LAST(n) AES_OP_W(_mm_aesenclast_si128, n) + +void TARGET_AES MY_FAST_CALL AesCbc_Decode_Intel(__m128i *p, __m128i *data, size_t numBlocks) +{ + __m128i iv = *p; + for (; numBlocks >= NUM_WAYS; numBlocks -= NUM_WAYS, data += NUM_WAYS) + { + UInt32 numRounds2 = *(const UInt32 *)(p + 1); + const __m128i *w = p + numRounds2 * 2; + __m128i m0, m1, m2; + { + const __m128i t = w[2]; + m0 = _mm_xor_si128(t, data[0]); + m1 = _mm_xor_si128(t, data[1]); + m2 = _mm_xor_si128(t, data[2]); + } + numRounds2--; + do + { + AES_DEC(1) + AES_DEC(0) + w -= 2; + } + while (--numRounds2 != 0); + AES_DEC(1) + AES_DEC_LAST(0) + + { + __m128i t; + t = _mm_xor_si128(m0, iv); iv = data[0]; data[0] = t; + t = _mm_xor_si128(m1, iv); iv = data[1]; data[1] = t; + t = _mm_xor_si128(m2, iv); iv = data[2]; data[2] = t; + } + } + for (; numBlocks != 0; numBlocks--, data++) + { + UInt32 numRounds2 = *(const UInt32 *)(p + 1); + const __m128i *w = p + numRounds2 * 2; + __m128i m = _mm_xor_si128(w[2], *data); + numRounds2--; + do + { + m = _mm_aesdec_si128(m, w[1]); + m = _mm_aesdec_si128(m, w[0]); + w -= 2; + } + while (--numRounds2 != 0); + m = _mm_aesdec_si128(m, w[1]); + m = _mm_aesdeclast_si128(m, w[0]); + + m = _mm_xor_si128(m, iv); + iv = *data; + *data = m; + } + *p = iv; +} + +void TARGET_AES MY_FAST_CALL AesCtr_Code_Intel(__m128i *p, __m128i *data, size_t numBlocks) +{ + __m128i ctr = *p; + __m128i one = _mm_set_epi64x(1, 0); + for (; numBlocks >= NUM_WAYS; numBlocks -= NUM_WAYS, data += NUM_WAYS) + { + UInt32 numRounds2 = *(const UInt32 *)(p + 1) - 1; + const __m128i *w = p; + __m128i m0, m1, m2; + { + const __m128i t = w[2]; + ctr = _mm_add_epi64(ctr, one); m0 = _mm_xor_si128(ctr, t); + ctr = _mm_add_epi64(ctr, one); m1 = _mm_xor_si128(ctr, t); + ctr = _mm_add_epi64(ctr, one); m2 = _mm_xor_si128(ctr, t); + } + w += 3; + do + { + AES_ENC(0) + AES_ENC(1) + w += 2; + } + while (--numRounds2 != 0); + AES_ENC(0) + AES_ENC_LAST(1) + data[0] = _mm_xor_si128(data[0], m0); + data[1] = _mm_xor_si128(data[1], m1); + data[2] = _mm_xor_si128(data[2], m2); + } + for (; numBlocks != 0; numBlocks--, data++) + { + UInt32 numRounds2 = *(const UInt32 *)(p + 1) - 1; + const __m128i *w = p; + __m128i m; + ctr = _mm_add_epi64(ctr, one); + m = _mm_xor_si128(ctr, p[2]); + w += 3; + do + { + m = _mm_aesenc_si128(m, w[0]); + m = _mm_aesenc_si128(m, w[1]); + w += 2; + } + while (--numRounds2 != 0); + m = _mm_aesenc_si128(m, w[0]); + m = _mm_aesenclast_si128(m, w[1]); + *data = _mm_xor_si128(*data, m); + } + *p = ctr; +} + +#else + +void MY_FAST_CALL AesCbc_Encode(UInt32 *ivAes, Byte *data, size_t numBlocks); +void MY_FAST_CALL AesCbc_Decode(UInt32 *ivAes, Byte *data, size_t numBlocks); +void MY_FAST_CALL AesCtr_Code(UInt32 *ivAes, Byte *data, size_t numBlocks); + +void MY_FAST_CALL AesCbc_Encode_Intel(UInt32 *p, Byte *data, size_t numBlocks) +{ + AesCbc_Encode(p, data, numBlocks); +} + +void MY_FAST_CALL AesCbc_Decode_Intel(UInt32 *p, Byte *data, size_t numBlocks) +{ + AesCbc_Decode(p, data, numBlocks); +} + +void MY_FAST_CALL AesCtr_Code_Intel(UInt32 *p, Byte *data, size_t numBlocks) +{ + AesCtr_Code(p, data, numBlocks); +} + +#endif diff --git a/contrib/libs/lzmasdk/Bra.c b/contrib/libs/lzmasdk/Bra.c index cbdcb290df..e04546ff30 100644 --- a/contrib/libs/lzmasdk/Bra.c +++ b/contrib/libs/lzmasdk/Bra.c @@ -1,230 +1,230 @@ -/* Bra.c -- Converters for RISC code
-2017-04-04 : Igor Pavlov : Public domain */
-
-#include "Precomp.h"
-
-#include "CpuArch.h"
-#include "Bra.h"
-
-SizeT ARM_Convert(Byte *data, SizeT size, UInt32 ip, int encoding)
-{
- Byte *p;
- const Byte *lim;
- size &= ~(size_t)3;
- ip += 4;
- p = data;
- lim = data + size;
-
- if (encoding)
-
- for (;;)
- {
- for (;;)
- {
- if (p >= lim)
- return p - data;
- p += 4;
- if (p[-1] == 0xEB)
- break;
- }
- {
- UInt32 v = GetUi32(p - 4);
- v <<= 2;
- v += ip + (UInt32)(p - data);
- v >>= 2;
- v &= 0x00FFFFFF;
- v |= 0xEB000000;
- SetUi32(p - 4, v);
- }
- }
-
- for (;;)
- {
- for (;;)
- {
- if (p >= lim)
- return p - data;
- p += 4;
- if (p[-1] == 0xEB)
- break;
- }
- {
- UInt32 v = GetUi32(p - 4);
- v <<= 2;
- v -= ip + (UInt32)(p - data);
- v >>= 2;
- v &= 0x00FFFFFF;
- v |= 0xEB000000;
- SetUi32(p - 4, v);
- }
- }
-}
-
-
-SizeT ARMT_Convert(Byte *data, SizeT size, UInt32 ip, int encoding)
-{
- Byte *p;
- const Byte *lim;
- size &= ~(size_t)1;
- p = data;
- lim = data + size - 4;
-
- if (encoding)
-
- for (;;)
- {
- UInt32 b1;
- for (;;)
- {
- UInt32 b3;
- if (p > lim)
- return p - data;
- b1 = p[1];
- b3 = p[3];
- p += 2;
- b1 ^= 8;
- if ((b3 & b1) >= 0xF8)
- break;
- }
- {
- UInt32 v =
- ((UInt32)b1 << 19)
- + (((UInt32)p[1] & 0x7) << 8)
- + (((UInt32)p[-2] << 11))
- + (p[0]);
-
- p += 2;
- {
- UInt32 cur = (ip + (UInt32)(p - data)) >> 1;
- v += cur;
- }
-
- p[-4] = (Byte)(v >> 11);
- p[-3] = (Byte)(0xF0 | ((v >> 19) & 0x7));
- p[-2] = (Byte)v;
- p[-1] = (Byte)(0xF8 | (v >> 8));
- }
- }
-
- for (;;)
- {
- UInt32 b1;
- for (;;)
- {
- UInt32 b3;
- if (p > lim)
- return p - data;
- b1 = p[1];
- b3 = p[3];
- p += 2;
- b1 ^= 8;
- if ((b3 & b1) >= 0xF8)
- break;
- }
- {
- UInt32 v =
- ((UInt32)b1 << 19)
- + (((UInt32)p[1] & 0x7) << 8)
- + (((UInt32)p[-2] << 11))
- + (p[0]);
-
- p += 2;
- {
- UInt32 cur = (ip + (UInt32)(p - data)) >> 1;
- v -= cur;
- }
-
- /*
- SetUi16(p - 4, (UInt16)(((v >> 11) & 0x7FF) | 0xF000));
- SetUi16(p - 2, (UInt16)(v | 0xF800));
- */
-
- p[-4] = (Byte)(v >> 11);
- p[-3] = (Byte)(0xF0 | ((v >> 19) & 0x7));
- p[-2] = (Byte)v;
- p[-1] = (Byte)(0xF8 | (v >> 8));
- }
- }
-}
-
-
-SizeT PPC_Convert(Byte *data, SizeT size, UInt32 ip, int encoding)
-{
- Byte *p;
- const Byte *lim;
- size &= ~(size_t)3;
- ip -= 4;
- p = data;
- lim = data + size;
-
- for (;;)
- {
- for (;;)
- {
- if (p >= lim)
- return p - data;
- p += 4;
- /* if ((v & 0xFC000003) == 0x48000001) */
- if ((p[-4] & 0xFC) == 0x48 && (p[-1] & 3) == 1)
- break;
- }
- {
- UInt32 v = GetBe32(p - 4);
- if (encoding)
- v += ip + (UInt32)(p - data);
- else
- v -= ip + (UInt32)(p - data);
- v &= 0x03FFFFFF;
- v |= 0x48000000;
- SetBe32(p - 4, v);
- }
- }
-}
-
-
-SizeT SPARC_Convert(Byte *data, SizeT size, UInt32 ip, int encoding)
-{
- Byte *p;
- const Byte *lim;
- size &= ~(size_t)3;
- ip -= 4;
- p = data;
- lim = data + size;
-
- for (;;)
- {
- for (;;)
- {
- if (p >= lim)
- return p - data;
- /*
- v = GetBe32(p);
- p += 4;
- m = v + ((UInt32)5 << 29);
- m ^= (UInt32)7 << 29;
- m += (UInt32)1 << 22;
- if ((m & ((UInt32)0x1FF << 23)) == 0)
- break;
- */
- p += 4;
- if ((p[-4] == 0x40 && (p[-3] & 0xC0) == 0) ||
- (p[-4] == 0x7F && (p[-3] >= 0xC0)))
- break;
- }
- {
- UInt32 v = GetBe32(p - 4);
- v <<= 2;
- if (encoding)
- v += ip + (UInt32)(p - data);
- else
- v -= ip + (UInt32)(p - data);
-
- v &= 0x01FFFFFF;
- v -= (UInt32)1 << 24;
- v ^= 0xFF000000;
- v >>= 2;
- v |= 0x40000000;
- SetBe32(p - 4, v);
- }
- }
-}
+/* Bra.c -- Converters for RISC code +2017-04-04 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include "CpuArch.h" +#include "Bra.h" + +SizeT ARM_Convert(Byte *data, SizeT size, UInt32 ip, int encoding) +{ + Byte *p; + const Byte *lim; + size &= ~(size_t)3; + ip += 4; + p = data; + lim = data + size; + + if (encoding) + + for (;;) + { + for (;;) + { + if (p >= lim) + return p - data; + p += 4; + if (p[-1] == 0xEB) + break; + } + { + UInt32 v = GetUi32(p - 4); + v <<= 2; + v += ip + (UInt32)(p - data); + v >>= 2; + v &= 0x00FFFFFF; + v |= 0xEB000000; + SetUi32(p - 4, v); + } + } + + for (;;) + { + for (;;) + { + if (p >= lim) + return p - data; + p += 4; + if (p[-1] == 0xEB) + break; + } + { + UInt32 v = GetUi32(p - 4); + v <<= 2; + v -= ip + (UInt32)(p - data); + v >>= 2; + v &= 0x00FFFFFF; + v |= 0xEB000000; + SetUi32(p - 4, v); + } + } +} + + +SizeT ARMT_Convert(Byte *data, SizeT size, UInt32 ip, int encoding) +{ + Byte *p; + const Byte *lim; + size &= ~(size_t)1; + p = data; + lim = data + size - 4; + + if (encoding) + + for (;;) + { + UInt32 b1; + for (;;) + { + UInt32 b3; + if (p > lim) + return p - data; + b1 = p[1]; + b3 = p[3]; + p += 2; + b1 ^= 8; + if ((b3 & b1) >= 0xF8) + break; + } + { + UInt32 v = + ((UInt32)b1 << 19) + + (((UInt32)p[1] & 0x7) << 8) + + (((UInt32)p[-2] << 11)) + + (p[0]); + + p += 2; + { + UInt32 cur = (ip + (UInt32)(p - data)) >> 1; + v += cur; + } + + p[-4] = (Byte)(v >> 11); + p[-3] = (Byte)(0xF0 | ((v >> 19) & 0x7)); + p[-2] = (Byte)v; + p[-1] = (Byte)(0xF8 | (v >> 8)); + } + } + + for (;;) + { + UInt32 b1; + for (;;) + { + UInt32 b3; + if (p > lim) + return p - data; + b1 = p[1]; + b3 = p[3]; + p += 2; + b1 ^= 8; + if ((b3 & b1) >= 0xF8) + break; + } + { + UInt32 v = + ((UInt32)b1 << 19) + + (((UInt32)p[1] & 0x7) << 8) + + (((UInt32)p[-2] << 11)) + + (p[0]); + + p += 2; + { + UInt32 cur = (ip + (UInt32)(p - data)) >> 1; + v -= cur; + } + + /* + SetUi16(p - 4, (UInt16)(((v >> 11) & 0x7FF) | 0xF000)); + SetUi16(p - 2, (UInt16)(v | 0xF800)); + */ + + p[-4] = (Byte)(v >> 11); + p[-3] = (Byte)(0xF0 | ((v >> 19) & 0x7)); + p[-2] = (Byte)v; + p[-1] = (Byte)(0xF8 | (v >> 8)); + } + } +} + + +SizeT PPC_Convert(Byte *data, SizeT size, UInt32 ip, int encoding) +{ + Byte *p; + const Byte *lim; + size &= ~(size_t)3; + ip -= 4; + p = data; + lim = data + size; + + for (;;) + { + for (;;) + { + if (p >= lim) + return p - data; + p += 4; + /* if ((v & 0xFC000003) == 0x48000001) */ + if ((p[-4] & 0xFC) == 0x48 && (p[-1] & 3) == 1) + break; + } + { + UInt32 v = GetBe32(p - 4); + if (encoding) + v += ip + (UInt32)(p - data); + else + v -= ip + (UInt32)(p - data); + v &= 0x03FFFFFF; + v |= 0x48000000; + SetBe32(p - 4, v); + } + } +} + + +SizeT SPARC_Convert(Byte *data, SizeT size, UInt32 ip, int encoding) +{ + Byte *p; + const Byte *lim; + size &= ~(size_t)3; + ip -= 4; + p = data; + lim = data + size; + + for (;;) + { + for (;;) + { + if (p >= lim) + return p - data; + /* + v = GetBe32(p); + p += 4; + m = v + ((UInt32)5 << 29); + m ^= (UInt32)7 << 29; + m += (UInt32)1 << 22; + if ((m & ((UInt32)0x1FF << 23)) == 0) + break; + */ + p += 4; + if ((p[-4] == 0x40 && (p[-3] & 0xC0) == 0) || + (p[-4] == 0x7F && (p[-3] >= 0xC0))) + break; + } + { + UInt32 v = GetBe32(p - 4); + v <<= 2; + if (encoding) + v += ip + (UInt32)(p - data); + else + v -= ip + (UInt32)(p - data); + + v &= 0x01FFFFFF; + v -= (UInt32)1 << 24; + v ^= 0xFF000000; + v >>= 2; + v |= 0x40000000; + SetBe32(p - 4, v); + } + } +} diff --git a/contrib/libs/lzmasdk/Bra.h b/contrib/libs/lzmasdk/Bra.h index aba8dce14f..f24c36ed6b 100644 --- a/contrib/libs/lzmasdk/Bra.h +++ b/contrib/libs/lzmasdk/Bra.h @@ -1,64 +1,64 @@ -/* Bra.h -- Branch converters for executables
-2013-01-18 : Igor Pavlov : Public domain */
-
-#ifndef __BRA_H
-#define __BRA_H
-
-#include "7zTypes.h"
-
-EXTERN_C_BEGIN
-
-/*
-These functions convert relative addresses to absolute addresses
-in CALL instructions to increase the compression ratio.
-
- In:
- data - data buffer
- size - size of data
- ip - current virtual Instruction Pinter (IP) value
- state - state variable for x86 converter
- encoding - 0 (for decoding), 1 (for encoding)
-
- Out:
- state - state variable for x86 converter
-
- Returns:
- The number of processed bytes. If you call these functions with multiple calls,
- you must start next call with first byte after block of processed bytes.
-
- Type Endian Alignment LookAhead
-
- x86 little 1 4
- ARMT little 2 2
- ARM little 4 0
- PPC big 4 0
- SPARC big 4 0
- IA64 little 16 0
-
- size must be >= Alignment + LookAhead, if it's not last block.
- If (size < Alignment + LookAhead), converter returns 0.
-
- Example:
-
- UInt32 ip = 0;
- for ()
- {
- ; size must be >= Alignment + LookAhead, if it's not last block
- SizeT processed = Convert(data, size, ip, 1);
- data += processed;
- size -= processed;
- ip += processed;
- }
-*/
-
-#define x86_Convert_Init(state) { state = 0; }
-SizeT x86_Convert(Byte *data, SizeT size, UInt32 ip, UInt32 *state, int encoding);
-SizeT ARM_Convert(Byte *data, SizeT size, UInt32 ip, int encoding);
-SizeT ARMT_Convert(Byte *data, SizeT size, UInt32 ip, int encoding);
-SizeT PPC_Convert(Byte *data, SizeT size, UInt32 ip, int encoding);
-SizeT SPARC_Convert(Byte *data, SizeT size, UInt32 ip, int encoding);
-SizeT IA64_Convert(Byte *data, SizeT size, UInt32 ip, int encoding);
-
-EXTERN_C_END
-
-#endif
+/* Bra.h -- Branch converters for executables +2013-01-18 : Igor Pavlov : Public domain */ + +#ifndef __BRA_H +#define __BRA_H + +#include "7zTypes.h" + +EXTERN_C_BEGIN + +/* +These functions convert relative addresses to absolute addresses +in CALL instructions to increase the compression ratio. + + In: + data - data buffer + size - size of data + ip - current virtual Instruction Pinter (IP) value + state - state variable for x86 converter + encoding - 0 (for decoding), 1 (for encoding) + + Out: + state - state variable for x86 converter + + Returns: + The number of processed bytes. If you call these functions with multiple calls, + you must start next call with first byte after block of processed bytes. + + Type Endian Alignment LookAhead + + x86 little 1 4 + ARMT little 2 2 + ARM little 4 0 + PPC big 4 0 + SPARC big 4 0 + IA64 little 16 0 + + size must be >= Alignment + LookAhead, if it's not last block. + If (size < Alignment + LookAhead), converter returns 0. + + Example: + + UInt32 ip = 0; + for () + { + ; size must be >= Alignment + LookAhead, if it's not last block + SizeT processed = Convert(data, size, ip, 1); + data += processed; + size -= processed; + ip += processed; + } +*/ + +#define x86_Convert_Init(state) { state = 0; } +SizeT x86_Convert(Byte *data, SizeT size, UInt32 ip, UInt32 *state, int encoding); +SizeT ARM_Convert(Byte *data, SizeT size, UInt32 ip, int encoding); +SizeT ARMT_Convert(Byte *data, SizeT size, UInt32 ip, int encoding); +SizeT PPC_Convert(Byte *data, SizeT size, UInt32 ip, int encoding); +SizeT SPARC_Convert(Byte *data, SizeT size, UInt32 ip, int encoding); +SizeT IA64_Convert(Byte *data, SizeT size, UInt32 ip, int encoding); + +EXTERN_C_END + +#endif diff --git a/contrib/libs/lzmasdk/Bra86.c b/contrib/libs/lzmasdk/Bra86.c index a6463c63ba..0a4747fe2e 100644 --- a/contrib/libs/lzmasdk/Bra86.c +++ b/contrib/libs/lzmasdk/Bra86.c @@ -1,82 +1,82 @@ -/* Bra86.c -- Converter for x86 code (BCJ)
-2017-04-03 : Igor Pavlov : Public domain */
-
-#include "Precomp.h"
-
-#include "Bra.h"
-
-#define Test86MSByte(b) ((((b) + 1) & 0xFE) == 0)
-
-SizeT x86_Convert(Byte *data, SizeT size, UInt32 ip, UInt32 *state, int encoding)
-{
- SizeT pos = 0;
- UInt32 mask = *state & 7;
- if (size < 5)
- return 0;
- size -= 4;
- ip += 5;
-
- for (;;)
- {
- Byte *p = data + pos;
- const Byte *limit = data + size;
- for (; p < limit; p++)
- if ((*p & 0xFE) == 0xE8)
- break;
-
- {
- SizeT d = (SizeT)(p - data - pos);
- pos = (SizeT)(p - data);
- if (p >= limit)
- {
- *state = (d > 2 ? 0 : mask >> (unsigned)d);
- return pos;
- }
- if (d > 2)
- mask = 0;
- else
- {
- mask >>= (unsigned)d;
- if (mask != 0 && (mask > 4 || mask == 3 || Test86MSByte(p[(size_t)(mask >> 1) + 1])))
- {
- mask = (mask >> 1) | 4;
- pos++;
- continue;
- }
- }
- }
-
- if (Test86MSByte(p[4]))
- {
- UInt32 v = ((UInt32)p[4] << 24) | ((UInt32)p[3] << 16) | ((UInt32)p[2] << 8) | ((UInt32)p[1]);
- UInt32 cur = ip + (UInt32)pos;
- pos += 5;
- if (encoding)
- v += cur;
- else
- v -= cur;
- if (mask != 0)
- {
- unsigned sh = (mask & 6) << 2;
- if (Test86MSByte((Byte)(v >> sh)))
- {
- v ^= (((UInt32)0x100 << sh) - 1);
- if (encoding)
- v += cur;
- else
- v -= cur;
- }
- mask = 0;
- }
- p[1] = (Byte)v;
- p[2] = (Byte)(v >> 8);
- p[3] = (Byte)(v >> 16);
- p[4] = (Byte)(0 - ((v >> 24) & 1));
- }
- else
- {
- mask = (mask >> 1) | 4;
- pos++;
- }
- }
-}
+/* Bra86.c -- Converter for x86 code (BCJ) +2017-04-03 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include "Bra.h" + +#define Test86MSByte(b) ((((b) + 1) & 0xFE) == 0) + +SizeT x86_Convert(Byte *data, SizeT size, UInt32 ip, UInt32 *state, int encoding) +{ + SizeT pos = 0; + UInt32 mask = *state & 7; + if (size < 5) + return 0; + size -= 4; + ip += 5; + + for (;;) + { + Byte *p = data + pos; + const Byte *limit = data + size; + for (; p < limit; p++) + if ((*p & 0xFE) == 0xE8) + break; + + { + SizeT d = (SizeT)(p - data - pos); + pos = (SizeT)(p - data); + if (p >= limit) + { + *state = (d > 2 ? 0 : mask >> (unsigned)d); + return pos; + } + if (d > 2) + mask = 0; + else + { + mask >>= (unsigned)d; + if (mask != 0 && (mask > 4 || mask == 3 || Test86MSByte(p[(size_t)(mask >> 1) + 1]))) + { + mask = (mask >> 1) | 4; + pos++; + continue; + } + } + } + + if (Test86MSByte(p[4])) + { + UInt32 v = ((UInt32)p[4] << 24) | ((UInt32)p[3] << 16) | ((UInt32)p[2] << 8) | ((UInt32)p[1]); + UInt32 cur = ip + (UInt32)pos; + pos += 5; + if (encoding) + v += cur; + else + v -= cur; + if (mask != 0) + { + unsigned sh = (mask & 6) << 2; + if (Test86MSByte((Byte)(v >> sh))) + { + v ^= (((UInt32)0x100 << sh) - 1); + if (encoding) + v += cur; + else + v -= cur; + } + mask = 0; + } + p[1] = (Byte)v; + p[2] = (Byte)(v >> 8); + p[3] = (Byte)(v >> 16); + p[4] = (Byte)(0 - ((v >> 24) & 1)); + } + else + { + mask = (mask >> 1) | 4; + pos++; + } + } +} diff --git a/contrib/libs/lzmasdk/BraIA64.c b/contrib/libs/lzmasdk/BraIA64.c index 2656907a0b..7cccd47b50 100644 --- a/contrib/libs/lzmasdk/BraIA64.c +++ b/contrib/libs/lzmasdk/BraIA64.c @@ -1,53 +1,53 @@ -/* BraIA64.c -- Converter for IA-64 code
-2017-01-26 : Igor Pavlov : Public domain */
-
-#include "Precomp.h"
-
-#include "CpuArch.h"
-#include "Bra.h"
-
-SizeT IA64_Convert(Byte *data, SizeT size, UInt32 ip, int encoding)
-{
- SizeT i;
- if (size < 16)
- return 0;
- size -= 16;
- i = 0;
- do
- {
- unsigned m = ((UInt32)0x334B0000 >> (data[i] & 0x1E)) & 3;
- if (m)
- {
- m++;
- do
- {
- Byte *p = data + (i + (size_t)m * 5 - 8);
- if (((p[3] >> m) & 15) == 5
- && (((p[-1] | ((UInt32)p[0] << 8)) >> m) & 0x70) == 0)
- {
- unsigned raw = GetUi32(p);
- unsigned v = raw >> m;
- v = (v & 0xFFFFF) | ((v & (1 << 23)) >> 3);
-
- v <<= 4;
- if (encoding)
- v += ip + (UInt32)i;
- else
- v -= ip + (UInt32)i;
- v >>= 4;
-
- v &= 0x1FFFFF;
- v += 0x700000;
- v &= 0x8FFFFF;
- raw &= ~((UInt32)0x8FFFFF << m);
- raw |= (v << m);
- SetUi32(p, raw);
- }
- }
- while (++m <= 4);
- }
- i += 16;
- }
- while (i <= size);
- return i;
-}
+/* BraIA64.c -- Converter for IA-64 code +2017-01-26 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include "CpuArch.h" +#include "Bra.h" + +SizeT IA64_Convert(Byte *data, SizeT size, UInt32 ip, int encoding) +{ + SizeT i; + if (size < 16) + return 0; + size -= 16; + i = 0; + do + { + unsigned m = ((UInt32)0x334B0000 >> (data[i] & 0x1E)) & 3; + if (m) + { + m++; + do + { + Byte *p = data + (i + (size_t)m * 5 - 8); + if (((p[3] >> m) & 15) == 5 + && (((p[-1] | ((UInt32)p[0] << 8)) >> m) & 0x70) == 0) + { + unsigned raw = GetUi32(p); + unsigned v = raw >> m; + v = (v & 0xFFFFF) | ((v & (1 << 23)) >> 3); + + v <<= 4; + if (encoding) + v += ip + (UInt32)i; + else + v -= ip + (UInt32)i; + v >>= 4; + + v &= 0x1FFFFF; + v += 0x700000; + v &= 0x8FFFFF; + raw &= ~((UInt32)0x8FFFFF << m); + raw |= (v << m); + SetUi32(p, raw); + } + } + while (++m <= 4); + } + i += 16; + } + while (i <= size); + return i; +} diff --git a/contrib/libs/lzmasdk/CpuArch.c b/contrib/libs/lzmasdk/CpuArch.c index ff1890e7fe..ca84088e38 100644 --- a/contrib/libs/lzmasdk/CpuArch.c +++ b/contrib/libs/lzmasdk/CpuArch.c @@ -1,218 +1,218 @@ -/* CpuArch.c -- CPU specific code
-2018-02-18: Igor Pavlov : Public domain */
-
-#include "Precomp.h"
-
-#include "CpuArch.h"
-
-#ifdef MY_CPU_X86_OR_AMD64
-
-#if (defined(_MSC_VER) && !defined(MY_CPU_AMD64)) || defined(__GNUC__)
-#define USE_ASM
-#endif
-
-#if !defined(USE_ASM) && _MSC_VER >= 1500
-#include <intrin.h>
-#endif
-
-#if defined(USE_ASM) && !defined(MY_CPU_AMD64)
-static UInt32 CheckFlag(UInt32 flag)
-{
- #ifdef _MSC_VER
- __asm pushfd;
- __asm pop EAX;
- __asm mov EDX, EAX;
- __asm xor EAX, flag;
- __asm push EAX;
- __asm popfd;
- __asm pushfd;
- __asm pop EAX;
- __asm xor EAX, EDX;
- __asm push EDX;
- __asm popfd;
- __asm and flag, EAX;
- #else
- __asm__ __volatile__ (
- "pushf\n\t"
- "pop %%EAX\n\t"
- "movl %%EAX,%%EDX\n\t"
- "xorl %0,%%EAX\n\t"
- "push %%EAX\n\t"
- "popf\n\t"
- "pushf\n\t"
- "pop %%EAX\n\t"
- "xorl %%EDX,%%EAX\n\t"
- "push %%EDX\n\t"
- "popf\n\t"
- "andl %%EAX, %0\n\t":
- "=c" (flag) : "c" (flag) :
- "%eax", "%edx");
- #endif
- return flag;
-}
-#define CHECK_CPUID_IS_SUPPORTED if (CheckFlag(1 << 18) == 0 || CheckFlag(1 << 21) == 0) return False;
-#else
-#define CHECK_CPUID_IS_SUPPORTED
-#endif
-
-void MyCPUID(UInt32 function, UInt32 *a, UInt32 *b, UInt32 *c, UInt32 *d)
-{
- #ifdef USE_ASM
-
- #ifdef _MSC_VER
-
- UInt32 a2, b2, c2, d2;
- __asm xor EBX, EBX;
- __asm xor ECX, ECX;
- __asm xor EDX, EDX;
- __asm mov EAX, function;
- __asm cpuid;
- __asm mov a2, EAX;
- __asm mov b2, EBX;
- __asm mov c2, ECX;
- __asm mov d2, EDX;
-
- *a = a2;
- *b = b2;
- *c = c2;
- *d = d2;
-
- #else
-
- __asm__ __volatile__ (
- #if defined(MY_CPU_AMD64) && defined(__PIC__)
- "mov %%rbx, %%rdi;"
- "cpuid;"
- "xchg %%rbx, %%rdi;"
- : "=a" (*a) ,
- "=D" (*b) ,
- #elif defined(MY_CPU_X86) && defined(__PIC__)
- "mov %%ebx, %%edi;"
- "cpuid;"
- "xchgl %%ebx, %%edi;"
- : "=a" (*a) ,
- "=D" (*b) ,
- #else
- "cpuid"
- : "=a" (*a) ,
- "=b" (*b) ,
- #endif
- "=c" (*c) ,
- "=d" (*d)
- : "0" (function)) ;
-
- #endif
-
- #else
-
- int CPUInfo[4];
- __cpuid(CPUInfo, function);
- *a = CPUInfo[0];
- *b = CPUInfo[1];
- *c = CPUInfo[2];
- *d = CPUInfo[3];
-
- #endif
-}
-
-BoolInt x86cpuid_CheckAndRead(Cx86cpuid *p)
-{
- CHECK_CPUID_IS_SUPPORTED
- MyCPUID(0, &p->maxFunc, &p->vendor[0], &p->vendor[2], &p->vendor[1]);
- MyCPUID(1, &p->ver, &p->b, &p->c, &p->d);
- return True;
-}
-
-static const UInt32 kVendors[][3] =
-{
- { 0x756E6547, 0x49656E69, 0x6C65746E},
- { 0x68747541, 0x69746E65, 0x444D4163},
- { 0x746E6543, 0x48727561, 0x736C7561}
-};
-
-int x86cpuid_GetFirm(const Cx86cpuid *p)
-{
- unsigned i;
- for (i = 0; i < sizeof(kVendors) / sizeof(kVendors[i]); i++)
- {
- const UInt32 *v = kVendors[i];
- if (v[0] == p->vendor[0] &&
- v[1] == p->vendor[1] &&
- v[2] == p->vendor[2])
- return (int)i;
- }
- return -1;
-}
-
-BoolInt CPU_Is_InOrder()
-{
- Cx86cpuid p;
- int firm;
- UInt32 family, model;
- if (!x86cpuid_CheckAndRead(&p))
- return True;
-
- family = x86cpuid_GetFamily(p.ver);
- model = x86cpuid_GetModel(p.ver);
-
- firm = x86cpuid_GetFirm(&p);
-
- switch (firm)
- {
- case CPU_FIRM_INTEL: return (family < 6 || (family == 6 && (
- /* In-Order Atom CPU */
- model == 0x1C /* 45 nm, N4xx, D4xx, N5xx, D5xx, 230, 330 */
- || model == 0x26 /* 45 nm, Z6xx */
- || model == 0x27 /* 32 nm, Z2460 */
- || model == 0x35 /* 32 nm, Z2760 */
- || model == 0x36 /* 32 nm, N2xxx, D2xxx */
- )));
- case CPU_FIRM_AMD: return (family < 5 || (family == 5 && (model < 6 || model == 0xA)));
- case CPU_FIRM_VIA: return (family < 6 || (family == 6 && model < 0xF));
- }
- return True;
-}
-
-#if !defined(MY_CPU_AMD64) && defined(_WIN32)
-#include <windows.h>
-static BoolInt CPU_Sys_Is_SSE_Supported()
-{
- OSVERSIONINFO vi;
- vi.dwOSVersionInfoSize = sizeof(vi);
- if (!GetVersionEx(&vi))
- return False;
- return (vi.dwMajorVersion >= 5);
-}
-#define CHECK_SYS_SSE_SUPPORT if (!CPU_Sys_Is_SSE_Supported()) return False;
-#else
-#define CHECK_SYS_SSE_SUPPORT
-#endif
-
-BoolInt CPU_Is_Aes_Supported()
-{
- Cx86cpuid p;
- CHECK_SYS_SSE_SUPPORT
- if (!x86cpuid_CheckAndRead(&p))
- return False;
- return (p.c >> 25) & 1;
-}
-
-BoolInt CPU_IsSupported_PageGB()
-{
- Cx86cpuid cpuid;
- if (!x86cpuid_CheckAndRead(&cpuid))
- return False;
- {
- UInt32 d[4] = { 0 };
- MyCPUID(0x80000000, &d[0], &d[1], &d[2], &d[3]);
- if (d[0] < 0x80000001)
- return False;
- }
- {
- UInt32 d[4] = { 0 };
- MyCPUID(0x80000001, &d[0], &d[1], &d[2], &d[3]);
- return (d[3] >> 26) & 1;
- }
-}
-
-#endif
+/* CpuArch.c -- CPU specific code +2018-02-18: Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include "CpuArch.h" + +#ifdef MY_CPU_X86_OR_AMD64 + +#if (defined(_MSC_VER) && !defined(MY_CPU_AMD64)) || defined(__GNUC__) +#define USE_ASM +#endif + +#if !defined(USE_ASM) && _MSC_VER >= 1500 +#include <intrin.h> +#endif + +#if defined(USE_ASM) && !defined(MY_CPU_AMD64) +static UInt32 CheckFlag(UInt32 flag) +{ + #ifdef _MSC_VER + __asm pushfd; + __asm pop EAX; + __asm mov EDX, EAX; + __asm xor EAX, flag; + __asm push EAX; + __asm popfd; + __asm pushfd; + __asm pop EAX; + __asm xor EAX, EDX; + __asm push EDX; + __asm popfd; + __asm and flag, EAX; + #else + __asm__ __volatile__ ( + "pushf\n\t" + "pop %%EAX\n\t" + "movl %%EAX,%%EDX\n\t" + "xorl %0,%%EAX\n\t" + "push %%EAX\n\t" + "popf\n\t" + "pushf\n\t" + "pop %%EAX\n\t" + "xorl %%EDX,%%EAX\n\t" + "push %%EDX\n\t" + "popf\n\t" + "andl %%EAX, %0\n\t": + "=c" (flag) : "c" (flag) : + "%eax", "%edx"); + #endif + return flag; +} +#define CHECK_CPUID_IS_SUPPORTED if (CheckFlag(1 << 18) == 0 || CheckFlag(1 << 21) == 0) return False; +#else +#define CHECK_CPUID_IS_SUPPORTED +#endif + +void MyCPUID(UInt32 function, UInt32 *a, UInt32 *b, UInt32 *c, UInt32 *d) +{ + #ifdef USE_ASM + + #ifdef _MSC_VER + + UInt32 a2, b2, c2, d2; + __asm xor EBX, EBX; + __asm xor ECX, ECX; + __asm xor EDX, EDX; + __asm mov EAX, function; + __asm cpuid; + __asm mov a2, EAX; + __asm mov b2, EBX; + __asm mov c2, ECX; + __asm mov d2, EDX; + + *a = a2; + *b = b2; + *c = c2; + *d = d2; + + #else + + __asm__ __volatile__ ( + #if defined(MY_CPU_AMD64) && defined(__PIC__) + "mov %%rbx, %%rdi;" + "cpuid;" + "xchg %%rbx, %%rdi;" + : "=a" (*a) , + "=D" (*b) , + #elif defined(MY_CPU_X86) && defined(__PIC__) + "mov %%ebx, %%edi;" + "cpuid;" + "xchgl %%ebx, %%edi;" + : "=a" (*a) , + "=D" (*b) , + #else + "cpuid" + : "=a" (*a) , + "=b" (*b) , + #endif + "=c" (*c) , + "=d" (*d) + : "0" (function)) ; + + #endif + + #else + + int CPUInfo[4]; + __cpuid(CPUInfo, function); + *a = CPUInfo[0]; + *b = CPUInfo[1]; + *c = CPUInfo[2]; + *d = CPUInfo[3]; + + #endif +} + +BoolInt x86cpuid_CheckAndRead(Cx86cpuid *p) +{ + CHECK_CPUID_IS_SUPPORTED + MyCPUID(0, &p->maxFunc, &p->vendor[0], &p->vendor[2], &p->vendor[1]); + MyCPUID(1, &p->ver, &p->b, &p->c, &p->d); + return True; +} + +static const UInt32 kVendors[][3] = +{ + { 0x756E6547, 0x49656E69, 0x6C65746E}, + { 0x68747541, 0x69746E65, 0x444D4163}, + { 0x746E6543, 0x48727561, 0x736C7561} +}; + +int x86cpuid_GetFirm(const Cx86cpuid *p) +{ + unsigned i; + for (i = 0; i < sizeof(kVendors) / sizeof(kVendors[i]); i++) + { + const UInt32 *v = kVendors[i]; + if (v[0] == p->vendor[0] && + v[1] == p->vendor[1] && + v[2] == p->vendor[2]) + return (int)i; + } + return -1; +} + +BoolInt CPU_Is_InOrder() +{ + Cx86cpuid p; + int firm; + UInt32 family, model; + if (!x86cpuid_CheckAndRead(&p)) + return True; + + family = x86cpuid_GetFamily(p.ver); + model = x86cpuid_GetModel(p.ver); + + firm = x86cpuid_GetFirm(&p); + + switch (firm) + { + case CPU_FIRM_INTEL: return (family < 6 || (family == 6 && ( + /* In-Order Atom CPU */ + model == 0x1C /* 45 nm, N4xx, D4xx, N5xx, D5xx, 230, 330 */ + || model == 0x26 /* 45 nm, Z6xx */ + || model == 0x27 /* 32 nm, Z2460 */ + || model == 0x35 /* 32 nm, Z2760 */ + || model == 0x36 /* 32 nm, N2xxx, D2xxx */ + ))); + case CPU_FIRM_AMD: return (family < 5 || (family == 5 && (model < 6 || model == 0xA))); + case CPU_FIRM_VIA: return (family < 6 || (family == 6 && model < 0xF)); + } + return True; +} + +#if !defined(MY_CPU_AMD64) && defined(_WIN32) +#include <windows.h> +static BoolInt CPU_Sys_Is_SSE_Supported() +{ + OSVERSIONINFO vi; + vi.dwOSVersionInfoSize = sizeof(vi); + if (!GetVersionEx(&vi)) + return False; + return (vi.dwMajorVersion >= 5); +} +#define CHECK_SYS_SSE_SUPPORT if (!CPU_Sys_Is_SSE_Supported()) return False; +#else +#define CHECK_SYS_SSE_SUPPORT +#endif + +BoolInt CPU_Is_Aes_Supported() +{ + Cx86cpuid p; + CHECK_SYS_SSE_SUPPORT + if (!x86cpuid_CheckAndRead(&p)) + return False; + return (p.c >> 25) & 1; +} + +BoolInt CPU_IsSupported_PageGB() +{ + Cx86cpuid cpuid; + if (!x86cpuid_CheckAndRead(&cpuid)) + return False; + { + UInt32 d[4] = { 0 }; + MyCPUID(0x80000000, &d[0], &d[1], &d[2], &d[3]); + if (d[0] < 0x80000001) + return False; + } + { + UInt32 d[4] = { 0 }; + MyCPUID(0x80000001, &d[0], &d[1], &d[2], &d[3]); + return (d[3] >> 26) & 1; + } +} + +#endif diff --git a/contrib/libs/lzmasdk/CpuArch.h b/contrib/libs/lzmasdk/CpuArch.h index 5f74c1c0cb..31ec712268 100644 --- a/contrib/libs/lzmasdk/CpuArch.h +++ b/contrib/libs/lzmasdk/CpuArch.h @@ -1,336 +1,336 @@ -/* CpuArch.h -- CPU specific code
-2018-02-18 : Igor Pavlov : Public domain */
-
-#ifndef __CPU_ARCH_H
-#define __CPU_ARCH_H
-
-#include "7zTypes.h"
-
-EXTERN_C_BEGIN
-
-/*
-MY_CPU_LE means that CPU is LITTLE ENDIAN.
-MY_CPU_BE means that CPU is BIG ENDIAN.
-If MY_CPU_LE and MY_CPU_BE are not defined, we don't know about ENDIANNESS of platform.
-
-MY_CPU_LE_UNALIGN means that CPU is LITTLE ENDIAN and CPU supports unaligned memory accesses.
-*/
-
-#if defined(_M_X64) \
- || defined(_M_AMD64) \
- || defined(__x86_64__) \
- || defined(__AMD64__) \
- || defined(__amd64__)
- #define MY_CPU_AMD64
- #ifdef __ILP32__
- #define MY_CPU_NAME "x32"
- #else
- #define MY_CPU_NAME "x64"
- #endif
- #define MY_CPU_64BIT
-#endif
-
-
-#if defined(_M_IX86) \
- || defined(__i386__)
- #define MY_CPU_X86
- #define MY_CPU_NAME "x86"
- #define MY_CPU_32BIT
-#endif
-
-
-#if defined(_M_ARM64) \
- || defined(__AARCH64EL__) \
- || defined(__AARCH64EB__) \
- || defined(__aarch64__)
- #define MY_CPU_ARM64
- #define MY_CPU_NAME "arm64"
- #define MY_CPU_64BIT
-#endif
-
-
-#if defined(_M_ARM) \
- || defined(_M_ARM_NT) \
- || defined(_M_ARMT) \
- || defined(__arm__) \
- || defined(__thumb__) \
- || defined(__ARMEL__) \
- || defined(__ARMEB__) \
- || defined(__THUMBEL__) \
- || defined(__THUMBEB__)
- #define MY_CPU_ARM
- #define MY_CPU_NAME "arm"
- #define MY_CPU_32BIT
-#endif
-
-
-#if defined(_M_IA64) \
- || defined(__ia64__)
- #define MY_CPU_IA64
- #define MY_CPU_NAME "ia64"
- #define MY_CPU_64BIT
-#endif
-
-
-#if defined(__mips64) \
- || defined(__mips64__) \
- || (defined(__mips) && (__mips == 64 || __mips == 4 || __mips == 3))
- #define MY_CPU_NAME "mips64"
- #define MY_CPU_64BIT
-#elif defined(__mips__)
- #define MY_CPU_NAME "mips"
- /* #define MY_CPU_32BIT */
-#endif
-
-
-#if defined(__ppc64__) \
- || defined(__powerpc64__)
- #ifdef __ILP32__
- #define MY_CPU_NAME "ppc64-32"
- #else
- #define MY_CPU_NAME "ppc64"
- #endif
- #define MY_CPU_64BIT
-#elif defined(__ppc__) \
- || defined(__powerpc__)
- #define MY_CPU_NAME "ppc"
- #define MY_CPU_32BIT
-#endif
-
-
-#if defined(__sparc64__)
- #define MY_CPU_NAME "sparc64"
- #define MY_CPU_64BIT
-#elif defined(__sparc__)
- #define MY_CPU_NAME "sparc"
- /* #define MY_CPU_32BIT */
-#endif
-
-
-#if defined(MY_CPU_X86) || defined(MY_CPU_AMD64)
-#define MY_CPU_X86_OR_AMD64
-#endif
-
-
-#ifdef _WIN32
-
- #ifdef MY_CPU_ARM
- #define MY_CPU_ARM_LE
- #endif
-
- #ifdef MY_CPU_ARM64
- #define MY_CPU_ARM64_LE
- #endif
-
- #ifdef _M_IA64
- #define MY_CPU_IA64_LE
- #endif
-
-#endif
-
-
-#if defined(MY_CPU_X86_OR_AMD64) \
- || defined(MY_CPU_ARM_LE) \
- || defined(MY_CPU_ARM64_LE) \
- || defined(MY_CPU_IA64_LE) \
- || defined(__LITTLE_ENDIAN__) \
- || defined(__ARMEL__) \
- || defined(__THUMBEL__) \
- || defined(__AARCH64EL__) \
- || defined(__MIPSEL__) \
- || defined(__MIPSEL) \
- || defined(_MIPSEL) \
- || defined(__BFIN__) \
- || (defined(__BYTE_ORDER__) && (__BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__))
- #define MY_CPU_LE
-#endif
-
-#if defined(__BIG_ENDIAN__) \
- || defined(__ARMEB__) \
- || defined(__THUMBEB__) \
- || defined(__AARCH64EB__) \
- || defined(__MIPSEB__) \
- || defined(__MIPSEB) \
- || defined(_MIPSEB) \
- || defined(__m68k__) \
- || defined(__s390__) \
- || defined(__s390x__) \
- || defined(__zarch__) \
- || (defined(__BYTE_ORDER__) && (__BYTE_ORDER__ == __ORDER_BIG_ENDIAN__))
- #define MY_CPU_BE
-#endif
-
-
-#if defined(MY_CPU_LE) && defined(MY_CPU_BE)
- #error Stop_Compiling_Bad_Endian
-#endif
-
-
-#if defined(MY_CPU_32BIT) && defined(MY_CPU_64BIT)
- #error Stop_Compiling_Bad_32_64_BIT
-#endif
-
-
-#ifndef MY_CPU_NAME
- #ifdef MY_CPU_LE
- #define MY_CPU_NAME "LE"
- #elif defined(MY_CPU_BE)
- #define MY_CPU_NAME "BE"
- #else
- /*
- #define MY_CPU_NAME ""
- */
- #endif
-#endif
-
-
-
-
-
-#ifdef MY_CPU_LE
- #if defined(MY_CPU_X86_OR_AMD64) \
- || defined(MY_CPU_ARM64) \
- || defined(__ARM_FEATURE_UNALIGNED)
- #define MY_CPU_LE_UNALIGN
- #endif
-#endif
-
-
-#ifdef MY_CPU_LE_UNALIGN
-
-#define GetUi16(p) (*(const UInt16 *)(const void *)(p))
-#define GetUi32(p) (*(const UInt32 *)(const void *)(p))
-#define GetUi64(p) (*(const UInt64 *)(const void *)(p))
-
-#define SetUi16(p, v) { *(UInt16 *)(p) = (v); }
-#define SetUi32(p, v) { *(UInt32 *)(p) = (v); }
-#define SetUi64(p, v) { *(UInt64 *)(p) = (v); }
-
-#else
-
-#define GetUi16(p) ( (UInt16) ( \
- ((const Byte *)(p))[0] | \
- ((UInt16)((const Byte *)(p))[1] << 8) ))
-
-#define GetUi32(p) ( \
- ((const Byte *)(p))[0] | \
- ((UInt32)((const Byte *)(p))[1] << 8) | \
- ((UInt32)((const Byte *)(p))[2] << 16) | \
- ((UInt32)((const Byte *)(p))[3] << 24))
-
-#define GetUi64(p) (GetUi32(p) | ((UInt64)GetUi32(((const Byte *)(p)) + 4) << 32))
-
-#define SetUi16(p, v) { Byte *_ppp_ = (Byte *)(p); UInt32 _vvv_ = (v); \
- _ppp_[0] = (Byte)_vvv_; \
- _ppp_[1] = (Byte)(_vvv_ >> 8); }
-
-#define SetUi32(p, v) { Byte *_ppp_ = (Byte *)(p); UInt32 _vvv_ = (v); \
- _ppp_[0] = (Byte)_vvv_; \
- _ppp_[1] = (Byte)(_vvv_ >> 8); \
- _ppp_[2] = (Byte)(_vvv_ >> 16); \
- _ppp_[3] = (Byte)(_vvv_ >> 24); }
-
-#define SetUi64(p, v) { Byte *_ppp2_ = (Byte *)(p); UInt64 _vvv2_ = (v); \
- SetUi32(_ppp2_ , (UInt32)_vvv2_); \
- SetUi32(_ppp2_ + 4, (UInt32)(_vvv2_ >> 32)); }
-
-#endif
-
-#ifdef __has_builtin
- #define MY__has_builtin(x) __has_builtin(x)
-#else
- #define MY__has_builtin(x) 0
-#endif
-
-#if defined(MY_CPU_LE_UNALIGN) && /* defined(_WIN64) && */ (_MSC_VER >= 1300)
-
-/* Note: we use bswap instruction, that is unsupported in 386 cpu */
-
-#include <stdlib.h>
-
-#pragma intrinsic(_byteswap_ushort)
-#pragma intrinsic(_byteswap_ulong)
-#pragma intrinsic(_byteswap_uint64)
-
-/* #define GetBe16(p) _byteswap_ushort(*(const UInt16 *)(const Byte *)(p)) */
-#define GetBe32(p) _byteswap_ulong(*(const UInt32 *)(const Byte *)(p))
-#define GetBe64(p) _byteswap_uint64(*(const UInt64 *)(const Byte *)(p))
-
-#define SetBe32(p, v) (*(UInt32 *)(void *)(p)) = _byteswap_ulong(v)
-
-#elif defined(MY_CPU_LE_UNALIGN) && ( \
- (defined(__GNUC__) && (__GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 3))) \
- || (defined(__clang__) && MY__has_builtin(__builtin_bswap16)) )
-
-/* #define GetBe16(p) __builtin_bswap16(*(const UInt16 *)(const Byte *)(p)) */
-#define GetBe32(p) __builtin_bswap32(*(const UInt32 *)(const Byte *)(p))
-#define GetBe64(p) __builtin_bswap64(*(const UInt64 *)(const Byte *)(p))
-
-#define SetBe32(p, v) (*(UInt32 *)(void *)(p)) = __builtin_bswap32(v)
-
-#else
-
-#define GetBe32(p) ( \
- ((UInt32)((const Byte *)(p))[0] << 24) | \
- ((UInt32)((const Byte *)(p))[1] << 16) | \
- ((UInt32)((const Byte *)(p))[2] << 8) | \
- ((const Byte *)(p))[3] )
-
-#define GetBe64(p) (((UInt64)GetBe32(p) << 32) | GetBe32(((const Byte *)(p)) + 4))
-
-#define SetBe32(p, v) { Byte *_ppp_ = (Byte *)(p); UInt32 _vvv_ = (v); \
- _ppp_[0] = (Byte)(_vvv_ >> 24); \
- _ppp_[1] = (Byte)(_vvv_ >> 16); \
- _ppp_[2] = (Byte)(_vvv_ >> 8); \
- _ppp_[3] = (Byte)_vvv_; }
-
-#endif
-
-
-#ifndef GetBe16
-
-#define GetBe16(p) ( (UInt16) ( \
- ((UInt16)((const Byte *)(p))[0] << 8) | \
- ((const Byte *)(p))[1] ))
-
-#endif
-
-
-
-#ifdef MY_CPU_X86_OR_AMD64
-
-typedef struct
-{
- UInt32 maxFunc;
- UInt32 vendor[3];
- UInt32 ver;
- UInt32 b;
- UInt32 c;
- UInt32 d;
-} Cx86cpuid;
-
-enum
-{
- CPU_FIRM_INTEL,
- CPU_FIRM_AMD,
- CPU_FIRM_VIA
-};
-
-void MyCPUID(UInt32 function, UInt32 *a, UInt32 *b, UInt32 *c, UInt32 *d);
-
-BoolInt x86cpuid_CheckAndRead(Cx86cpuid *p);
-int x86cpuid_GetFirm(const Cx86cpuid *p);
-
-#define x86cpuid_GetFamily(ver) (((ver >> 16) & 0xFF0) | ((ver >> 8) & 0xF))
-#define x86cpuid_GetModel(ver) (((ver >> 12) & 0xF0) | ((ver >> 4) & 0xF))
-#define x86cpuid_GetStepping(ver) (ver & 0xF)
-
-BoolInt CPU_Is_InOrder();
-BoolInt CPU_Is_Aes_Supported();
-BoolInt CPU_IsSupported_PageGB();
-
-#endif
-
-EXTERN_C_END
-
-#endif
+/* CpuArch.h -- CPU specific code +2018-02-18 : Igor Pavlov : Public domain */ + +#ifndef __CPU_ARCH_H +#define __CPU_ARCH_H + +#include "7zTypes.h" + +EXTERN_C_BEGIN + +/* +MY_CPU_LE means that CPU is LITTLE ENDIAN. +MY_CPU_BE means that CPU is BIG ENDIAN. +If MY_CPU_LE and MY_CPU_BE are not defined, we don't know about ENDIANNESS of platform. + +MY_CPU_LE_UNALIGN means that CPU is LITTLE ENDIAN and CPU supports unaligned memory accesses. +*/ + +#if defined(_M_X64) \ + || defined(_M_AMD64) \ + || defined(__x86_64__) \ + || defined(__AMD64__) \ + || defined(__amd64__) + #define MY_CPU_AMD64 + #ifdef __ILP32__ + #define MY_CPU_NAME "x32" + #else + #define MY_CPU_NAME "x64" + #endif + #define MY_CPU_64BIT +#endif + + +#if defined(_M_IX86) \ + || defined(__i386__) + #define MY_CPU_X86 + #define MY_CPU_NAME "x86" + #define MY_CPU_32BIT +#endif + + +#if defined(_M_ARM64) \ + || defined(__AARCH64EL__) \ + || defined(__AARCH64EB__) \ + || defined(__aarch64__) + #define MY_CPU_ARM64 + #define MY_CPU_NAME "arm64" + #define MY_CPU_64BIT +#endif + + +#if defined(_M_ARM) \ + || defined(_M_ARM_NT) \ + || defined(_M_ARMT) \ + || defined(__arm__) \ + || defined(__thumb__) \ + || defined(__ARMEL__) \ + || defined(__ARMEB__) \ + || defined(__THUMBEL__) \ + || defined(__THUMBEB__) + #define MY_CPU_ARM + #define MY_CPU_NAME "arm" + #define MY_CPU_32BIT +#endif + + +#if defined(_M_IA64) \ + || defined(__ia64__) + #define MY_CPU_IA64 + #define MY_CPU_NAME "ia64" + #define MY_CPU_64BIT +#endif + + +#if defined(__mips64) \ + || defined(__mips64__) \ + || (defined(__mips) && (__mips == 64 || __mips == 4 || __mips == 3)) + #define MY_CPU_NAME "mips64" + #define MY_CPU_64BIT +#elif defined(__mips__) + #define MY_CPU_NAME "mips" + /* #define MY_CPU_32BIT */ +#endif + + +#if defined(__ppc64__) \ + || defined(__powerpc64__) + #ifdef __ILP32__ + #define MY_CPU_NAME "ppc64-32" + #else + #define MY_CPU_NAME "ppc64" + #endif + #define MY_CPU_64BIT +#elif defined(__ppc__) \ + || defined(__powerpc__) + #define MY_CPU_NAME "ppc" + #define MY_CPU_32BIT +#endif + + +#if defined(__sparc64__) + #define MY_CPU_NAME "sparc64" + #define MY_CPU_64BIT +#elif defined(__sparc__) + #define MY_CPU_NAME "sparc" + /* #define MY_CPU_32BIT */ +#endif + + +#if defined(MY_CPU_X86) || defined(MY_CPU_AMD64) +#define MY_CPU_X86_OR_AMD64 +#endif + + +#ifdef _WIN32 + + #ifdef MY_CPU_ARM + #define MY_CPU_ARM_LE + #endif + + #ifdef MY_CPU_ARM64 + #define MY_CPU_ARM64_LE + #endif + + #ifdef _M_IA64 + #define MY_CPU_IA64_LE + #endif + +#endif + + +#if defined(MY_CPU_X86_OR_AMD64) \ + || defined(MY_CPU_ARM_LE) \ + || defined(MY_CPU_ARM64_LE) \ + || defined(MY_CPU_IA64_LE) \ + || defined(__LITTLE_ENDIAN__) \ + || defined(__ARMEL__) \ + || defined(__THUMBEL__) \ + || defined(__AARCH64EL__) \ + || defined(__MIPSEL__) \ + || defined(__MIPSEL) \ + || defined(_MIPSEL) \ + || defined(__BFIN__) \ + || (defined(__BYTE_ORDER__) && (__BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__)) + #define MY_CPU_LE +#endif + +#if defined(__BIG_ENDIAN__) \ + || defined(__ARMEB__) \ + || defined(__THUMBEB__) \ + || defined(__AARCH64EB__) \ + || defined(__MIPSEB__) \ + || defined(__MIPSEB) \ + || defined(_MIPSEB) \ + || defined(__m68k__) \ + || defined(__s390__) \ + || defined(__s390x__) \ + || defined(__zarch__) \ + || (defined(__BYTE_ORDER__) && (__BYTE_ORDER__ == __ORDER_BIG_ENDIAN__)) + #define MY_CPU_BE +#endif + + +#if defined(MY_CPU_LE) && defined(MY_CPU_BE) + #error Stop_Compiling_Bad_Endian +#endif + + +#if defined(MY_CPU_32BIT) && defined(MY_CPU_64BIT) + #error Stop_Compiling_Bad_32_64_BIT +#endif + + +#ifndef MY_CPU_NAME + #ifdef MY_CPU_LE + #define MY_CPU_NAME "LE" + #elif defined(MY_CPU_BE) + #define MY_CPU_NAME "BE" + #else + /* + #define MY_CPU_NAME "" + */ + #endif +#endif + + + + + +#ifdef MY_CPU_LE + #if defined(MY_CPU_X86_OR_AMD64) \ + || defined(MY_CPU_ARM64) \ + || defined(__ARM_FEATURE_UNALIGNED) + #define MY_CPU_LE_UNALIGN + #endif +#endif + + +#ifdef MY_CPU_LE_UNALIGN + +#define GetUi16(p) (*(const UInt16 *)(const void *)(p)) +#define GetUi32(p) (*(const UInt32 *)(const void *)(p)) +#define GetUi64(p) (*(const UInt64 *)(const void *)(p)) + +#define SetUi16(p, v) { *(UInt16 *)(p) = (v); } +#define SetUi32(p, v) { *(UInt32 *)(p) = (v); } +#define SetUi64(p, v) { *(UInt64 *)(p) = (v); } + +#else + +#define GetUi16(p) ( (UInt16) ( \ + ((const Byte *)(p))[0] | \ + ((UInt16)((const Byte *)(p))[1] << 8) )) + +#define GetUi32(p) ( \ + ((const Byte *)(p))[0] | \ + ((UInt32)((const Byte *)(p))[1] << 8) | \ + ((UInt32)((const Byte *)(p))[2] << 16) | \ + ((UInt32)((const Byte *)(p))[3] << 24)) + +#define GetUi64(p) (GetUi32(p) | ((UInt64)GetUi32(((const Byte *)(p)) + 4) << 32)) + +#define SetUi16(p, v) { Byte *_ppp_ = (Byte *)(p); UInt32 _vvv_ = (v); \ + _ppp_[0] = (Byte)_vvv_; \ + _ppp_[1] = (Byte)(_vvv_ >> 8); } + +#define SetUi32(p, v) { Byte *_ppp_ = (Byte *)(p); UInt32 _vvv_ = (v); \ + _ppp_[0] = (Byte)_vvv_; \ + _ppp_[1] = (Byte)(_vvv_ >> 8); \ + _ppp_[2] = (Byte)(_vvv_ >> 16); \ + _ppp_[3] = (Byte)(_vvv_ >> 24); } + +#define SetUi64(p, v) { Byte *_ppp2_ = (Byte *)(p); UInt64 _vvv2_ = (v); \ + SetUi32(_ppp2_ , (UInt32)_vvv2_); \ + SetUi32(_ppp2_ + 4, (UInt32)(_vvv2_ >> 32)); } + +#endif + +#ifdef __has_builtin + #define MY__has_builtin(x) __has_builtin(x) +#else + #define MY__has_builtin(x) 0 +#endif + +#if defined(MY_CPU_LE_UNALIGN) && /* defined(_WIN64) && */ (_MSC_VER >= 1300) + +/* Note: we use bswap instruction, that is unsupported in 386 cpu */ + +#include <stdlib.h> + +#pragma intrinsic(_byteswap_ushort) +#pragma intrinsic(_byteswap_ulong) +#pragma intrinsic(_byteswap_uint64) + +/* #define GetBe16(p) _byteswap_ushort(*(const UInt16 *)(const Byte *)(p)) */ +#define GetBe32(p) _byteswap_ulong(*(const UInt32 *)(const Byte *)(p)) +#define GetBe64(p) _byteswap_uint64(*(const UInt64 *)(const Byte *)(p)) + +#define SetBe32(p, v) (*(UInt32 *)(void *)(p)) = _byteswap_ulong(v) + +#elif defined(MY_CPU_LE_UNALIGN) && ( \ + (defined(__GNUC__) && (__GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 3))) \ + || (defined(__clang__) && MY__has_builtin(__builtin_bswap16)) ) + +/* #define GetBe16(p) __builtin_bswap16(*(const UInt16 *)(const Byte *)(p)) */ +#define GetBe32(p) __builtin_bswap32(*(const UInt32 *)(const Byte *)(p)) +#define GetBe64(p) __builtin_bswap64(*(const UInt64 *)(const Byte *)(p)) + +#define SetBe32(p, v) (*(UInt32 *)(void *)(p)) = __builtin_bswap32(v) + +#else + +#define GetBe32(p) ( \ + ((UInt32)((const Byte *)(p))[0] << 24) | \ + ((UInt32)((const Byte *)(p))[1] << 16) | \ + ((UInt32)((const Byte *)(p))[2] << 8) | \ + ((const Byte *)(p))[3] ) + +#define GetBe64(p) (((UInt64)GetBe32(p) << 32) | GetBe32(((const Byte *)(p)) + 4)) + +#define SetBe32(p, v) { Byte *_ppp_ = (Byte *)(p); UInt32 _vvv_ = (v); \ + _ppp_[0] = (Byte)(_vvv_ >> 24); \ + _ppp_[1] = (Byte)(_vvv_ >> 16); \ + _ppp_[2] = (Byte)(_vvv_ >> 8); \ + _ppp_[3] = (Byte)_vvv_; } + +#endif + + +#ifndef GetBe16 + +#define GetBe16(p) ( (UInt16) ( \ + ((UInt16)((const Byte *)(p))[0] << 8) | \ + ((const Byte *)(p))[1] )) + +#endif + + + +#ifdef MY_CPU_X86_OR_AMD64 + +typedef struct +{ + UInt32 maxFunc; + UInt32 vendor[3]; + UInt32 ver; + UInt32 b; + UInt32 c; + UInt32 d; +} Cx86cpuid; + +enum +{ + CPU_FIRM_INTEL, + CPU_FIRM_AMD, + CPU_FIRM_VIA +}; + +void MyCPUID(UInt32 function, UInt32 *a, UInt32 *b, UInt32 *c, UInt32 *d); + +BoolInt x86cpuid_CheckAndRead(Cx86cpuid *p); +int x86cpuid_GetFirm(const Cx86cpuid *p); + +#define x86cpuid_GetFamily(ver) (((ver >> 16) & 0xFF0) | ((ver >> 8) & 0xF)) +#define x86cpuid_GetModel(ver) (((ver >> 12) & 0xF0) | ((ver >> 4) & 0xF)) +#define x86cpuid_GetStepping(ver) (ver & 0xF) + +BoolInt CPU_Is_InOrder(); +BoolInt CPU_Is_Aes_Supported(); +BoolInt CPU_IsSupported_PageGB(); + +#endif + +EXTERN_C_END + +#endif diff --git a/contrib/libs/lzmasdk/Lzma2Dec.c b/contrib/libs/lzmasdk/Lzma2Dec.c index 2e631051ba..9d9517eb1e 100644 --- a/contrib/libs/lzmasdk/Lzma2Dec.c +++ b/contrib/libs/lzmasdk/Lzma2Dec.c @@ -1,488 +1,488 @@ -/* Lzma2Dec.c -- LZMA2 Decoder
-2019-02-02 : Igor Pavlov : Public domain */
-
-/* #define SHOW_DEBUG_INFO */
-
-#include "Precomp.h"
-
-#ifdef SHOW_DEBUG_INFO
-#include <stdio.h>
-#endif
-
-#include <string.h>
-
-#include "Lzma2Dec.h"
-
-/*
-00000000 - End of data
-00000001 U U - Uncompressed, reset dic, need reset state and set new prop
-00000010 U U - Uncompressed, no reset
-100uuuuu U U P P - LZMA, no reset
-101uuuuu U U P P - LZMA, reset state
-110uuuuu U U P P S - LZMA, reset state + set new prop
-111uuuuu U U P P S - LZMA, reset state + set new prop, reset dic
-
- u, U - Unpack Size
- P - Pack Size
- S - Props
-*/
-
-#define LZMA2_CONTROL_COPY_RESET_DIC 1
-
-#define LZMA2_IS_UNCOMPRESSED_STATE(p) (((p)->control & (1 << 7)) == 0)
-
-#define LZMA2_LCLP_MAX 4
-#define LZMA2_DIC_SIZE_FROM_PROP(p) (((UInt32)2 | ((p) & 1)) << ((p) / 2 + 11))
-
-#ifdef SHOW_DEBUG_INFO
-#define PRF(x) x
-#else
-#define PRF(x)
-#endif
-
-typedef enum
-{
- LZMA2_STATE_CONTROL,
- LZMA2_STATE_UNPACK0,
- LZMA2_STATE_UNPACK1,
- LZMA2_STATE_PACK0,
- LZMA2_STATE_PACK1,
- LZMA2_STATE_PROP,
- LZMA2_STATE_DATA,
- LZMA2_STATE_DATA_CONT,
- LZMA2_STATE_FINISHED,
- LZMA2_STATE_ERROR
-} ELzma2State;
-
-static SRes Lzma2Dec_GetOldProps(Byte prop, Byte *props)
-{
- UInt32 dicSize;
- if (prop > 40)
- return SZ_ERROR_UNSUPPORTED;
- dicSize = (prop == 40) ? 0xFFFFFFFF : LZMA2_DIC_SIZE_FROM_PROP(prop);
- props[0] = (Byte)LZMA2_LCLP_MAX;
- props[1] = (Byte)(dicSize);
- props[2] = (Byte)(dicSize >> 8);
- props[3] = (Byte)(dicSize >> 16);
- props[4] = (Byte)(dicSize >> 24);
- return SZ_OK;
-}
-
-SRes Lzma2Dec_AllocateProbs(CLzma2Dec *p, Byte prop, ISzAllocPtr alloc)
-{
- Byte props[LZMA_PROPS_SIZE];
- RINOK(Lzma2Dec_GetOldProps(prop, props));
- return LzmaDec_AllocateProbs(&p->decoder, props, LZMA_PROPS_SIZE, alloc);
-}
-
-SRes Lzma2Dec_Allocate(CLzma2Dec *p, Byte prop, ISzAllocPtr alloc)
-{
- Byte props[LZMA_PROPS_SIZE];
- RINOK(Lzma2Dec_GetOldProps(prop, props));
- return LzmaDec_Allocate(&p->decoder, props, LZMA_PROPS_SIZE, alloc);
-}
-
-void Lzma2Dec_Init(CLzma2Dec *p)
-{
- p->state = LZMA2_STATE_CONTROL;
- p->needInitLevel = 0xE0;
- p->isExtraMode = False;
- p->unpackSize = 0;
-
- // p->decoder.dicPos = 0; // we can use it instead of full init
- LzmaDec_Init(&p->decoder);
-}
-
-static ELzma2State Lzma2Dec_UpdateState(CLzma2Dec *p, Byte b)
-{
- switch (p->state)
- {
- case LZMA2_STATE_CONTROL:
- p->isExtraMode = False;
- p->control = b;
- PRF(printf("\n %8X", (unsigned)p->decoder.dicPos));
- PRF(printf(" %02X", (unsigned)b));
- if (b == 0)
- return LZMA2_STATE_FINISHED;
- if (LZMA2_IS_UNCOMPRESSED_STATE(p))
- {
- if (b == LZMA2_CONTROL_COPY_RESET_DIC)
- p->needInitLevel = 0xC0;
- else if (b > 2 || p->needInitLevel == 0xE0)
- return LZMA2_STATE_ERROR;
- }
- else
- {
- if (b < p->needInitLevel)
- return LZMA2_STATE_ERROR;
- p->needInitLevel = 0;
- p->unpackSize = (UInt32)(b & 0x1F) << 16;
- }
- return LZMA2_STATE_UNPACK0;
-
- case LZMA2_STATE_UNPACK0:
- p->unpackSize |= (UInt32)b << 8;
- return LZMA2_STATE_UNPACK1;
-
- case LZMA2_STATE_UNPACK1:
- p->unpackSize |= (UInt32)b;
- p->unpackSize++;
- PRF(printf(" %7u", (unsigned)p->unpackSize));
- return LZMA2_IS_UNCOMPRESSED_STATE(p) ? LZMA2_STATE_DATA : LZMA2_STATE_PACK0;
-
- case LZMA2_STATE_PACK0:
- p->packSize = (UInt32)b << 8;
- return LZMA2_STATE_PACK1;
-
- case LZMA2_STATE_PACK1:
- p->packSize |= (UInt32)b;
- p->packSize++;
- // if (p->packSize < 5) return LZMA2_STATE_ERROR;
- PRF(printf(" %5u", (unsigned)p->packSize));
- return (p->control & 0x40) ? LZMA2_STATE_PROP : LZMA2_STATE_DATA;
-
- case LZMA2_STATE_PROP:
- {
- unsigned lc, lp;
- if (b >= (9 * 5 * 5))
- return LZMA2_STATE_ERROR;
- lc = b % 9;
- b /= 9;
- p->decoder.prop.pb = (Byte)(b / 5);
- lp = b % 5;
- if (lc + lp > LZMA2_LCLP_MAX)
- return LZMA2_STATE_ERROR;
- p->decoder.prop.lc = (Byte)lc;
- p->decoder.prop.lp = (Byte)lp;
- return LZMA2_STATE_DATA;
- }
- }
- return LZMA2_STATE_ERROR;
-}
-
-static void LzmaDec_UpdateWithUncompressed(CLzmaDec *p, const Byte *src, SizeT size)
-{
- memcpy(p->dic + p->dicPos, src, size);
- p->dicPos += size;
- if (p->checkDicSize == 0 && p->prop.dicSize - p->processedPos <= size)
- p->checkDicSize = p->prop.dicSize;
- p->processedPos += (UInt32)size;
-}
-
-void LzmaDec_InitDicAndState(CLzmaDec *p, BoolInt initDic, BoolInt initState);
-
-
-SRes Lzma2Dec_DecodeToDic(CLzma2Dec *p, SizeT dicLimit,
- const Byte *src, SizeT *srcLen, ELzmaFinishMode finishMode, ELzmaStatus *status)
-{
- SizeT inSize = *srcLen;
- *srcLen = 0;
- *status = LZMA_STATUS_NOT_SPECIFIED;
-
- while (p->state != LZMA2_STATE_ERROR)
- {
- SizeT dicPos;
-
- if (p->state == LZMA2_STATE_FINISHED)
- {
- *status = LZMA_STATUS_FINISHED_WITH_MARK;
- return SZ_OK;
- }
-
- dicPos = p->decoder.dicPos;
-
- if (dicPos == dicLimit && finishMode == LZMA_FINISH_ANY)
- {
- *status = LZMA_STATUS_NOT_FINISHED;
- return SZ_OK;
- }
-
- if (p->state != LZMA2_STATE_DATA && p->state != LZMA2_STATE_DATA_CONT)
- {
- if (*srcLen == inSize)
- {
- *status = LZMA_STATUS_NEEDS_MORE_INPUT;
- return SZ_OK;
- }
- (*srcLen)++;
- p->state = Lzma2Dec_UpdateState(p, *src++);
- if (dicPos == dicLimit && p->state != LZMA2_STATE_FINISHED)
- break;
- continue;
- }
-
- {
- SizeT inCur = inSize - *srcLen;
- SizeT outCur = dicLimit - dicPos;
- ELzmaFinishMode curFinishMode = LZMA_FINISH_ANY;
-
- if (outCur >= p->unpackSize)
- {
- outCur = (SizeT)p->unpackSize;
- curFinishMode = LZMA_FINISH_END;
- }
-
- if (LZMA2_IS_UNCOMPRESSED_STATE(p))
- {
- if (inCur == 0)
- {
- *status = LZMA_STATUS_NEEDS_MORE_INPUT;
- return SZ_OK;
- }
-
- if (p->state == LZMA2_STATE_DATA)
- {
- BoolInt initDic = (p->control == LZMA2_CONTROL_COPY_RESET_DIC);
- LzmaDec_InitDicAndState(&p->decoder, initDic, False);
- }
-
- if (inCur > outCur)
- inCur = outCur;
- if (inCur == 0)
- break;
-
- LzmaDec_UpdateWithUncompressed(&p->decoder, src, inCur);
-
- src += inCur;
- *srcLen += inCur;
- p->unpackSize -= (UInt32)inCur;
- p->state = (p->unpackSize == 0) ? LZMA2_STATE_CONTROL : LZMA2_STATE_DATA_CONT;
- }
- else
- {
- SRes res;
-
- if (p->state == LZMA2_STATE_DATA)
- {
- BoolInt initDic = (p->control >= 0xE0);
- BoolInt initState = (p->control >= 0xA0);
- LzmaDec_InitDicAndState(&p->decoder, initDic, initState);
- p->state = LZMA2_STATE_DATA_CONT;
- }
-
- if (inCur > p->packSize)
- inCur = (SizeT)p->packSize;
-
- res = LzmaDec_DecodeToDic(&p->decoder, dicPos + outCur, src, &inCur, curFinishMode, status);
-
- src += inCur;
- *srcLen += inCur;
- p->packSize -= (UInt32)inCur;
- outCur = p->decoder.dicPos - dicPos;
- p->unpackSize -= (UInt32)outCur;
-
- if (res != 0)
- break;
-
- if (*status == LZMA_STATUS_NEEDS_MORE_INPUT)
- {
- if (p->packSize == 0)
- break;
- return SZ_OK;
- }
-
- if (inCur == 0 && outCur == 0)
- {
- if (*status != LZMA_STATUS_MAYBE_FINISHED_WITHOUT_MARK
- || p->unpackSize != 0
- || p->packSize != 0)
- break;
- p->state = LZMA2_STATE_CONTROL;
- }
-
- *status = LZMA_STATUS_NOT_SPECIFIED;
- }
- }
- }
-
- *status = LZMA_STATUS_NOT_SPECIFIED;
- p->state = LZMA2_STATE_ERROR;
- return SZ_ERROR_DATA;
-}
-
-
-
-
-ELzma2ParseStatus Lzma2Dec_Parse(CLzma2Dec *p,
- SizeT outSize,
- const Byte *src, SizeT *srcLen,
- int checkFinishBlock)
-{
- SizeT inSize = *srcLen;
- *srcLen = 0;
-
- while (p->state != LZMA2_STATE_ERROR)
- {
- if (p->state == LZMA2_STATE_FINISHED)
- return (ELzma2ParseStatus)LZMA_STATUS_FINISHED_WITH_MARK;
-
- if (outSize == 0 && !checkFinishBlock)
- return (ELzma2ParseStatus)LZMA_STATUS_NOT_FINISHED;
-
- if (p->state != LZMA2_STATE_DATA && p->state != LZMA2_STATE_DATA_CONT)
- {
- if (*srcLen == inSize)
- return (ELzma2ParseStatus)LZMA_STATUS_NEEDS_MORE_INPUT;
- (*srcLen)++;
-
- p->state = Lzma2Dec_UpdateState(p, *src++);
-
- if (p->state == LZMA2_STATE_UNPACK0)
- {
- // if (p->decoder.dicPos != 0)
- if (p->control == LZMA2_CONTROL_COPY_RESET_DIC || p->control >= 0xE0)
- return LZMA2_PARSE_STATUS_NEW_BLOCK;
- // if (outSize == 0) return LZMA_STATUS_NOT_FINISHED;
- }
-
- // The following code can be commented.
- // It's not big problem, if we read additional input bytes.
- // It will be stopped later in LZMA2_STATE_DATA / LZMA2_STATE_DATA_CONT state.
-
- if (outSize == 0 && p->state != LZMA2_STATE_FINISHED)
- {
- // checkFinishBlock is true. So we expect that block must be finished,
- // We can return LZMA_STATUS_NOT_SPECIFIED or LZMA_STATUS_NOT_FINISHED here
- // break;
- return (ELzma2ParseStatus)LZMA_STATUS_NOT_FINISHED;
- }
-
- if (p->state == LZMA2_STATE_DATA)
- return LZMA2_PARSE_STATUS_NEW_CHUNK;
-
- continue;
- }
-
- if (outSize == 0)
- return (ELzma2ParseStatus)LZMA_STATUS_NOT_FINISHED;
-
- {
- SizeT inCur = inSize - *srcLen;
-
- if (LZMA2_IS_UNCOMPRESSED_STATE(p))
- {
- if (inCur == 0)
- return (ELzma2ParseStatus)LZMA_STATUS_NEEDS_MORE_INPUT;
- if (inCur > p->unpackSize)
- inCur = p->unpackSize;
- if (inCur > outSize)
- inCur = outSize;
- p->decoder.dicPos += inCur;
- src += inCur;
- *srcLen += inCur;
- outSize -= inCur;
- p->unpackSize -= (UInt32)inCur;
- p->state = (p->unpackSize == 0) ? LZMA2_STATE_CONTROL : LZMA2_STATE_DATA_CONT;
- }
- else
- {
- p->isExtraMode = True;
-
- if (inCur == 0)
- {
- if (p->packSize != 0)
- return (ELzma2ParseStatus)LZMA_STATUS_NEEDS_MORE_INPUT;
- }
- else if (p->state == LZMA2_STATE_DATA)
- {
- p->state = LZMA2_STATE_DATA_CONT;
- if (*src != 0)
- {
- // first byte of lzma chunk must be Zero
- *srcLen += 1;
- p->packSize--;
- break;
- }
- }
-
- if (inCur > p->packSize)
- inCur = (SizeT)p->packSize;
-
- src += inCur;
- *srcLen += inCur;
- p->packSize -= (UInt32)inCur;
-
- if (p->packSize == 0)
- {
- SizeT rem = outSize;
- if (rem > p->unpackSize)
- rem = p->unpackSize;
- p->decoder.dicPos += rem;
- p->unpackSize -= (UInt32)rem;
- outSize -= rem;
- if (p->unpackSize == 0)
- p->state = LZMA2_STATE_CONTROL;
- }
- }
- }
- }
-
- p->state = LZMA2_STATE_ERROR;
- return (ELzma2ParseStatus)LZMA_STATUS_NOT_SPECIFIED;
-}
-
-
-
-
-SRes Lzma2Dec_DecodeToBuf(CLzma2Dec *p, Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen, ELzmaFinishMode finishMode, ELzmaStatus *status)
-{
- SizeT outSize = *destLen, inSize = *srcLen;
- *srcLen = *destLen = 0;
-
- for (;;)
- {
- SizeT inCur = inSize, outCur, dicPos;
- ELzmaFinishMode curFinishMode;
- SRes res;
-
- if (p->decoder.dicPos == p->decoder.dicBufSize)
- p->decoder.dicPos = 0;
- dicPos = p->decoder.dicPos;
- curFinishMode = LZMA_FINISH_ANY;
- outCur = p->decoder.dicBufSize - dicPos;
-
- if (outCur >= outSize)
- {
- outCur = outSize;
- curFinishMode = finishMode;
- }
-
- res = Lzma2Dec_DecodeToDic(p, dicPos + outCur, src, &inCur, curFinishMode, status);
-
- src += inCur;
- inSize -= inCur;
- *srcLen += inCur;
- outCur = p->decoder.dicPos - dicPos;
- memcpy(dest, p->decoder.dic + dicPos, outCur);
- dest += outCur;
- outSize -= outCur;
- *destLen += outCur;
- if (res != 0)
- return res;
- if (outCur == 0 || outSize == 0)
- return SZ_OK;
- }
-}
-
-
-SRes Lzma2Decode(Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen,
- Byte prop, ELzmaFinishMode finishMode, ELzmaStatus *status, ISzAllocPtr alloc)
-{
- CLzma2Dec p;
- SRes res;
- SizeT outSize = *destLen, inSize = *srcLen;
- *destLen = *srcLen = 0;
- *status = LZMA_STATUS_NOT_SPECIFIED;
- Lzma2Dec_Construct(&p);
- RINOK(Lzma2Dec_AllocateProbs(&p, prop, alloc));
- p.decoder.dic = dest;
- p.decoder.dicBufSize = outSize;
- Lzma2Dec_Init(&p);
- *srcLen = inSize;
- res = Lzma2Dec_DecodeToDic(&p, outSize, src, srcLen, finishMode, status);
- *destLen = p.decoder.dicPos;
- if (res == SZ_OK && *status == LZMA_STATUS_NEEDS_MORE_INPUT)
- res = SZ_ERROR_INPUT_EOF;
- Lzma2Dec_FreeProbs(&p, alloc);
- return res;
-}
+/* Lzma2Dec.c -- LZMA2 Decoder +2019-02-02 : Igor Pavlov : Public domain */ + +/* #define SHOW_DEBUG_INFO */ + +#include "Precomp.h" + +#ifdef SHOW_DEBUG_INFO +#include <stdio.h> +#endif + +#include <string.h> + +#include "Lzma2Dec.h" + +/* +00000000 - End of data +00000001 U U - Uncompressed, reset dic, need reset state and set new prop +00000010 U U - Uncompressed, no reset +100uuuuu U U P P - LZMA, no reset +101uuuuu U U P P - LZMA, reset state +110uuuuu U U P P S - LZMA, reset state + set new prop +111uuuuu U U P P S - LZMA, reset state + set new prop, reset dic + + u, U - Unpack Size + P - Pack Size + S - Props +*/ + +#define LZMA2_CONTROL_COPY_RESET_DIC 1 + +#define LZMA2_IS_UNCOMPRESSED_STATE(p) (((p)->control & (1 << 7)) == 0) + +#define LZMA2_LCLP_MAX 4 +#define LZMA2_DIC_SIZE_FROM_PROP(p) (((UInt32)2 | ((p) & 1)) << ((p) / 2 + 11)) + +#ifdef SHOW_DEBUG_INFO +#define PRF(x) x +#else +#define PRF(x) +#endif + +typedef enum +{ + LZMA2_STATE_CONTROL, + LZMA2_STATE_UNPACK0, + LZMA2_STATE_UNPACK1, + LZMA2_STATE_PACK0, + LZMA2_STATE_PACK1, + LZMA2_STATE_PROP, + LZMA2_STATE_DATA, + LZMA2_STATE_DATA_CONT, + LZMA2_STATE_FINISHED, + LZMA2_STATE_ERROR +} ELzma2State; + +static SRes Lzma2Dec_GetOldProps(Byte prop, Byte *props) +{ + UInt32 dicSize; + if (prop > 40) + return SZ_ERROR_UNSUPPORTED; + dicSize = (prop == 40) ? 0xFFFFFFFF : LZMA2_DIC_SIZE_FROM_PROP(prop); + props[0] = (Byte)LZMA2_LCLP_MAX; + props[1] = (Byte)(dicSize); + props[2] = (Byte)(dicSize >> 8); + props[3] = (Byte)(dicSize >> 16); + props[4] = (Byte)(dicSize >> 24); + return SZ_OK; +} + +SRes Lzma2Dec_AllocateProbs(CLzma2Dec *p, Byte prop, ISzAllocPtr alloc) +{ + Byte props[LZMA_PROPS_SIZE]; + RINOK(Lzma2Dec_GetOldProps(prop, props)); + return LzmaDec_AllocateProbs(&p->decoder, props, LZMA_PROPS_SIZE, alloc); +} + +SRes Lzma2Dec_Allocate(CLzma2Dec *p, Byte prop, ISzAllocPtr alloc) +{ + Byte props[LZMA_PROPS_SIZE]; + RINOK(Lzma2Dec_GetOldProps(prop, props)); + return LzmaDec_Allocate(&p->decoder, props, LZMA_PROPS_SIZE, alloc); +} + +void Lzma2Dec_Init(CLzma2Dec *p) +{ + p->state = LZMA2_STATE_CONTROL; + p->needInitLevel = 0xE0; + p->isExtraMode = False; + p->unpackSize = 0; + + // p->decoder.dicPos = 0; // we can use it instead of full init + LzmaDec_Init(&p->decoder); +} + +static ELzma2State Lzma2Dec_UpdateState(CLzma2Dec *p, Byte b) +{ + switch (p->state) + { + case LZMA2_STATE_CONTROL: + p->isExtraMode = False; + p->control = b; + PRF(printf("\n %8X", (unsigned)p->decoder.dicPos)); + PRF(printf(" %02X", (unsigned)b)); + if (b == 0) + return LZMA2_STATE_FINISHED; + if (LZMA2_IS_UNCOMPRESSED_STATE(p)) + { + if (b == LZMA2_CONTROL_COPY_RESET_DIC) + p->needInitLevel = 0xC0; + else if (b > 2 || p->needInitLevel == 0xE0) + return LZMA2_STATE_ERROR; + } + else + { + if (b < p->needInitLevel) + return LZMA2_STATE_ERROR; + p->needInitLevel = 0; + p->unpackSize = (UInt32)(b & 0x1F) << 16; + } + return LZMA2_STATE_UNPACK0; + + case LZMA2_STATE_UNPACK0: + p->unpackSize |= (UInt32)b << 8; + return LZMA2_STATE_UNPACK1; + + case LZMA2_STATE_UNPACK1: + p->unpackSize |= (UInt32)b; + p->unpackSize++; + PRF(printf(" %7u", (unsigned)p->unpackSize)); + return LZMA2_IS_UNCOMPRESSED_STATE(p) ? LZMA2_STATE_DATA : LZMA2_STATE_PACK0; + + case LZMA2_STATE_PACK0: + p->packSize = (UInt32)b << 8; + return LZMA2_STATE_PACK1; + + case LZMA2_STATE_PACK1: + p->packSize |= (UInt32)b; + p->packSize++; + // if (p->packSize < 5) return LZMA2_STATE_ERROR; + PRF(printf(" %5u", (unsigned)p->packSize)); + return (p->control & 0x40) ? LZMA2_STATE_PROP : LZMA2_STATE_DATA; + + case LZMA2_STATE_PROP: + { + unsigned lc, lp; + if (b >= (9 * 5 * 5)) + return LZMA2_STATE_ERROR; + lc = b % 9; + b /= 9; + p->decoder.prop.pb = (Byte)(b / 5); + lp = b % 5; + if (lc + lp > LZMA2_LCLP_MAX) + return LZMA2_STATE_ERROR; + p->decoder.prop.lc = (Byte)lc; + p->decoder.prop.lp = (Byte)lp; + return LZMA2_STATE_DATA; + } + } + return LZMA2_STATE_ERROR; +} + +static void LzmaDec_UpdateWithUncompressed(CLzmaDec *p, const Byte *src, SizeT size) +{ + memcpy(p->dic + p->dicPos, src, size); + p->dicPos += size; + if (p->checkDicSize == 0 && p->prop.dicSize - p->processedPos <= size) + p->checkDicSize = p->prop.dicSize; + p->processedPos += (UInt32)size; +} + +void LzmaDec_InitDicAndState(CLzmaDec *p, BoolInt initDic, BoolInt initState); + + +SRes Lzma2Dec_DecodeToDic(CLzma2Dec *p, SizeT dicLimit, + const Byte *src, SizeT *srcLen, ELzmaFinishMode finishMode, ELzmaStatus *status) +{ + SizeT inSize = *srcLen; + *srcLen = 0; + *status = LZMA_STATUS_NOT_SPECIFIED; + + while (p->state != LZMA2_STATE_ERROR) + { + SizeT dicPos; + + if (p->state == LZMA2_STATE_FINISHED) + { + *status = LZMA_STATUS_FINISHED_WITH_MARK; + return SZ_OK; + } + + dicPos = p->decoder.dicPos; + + if (dicPos == dicLimit && finishMode == LZMA_FINISH_ANY) + { + *status = LZMA_STATUS_NOT_FINISHED; + return SZ_OK; + } + + if (p->state != LZMA2_STATE_DATA && p->state != LZMA2_STATE_DATA_CONT) + { + if (*srcLen == inSize) + { + *status = LZMA_STATUS_NEEDS_MORE_INPUT; + return SZ_OK; + } + (*srcLen)++; + p->state = Lzma2Dec_UpdateState(p, *src++); + if (dicPos == dicLimit && p->state != LZMA2_STATE_FINISHED) + break; + continue; + } + + { + SizeT inCur = inSize - *srcLen; + SizeT outCur = dicLimit - dicPos; + ELzmaFinishMode curFinishMode = LZMA_FINISH_ANY; + + if (outCur >= p->unpackSize) + { + outCur = (SizeT)p->unpackSize; + curFinishMode = LZMA_FINISH_END; + } + + if (LZMA2_IS_UNCOMPRESSED_STATE(p)) + { + if (inCur == 0) + { + *status = LZMA_STATUS_NEEDS_MORE_INPUT; + return SZ_OK; + } + + if (p->state == LZMA2_STATE_DATA) + { + BoolInt initDic = (p->control == LZMA2_CONTROL_COPY_RESET_DIC); + LzmaDec_InitDicAndState(&p->decoder, initDic, False); + } + + if (inCur > outCur) + inCur = outCur; + if (inCur == 0) + break; + + LzmaDec_UpdateWithUncompressed(&p->decoder, src, inCur); + + src += inCur; + *srcLen += inCur; + p->unpackSize -= (UInt32)inCur; + p->state = (p->unpackSize == 0) ? LZMA2_STATE_CONTROL : LZMA2_STATE_DATA_CONT; + } + else + { + SRes res; + + if (p->state == LZMA2_STATE_DATA) + { + BoolInt initDic = (p->control >= 0xE0); + BoolInt initState = (p->control >= 0xA0); + LzmaDec_InitDicAndState(&p->decoder, initDic, initState); + p->state = LZMA2_STATE_DATA_CONT; + } + + if (inCur > p->packSize) + inCur = (SizeT)p->packSize; + + res = LzmaDec_DecodeToDic(&p->decoder, dicPos + outCur, src, &inCur, curFinishMode, status); + + src += inCur; + *srcLen += inCur; + p->packSize -= (UInt32)inCur; + outCur = p->decoder.dicPos - dicPos; + p->unpackSize -= (UInt32)outCur; + + if (res != 0) + break; + + if (*status == LZMA_STATUS_NEEDS_MORE_INPUT) + { + if (p->packSize == 0) + break; + return SZ_OK; + } + + if (inCur == 0 && outCur == 0) + { + if (*status != LZMA_STATUS_MAYBE_FINISHED_WITHOUT_MARK + || p->unpackSize != 0 + || p->packSize != 0) + break; + p->state = LZMA2_STATE_CONTROL; + } + + *status = LZMA_STATUS_NOT_SPECIFIED; + } + } + } + + *status = LZMA_STATUS_NOT_SPECIFIED; + p->state = LZMA2_STATE_ERROR; + return SZ_ERROR_DATA; +} + + + + +ELzma2ParseStatus Lzma2Dec_Parse(CLzma2Dec *p, + SizeT outSize, + const Byte *src, SizeT *srcLen, + int checkFinishBlock) +{ + SizeT inSize = *srcLen; + *srcLen = 0; + + while (p->state != LZMA2_STATE_ERROR) + { + if (p->state == LZMA2_STATE_FINISHED) + return (ELzma2ParseStatus)LZMA_STATUS_FINISHED_WITH_MARK; + + if (outSize == 0 && !checkFinishBlock) + return (ELzma2ParseStatus)LZMA_STATUS_NOT_FINISHED; + + if (p->state != LZMA2_STATE_DATA && p->state != LZMA2_STATE_DATA_CONT) + { + if (*srcLen == inSize) + return (ELzma2ParseStatus)LZMA_STATUS_NEEDS_MORE_INPUT; + (*srcLen)++; + + p->state = Lzma2Dec_UpdateState(p, *src++); + + if (p->state == LZMA2_STATE_UNPACK0) + { + // if (p->decoder.dicPos != 0) + if (p->control == LZMA2_CONTROL_COPY_RESET_DIC || p->control >= 0xE0) + return LZMA2_PARSE_STATUS_NEW_BLOCK; + // if (outSize == 0) return LZMA_STATUS_NOT_FINISHED; + } + + // The following code can be commented. + // It's not big problem, if we read additional input bytes. + // It will be stopped later in LZMA2_STATE_DATA / LZMA2_STATE_DATA_CONT state. + + if (outSize == 0 && p->state != LZMA2_STATE_FINISHED) + { + // checkFinishBlock is true. So we expect that block must be finished, + // We can return LZMA_STATUS_NOT_SPECIFIED or LZMA_STATUS_NOT_FINISHED here + // break; + return (ELzma2ParseStatus)LZMA_STATUS_NOT_FINISHED; + } + + if (p->state == LZMA2_STATE_DATA) + return LZMA2_PARSE_STATUS_NEW_CHUNK; + + continue; + } + + if (outSize == 0) + return (ELzma2ParseStatus)LZMA_STATUS_NOT_FINISHED; + + { + SizeT inCur = inSize - *srcLen; + + if (LZMA2_IS_UNCOMPRESSED_STATE(p)) + { + if (inCur == 0) + return (ELzma2ParseStatus)LZMA_STATUS_NEEDS_MORE_INPUT; + if (inCur > p->unpackSize) + inCur = p->unpackSize; + if (inCur > outSize) + inCur = outSize; + p->decoder.dicPos += inCur; + src += inCur; + *srcLen += inCur; + outSize -= inCur; + p->unpackSize -= (UInt32)inCur; + p->state = (p->unpackSize == 0) ? LZMA2_STATE_CONTROL : LZMA2_STATE_DATA_CONT; + } + else + { + p->isExtraMode = True; + + if (inCur == 0) + { + if (p->packSize != 0) + return (ELzma2ParseStatus)LZMA_STATUS_NEEDS_MORE_INPUT; + } + else if (p->state == LZMA2_STATE_DATA) + { + p->state = LZMA2_STATE_DATA_CONT; + if (*src != 0) + { + // first byte of lzma chunk must be Zero + *srcLen += 1; + p->packSize--; + break; + } + } + + if (inCur > p->packSize) + inCur = (SizeT)p->packSize; + + src += inCur; + *srcLen += inCur; + p->packSize -= (UInt32)inCur; + + if (p->packSize == 0) + { + SizeT rem = outSize; + if (rem > p->unpackSize) + rem = p->unpackSize; + p->decoder.dicPos += rem; + p->unpackSize -= (UInt32)rem; + outSize -= rem; + if (p->unpackSize == 0) + p->state = LZMA2_STATE_CONTROL; + } + } + } + } + + p->state = LZMA2_STATE_ERROR; + return (ELzma2ParseStatus)LZMA_STATUS_NOT_SPECIFIED; +} + + + + +SRes Lzma2Dec_DecodeToBuf(CLzma2Dec *p, Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen, ELzmaFinishMode finishMode, ELzmaStatus *status) +{ + SizeT outSize = *destLen, inSize = *srcLen; + *srcLen = *destLen = 0; + + for (;;) + { + SizeT inCur = inSize, outCur, dicPos; + ELzmaFinishMode curFinishMode; + SRes res; + + if (p->decoder.dicPos == p->decoder.dicBufSize) + p->decoder.dicPos = 0; + dicPos = p->decoder.dicPos; + curFinishMode = LZMA_FINISH_ANY; + outCur = p->decoder.dicBufSize - dicPos; + + if (outCur >= outSize) + { + outCur = outSize; + curFinishMode = finishMode; + } + + res = Lzma2Dec_DecodeToDic(p, dicPos + outCur, src, &inCur, curFinishMode, status); + + src += inCur; + inSize -= inCur; + *srcLen += inCur; + outCur = p->decoder.dicPos - dicPos; + memcpy(dest, p->decoder.dic + dicPos, outCur); + dest += outCur; + outSize -= outCur; + *destLen += outCur; + if (res != 0) + return res; + if (outCur == 0 || outSize == 0) + return SZ_OK; + } +} + + +SRes Lzma2Decode(Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen, + Byte prop, ELzmaFinishMode finishMode, ELzmaStatus *status, ISzAllocPtr alloc) +{ + CLzma2Dec p; + SRes res; + SizeT outSize = *destLen, inSize = *srcLen; + *destLen = *srcLen = 0; + *status = LZMA_STATUS_NOT_SPECIFIED; + Lzma2Dec_Construct(&p); + RINOK(Lzma2Dec_AllocateProbs(&p, prop, alloc)); + p.decoder.dic = dest; + p.decoder.dicBufSize = outSize; + Lzma2Dec_Init(&p); + *srcLen = inSize; + res = Lzma2Dec_DecodeToDic(&p, outSize, src, srcLen, finishMode, status); + *destLen = p.decoder.dicPos; + if (res == SZ_OK && *status == LZMA_STATUS_NEEDS_MORE_INPUT) + res = SZ_ERROR_INPUT_EOF; + Lzma2Dec_FreeProbs(&p, alloc); + return res; +} diff --git a/contrib/libs/lzmasdk/Lzma2Dec.h b/contrib/libs/lzmasdk/Lzma2Dec.h index da50387250..74948d910b 100644 --- a/contrib/libs/lzmasdk/Lzma2Dec.h +++ b/contrib/libs/lzmasdk/Lzma2Dec.h @@ -1,120 +1,120 @@ -/* Lzma2Dec.h -- LZMA2 Decoder
-2018-02-19 : Igor Pavlov : Public domain */
-
-#ifndef __LZMA2_DEC_H
-#define __LZMA2_DEC_H
-
-#include "LzmaDec.h"
-
-EXTERN_C_BEGIN
-
-/* ---------- State Interface ---------- */
-
-typedef struct
-{
- unsigned state;
- Byte control;
- Byte needInitLevel;
- Byte isExtraMode;
- Byte _pad_;
- UInt32 packSize;
- UInt32 unpackSize;
- CLzmaDec decoder;
-} CLzma2Dec;
-
-#define Lzma2Dec_Construct(p) LzmaDec_Construct(&(p)->decoder)
-#define Lzma2Dec_FreeProbs(p, alloc) LzmaDec_FreeProbs(&(p)->decoder, alloc)
-#define Lzma2Dec_Free(p, alloc) LzmaDec_Free(&(p)->decoder, alloc)
-
-SRes Lzma2Dec_AllocateProbs(CLzma2Dec *p, Byte prop, ISzAllocPtr alloc);
-SRes Lzma2Dec_Allocate(CLzma2Dec *p, Byte prop, ISzAllocPtr alloc);
-void Lzma2Dec_Init(CLzma2Dec *p);
-
-/*
-finishMode:
- It has meaning only if the decoding reaches output limit (*destLen or dicLimit).
- LZMA_FINISH_ANY - use smallest number of input bytes
- LZMA_FINISH_END - read EndOfStream marker after decoding
-
-Returns:
- SZ_OK
- status:
- LZMA_STATUS_FINISHED_WITH_MARK
- LZMA_STATUS_NOT_FINISHED
- LZMA_STATUS_NEEDS_MORE_INPUT
- SZ_ERROR_DATA - Data error
-*/
-
-SRes Lzma2Dec_DecodeToDic(CLzma2Dec *p, SizeT dicLimit,
- const Byte *src, SizeT *srcLen, ELzmaFinishMode finishMode, ELzmaStatus *status);
-
-SRes Lzma2Dec_DecodeToBuf(CLzma2Dec *p, Byte *dest, SizeT *destLen,
- const Byte *src, SizeT *srcLen, ELzmaFinishMode finishMode, ELzmaStatus *status);
-
-
-/* ---------- LZMA2 block and chunk parsing ---------- */
-
-/*
-Lzma2Dec_Parse() parses compressed data stream up to next independent block or next chunk data.
-It can return LZMA_STATUS_* code or LZMA2_PARSE_STATUS_* code:
- - LZMA2_PARSE_STATUS_NEW_BLOCK - there is new block, and 1 additional byte (control byte of next block header) was read from input.
- - LZMA2_PARSE_STATUS_NEW_CHUNK - there is new chunk, and only lzma2 header of new chunk was read.
- CLzma2Dec::unpackSize contains unpack size of that chunk
-*/
-
-typedef enum
-{
-/*
- LZMA_STATUS_NOT_SPECIFIED // data error
- LZMA_STATUS_FINISHED_WITH_MARK
- LZMA_STATUS_NOT_FINISHED //
- LZMA_STATUS_NEEDS_MORE_INPUT
- LZMA_STATUS_MAYBE_FINISHED_WITHOUT_MARK // unused
-*/
- LZMA2_PARSE_STATUS_NEW_BLOCK = LZMA_STATUS_MAYBE_FINISHED_WITHOUT_MARK + 1,
- LZMA2_PARSE_STATUS_NEW_CHUNK
-} ELzma2ParseStatus;
-
-ELzma2ParseStatus Lzma2Dec_Parse(CLzma2Dec *p,
- SizeT outSize, // output size
- const Byte *src, SizeT *srcLen,
- int checkFinishBlock // set (checkFinishBlock = 1), if it must read full input data, if decoder.dicPos reaches blockMax position.
- );
-
-/*
-LZMA2 parser doesn't decode LZMA chunks, so we must read
- full input LZMA chunk to decode some part of LZMA chunk.
-
-Lzma2Dec_GetUnpackExtra() returns the value that shows
- max possible number of output bytes that can be output by decoder
- at current input positon.
-*/
-
-#define Lzma2Dec_GetUnpackExtra(p) ((p)->isExtraMode ? (p)->unpackSize : 0);
-
-
-/* ---------- One Call Interface ---------- */
-
-/*
-finishMode:
- It has meaning only if the decoding reaches output limit (*destLen).
- LZMA_FINISH_ANY - use smallest number of input bytes
- LZMA_FINISH_END - read EndOfStream marker after decoding
-
-Returns:
- SZ_OK
- status:
- LZMA_STATUS_FINISHED_WITH_MARK
- LZMA_STATUS_NOT_FINISHED
- SZ_ERROR_DATA - Data error
- SZ_ERROR_MEM - Memory allocation error
- SZ_ERROR_UNSUPPORTED - Unsupported properties
- SZ_ERROR_INPUT_EOF - It needs more bytes in input buffer (src).
-*/
-
-SRes Lzma2Decode(Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen,
- Byte prop, ELzmaFinishMode finishMode, ELzmaStatus *status, ISzAllocPtr alloc);
-
-EXTERN_C_END
-
-#endif
+/* Lzma2Dec.h -- LZMA2 Decoder +2018-02-19 : Igor Pavlov : Public domain */ + +#ifndef __LZMA2_DEC_H +#define __LZMA2_DEC_H + +#include "LzmaDec.h" + +EXTERN_C_BEGIN + +/* ---------- State Interface ---------- */ + +typedef struct +{ + unsigned state; + Byte control; + Byte needInitLevel; + Byte isExtraMode; + Byte _pad_; + UInt32 packSize; + UInt32 unpackSize; + CLzmaDec decoder; +} CLzma2Dec; + +#define Lzma2Dec_Construct(p) LzmaDec_Construct(&(p)->decoder) +#define Lzma2Dec_FreeProbs(p, alloc) LzmaDec_FreeProbs(&(p)->decoder, alloc) +#define Lzma2Dec_Free(p, alloc) LzmaDec_Free(&(p)->decoder, alloc) + +SRes Lzma2Dec_AllocateProbs(CLzma2Dec *p, Byte prop, ISzAllocPtr alloc); +SRes Lzma2Dec_Allocate(CLzma2Dec *p, Byte prop, ISzAllocPtr alloc); +void Lzma2Dec_Init(CLzma2Dec *p); + +/* +finishMode: + It has meaning only if the decoding reaches output limit (*destLen or dicLimit). + LZMA_FINISH_ANY - use smallest number of input bytes + LZMA_FINISH_END - read EndOfStream marker after decoding + +Returns: + SZ_OK + status: + LZMA_STATUS_FINISHED_WITH_MARK + LZMA_STATUS_NOT_FINISHED + LZMA_STATUS_NEEDS_MORE_INPUT + SZ_ERROR_DATA - Data error +*/ + +SRes Lzma2Dec_DecodeToDic(CLzma2Dec *p, SizeT dicLimit, + const Byte *src, SizeT *srcLen, ELzmaFinishMode finishMode, ELzmaStatus *status); + +SRes Lzma2Dec_DecodeToBuf(CLzma2Dec *p, Byte *dest, SizeT *destLen, + const Byte *src, SizeT *srcLen, ELzmaFinishMode finishMode, ELzmaStatus *status); + + +/* ---------- LZMA2 block and chunk parsing ---------- */ + +/* +Lzma2Dec_Parse() parses compressed data stream up to next independent block or next chunk data. +It can return LZMA_STATUS_* code or LZMA2_PARSE_STATUS_* code: + - LZMA2_PARSE_STATUS_NEW_BLOCK - there is new block, and 1 additional byte (control byte of next block header) was read from input. + - LZMA2_PARSE_STATUS_NEW_CHUNK - there is new chunk, and only lzma2 header of new chunk was read. + CLzma2Dec::unpackSize contains unpack size of that chunk +*/ + +typedef enum +{ +/* + LZMA_STATUS_NOT_SPECIFIED // data error + LZMA_STATUS_FINISHED_WITH_MARK + LZMA_STATUS_NOT_FINISHED // + LZMA_STATUS_NEEDS_MORE_INPUT + LZMA_STATUS_MAYBE_FINISHED_WITHOUT_MARK // unused +*/ + LZMA2_PARSE_STATUS_NEW_BLOCK = LZMA_STATUS_MAYBE_FINISHED_WITHOUT_MARK + 1, + LZMA2_PARSE_STATUS_NEW_CHUNK +} ELzma2ParseStatus; + +ELzma2ParseStatus Lzma2Dec_Parse(CLzma2Dec *p, + SizeT outSize, // output size + const Byte *src, SizeT *srcLen, + int checkFinishBlock // set (checkFinishBlock = 1), if it must read full input data, if decoder.dicPos reaches blockMax position. + ); + +/* +LZMA2 parser doesn't decode LZMA chunks, so we must read + full input LZMA chunk to decode some part of LZMA chunk. + +Lzma2Dec_GetUnpackExtra() returns the value that shows + max possible number of output bytes that can be output by decoder + at current input positon. +*/ + +#define Lzma2Dec_GetUnpackExtra(p) ((p)->isExtraMode ? (p)->unpackSize : 0); + + +/* ---------- One Call Interface ---------- */ + +/* +finishMode: + It has meaning only if the decoding reaches output limit (*destLen). + LZMA_FINISH_ANY - use smallest number of input bytes + LZMA_FINISH_END - read EndOfStream marker after decoding + +Returns: + SZ_OK + status: + LZMA_STATUS_FINISHED_WITH_MARK + LZMA_STATUS_NOT_FINISHED + SZ_ERROR_DATA - Data error + SZ_ERROR_MEM - Memory allocation error + SZ_ERROR_UNSUPPORTED - Unsupported properties + SZ_ERROR_INPUT_EOF - It needs more bytes in input buffer (src). +*/ + +SRes Lzma2Decode(Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen, + Byte prop, ELzmaFinishMode finishMode, ELzmaStatus *status, ISzAllocPtr alloc); + +EXTERN_C_END + +#endif diff --git a/contrib/libs/lzmasdk/Lzma2Enc.c b/contrib/libs/lzmasdk/Lzma2Enc.c index d541477527..a6fb10daac 100644 --- a/contrib/libs/lzmasdk/Lzma2Enc.c +++ b/contrib/libs/lzmasdk/Lzma2Enc.c @@ -1,803 +1,803 @@ -/* Lzma2Enc.c -- LZMA2 Encoder
-2018-07-04 : Igor Pavlov : Public domain */
-
-#include "Precomp.h"
-
-#include <string.h>
-
-/* #define _7ZIP_ST */
-
-#include "Lzma2Enc.h"
-
-#ifndef _7ZIP_ST
-#include "MtCoder.h"
-#else
-#define MTCODER__THREADS_MAX 1
-#endif
-
-#define LZMA2_CONTROL_LZMA (1 << 7)
-#define LZMA2_CONTROL_COPY_NO_RESET 2
-#define LZMA2_CONTROL_COPY_RESET_DIC 1
-#define LZMA2_CONTROL_EOF 0
-
-#define LZMA2_LCLP_MAX 4
-
-#define LZMA2_DIC_SIZE_FROM_PROP(p) (((UInt32)2 | ((p) & 1)) << ((p) / 2 + 11))
-
-#define LZMA2_PACK_SIZE_MAX (1 << 16)
-#define LZMA2_COPY_CHUNK_SIZE LZMA2_PACK_SIZE_MAX
-#define LZMA2_UNPACK_SIZE_MAX (1 << 21)
-#define LZMA2_KEEP_WINDOW_SIZE LZMA2_UNPACK_SIZE_MAX
-
-#define LZMA2_CHUNK_SIZE_COMPRESSED_MAX ((1 << 16) + 16)
-
-
-#define PRF(x) /* x */
-
-
-/* ---------- CLimitedSeqInStream ---------- */
-
-typedef struct
-{
- ISeqInStream vt;
- ISeqInStream *realStream;
- UInt64 limit;
- UInt64 processed;
- int finished;
-} CLimitedSeqInStream;
-
-static void LimitedSeqInStream_Init(CLimitedSeqInStream *p)
-{
- p->limit = (UInt64)(Int64)-1;
- p->processed = 0;
- p->finished = 0;
-}
-
-static SRes LimitedSeqInStream_Read(const ISeqInStream *pp, void *data, size_t *size)
-{
- CLimitedSeqInStream *p = CONTAINER_FROM_VTBL(pp, CLimitedSeqInStream, vt);
- size_t size2 = *size;
- SRes res = SZ_OK;
-
- if (p->limit != (UInt64)(Int64)-1)
- {
- UInt64 rem = p->limit - p->processed;
- if (size2 > rem)
- size2 = (size_t)rem;
- }
- if (size2 != 0)
- {
- res = ISeqInStream_Read(p->realStream, data, &size2);
- p->finished = (size2 == 0 ? 1 : 0);
- p->processed += size2;
- }
- *size = size2;
- return res;
-}
-
-
-/* ---------- CLzma2EncInt ---------- */
-
-typedef struct
-{
- CLzmaEncHandle enc;
- Byte propsAreSet;
- Byte propsByte;
- Byte needInitState;
- Byte needInitProp;
- UInt64 srcPos;
-} CLzma2EncInt;
-
-
-static SRes Lzma2EncInt_InitStream(CLzma2EncInt *p, const CLzma2EncProps *props)
-{
- if (!p->propsAreSet)
- {
- SizeT propsSize = LZMA_PROPS_SIZE;
- Byte propsEncoded[LZMA_PROPS_SIZE];
- RINOK(LzmaEnc_SetProps(p->enc, &props->lzmaProps));
- RINOK(LzmaEnc_WriteProperties(p->enc, propsEncoded, &propsSize));
- p->propsByte = propsEncoded[0];
- p->propsAreSet = True;
- }
- return SZ_OK;
-}
-
-static void Lzma2EncInt_InitBlock(CLzma2EncInt *p)
-{
- p->srcPos = 0;
- p->needInitState = True;
- p->needInitProp = True;
-}
-
-
-SRes LzmaEnc_PrepareForLzma2(CLzmaEncHandle pp, ISeqInStream *inStream, UInt32 keepWindowSize,
- ISzAllocPtr alloc, ISzAllocPtr allocBig);
-SRes LzmaEnc_MemPrepare(CLzmaEncHandle pp, const Byte *src, SizeT srcLen,
- UInt32 keepWindowSize, ISzAllocPtr alloc, ISzAllocPtr allocBig);
-SRes LzmaEnc_CodeOneMemBlock(CLzmaEncHandle pp, BoolInt reInit,
- Byte *dest, size_t *destLen, UInt32 desiredPackSize, UInt32 *unpackSize);
-const Byte *LzmaEnc_GetCurBuf(CLzmaEncHandle pp);
-void LzmaEnc_Finish(CLzmaEncHandle pp);
-void LzmaEnc_SaveState(CLzmaEncHandle pp);
-void LzmaEnc_RestoreState(CLzmaEncHandle pp);
-
-/*
-UInt32 LzmaEnc_GetNumAvailableBytes(CLzmaEncHandle pp);
-*/
-
-static SRes Lzma2EncInt_EncodeSubblock(CLzma2EncInt *p, Byte *outBuf,
- size_t *packSizeRes, ISeqOutStream *outStream)
-{
- size_t packSizeLimit = *packSizeRes;
- size_t packSize = packSizeLimit;
- UInt32 unpackSize = LZMA2_UNPACK_SIZE_MAX;
- unsigned lzHeaderSize = 5 + (p->needInitProp ? 1 : 0);
- BoolInt useCopyBlock;
- SRes res;
-
- *packSizeRes = 0;
- if (packSize < lzHeaderSize)
- return SZ_ERROR_OUTPUT_EOF;
- packSize -= lzHeaderSize;
-
- LzmaEnc_SaveState(p->enc);
- res = LzmaEnc_CodeOneMemBlock(p->enc, p->needInitState,
- outBuf + lzHeaderSize, &packSize, LZMA2_PACK_SIZE_MAX, &unpackSize);
-
- PRF(printf("\npackSize = %7d unpackSize = %7d ", packSize, unpackSize));
-
- if (unpackSize == 0)
- return res;
-
- if (res == SZ_OK)
- useCopyBlock = (packSize + 2 >= unpackSize || packSize > (1 << 16));
- else
- {
- if (res != SZ_ERROR_OUTPUT_EOF)
- return res;
- res = SZ_OK;
- useCopyBlock = True;
- }
-
- if (useCopyBlock)
- {
- size_t destPos = 0;
- PRF(printf("################# COPY "));
-
- while (unpackSize > 0)
- {
- UInt32 u = (unpackSize < LZMA2_COPY_CHUNK_SIZE) ? unpackSize : LZMA2_COPY_CHUNK_SIZE;
- if (packSizeLimit - destPos < u + 3)
- return SZ_ERROR_OUTPUT_EOF;
- outBuf[destPos++] = (Byte)(p->srcPos == 0 ? LZMA2_CONTROL_COPY_RESET_DIC : LZMA2_CONTROL_COPY_NO_RESET);
- outBuf[destPos++] = (Byte)((u - 1) >> 8);
- outBuf[destPos++] = (Byte)(u - 1);
- memcpy(outBuf + destPos, LzmaEnc_GetCurBuf(p->enc) - unpackSize, u);
- unpackSize -= u;
- destPos += u;
- p->srcPos += u;
-
- if (outStream)
- {
- *packSizeRes += destPos;
- if (ISeqOutStream_Write(outStream, outBuf, destPos) != destPos)
- return SZ_ERROR_WRITE;
- destPos = 0;
- }
- else
- *packSizeRes = destPos;
- /* needInitState = True; */
- }
-
- LzmaEnc_RestoreState(p->enc);
- return SZ_OK;
- }
-
- {
- size_t destPos = 0;
- UInt32 u = unpackSize - 1;
- UInt32 pm = (UInt32)(packSize - 1);
- unsigned mode = (p->srcPos == 0) ? 3 : (p->needInitState ? (p->needInitProp ? 2 : 1) : 0);
-
- PRF(printf(" "));
-
- outBuf[destPos++] = (Byte)(LZMA2_CONTROL_LZMA | (mode << 5) | ((u >> 16) & 0x1F));
- outBuf[destPos++] = (Byte)(u >> 8);
- outBuf[destPos++] = (Byte)u;
- outBuf[destPos++] = (Byte)(pm >> 8);
- outBuf[destPos++] = (Byte)pm;
-
- if (p->needInitProp)
- outBuf[destPos++] = p->propsByte;
-
- p->needInitProp = False;
- p->needInitState = False;
- destPos += packSize;
- p->srcPos += unpackSize;
-
- if (outStream)
- if (ISeqOutStream_Write(outStream, outBuf, destPos) != destPos)
- return SZ_ERROR_WRITE;
-
- *packSizeRes = destPos;
- return SZ_OK;
- }
-}
-
-
-/* ---------- Lzma2 Props ---------- */
-
-void Lzma2EncProps_Init(CLzma2EncProps *p)
-{
- LzmaEncProps_Init(&p->lzmaProps);
- p->blockSize = LZMA2_ENC_PROPS__BLOCK_SIZE__AUTO;
- p->numBlockThreads_Reduced = -1;
- p->numBlockThreads_Max = -1;
- p->numTotalThreads = -1;
-}
-
-void Lzma2EncProps_Normalize(CLzma2EncProps *p)
-{
- UInt64 fileSize;
- int t1, t1n, t2, t2r, t3;
- {
- CLzmaEncProps lzmaProps = p->lzmaProps;
- LzmaEncProps_Normalize(&lzmaProps);
- t1n = lzmaProps.numThreads;
- }
-
- t1 = p->lzmaProps.numThreads;
- t2 = p->numBlockThreads_Max;
- t3 = p->numTotalThreads;
-
- if (t2 > MTCODER__THREADS_MAX)
- t2 = MTCODER__THREADS_MAX;
-
- if (t3 <= 0)
- {
- if (t2 <= 0)
- t2 = 1;
- t3 = t1n * t2;
- }
- else if (t2 <= 0)
- {
- t2 = t3 / t1n;
- if (t2 == 0)
- {
- t1 = 1;
- t2 = t3;
- }
- if (t2 > MTCODER__THREADS_MAX)
- t2 = MTCODER__THREADS_MAX;
- }
- else if (t1 <= 0)
- {
- t1 = t3 / t2;
- if (t1 == 0)
- t1 = 1;
- }
- else
- t3 = t1n * t2;
-
- p->lzmaProps.numThreads = t1;
-
- t2r = t2;
-
- fileSize = p->lzmaProps.reduceSize;
-
- if ( p->blockSize != LZMA2_ENC_PROPS__BLOCK_SIZE__SOLID
- && p->blockSize != LZMA2_ENC_PROPS__BLOCK_SIZE__AUTO
- && (p->blockSize < fileSize || fileSize == (UInt64)(Int64)-1))
- p->lzmaProps.reduceSize = p->blockSize;
-
- LzmaEncProps_Normalize(&p->lzmaProps);
-
- p->lzmaProps.reduceSize = fileSize;
-
- t1 = p->lzmaProps.numThreads;
-
- if (p->blockSize == LZMA2_ENC_PROPS__BLOCK_SIZE__SOLID)
- {
- t2r = t2 = 1;
- t3 = t1;
- }
- else if (p->blockSize == LZMA2_ENC_PROPS__BLOCK_SIZE__AUTO && t2 <= 1)
- {
- /* if there is no block multi-threading, we use SOLID block */
- p->blockSize = LZMA2_ENC_PROPS__BLOCK_SIZE__SOLID;
- }
- else
- {
- if (p->blockSize == LZMA2_ENC_PROPS__BLOCK_SIZE__AUTO)
- {
- const UInt32 kMinSize = (UInt32)1 << 20;
- const UInt32 kMaxSize = (UInt32)1 << 28;
- const UInt32 dictSize = p->lzmaProps.dictSize;
- UInt64 blockSize = (UInt64)dictSize << 2;
- if (blockSize < kMinSize) blockSize = kMinSize;
- if (blockSize > kMaxSize) blockSize = kMaxSize;
- if (blockSize < dictSize) blockSize = dictSize;
- blockSize += (kMinSize - 1);
- blockSize &= ~(UInt64)(kMinSize - 1);
- p->blockSize = blockSize;
- }
-
- if (t2 > 1 && fileSize != (UInt64)(Int64)-1)
- {
- UInt64 numBlocks = fileSize / p->blockSize;
- if (numBlocks * p->blockSize != fileSize)
- numBlocks++;
- if (numBlocks < (unsigned)t2)
- {
- t2r = (unsigned)numBlocks;
- if (t2r == 0)
- t2r = 1;
- t3 = t1 * t2r;
- }
- }
- }
-
- p->numBlockThreads_Max = t2;
- p->numBlockThreads_Reduced = t2r;
- p->numTotalThreads = t3;
-}
-
-
-static SRes Progress(ICompressProgress *p, UInt64 inSize, UInt64 outSize)
-{
- return (p && ICompressProgress_Progress(p, inSize, outSize) != SZ_OK) ? SZ_ERROR_PROGRESS : SZ_OK;
-}
-
-
-/* ---------- Lzma2 ---------- */
-
-typedef struct
-{
- Byte propEncoded;
- CLzma2EncProps props;
- UInt64 expectedDataSize;
-
- Byte *tempBufLzma;
-
- ISzAllocPtr alloc;
- ISzAllocPtr allocBig;
-
- CLzma2EncInt coders[MTCODER__THREADS_MAX];
-
- #ifndef _7ZIP_ST
-
- ISeqOutStream *outStream;
- Byte *outBuf;
- size_t outBuf_Rem; /* remainder in outBuf */
-
- size_t outBufSize; /* size of allocated outBufs[i] */
- size_t outBufsDataSizes[MTCODER__BLOCKS_MAX];
- BoolInt mtCoder_WasConstructed;
- CMtCoder mtCoder;
- Byte *outBufs[MTCODER__BLOCKS_MAX];
-
- #endif
-
-} CLzma2Enc;
-
-
-
-CLzma2EncHandle Lzma2Enc_Create(ISzAllocPtr alloc, ISzAllocPtr allocBig)
-{
- CLzma2Enc *p = (CLzma2Enc *)ISzAlloc_Alloc(alloc, sizeof(CLzma2Enc));
- if (!p)
- return NULL;
- Lzma2EncProps_Init(&p->props);
- Lzma2EncProps_Normalize(&p->props);
- p->expectedDataSize = (UInt64)(Int64)-1;
- p->tempBufLzma = NULL;
- p->alloc = alloc;
- p->allocBig = allocBig;
- {
- unsigned i;
- for (i = 0; i < MTCODER__THREADS_MAX; i++)
- p->coders[i].enc = NULL;
- }
-
- #ifndef _7ZIP_ST
- p->mtCoder_WasConstructed = False;
- {
- unsigned i;
- for (i = 0; i < MTCODER__BLOCKS_MAX; i++)
- p->outBufs[i] = NULL;
- p->outBufSize = 0;
- }
- #endif
-
- return p;
-}
-
-
-#ifndef _7ZIP_ST
-
-static void Lzma2Enc_FreeOutBufs(CLzma2Enc *p)
-{
- unsigned i;
- for (i = 0; i < MTCODER__BLOCKS_MAX; i++)
- if (p->outBufs[i])
- {
- ISzAlloc_Free(p->alloc, p->outBufs[i]);
- p->outBufs[i] = NULL;
- }
- p->outBufSize = 0;
-}
-
-#endif
-
-
-void Lzma2Enc_Destroy(CLzma2EncHandle pp)
-{
- CLzma2Enc *p = (CLzma2Enc *)pp;
- unsigned i;
- for (i = 0; i < MTCODER__THREADS_MAX; i++)
- {
- CLzma2EncInt *t = &p->coders[i];
- if (t->enc)
- {
- LzmaEnc_Destroy(t->enc, p->alloc, p->allocBig);
- t->enc = NULL;
- }
- }
-
-
- #ifndef _7ZIP_ST
- if (p->mtCoder_WasConstructed)
- {
- MtCoder_Destruct(&p->mtCoder);
- p->mtCoder_WasConstructed = False;
- }
- Lzma2Enc_FreeOutBufs(p);
- #endif
-
- ISzAlloc_Free(p->alloc, p->tempBufLzma);
- p->tempBufLzma = NULL;
-
- ISzAlloc_Free(p->alloc, pp);
-}
-
-
-SRes Lzma2Enc_SetProps(CLzma2EncHandle pp, const CLzma2EncProps *props)
-{
- CLzma2Enc *p = (CLzma2Enc *)pp;
- CLzmaEncProps lzmaProps = props->lzmaProps;
- LzmaEncProps_Normalize(&lzmaProps);
- if (lzmaProps.lc + lzmaProps.lp > LZMA2_LCLP_MAX)
- return SZ_ERROR_PARAM;
- p->props = *props;
- Lzma2EncProps_Normalize(&p->props);
- return SZ_OK;
-}
-
-
-void Lzma2Enc_SetDataSize(CLzmaEncHandle pp, UInt64 expectedDataSiize)
-{
- CLzma2Enc *p = (CLzma2Enc *)pp;
- p->expectedDataSize = expectedDataSiize;
-}
-
-
-Byte Lzma2Enc_WriteProperties(CLzma2EncHandle pp)
-{
- CLzma2Enc *p = (CLzma2Enc *)pp;
- unsigned i;
- UInt32 dicSize = LzmaEncProps_GetDictSize(&p->props.lzmaProps);
- for (i = 0; i < 40; i++)
- if (dicSize <= LZMA2_DIC_SIZE_FROM_PROP(i))
- break;
- return (Byte)i;
-}
-
-
-static SRes Lzma2Enc_EncodeMt1(
- CLzma2Enc *me,
- CLzma2EncInt *p,
- ISeqOutStream *outStream,
- Byte *outBuf, size_t *outBufSize,
- ISeqInStream *inStream,
- const Byte *inData, size_t inDataSize,
- int finished,
- ICompressProgress *progress)
-{
- UInt64 unpackTotal = 0;
- UInt64 packTotal = 0;
- size_t outLim = 0;
- CLimitedSeqInStream limitedInStream;
-
- if (outBuf)
- {
- outLim = *outBufSize;
- *outBufSize = 0;
- }
-
- if (!p->enc)
- {
- p->propsAreSet = False;
- p->enc = LzmaEnc_Create(me->alloc);
- if (!p->enc)
- return SZ_ERROR_MEM;
- }
-
- limitedInStream.realStream = inStream;
- if (inStream)
- {
- limitedInStream.vt.Read = LimitedSeqInStream_Read;
- }
-
- if (!outBuf)
- {
- // outStream version works only in one thread. So we use CLzma2Enc::tempBufLzma
- if (!me->tempBufLzma)
- {
- me->tempBufLzma = (Byte *)ISzAlloc_Alloc(me->alloc, LZMA2_CHUNK_SIZE_COMPRESSED_MAX);
- if (!me->tempBufLzma)
- return SZ_ERROR_MEM;
- }
- }
-
- RINOK(Lzma2EncInt_InitStream(p, &me->props));
-
- for (;;)
- {
- SRes res = SZ_OK;
- size_t inSizeCur = 0;
-
- Lzma2EncInt_InitBlock(p);
-
- LimitedSeqInStream_Init(&limitedInStream);
- limitedInStream.limit = me->props.blockSize;
-
- if (inStream)
- {
- UInt64 expected = (UInt64)(Int64)-1;
- // inStream version works only in one thread. So we use CLzma2Enc::expectedDataSize
- if (me->expectedDataSize != (UInt64)(Int64)-1
- && me->expectedDataSize >= unpackTotal)
- expected = me->expectedDataSize - unpackTotal;
- if (me->props.blockSize != LZMA2_ENC_PROPS__BLOCK_SIZE__SOLID
- && expected > me->props.blockSize)
- expected = (size_t)me->props.blockSize;
-
- LzmaEnc_SetDataSize(p->enc, expected);
-
- RINOK(LzmaEnc_PrepareForLzma2(p->enc,
- &limitedInStream.vt,
- LZMA2_KEEP_WINDOW_SIZE,
- me->alloc,
- me->allocBig));
- }
- else
- {
- inSizeCur = inDataSize - (size_t)unpackTotal;
- if (me->props.blockSize != LZMA2_ENC_PROPS__BLOCK_SIZE__SOLID
- && inSizeCur > me->props.blockSize)
- inSizeCur = (size_t)me->props.blockSize;
-
- // LzmaEnc_SetDataSize(p->enc, inSizeCur);
-
- RINOK(LzmaEnc_MemPrepare(p->enc,
- inData + (size_t)unpackTotal, inSizeCur,
- LZMA2_KEEP_WINDOW_SIZE,
- me->alloc,
- me->allocBig));
- }
-
- for (;;)
- {
- size_t packSize = LZMA2_CHUNK_SIZE_COMPRESSED_MAX;
- if (outBuf)
- packSize = outLim - (size_t)packTotal;
-
- res = Lzma2EncInt_EncodeSubblock(p,
- outBuf ? outBuf + (size_t)packTotal : me->tempBufLzma, &packSize,
- outBuf ? NULL : outStream);
-
- if (res != SZ_OK)
- break;
-
- packTotal += packSize;
- if (outBuf)
- *outBufSize = (size_t)packTotal;
-
- res = Progress(progress, unpackTotal + p->srcPos, packTotal);
- if (res != SZ_OK)
- break;
-
- /*
- if (LzmaEnc_GetNumAvailableBytes(p->enc) == 0)
- break;
- */
-
- if (packSize == 0)
- break;
- }
-
- LzmaEnc_Finish(p->enc);
-
- unpackTotal += p->srcPos;
-
- RINOK(res);
-
- if (p->srcPos != (inStream ? limitedInStream.processed : inSizeCur))
- return SZ_ERROR_FAIL;
-
- if (inStream ? limitedInStream.finished : (unpackTotal == inDataSize))
- {
- if (finished)
- {
- if (outBuf)
- {
- size_t destPos = *outBufSize;
- if (destPos >= outLim)
- return SZ_ERROR_OUTPUT_EOF;
- outBuf[destPos] = 0;
- *outBufSize = destPos + 1;
- }
- else
- {
- Byte b = 0;
- if (ISeqOutStream_Write(outStream, &b, 1) != 1)
- return SZ_ERROR_WRITE;
- }
- }
- return SZ_OK;
- }
- }
-}
-
-
-
-#ifndef _7ZIP_ST
-
-static SRes Lzma2Enc_MtCallback_Code(void *pp, unsigned coderIndex, unsigned outBufIndex,
- const Byte *src, size_t srcSize, int finished)
-{
- CLzma2Enc *me = (CLzma2Enc *)pp;
- size_t destSize = me->outBufSize;
- SRes res;
- CMtProgressThunk progressThunk;
-
- Byte *dest = me->outBufs[outBufIndex];
-
- me->outBufsDataSizes[outBufIndex] = 0;
-
- if (!dest)
- {
- dest = (Byte *)ISzAlloc_Alloc(me->alloc, me->outBufSize);
- if (!dest)
- return SZ_ERROR_MEM;
- me->outBufs[outBufIndex] = dest;
- }
-
- MtProgressThunk_CreateVTable(&progressThunk);
- progressThunk.mtProgress = &me->mtCoder.mtProgress;
- progressThunk.inSize = 0;
- progressThunk.outSize = 0;
-
- res = Lzma2Enc_EncodeMt1(me,
- &me->coders[coderIndex],
- NULL, dest, &destSize,
- NULL, src, srcSize,
- finished,
- &progressThunk.vt);
-
- me->outBufsDataSizes[outBufIndex] = destSize;
-
- return res;
-}
-
-
-static SRes Lzma2Enc_MtCallback_Write(void *pp, unsigned outBufIndex)
-{
- CLzma2Enc *me = (CLzma2Enc *)pp;
- size_t size = me->outBufsDataSizes[outBufIndex];
- const Byte *data = me->outBufs[outBufIndex];
-
- if (me->outStream)
- return ISeqOutStream_Write(me->outStream, data, size) == size ? SZ_OK : SZ_ERROR_WRITE;
-
- if (size > me->outBuf_Rem)
- return SZ_ERROR_OUTPUT_EOF;
- memcpy(me->outBuf, data, size);
- me->outBuf_Rem -= size;
- me->outBuf += size;
- return SZ_OK;
-}
-
-#endif
-
-
-
-SRes Lzma2Enc_Encode2(CLzma2EncHandle pp,
- ISeqOutStream *outStream,
- Byte *outBuf, size_t *outBufSize,
- ISeqInStream *inStream,
- const Byte *inData, size_t inDataSize,
- ICompressProgress *progress)
-{
- CLzma2Enc *p = (CLzma2Enc *)pp;
-
- if (inStream && inData)
- return SZ_ERROR_PARAM;
-
- if (outStream && outBuf)
- return SZ_ERROR_PARAM;
-
- {
- unsigned i;
- for (i = 0; i < MTCODER__THREADS_MAX; i++)
- p->coders[i].propsAreSet = False;
- }
-
- #ifndef _7ZIP_ST
-
- if (p->props.numBlockThreads_Reduced > 1)
- {
- IMtCoderCallback2 vt;
-
- if (!p->mtCoder_WasConstructed)
- {
- p->mtCoder_WasConstructed = True;
- MtCoder_Construct(&p->mtCoder);
- }
-
- vt.Code = Lzma2Enc_MtCallback_Code;
- vt.Write = Lzma2Enc_MtCallback_Write;
-
- p->outStream = outStream;
- p->outBuf = NULL;
- p->outBuf_Rem = 0;
- if (!outStream)
- {
- p->outBuf = outBuf;
- p->outBuf_Rem = *outBufSize;
- *outBufSize = 0;
- }
-
- p->mtCoder.allocBig = p->allocBig;
- p->mtCoder.progress = progress;
- p->mtCoder.inStream = inStream;
- p->mtCoder.inData = inData;
- p->mtCoder.inDataSize = inDataSize;
- p->mtCoder.mtCallback = &vt;
- p->mtCoder.mtCallbackObject = p;
-
- p->mtCoder.blockSize = (size_t)p->props.blockSize;
- if (p->mtCoder.blockSize != p->props.blockSize)
- return SZ_ERROR_PARAM; /* SZ_ERROR_MEM */
-
- {
- size_t destBlockSize = p->mtCoder.blockSize + (p->mtCoder.blockSize >> 10) + 16;
- if (destBlockSize < p->mtCoder.blockSize)
- return SZ_ERROR_PARAM;
- if (p->outBufSize != destBlockSize)
- Lzma2Enc_FreeOutBufs(p);
- p->outBufSize = destBlockSize;
- }
-
- p->mtCoder.numThreadsMax = p->props.numBlockThreads_Max;
- p->mtCoder.expectedDataSize = p->expectedDataSize;
-
- {
- SRes res = MtCoder_Code(&p->mtCoder);
- if (!outStream)
- *outBufSize = p->outBuf - outBuf;
- return res;
- }
- }
-
- #endif
-
-
- return Lzma2Enc_EncodeMt1(p,
- &p->coders[0],
- outStream, outBuf, outBufSize,
- inStream, inData, inDataSize,
- True, /* finished */
- progress);
-}
+/* Lzma2Enc.c -- LZMA2 Encoder +2018-07-04 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include <string.h> + +/* #define _7ZIP_ST */ + +#include "Lzma2Enc.h" + +#ifndef _7ZIP_ST +#include "MtCoder.h" +#else +#define MTCODER__THREADS_MAX 1 +#endif + +#define LZMA2_CONTROL_LZMA (1 << 7) +#define LZMA2_CONTROL_COPY_NO_RESET 2 +#define LZMA2_CONTROL_COPY_RESET_DIC 1 +#define LZMA2_CONTROL_EOF 0 + +#define LZMA2_LCLP_MAX 4 + +#define LZMA2_DIC_SIZE_FROM_PROP(p) (((UInt32)2 | ((p) & 1)) << ((p) / 2 + 11)) + +#define LZMA2_PACK_SIZE_MAX (1 << 16) +#define LZMA2_COPY_CHUNK_SIZE LZMA2_PACK_SIZE_MAX +#define LZMA2_UNPACK_SIZE_MAX (1 << 21) +#define LZMA2_KEEP_WINDOW_SIZE LZMA2_UNPACK_SIZE_MAX + +#define LZMA2_CHUNK_SIZE_COMPRESSED_MAX ((1 << 16) + 16) + + +#define PRF(x) /* x */ + + +/* ---------- CLimitedSeqInStream ---------- */ + +typedef struct +{ + ISeqInStream vt; + ISeqInStream *realStream; + UInt64 limit; + UInt64 processed; + int finished; +} CLimitedSeqInStream; + +static void LimitedSeqInStream_Init(CLimitedSeqInStream *p) +{ + p->limit = (UInt64)(Int64)-1; + p->processed = 0; + p->finished = 0; +} + +static SRes LimitedSeqInStream_Read(const ISeqInStream *pp, void *data, size_t *size) +{ + CLimitedSeqInStream *p = CONTAINER_FROM_VTBL(pp, CLimitedSeqInStream, vt); + size_t size2 = *size; + SRes res = SZ_OK; + + if (p->limit != (UInt64)(Int64)-1) + { + UInt64 rem = p->limit - p->processed; + if (size2 > rem) + size2 = (size_t)rem; + } + if (size2 != 0) + { + res = ISeqInStream_Read(p->realStream, data, &size2); + p->finished = (size2 == 0 ? 1 : 0); + p->processed += size2; + } + *size = size2; + return res; +} + + +/* ---------- CLzma2EncInt ---------- */ + +typedef struct +{ + CLzmaEncHandle enc; + Byte propsAreSet; + Byte propsByte; + Byte needInitState; + Byte needInitProp; + UInt64 srcPos; +} CLzma2EncInt; + + +static SRes Lzma2EncInt_InitStream(CLzma2EncInt *p, const CLzma2EncProps *props) +{ + if (!p->propsAreSet) + { + SizeT propsSize = LZMA_PROPS_SIZE; + Byte propsEncoded[LZMA_PROPS_SIZE]; + RINOK(LzmaEnc_SetProps(p->enc, &props->lzmaProps)); + RINOK(LzmaEnc_WriteProperties(p->enc, propsEncoded, &propsSize)); + p->propsByte = propsEncoded[0]; + p->propsAreSet = True; + } + return SZ_OK; +} + +static void Lzma2EncInt_InitBlock(CLzma2EncInt *p) +{ + p->srcPos = 0; + p->needInitState = True; + p->needInitProp = True; +} + + +SRes LzmaEnc_PrepareForLzma2(CLzmaEncHandle pp, ISeqInStream *inStream, UInt32 keepWindowSize, + ISzAllocPtr alloc, ISzAllocPtr allocBig); +SRes LzmaEnc_MemPrepare(CLzmaEncHandle pp, const Byte *src, SizeT srcLen, + UInt32 keepWindowSize, ISzAllocPtr alloc, ISzAllocPtr allocBig); +SRes LzmaEnc_CodeOneMemBlock(CLzmaEncHandle pp, BoolInt reInit, + Byte *dest, size_t *destLen, UInt32 desiredPackSize, UInt32 *unpackSize); +const Byte *LzmaEnc_GetCurBuf(CLzmaEncHandle pp); +void LzmaEnc_Finish(CLzmaEncHandle pp); +void LzmaEnc_SaveState(CLzmaEncHandle pp); +void LzmaEnc_RestoreState(CLzmaEncHandle pp); + +/* +UInt32 LzmaEnc_GetNumAvailableBytes(CLzmaEncHandle pp); +*/ + +static SRes Lzma2EncInt_EncodeSubblock(CLzma2EncInt *p, Byte *outBuf, + size_t *packSizeRes, ISeqOutStream *outStream) +{ + size_t packSizeLimit = *packSizeRes; + size_t packSize = packSizeLimit; + UInt32 unpackSize = LZMA2_UNPACK_SIZE_MAX; + unsigned lzHeaderSize = 5 + (p->needInitProp ? 1 : 0); + BoolInt useCopyBlock; + SRes res; + + *packSizeRes = 0; + if (packSize < lzHeaderSize) + return SZ_ERROR_OUTPUT_EOF; + packSize -= lzHeaderSize; + + LzmaEnc_SaveState(p->enc); + res = LzmaEnc_CodeOneMemBlock(p->enc, p->needInitState, + outBuf + lzHeaderSize, &packSize, LZMA2_PACK_SIZE_MAX, &unpackSize); + + PRF(printf("\npackSize = %7d unpackSize = %7d ", packSize, unpackSize)); + + if (unpackSize == 0) + return res; + + if (res == SZ_OK) + useCopyBlock = (packSize + 2 >= unpackSize || packSize > (1 << 16)); + else + { + if (res != SZ_ERROR_OUTPUT_EOF) + return res; + res = SZ_OK; + useCopyBlock = True; + } + + if (useCopyBlock) + { + size_t destPos = 0; + PRF(printf("################# COPY ")); + + while (unpackSize > 0) + { + UInt32 u = (unpackSize < LZMA2_COPY_CHUNK_SIZE) ? unpackSize : LZMA2_COPY_CHUNK_SIZE; + if (packSizeLimit - destPos < u + 3) + return SZ_ERROR_OUTPUT_EOF; + outBuf[destPos++] = (Byte)(p->srcPos == 0 ? LZMA2_CONTROL_COPY_RESET_DIC : LZMA2_CONTROL_COPY_NO_RESET); + outBuf[destPos++] = (Byte)((u - 1) >> 8); + outBuf[destPos++] = (Byte)(u - 1); + memcpy(outBuf + destPos, LzmaEnc_GetCurBuf(p->enc) - unpackSize, u); + unpackSize -= u; + destPos += u; + p->srcPos += u; + + if (outStream) + { + *packSizeRes += destPos; + if (ISeqOutStream_Write(outStream, outBuf, destPos) != destPos) + return SZ_ERROR_WRITE; + destPos = 0; + } + else + *packSizeRes = destPos; + /* needInitState = True; */ + } + + LzmaEnc_RestoreState(p->enc); + return SZ_OK; + } + + { + size_t destPos = 0; + UInt32 u = unpackSize - 1; + UInt32 pm = (UInt32)(packSize - 1); + unsigned mode = (p->srcPos == 0) ? 3 : (p->needInitState ? (p->needInitProp ? 2 : 1) : 0); + + PRF(printf(" ")); + + outBuf[destPos++] = (Byte)(LZMA2_CONTROL_LZMA | (mode << 5) | ((u >> 16) & 0x1F)); + outBuf[destPos++] = (Byte)(u >> 8); + outBuf[destPos++] = (Byte)u; + outBuf[destPos++] = (Byte)(pm >> 8); + outBuf[destPos++] = (Byte)pm; + + if (p->needInitProp) + outBuf[destPos++] = p->propsByte; + + p->needInitProp = False; + p->needInitState = False; + destPos += packSize; + p->srcPos += unpackSize; + + if (outStream) + if (ISeqOutStream_Write(outStream, outBuf, destPos) != destPos) + return SZ_ERROR_WRITE; + + *packSizeRes = destPos; + return SZ_OK; + } +} + + +/* ---------- Lzma2 Props ---------- */ + +void Lzma2EncProps_Init(CLzma2EncProps *p) +{ + LzmaEncProps_Init(&p->lzmaProps); + p->blockSize = LZMA2_ENC_PROPS__BLOCK_SIZE__AUTO; + p->numBlockThreads_Reduced = -1; + p->numBlockThreads_Max = -1; + p->numTotalThreads = -1; +} + +void Lzma2EncProps_Normalize(CLzma2EncProps *p) +{ + UInt64 fileSize; + int t1, t1n, t2, t2r, t3; + { + CLzmaEncProps lzmaProps = p->lzmaProps; + LzmaEncProps_Normalize(&lzmaProps); + t1n = lzmaProps.numThreads; + } + + t1 = p->lzmaProps.numThreads; + t2 = p->numBlockThreads_Max; + t3 = p->numTotalThreads; + + if (t2 > MTCODER__THREADS_MAX) + t2 = MTCODER__THREADS_MAX; + + if (t3 <= 0) + { + if (t2 <= 0) + t2 = 1; + t3 = t1n * t2; + } + else if (t2 <= 0) + { + t2 = t3 / t1n; + if (t2 == 0) + { + t1 = 1; + t2 = t3; + } + if (t2 > MTCODER__THREADS_MAX) + t2 = MTCODER__THREADS_MAX; + } + else if (t1 <= 0) + { + t1 = t3 / t2; + if (t1 == 0) + t1 = 1; + } + else + t3 = t1n * t2; + + p->lzmaProps.numThreads = t1; + + t2r = t2; + + fileSize = p->lzmaProps.reduceSize; + + if ( p->blockSize != LZMA2_ENC_PROPS__BLOCK_SIZE__SOLID + && p->blockSize != LZMA2_ENC_PROPS__BLOCK_SIZE__AUTO + && (p->blockSize < fileSize || fileSize == (UInt64)(Int64)-1)) + p->lzmaProps.reduceSize = p->blockSize; + + LzmaEncProps_Normalize(&p->lzmaProps); + + p->lzmaProps.reduceSize = fileSize; + + t1 = p->lzmaProps.numThreads; + + if (p->blockSize == LZMA2_ENC_PROPS__BLOCK_SIZE__SOLID) + { + t2r = t2 = 1; + t3 = t1; + } + else if (p->blockSize == LZMA2_ENC_PROPS__BLOCK_SIZE__AUTO && t2 <= 1) + { + /* if there is no block multi-threading, we use SOLID block */ + p->blockSize = LZMA2_ENC_PROPS__BLOCK_SIZE__SOLID; + } + else + { + if (p->blockSize == LZMA2_ENC_PROPS__BLOCK_SIZE__AUTO) + { + const UInt32 kMinSize = (UInt32)1 << 20; + const UInt32 kMaxSize = (UInt32)1 << 28; + const UInt32 dictSize = p->lzmaProps.dictSize; + UInt64 blockSize = (UInt64)dictSize << 2; + if (blockSize < kMinSize) blockSize = kMinSize; + if (blockSize > kMaxSize) blockSize = kMaxSize; + if (blockSize < dictSize) blockSize = dictSize; + blockSize += (kMinSize - 1); + blockSize &= ~(UInt64)(kMinSize - 1); + p->blockSize = blockSize; + } + + if (t2 > 1 && fileSize != (UInt64)(Int64)-1) + { + UInt64 numBlocks = fileSize / p->blockSize; + if (numBlocks * p->blockSize != fileSize) + numBlocks++; + if (numBlocks < (unsigned)t2) + { + t2r = (unsigned)numBlocks; + if (t2r == 0) + t2r = 1; + t3 = t1 * t2r; + } + } + } + + p->numBlockThreads_Max = t2; + p->numBlockThreads_Reduced = t2r; + p->numTotalThreads = t3; +} + + +static SRes Progress(ICompressProgress *p, UInt64 inSize, UInt64 outSize) +{ + return (p && ICompressProgress_Progress(p, inSize, outSize) != SZ_OK) ? SZ_ERROR_PROGRESS : SZ_OK; +} + + +/* ---------- Lzma2 ---------- */ + +typedef struct +{ + Byte propEncoded; + CLzma2EncProps props; + UInt64 expectedDataSize; + + Byte *tempBufLzma; + + ISzAllocPtr alloc; + ISzAllocPtr allocBig; + + CLzma2EncInt coders[MTCODER__THREADS_MAX]; + + #ifndef _7ZIP_ST + + ISeqOutStream *outStream; + Byte *outBuf; + size_t outBuf_Rem; /* remainder in outBuf */ + + size_t outBufSize; /* size of allocated outBufs[i] */ + size_t outBufsDataSizes[MTCODER__BLOCKS_MAX]; + BoolInt mtCoder_WasConstructed; + CMtCoder mtCoder; + Byte *outBufs[MTCODER__BLOCKS_MAX]; + + #endif + +} CLzma2Enc; + + + +CLzma2EncHandle Lzma2Enc_Create(ISzAllocPtr alloc, ISzAllocPtr allocBig) +{ + CLzma2Enc *p = (CLzma2Enc *)ISzAlloc_Alloc(alloc, sizeof(CLzma2Enc)); + if (!p) + return NULL; + Lzma2EncProps_Init(&p->props); + Lzma2EncProps_Normalize(&p->props); + p->expectedDataSize = (UInt64)(Int64)-1; + p->tempBufLzma = NULL; + p->alloc = alloc; + p->allocBig = allocBig; + { + unsigned i; + for (i = 0; i < MTCODER__THREADS_MAX; i++) + p->coders[i].enc = NULL; + } + + #ifndef _7ZIP_ST + p->mtCoder_WasConstructed = False; + { + unsigned i; + for (i = 0; i < MTCODER__BLOCKS_MAX; i++) + p->outBufs[i] = NULL; + p->outBufSize = 0; + } + #endif + + return p; +} + + +#ifndef _7ZIP_ST + +static void Lzma2Enc_FreeOutBufs(CLzma2Enc *p) +{ + unsigned i; + for (i = 0; i < MTCODER__BLOCKS_MAX; i++) + if (p->outBufs[i]) + { + ISzAlloc_Free(p->alloc, p->outBufs[i]); + p->outBufs[i] = NULL; + } + p->outBufSize = 0; +} + +#endif + + +void Lzma2Enc_Destroy(CLzma2EncHandle pp) +{ + CLzma2Enc *p = (CLzma2Enc *)pp; + unsigned i; + for (i = 0; i < MTCODER__THREADS_MAX; i++) + { + CLzma2EncInt *t = &p->coders[i]; + if (t->enc) + { + LzmaEnc_Destroy(t->enc, p->alloc, p->allocBig); + t->enc = NULL; + } + } + + + #ifndef _7ZIP_ST + if (p->mtCoder_WasConstructed) + { + MtCoder_Destruct(&p->mtCoder); + p->mtCoder_WasConstructed = False; + } + Lzma2Enc_FreeOutBufs(p); + #endif + + ISzAlloc_Free(p->alloc, p->tempBufLzma); + p->tempBufLzma = NULL; + + ISzAlloc_Free(p->alloc, pp); +} + + +SRes Lzma2Enc_SetProps(CLzma2EncHandle pp, const CLzma2EncProps *props) +{ + CLzma2Enc *p = (CLzma2Enc *)pp; + CLzmaEncProps lzmaProps = props->lzmaProps; + LzmaEncProps_Normalize(&lzmaProps); + if (lzmaProps.lc + lzmaProps.lp > LZMA2_LCLP_MAX) + return SZ_ERROR_PARAM; + p->props = *props; + Lzma2EncProps_Normalize(&p->props); + return SZ_OK; +} + + +void Lzma2Enc_SetDataSize(CLzmaEncHandle pp, UInt64 expectedDataSiize) +{ + CLzma2Enc *p = (CLzma2Enc *)pp; + p->expectedDataSize = expectedDataSiize; +} + + +Byte Lzma2Enc_WriteProperties(CLzma2EncHandle pp) +{ + CLzma2Enc *p = (CLzma2Enc *)pp; + unsigned i; + UInt32 dicSize = LzmaEncProps_GetDictSize(&p->props.lzmaProps); + for (i = 0; i < 40; i++) + if (dicSize <= LZMA2_DIC_SIZE_FROM_PROP(i)) + break; + return (Byte)i; +} + + +static SRes Lzma2Enc_EncodeMt1( + CLzma2Enc *me, + CLzma2EncInt *p, + ISeqOutStream *outStream, + Byte *outBuf, size_t *outBufSize, + ISeqInStream *inStream, + const Byte *inData, size_t inDataSize, + int finished, + ICompressProgress *progress) +{ + UInt64 unpackTotal = 0; + UInt64 packTotal = 0; + size_t outLim = 0; + CLimitedSeqInStream limitedInStream; + + if (outBuf) + { + outLim = *outBufSize; + *outBufSize = 0; + } + + if (!p->enc) + { + p->propsAreSet = False; + p->enc = LzmaEnc_Create(me->alloc); + if (!p->enc) + return SZ_ERROR_MEM; + } + + limitedInStream.realStream = inStream; + if (inStream) + { + limitedInStream.vt.Read = LimitedSeqInStream_Read; + } + + if (!outBuf) + { + // outStream version works only in one thread. So we use CLzma2Enc::tempBufLzma + if (!me->tempBufLzma) + { + me->tempBufLzma = (Byte *)ISzAlloc_Alloc(me->alloc, LZMA2_CHUNK_SIZE_COMPRESSED_MAX); + if (!me->tempBufLzma) + return SZ_ERROR_MEM; + } + } + + RINOK(Lzma2EncInt_InitStream(p, &me->props)); + + for (;;) + { + SRes res = SZ_OK; + size_t inSizeCur = 0; + + Lzma2EncInt_InitBlock(p); + + LimitedSeqInStream_Init(&limitedInStream); + limitedInStream.limit = me->props.blockSize; + + if (inStream) + { + UInt64 expected = (UInt64)(Int64)-1; + // inStream version works only in one thread. So we use CLzma2Enc::expectedDataSize + if (me->expectedDataSize != (UInt64)(Int64)-1 + && me->expectedDataSize >= unpackTotal) + expected = me->expectedDataSize - unpackTotal; + if (me->props.blockSize != LZMA2_ENC_PROPS__BLOCK_SIZE__SOLID + && expected > me->props.blockSize) + expected = (size_t)me->props.blockSize; + + LzmaEnc_SetDataSize(p->enc, expected); + + RINOK(LzmaEnc_PrepareForLzma2(p->enc, + &limitedInStream.vt, + LZMA2_KEEP_WINDOW_SIZE, + me->alloc, + me->allocBig)); + } + else + { + inSizeCur = inDataSize - (size_t)unpackTotal; + if (me->props.blockSize != LZMA2_ENC_PROPS__BLOCK_SIZE__SOLID + && inSizeCur > me->props.blockSize) + inSizeCur = (size_t)me->props.blockSize; + + // LzmaEnc_SetDataSize(p->enc, inSizeCur); + + RINOK(LzmaEnc_MemPrepare(p->enc, + inData + (size_t)unpackTotal, inSizeCur, + LZMA2_KEEP_WINDOW_SIZE, + me->alloc, + me->allocBig)); + } + + for (;;) + { + size_t packSize = LZMA2_CHUNK_SIZE_COMPRESSED_MAX; + if (outBuf) + packSize = outLim - (size_t)packTotal; + + res = Lzma2EncInt_EncodeSubblock(p, + outBuf ? outBuf + (size_t)packTotal : me->tempBufLzma, &packSize, + outBuf ? NULL : outStream); + + if (res != SZ_OK) + break; + + packTotal += packSize; + if (outBuf) + *outBufSize = (size_t)packTotal; + + res = Progress(progress, unpackTotal + p->srcPos, packTotal); + if (res != SZ_OK) + break; + + /* + if (LzmaEnc_GetNumAvailableBytes(p->enc) == 0) + break; + */ + + if (packSize == 0) + break; + } + + LzmaEnc_Finish(p->enc); + + unpackTotal += p->srcPos; + + RINOK(res); + + if (p->srcPos != (inStream ? limitedInStream.processed : inSizeCur)) + return SZ_ERROR_FAIL; + + if (inStream ? limitedInStream.finished : (unpackTotal == inDataSize)) + { + if (finished) + { + if (outBuf) + { + size_t destPos = *outBufSize; + if (destPos >= outLim) + return SZ_ERROR_OUTPUT_EOF; + outBuf[destPos] = 0; + *outBufSize = destPos + 1; + } + else + { + Byte b = 0; + if (ISeqOutStream_Write(outStream, &b, 1) != 1) + return SZ_ERROR_WRITE; + } + } + return SZ_OK; + } + } +} + + + +#ifndef _7ZIP_ST + +static SRes Lzma2Enc_MtCallback_Code(void *pp, unsigned coderIndex, unsigned outBufIndex, + const Byte *src, size_t srcSize, int finished) +{ + CLzma2Enc *me = (CLzma2Enc *)pp; + size_t destSize = me->outBufSize; + SRes res; + CMtProgressThunk progressThunk; + + Byte *dest = me->outBufs[outBufIndex]; + + me->outBufsDataSizes[outBufIndex] = 0; + + if (!dest) + { + dest = (Byte *)ISzAlloc_Alloc(me->alloc, me->outBufSize); + if (!dest) + return SZ_ERROR_MEM; + me->outBufs[outBufIndex] = dest; + } + + MtProgressThunk_CreateVTable(&progressThunk); + progressThunk.mtProgress = &me->mtCoder.mtProgress; + progressThunk.inSize = 0; + progressThunk.outSize = 0; + + res = Lzma2Enc_EncodeMt1(me, + &me->coders[coderIndex], + NULL, dest, &destSize, + NULL, src, srcSize, + finished, + &progressThunk.vt); + + me->outBufsDataSizes[outBufIndex] = destSize; + + return res; +} + + +static SRes Lzma2Enc_MtCallback_Write(void *pp, unsigned outBufIndex) +{ + CLzma2Enc *me = (CLzma2Enc *)pp; + size_t size = me->outBufsDataSizes[outBufIndex]; + const Byte *data = me->outBufs[outBufIndex]; + + if (me->outStream) + return ISeqOutStream_Write(me->outStream, data, size) == size ? SZ_OK : SZ_ERROR_WRITE; + + if (size > me->outBuf_Rem) + return SZ_ERROR_OUTPUT_EOF; + memcpy(me->outBuf, data, size); + me->outBuf_Rem -= size; + me->outBuf += size; + return SZ_OK; +} + +#endif + + + +SRes Lzma2Enc_Encode2(CLzma2EncHandle pp, + ISeqOutStream *outStream, + Byte *outBuf, size_t *outBufSize, + ISeqInStream *inStream, + const Byte *inData, size_t inDataSize, + ICompressProgress *progress) +{ + CLzma2Enc *p = (CLzma2Enc *)pp; + + if (inStream && inData) + return SZ_ERROR_PARAM; + + if (outStream && outBuf) + return SZ_ERROR_PARAM; + + { + unsigned i; + for (i = 0; i < MTCODER__THREADS_MAX; i++) + p->coders[i].propsAreSet = False; + } + + #ifndef _7ZIP_ST + + if (p->props.numBlockThreads_Reduced > 1) + { + IMtCoderCallback2 vt; + + if (!p->mtCoder_WasConstructed) + { + p->mtCoder_WasConstructed = True; + MtCoder_Construct(&p->mtCoder); + } + + vt.Code = Lzma2Enc_MtCallback_Code; + vt.Write = Lzma2Enc_MtCallback_Write; + + p->outStream = outStream; + p->outBuf = NULL; + p->outBuf_Rem = 0; + if (!outStream) + { + p->outBuf = outBuf; + p->outBuf_Rem = *outBufSize; + *outBufSize = 0; + } + + p->mtCoder.allocBig = p->allocBig; + p->mtCoder.progress = progress; + p->mtCoder.inStream = inStream; + p->mtCoder.inData = inData; + p->mtCoder.inDataSize = inDataSize; + p->mtCoder.mtCallback = &vt; + p->mtCoder.mtCallbackObject = p; + + p->mtCoder.blockSize = (size_t)p->props.blockSize; + if (p->mtCoder.blockSize != p->props.blockSize) + return SZ_ERROR_PARAM; /* SZ_ERROR_MEM */ + + { + size_t destBlockSize = p->mtCoder.blockSize + (p->mtCoder.blockSize >> 10) + 16; + if (destBlockSize < p->mtCoder.blockSize) + return SZ_ERROR_PARAM; + if (p->outBufSize != destBlockSize) + Lzma2Enc_FreeOutBufs(p); + p->outBufSize = destBlockSize; + } + + p->mtCoder.numThreadsMax = p->props.numBlockThreads_Max; + p->mtCoder.expectedDataSize = p->expectedDataSize; + + { + SRes res = MtCoder_Code(&p->mtCoder); + if (!outStream) + *outBufSize = p->outBuf - outBuf; + return res; + } + } + + #endif + + + return Lzma2Enc_EncodeMt1(p, + &p->coders[0], + outStream, outBuf, outBufSize, + inStream, inData, inDataSize, + True, /* finished */ + progress); +} diff --git a/contrib/libs/lzmasdk/Lzma2Enc.h b/contrib/libs/lzmasdk/Lzma2Enc.h index 65f2dd145d..b67986ce51 100644 --- a/contrib/libs/lzmasdk/Lzma2Enc.h +++ b/contrib/libs/lzmasdk/Lzma2Enc.h @@ -1,55 +1,55 @@ -/* Lzma2Enc.h -- LZMA2 Encoder
-2017-07-27 : Igor Pavlov : Public domain */
-
-#ifndef __LZMA2_ENC_H
-#define __LZMA2_ENC_H
-
-#include "LzmaEnc.h"
-
-EXTERN_C_BEGIN
-
-#define LZMA2_ENC_PROPS__BLOCK_SIZE__AUTO 0
-#define LZMA2_ENC_PROPS__BLOCK_SIZE__SOLID ((UInt64)(Int64)-1)
-
-typedef struct
-{
- CLzmaEncProps lzmaProps;
- UInt64 blockSize;
- int numBlockThreads_Reduced;
- int numBlockThreads_Max;
- int numTotalThreads;
-} CLzma2EncProps;
-
-void Lzma2EncProps_Init(CLzma2EncProps *p);
-void Lzma2EncProps_Normalize(CLzma2EncProps *p);
-
-/* ---------- CLzmaEnc2Handle Interface ---------- */
-
-/* Lzma2Enc_* functions can return the following exit codes:
-SRes:
- SZ_OK - OK
- SZ_ERROR_MEM - Memory allocation error
- SZ_ERROR_PARAM - Incorrect paramater in props
- SZ_ERROR_WRITE - ISeqOutStream write callback error
- SZ_ERROR_OUTPUT_EOF - output buffer overflow - version with (Byte *) output
- SZ_ERROR_PROGRESS - some break from progress callback
- SZ_ERROR_THREAD - error in multithreading functions (only for Mt version)
-*/
-
-typedef void * CLzma2EncHandle;
-
-CLzma2EncHandle Lzma2Enc_Create(ISzAllocPtr alloc, ISzAllocPtr allocBig);
-void Lzma2Enc_Destroy(CLzma2EncHandle p);
-SRes Lzma2Enc_SetProps(CLzma2EncHandle p, const CLzma2EncProps *props);
-void Lzma2Enc_SetDataSize(CLzma2EncHandle p, UInt64 expectedDataSiize);
-Byte Lzma2Enc_WriteProperties(CLzma2EncHandle p);
-SRes Lzma2Enc_Encode2(CLzma2EncHandle p,
- ISeqOutStream *outStream,
- Byte *outBuf, size_t *outBufSize,
- ISeqInStream *inStream,
- const Byte *inData, size_t inDataSize,
- ICompressProgress *progress);
-
-EXTERN_C_END
-
-#endif
+/* Lzma2Enc.h -- LZMA2 Encoder +2017-07-27 : Igor Pavlov : Public domain */ + +#ifndef __LZMA2_ENC_H +#define __LZMA2_ENC_H + +#include "LzmaEnc.h" + +EXTERN_C_BEGIN + +#define LZMA2_ENC_PROPS__BLOCK_SIZE__AUTO 0 +#define LZMA2_ENC_PROPS__BLOCK_SIZE__SOLID ((UInt64)(Int64)-1) + +typedef struct +{ + CLzmaEncProps lzmaProps; + UInt64 blockSize; + int numBlockThreads_Reduced; + int numBlockThreads_Max; + int numTotalThreads; +} CLzma2EncProps; + +void Lzma2EncProps_Init(CLzma2EncProps *p); +void Lzma2EncProps_Normalize(CLzma2EncProps *p); + +/* ---------- CLzmaEnc2Handle Interface ---------- */ + +/* Lzma2Enc_* functions can return the following exit codes: +SRes: + SZ_OK - OK + SZ_ERROR_MEM - Memory allocation error + SZ_ERROR_PARAM - Incorrect paramater in props + SZ_ERROR_WRITE - ISeqOutStream write callback error + SZ_ERROR_OUTPUT_EOF - output buffer overflow - version with (Byte *) output + SZ_ERROR_PROGRESS - some break from progress callback + SZ_ERROR_THREAD - error in multithreading functions (only for Mt version) +*/ + +typedef void * CLzma2EncHandle; + +CLzma2EncHandle Lzma2Enc_Create(ISzAllocPtr alloc, ISzAllocPtr allocBig); +void Lzma2Enc_Destroy(CLzma2EncHandle p); +SRes Lzma2Enc_SetProps(CLzma2EncHandle p, const CLzma2EncProps *props); +void Lzma2Enc_SetDataSize(CLzma2EncHandle p, UInt64 expectedDataSiize); +Byte Lzma2Enc_WriteProperties(CLzma2EncHandle p); +SRes Lzma2Enc_Encode2(CLzma2EncHandle p, + ISeqOutStream *outStream, + Byte *outBuf, size_t *outBufSize, + ISeqInStream *inStream, + const Byte *inData, size_t inDataSize, + ICompressProgress *progress); + +EXTERN_C_END + +#endif diff --git a/contrib/libs/lzmasdk/LzmaEnc.c b/contrib/libs/lzmasdk/LzmaEnc.c index a5a087f556..3f2ce1a8bc 100644 --- a/contrib/libs/lzmasdk/LzmaEnc.c +++ b/contrib/libs/lzmasdk/LzmaEnc.c @@ -1,2984 +1,2984 @@ -/* LzmaEnc.c -- LZMA Encoder
-2019-01-10: Igor Pavlov : Public domain */
-
-#include "Precomp.h"
-
-#include <string.h>
-
-/* #define SHOW_STAT */
-/* #define SHOW_STAT2 */
-
-#if defined(SHOW_STAT) || defined(SHOW_STAT2)
-#include <stdio.h>
-#endif
-
-#include "LzmaEnc.h"
-
-#include "LzFind.h"
-#ifndef _7ZIP_ST
-#include "LzFindMt.h"
-#endif
-
-#ifdef SHOW_STAT
-static unsigned g_STAT_OFFSET = 0;
-#endif
-
-#define kLzmaMaxHistorySize ((UInt32)3 << 29)
-/* #define kLzmaMaxHistorySize ((UInt32)7 << 29) */
-
-#define kNumTopBits 24
-#define kTopValue ((UInt32)1 << kNumTopBits)
-
-#define kNumBitModelTotalBits 11
-#define kBitModelTotal (1 << kNumBitModelTotalBits)
-#define kNumMoveBits 5
-#define kProbInitValue (kBitModelTotal >> 1)
-
-#define kNumMoveReducingBits 4
-#define kNumBitPriceShiftBits 4
-#define kBitPrice (1 << kNumBitPriceShiftBits)
-
-#define REP_LEN_COUNT 64
-
-void LzmaEncProps_Init(CLzmaEncProps *p)
-{
- p->level = 5;
- p->dictSize = p->mc = 0;
- p->reduceSize = (UInt64)(Int64)-1;
- p->lc = p->lp = p->pb = p->algo = p->fb = p->btMode = p->numHashBytes = p->numThreads = -1;
- p->writeEndMark = 0;
-}
-
-void LzmaEncProps_Normalize(CLzmaEncProps *p)
-{
- int level = p->level;
- if (level < 0) level = 5;
- p->level = level;
-
- if (p->dictSize == 0) p->dictSize = (level <= 5 ? (1 << (level * 2 + 14)) : (level <= 7 ? (1 << 25) : (1 << 26)));
- if (p->dictSize > p->reduceSize)
- {
- unsigned i;
- UInt32 reduceSize = (UInt32)p->reduceSize;
- for (i = 11; i <= 30; i++)
- {
- if (reduceSize <= ((UInt32)2 << i)) { p->dictSize = ((UInt32)2 << i); break; }
- if (reduceSize <= ((UInt32)3 << i)) { p->dictSize = ((UInt32)3 << i); break; }
- }
- }
-
- if (p->lc < 0) p->lc = 3;
- if (p->lp < 0) p->lp = 0;
- if (p->pb < 0) p->pb = 2;
-
- if (p->algo < 0) p->algo = (level < 5 ? 0 : 1);
- if (p->fb < 0) p->fb = (level < 7 ? 32 : 64);
- if (p->btMode < 0) p->btMode = (p->algo == 0 ? 0 : 1);
- if (p->numHashBytes < 0) p->numHashBytes = 4;
- if (p->mc == 0) p->mc = (16 + (p->fb >> 1)) >> (p->btMode ? 0 : 1);
-
- if (p->numThreads < 0)
- p->numThreads =
- #ifndef _7ZIP_ST
- ((p->btMode && p->algo) ? 2 : 1);
- #else
- 1;
- #endif
-}
-
-UInt32 LzmaEncProps_GetDictSize(const CLzmaEncProps *props2)
-{
- CLzmaEncProps props = *props2;
- LzmaEncProps_Normalize(&props);
- return props.dictSize;
-}
-
-#if (_MSC_VER >= 1400)
-/* BSR code is fast for some new CPUs */
-/* #define LZMA_LOG_BSR */
-#endif
-
-#ifdef LZMA_LOG_BSR
-
-#define kDicLogSizeMaxCompress 32
-
-#define BSR2_RET(pos, res) { unsigned long zz; _BitScanReverse(&zz, (pos)); res = (zz + zz) + ((pos >> (zz - 1)) & 1); }
-
-static unsigned GetPosSlot1(UInt32 pos)
-{
- unsigned res;
- BSR2_RET(pos, res);
- return res;
-}
-#define GetPosSlot2(pos, res) { BSR2_RET(pos, res); }
-#define GetPosSlot(pos, res) { if (pos < 2) res = pos; else BSR2_RET(pos, res); }
-
-#else
-
-#define kNumLogBits (9 + sizeof(size_t) / 2)
-/* #define kNumLogBits (11 + sizeof(size_t) / 8 * 3) */
-
-#define kDicLogSizeMaxCompress ((kNumLogBits - 1) * 2 + 7)
-
-static void LzmaEnc_FastPosInit(Byte *g_FastPos)
-{
- unsigned slot;
- g_FastPos[0] = 0;
- g_FastPos[1] = 1;
- g_FastPos += 2;
-
- for (slot = 2; slot < kNumLogBits * 2; slot++)
- {
- size_t k = ((size_t)1 << ((slot >> 1) - 1));
- size_t j;
- for (j = 0; j < k; j++)
- g_FastPos[j] = (Byte)slot;
- g_FastPos += k;
- }
-}
-
-/* we can use ((limit - pos) >> 31) only if (pos < ((UInt32)1 << 31)) */
-/*
-#define BSR2_RET(pos, res) { unsigned zz = 6 + ((kNumLogBits - 1) & \
- (0 - (((((UInt32)1 << (kNumLogBits + 6)) - 1) - pos) >> 31))); \
- res = p->g_FastPos[pos >> zz] + (zz * 2); }
-*/
-
-/*
-#define BSR2_RET(pos, res) { unsigned zz = 6 + ((kNumLogBits - 1) & \
- (0 - (((((UInt32)1 << (kNumLogBits)) - 1) - (pos >> 6)) >> 31))); \
- res = p->g_FastPos[pos >> zz] + (zz * 2); }
-*/
-
-#define BSR2_RET(pos, res) { unsigned zz = (pos < (1 << (kNumLogBits + 6))) ? 6 : 6 + kNumLogBits - 1; \
- res = p->g_FastPos[pos >> zz] + (zz * 2); }
-
-/*
-#define BSR2_RET(pos, res) { res = (pos < (1 << (kNumLogBits + 6))) ? \
- p->g_FastPos[pos >> 6] + 12 : \
- p->g_FastPos[pos >> (6 + kNumLogBits - 1)] + (6 + (kNumLogBits - 1)) * 2; }
-*/
-
-#define GetPosSlot1(pos) p->g_FastPos[pos]
-#define GetPosSlot2(pos, res) { BSR2_RET(pos, res); }
-#define GetPosSlot(pos, res) { if (pos < kNumFullDistances) res = p->g_FastPos[pos & (kNumFullDistances - 1)]; else BSR2_RET(pos, res); }
-
-#endif
-
-
-#define LZMA_NUM_REPS 4
-
-typedef UInt16 CState;
-typedef UInt16 CExtra;
-
-typedef struct
-{
- UInt32 price;
- CState state;
- CExtra extra;
- // 0 : normal
- // 1 : LIT : MATCH
- // > 1 : MATCH (extra-1) : LIT : REP0 (len)
- UInt32 len;
- UInt32 dist;
- UInt32 reps[LZMA_NUM_REPS];
-} COptimal;
-
-
-// 18.06
-#define kNumOpts (1 << 11)
-#define kPackReserve (kNumOpts * 8)
-// #define kNumOpts (1 << 12)
-// #define kPackReserve (1 + kNumOpts * 2)
-
-#define kNumLenToPosStates 4
-#define kNumPosSlotBits 6
-#define kDicLogSizeMin 0
-#define kDicLogSizeMax 32
-#define kDistTableSizeMax (kDicLogSizeMax * 2)
-
-#define kNumAlignBits 4
-#define kAlignTableSize (1 << kNumAlignBits)
-#define kAlignMask (kAlignTableSize - 1)
-
-#define kStartPosModelIndex 4
-#define kEndPosModelIndex 14
-#define kNumFullDistances (1 << (kEndPosModelIndex >> 1))
-
-typedef
-#ifdef _LZMA_PROB32
- UInt32
-#else
- UInt16
-#endif
- CLzmaProb;
-
-#define LZMA_PB_MAX 4
-#define LZMA_LC_MAX 8
-#define LZMA_LP_MAX 4
-
-#define LZMA_NUM_PB_STATES_MAX (1 << LZMA_PB_MAX)
-
-#define kLenNumLowBits 3
-#define kLenNumLowSymbols (1 << kLenNumLowBits)
-#define kLenNumHighBits 8
-#define kLenNumHighSymbols (1 << kLenNumHighBits)
-#define kLenNumSymbolsTotal (kLenNumLowSymbols * 2 + kLenNumHighSymbols)
-
-#define LZMA_MATCH_LEN_MIN 2
-#define LZMA_MATCH_LEN_MAX (LZMA_MATCH_LEN_MIN + kLenNumSymbolsTotal - 1)
-
-#define kNumStates 12
-
-
-typedef struct
-{
- CLzmaProb low[LZMA_NUM_PB_STATES_MAX << (kLenNumLowBits + 1)];
- CLzmaProb high[kLenNumHighSymbols];
-} CLenEnc;
-
-
-typedef struct
-{
- unsigned tableSize;
- UInt32 prices[LZMA_NUM_PB_STATES_MAX][kLenNumSymbolsTotal];
- // UInt32 prices1[LZMA_NUM_PB_STATES_MAX][kLenNumLowSymbols * 2];
- // UInt32 prices2[kLenNumSymbolsTotal];
-} CLenPriceEnc;
-
-#define GET_PRICE_LEN(p, posState, len) \
- ((p)->prices[posState][(size_t)(len) - LZMA_MATCH_LEN_MIN])
-
-/*
-#define GET_PRICE_LEN(p, posState, len) \
- ((p)->prices2[(size_t)(len) - 2] + ((p)->prices1[posState][((len) - 2) & (kLenNumLowSymbols * 2 - 1)] & (((len) - 2 - kLenNumLowSymbols * 2) >> 9)))
-*/
-
-typedef struct
-{
- UInt32 range;
- unsigned cache;
- UInt64 low;
- UInt64 cacheSize;
- Byte *buf;
- Byte *bufLim;
- Byte *bufBase;
- ISeqOutStream *outStream;
- UInt64 processed;
- SRes res;
-} CRangeEnc;
-
-
-typedef struct
-{
- CLzmaProb *litProbs;
-
- unsigned state;
- UInt32 reps[LZMA_NUM_REPS];
-
- CLzmaProb posAlignEncoder[1 << kNumAlignBits];
- CLzmaProb isRep[kNumStates];
- CLzmaProb isRepG0[kNumStates];
- CLzmaProb isRepG1[kNumStates];
- CLzmaProb isRepG2[kNumStates];
- CLzmaProb isMatch[kNumStates][LZMA_NUM_PB_STATES_MAX];
- CLzmaProb isRep0Long[kNumStates][LZMA_NUM_PB_STATES_MAX];
-
- CLzmaProb posSlotEncoder[kNumLenToPosStates][1 << kNumPosSlotBits];
- CLzmaProb posEncoders[kNumFullDistances];
-
- CLenEnc lenProbs;
- CLenEnc repLenProbs;
-
-} CSaveState;
-
-
-typedef UInt32 CProbPrice;
-
-
-typedef struct
-{
- void *matchFinderObj;
- IMatchFinder matchFinder;
-
- unsigned optCur;
- unsigned optEnd;
-
- unsigned longestMatchLen;
- unsigned numPairs;
- UInt32 numAvail;
-
- unsigned state;
- unsigned numFastBytes;
- unsigned additionalOffset;
- UInt32 reps[LZMA_NUM_REPS];
- unsigned lpMask, pbMask;
- CLzmaProb *litProbs;
- CRangeEnc rc;
-
- UInt32 backRes;
-
- unsigned lc, lp, pb;
- unsigned lclp;
-
- BoolInt fastMode;
- BoolInt writeEndMark;
- BoolInt finished;
- BoolInt multiThread;
- BoolInt needInit;
- // BoolInt _maxMode;
-
- UInt64 nowPos64;
-
- unsigned matchPriceCount;
- // unsigned alignPriceCount;
- int repLenEncCounter;
-
- unsigned distTableSize;
-
- UInt32 dictSize;
- SRes result;
-
- #ifndef _7ZIP_ST
- BoolInt mtMode;
- // begin of CMatchFinderMt is used in LZ thread
- CMatchFinderMt matchFinderMt;
- // end of CMatchFinderMt is used in BT and HASH threads
- #endif
-
- CMatchFinder matchFinderBase;
-
- #ifndef _7ZIP_ST
- Byte pad[128];
- #endif
-
- // LZ thread
- CProbPrice ProbPrices[kBitModelTotal >> kNumMoveReducingBits];
-
- UInt32 matches[LZMA_MATCH_LEN_MAX * 2 + 2 + 1];
-
- UInt32 alignPrices[kAlignTableSize];
- UInt32 posSlotPrices[kNumLenToPosStates][kDistTableSizeMax];
- UInt32 distancesPrices[kNumLenToPosStates][kNumFullDistances];
-
- CLzmaProb posAlignEncoder[1 << kNumAlignBits];
- CLzmaProb isRep[kNumStates];
- CLzmaProb isRepG0[kNumStates];
- CLzmaProb isRepG1[kNumStates];
- CLzmaProb isRepG2[kNumStates];
- CLzmaProb isMatch[kNumStates][LZMA_NUM_PB_STATES_MAX];
- CLzmaProb isRep0Long[kNumStates][LZMA_NUM_PB_STATES_MAX];
- CLzmaProb posSlotEncoder[kNumLenToPosStates][1 << kNumPosSlotBits];
- CLzmaProb posEncoders[kNumFullDistances];
-
- CLenEnc lenProbs;
- CLenEnc repLenProbs;
-
- #ifndef LZMA_LOG_BSR
- Byte g_FastPos[1 << kNumLogBits];
- #endif
-
- CLenPriceEnc lenEnc;
- CLenPriceEnc repLenEnc;
-
- COptimal opt[kNumOpts];
-
- CSaveState saveState;
-
- #ifndef _7ZIP_ST
- Byte pad2[128];
- #endif
-} CLzmaEnc;
-
-
-
-#define COPY_ARR(dest, src, arr) memcpy(dest->arr, src->arr, sizeof(src->arr));
-
-void LzmaEnc_SaveState(CLzmaEncHandle pp)
-{
- CLzmaEnc *p = (CLzmaEnc *)pp;
- CSaveState *dest = &p->saveState;
-
- dest->state = p->state;
-
- dest->lenProbs = p->lenProbs;
- dest->repLenProbs = p->repLenProbs;
-
- COPY_ARR(dest, p, reps);
-
- COPY_ARR(dest, p, posAlignEncoder);
- COPY_ARR(dest, p, isRep);
- COPY_ARR(dest, p, isRepG0);
- COPY_ARR(dest, p, isRepG1);
- COPY_ARR(dest, p, isRepG2);
- COPY_ARR(dest, p, isMatch);
- COPY_ARR(dest, p, isRep0Long);
- COPY_ARR(dest, p, posSlotEncoder);
- COPY_ARR(dest, p, posEncoders);
-
- memcpy(dest->litProbs, p->litProbs, ((UInt32)0x300 << p->lclp) * sizeof(CLzmaProb));
-}
-
-
-void LzmaEnc_RestoreState(CLzmaEncHandle pp)
-{
- CLzmaEnc *dest = (CLzmaEnc *)pp;
- const CSaveState *p = &dest->saveState;
-
- dest->state = p->state;
-
- dest->lenProbs = p->lenProbs;
- dest->repLenProbs = p->repLenProbs;
-
- COPY_ARR(dest, p, reps);
-
- COPY_ARR(dest, p, posAlignEncoder);
- COPY_ARR(dest, p, isRep);
- COPY_ARR(dest, p, isRepG0);
- COPY_ARR(dest, p, isRepG1);
- COPY_ARR(dest, p, isRepG2);
- COPY_ARR(dest, p, isMatch);
- COPY_ARR(dest, p, isRep0Long);
- COPY_ARR(dest, p, posSlotEncoder);
- COPY_ARR(dest, p, posEncoders);
-
- memcpy(dest->litProbs, p->litProbs, ((UInt32)0x300 << dest->lclp) * sizeof(CLzmaProb));
-}
-
-
-
-SRes LzmaEnc_SetProps(CLzmaEncHandle pp, const CLzmaEncProps *props2)
-{
- CLzmaEnc *p = (CLzmaEnc *)pp;
- CLzmaEncProps props = *props2;
- LzmaEncProps_Normalize(&props);
-
- if (props.lc > LZMA_LC_MAX
- || props.lp > LZMA_LP_MAX
- || props.pb > LZMA_PB_MAX
- || props.dictSize > ((UInt64)1 << kDicLogSizeMaxCompress)
- || props.dictSize > kLzmaMaxHistorySize)
- return SZ_ERROR_PARAM;
-
- p->dictSize = props.dictSize;
- {
- unsigned fb = props.fb;
- if (fb < 5)
- fb = 5;
- if (fb > LZMA_MATCH_LEN_MAX)
- fb = LZMA_MATCH_LEN_MAX;
- p->numFastBytes = fb;
- }
- p->lc = props.lc;
- p->lp = props.lp;
- p->pb = props.pb;
- p->fastMode = (props.algo == 0);
- // p->_maxMode = True;
- p->matchFinderBase.btMode = (Byte)(props.btMode ? 1 : 0);
- {
- unsigned numHashBytes = 4;
- if (props.btMode)
- {
- if (props.numHashBytes < 2)
- numHashBytes = 2;
- else if (props.numHashBytes < 4)
- numHashBytes = props.numHashBytes;
- }
- p->matchFinderBase.numHashBytes = numHashBytes;
- }
-
- p->matchFinderBase.cutValue = props.mc;
-
- p->writeEndMark = props.writeEndMark;
-
- #ifndef _7ZIP_ST
- /*
- if (newMultiThread != _multiThread)
- {
- ReleaseMatchFinder();
- _multiThread = newMultiThread;
- }
- */
- p->multiThread = (props.numThreads > 1);
- #endif
-
- return SZ_OK;
-}
-
-
-void LzmaEnc_SetDataSize(CLzmaEncHandle pp, UInt64 expectedDataSiize)
-{
- CLzmaEnc *p = (CLzmaEnc *)pp;
- p->matchFinderBase.expectedDataSize = expectedDataSiize;
-}
-
-
-#define kState_Start 0
-#define kState_LitAfterMatch 4
-#define kState_LitAfterRep 5
-#define kState_MatchAfterLit 7
-#define kState_RepAfterLit 8
-
-static const Byte kLiteralNextStates[kNumStates] = {0, 0, 0, 0, 1, 2, 3, 4, 5, 6, 4, 5};
-static const Byte kMatchNextStates[kNumStates] = {7, 7, 7, 7, 7, 7, 7, 10, 10, 10, 10, 10};
-static const Byte kRepNextStates[kNumStates] = {8, 8, 8, 8, 8, 8, 8, 11, 11, 11, 11, 11};
-static const Byte kShortRepNextStates[kNumStates]= {9, 9, 9, 9, 9, 9, 9, 11, 11, 11, 11, 11};
-
-#define IsLitState(s) ((s) < 7)
-#define GetLenToPosState2(len) (((len) < kNumLenToPosStates - 1) ? (len) : kNumLenToPosStates - 1)
-#define GetLenToPosState(len) (((len) < kNumLenToPosStates + 1) ? (len) - 2 : kNumLenToPosStates - 1)
-
-#define kInfinityPrice (1 << 30)
-
-static void RangeEnc_Construct(CRangeEnc *p)
-{
- p->outStream = NULL;
- p->bufBase = NULL;
-}
-
-#define RangeEnc_GetProcessed(p) ((p)->processed + ((p)->buf - (p)->bufBase) + (p)->cacheSize)
-#define RangeEnc_GetProcessed_sizet(p) ((size_t)(p)->processed + ((p)->buf - (p)->bufBase) + (size_t)(p)->cacheSize)
-
-#define RC_BUF_SIZE (1 << 16)
-
-static int RangeEnc_Alloc(CRangeEnc *p, ISzAllocPtr alloc)
-{
- if (!p->bufBase)
- {
- p->bufBase = (Byte *)ISzAlloc_Alloc(alloc, RC_BUF_SIZE);
- if (!p->bufBase)
- return 0;
- p->bufLim = p->bufBase + RC_BUF_SIZE;
- }
- return 1;
-}
-
-static void RangeEnc_Free(CRangeEnc *p, ISzAllocPtr alloc)
-{
- ISzAlloc_Free(alloc, p->bufBase);
- p->bufBase = 0;
-}
-
-static void RangeEnc_Init(CRangeEnc *p)
-{
- /* Stream.Init(); */
- p->range = 0xFFFFFFFF;
- p->cache = 0;
- p->low = 0;
- p->cacheSize = 0;
-
- p->buf = p->bufBase;
-
- p->processed = 0;
- p->res = SZ_OK;
-}
-
-MY_NO_INLINE static void RangeEnc_FlushStream(CRangeEnc *p)
-{
- size_t num;
- if (p->res != SZ_OK)
- return;
- num = p->buf - p->bufBase;
- if (num != ISeqOutStream_Write(p->outStream, p->bufBase, num))
- p->res = SZ_ERROR_WRITE;
- p->processed += num;
- p->buf = p->bufBase;
-}
-
-MY_NO_INLINE static void MY_FAST_CALL RangeEnc_ShiftLow(CRangeEnc *p)
-{
- UInt32 low = (UInt32)p->low;
- unsigned high = (unsigned)(p->low >> 32);
- p->low = (UInt32)(low << 8);
- if (low < (UInt32)0xFF000000 || high != 0)
- {
- {
- Byte *buf = p->buf;
- *buf++ = (Byte)(p->cache + high);
- p->cache = (unsigned)(low >> 24);
- p->buf = buf;
- if (buf == p->bufLim)
- RangeEnc_FlushStream(p);
- if (p->cacheSize == 0)
- return;
- }
- high += 0xFF;
- for (;;)
- {
- Byte *buf = p->buf;
- *buf++ = (Byte)(high);
- p->buf = buf;
- if (buf == p->bufLim)
- RangeEnc_FlushStream(p);
- if (--p->cacheSize == 0)
- return;
- }
- }
- p->cacheSize++;
-}
-
-static void RangeEnc_FlushData(CRangeEnc *p)
-{
- int i;
- for (i = 0; i < 5; i++)
- RangeEnc_ShiftLow(p);
-}
-
-#define RC_NORM(p) if (range < kTopValue) { range <<= 8; RangeEnc_ShiftLow(p); }
-
-#define RC_BIT_PRE(p, prob) \
- ttt = *(prob); \
- newBound = (range >> kNumBitModelTotalBits) * ttt;
-
-// #define _LZMA_ENC_USE_BRANCH
-
-#ifdef _LZMA_ENC_USE_BRANCH
-
-#define RC_BIT(p, prob, bit) { \
- RC_BIT_PRE(p, prob) \
- if (bit == 0) { range = newBound; ttt += (kBitModelTotal - ttt) >> kNumMoveBits; } \
- else { (p)->low += newBound; range -= newBound; ttt -= ttt >> kNumMoveBits; } \
- *(prob) = (CLzmaProb)ttt; \
- RC_NORM(p) \
- }
-
-#else
-
-#define RC_BIT(p, prob, bit) { \
- UInt32 mask; \
- RC_BIT_PRE(p, prob) \
- mask = 0 - (UInt32)bit; \
- range &= mask; \
- mask &= newBound; \
- range -= mask; \
- (p)->low += mask; \
- mask = (UInt32)bit - 1; \
- range += newBound & mask; \
- mask &= (kBitModelTotal - ((1 << kNumMoveBits) - 1)); \
- mask += ((1 << kNumMoveBits) - 1); \
- ttt += (Int32)(mask - ttt) >> kNumMoveBits; \
- *(prob) = (CLzmaProb)ttt; \
- RC_NORM(p) \
- }
-
-#endif
-
-
-
-
-#define RC_BIT_0_BASE(p, prob) \
- range = newBound; *(prob) = (CLzmaProb)(ttt + ((kBitModelTotal - ttt) >> kNumMoveBits));
-
-#define RC_BIT_1_BASE(p, prob) \
- range -= newBound; (p)->low += newBound; *(prob) = (CLzmaProb)(ttt - (ttt >> kNumMoveBits)); \
-
-#define RC_BIT_0(p, prob) \
- RC_BIT_0_BASE(p, prob) \
- RC_NORM(p)
-
-#define RC_BIT_1(p, prob) \
- RC_BIT_1_BASE(p, prob) \
- RC_NORM(p)
-
-static void RangeEnc_EncodeBit_0(CRangeEnc *p, CLzmaProb *prob)
-{
- UInt32 range, ttt, newBound;
- range = p->range;
- RC_BIT_PRE(p, prob)
- RC_BIT_0(p, prob)
- p->range = range;
-}
-
-static void LitEnc_Encode(CRangeEnc *p, CLzmaProb *probs, UInt32 sym)
-{
- UInt32 range = p->range;
- sym |= 0x100;
- do
- {
- UInt32 ttt, newBound;
- // RangeEnc_EncodeBit(p, probs + (sym >> 8), (sym >> 7) & 1);
- CLzmaProb *prob = probs + (sym >> 8);
- UInt32 bit = (sym >> 7) & 1;
- sym <<= 1;
- RC_BIT(p, prob, bit);
- }
- while (sym < 0x10000);
- p->range = range;
-}
-
-static void LitEnc_EncodeMatched(CRangeEnc *p, CLzmaProb *probs, UInt32 sym, UInt32 matchByte)
-{
- UInt32 range = p->range;
- UInt32 offs = 0x100;
- sym |= 0x100;
- do
- {
- UInt32 ttt, newBound;
- CLzmaProb *prob;
- UInt32 bit;
- matchByte <<= 1;
- // RangeEnc_EncodeBit(p, probs + (offs + (matchByte & offs) + (sym >> 8)), (sym >> 7) & 1);
- prob = probs + (offs + (matchByte & offs) + (sym >> 8));
- bit = (sym >> 7) & 1;
- sym <<= 1;
- offs &= ~(matchByte ^ sym);
- RC_BIT(p, prob, bit);
- }
- while (sym < 0x10000);
- p->range = range;
-}
-
-
-
-static void LzmaEnc_InitPriceTables(CProbPrice *ProbPrices)
-{
- UInt32 i;
- for (i = 0; i < (kBitModelTotal >> kNumMoveReducingBits); i++)
- {
- const unsigned kCyclesBits = kNumBitPriceShiftBits;
- UInt32 w = (i << kNumMoveReducingBits) + (1 << (kNumMoveReducingBits - 1));
- unsigned bitCount = 0;
- unsigned j;
- for (j = 0; j < kCyclesBits; j++)
- {
- w = w * w;
- bitCount <<= 1;
- while (w >= ((UInt32)1 << 16))
- {
- w >>= 1;
- bitCount++;
- }
- }
- ProbPrices[i] = (CProbPrice)((kNumBitModelTotalBits << kCyclesBits) - 15 - bitCount);
- // printf("\n%3d: %5d", i, ProbPrices[i]);
- }
-}
-
-
-#define GET_PRICE(prob, bit) \
- p->ProbPrices[((prob) ^ (unsigned)(((-(int)(bit))) & (kBitModelTotal - 1))) >> kNumMoveReducingBits];
-
-#define GET_PRICEa(prob, bit) \
- ProbPrices[((prob) ^ (unsigned)((-((int)(bit))) & (kBitModelTotal - 1))) >> kNumMoveReducingBits];
-
-#define GET_PRICE_0(prob) p->ProbPrices[(prob) >> kNumMoveReducingBits]
-#define GET_PRICE_1(prob) p->ProbPrices[((prob) ^ (kBitModelTotal - 1)) >> kNumMoveReducingBits]
-
-#define GET_PRICEa_0(prob) ProbPrices[(prob) >> kNumMoveReducingBits]
-#define GET_PRICEa_1(prob) ProbPrices[((prob) ^ (kBitModelTotal - 1)) >> kNumMoveReducingBits]
-
-
-static UInt32 LitEnc_GetPrice(const CLzmaProb *probs, UInt32 sym, const CProbPrice *ProbPrices)
-{
- UInt32 price = 0;
- sym |= 0x100;
- do
- {
- unsigned bit = sym & 1;
- sym >>= 1;
- price += GET_PRICEa(probs[sym], bit);
- }
- while (sym >= 2);
- return price;
-}
-
-
-static UInt32 LitEnc_Matched_GetPrice(const CLzmaProb *probs, UInt32 sym, UInt32 matchByte, const CProbPrice *ProbPrices)
-{
- UInt32 price = 0;
- UInt32 offs = 0x100;
- sym |= 0x100;
- do
- {
- matchByte <<= 1;
- price += GET_PRICEa(probs[offs + (matchByte & offs) + (sym >> 8)], (sym >> 7) & 1);
- sym <<= 1;
- offs &= ~(matchByte ^ sym);
- }
- while (sym < 0x10000);
- return price;
-}
-
-
-static void RcTree_ReverseEncode(CRangeEnc *rc, CLzmaProb *probs, unsigned numBits, unsigned sym)
-{
- UInt32 range = rc->range;
- unsigned m = 1;
- do
- {
- UInt32 ttt, newBound;
- unsigned bit = sym & 1;
- // RangeEnc_EncodeBit(rc, probs + m, bit);
- sym >>= 1;
- RC_BIT(rc, probs + m, bit);
- m = (m << 1) | bit;
- }
- while (--numBits);
- rc->range = range;
-}
-
-
-
-static void LenEnc_Init(CLenEnc *p)
-{
- unsigned i;
- for (i = 0; i < (LZMA_NUM_PB_STATES_MAX << (kLenNumLowBits + 1)); i++)
- p->low[i] = kProbInitValue;
- for (i = 0; i < kLenNumHighSymbols; i++)
- p->high[i] = kProbInitValue;
-}
-
-static void LenEnc_Encode(CLenEnc *p, CRangeEnc *rc, unsigned sym, unsigned posState)
-{
- UInt32 range, ttt, newBound;
- CLzmaProb *probs = p->low;
- range = rc->range;
- RC_BIT_PRE(rc, probs);
- if (sym >= kLenNumLowSymbols)
- {
- RC_BIT_1(rc, probs);
- probs += kLenNumLowSymbols;
- RC_BIT_PRE(rc, probs);
- if (sym >= kLenNumLowSymbols * 2)
- {
- RC_BIT_1(rc, probs);
- rc->range = range;
- // RcTree_Encode(rc, p->high, kLenNumHighBits, sym - kLenNumLowSymbols * 2);
- LitEnc_Encode(rc, p->high, sym - kLenNumLowSymbols * 2);
- return;
- }
- sym -= kLenNumLowSymbols;
- }
-
- // RcTree_Encode(rc, probs + (posState << kLenNumLowBits), kLenNumLowBits, sym);
- {
- unsigned m;
- unsigned bit;
- RC_BIT_0(rc, probs);
- probs += (posState << (1 + kLenNumLowBits));
- bit = (sym >> 2) ; RC_BIT(rc, probs + 1, bit); m = (1 << 1) + bit;
- bit = (sym >> 1) & 1; RC_BIT(rc, probs + m, bit); m = (m << 1) + bit;
- bit = sym & 1; RC_BIT(rc, probs + m, bit);
- rc->range = range;
- }
-}
-
-static void SetPrices_3(const CLzmaProb *probs, UInt32 startPrice, UInt32 *prices, const CProbPrice *ProbPrices)
-{
- unsigned i;
- for (i = 0; i < 8; i += 2)
- {
- UInt32 price = startPrice;
- UInt32 prob;
- price += GET_PRICEa(probs[1 ], (i >> 2));
- price += GET_PRICEa(probs[2 + (i >> 2)], (i >> 1) & 1);
- prob = probs[4 + (i >> 1)];
- prices[i ] = price + GET_PRICEa_0(prob);
- prices[i + 1] = price + GET_PRICEa_1(prob);
- }
-}
-
-
-MY_NO_INLINE static void MY_FAST_CALL LenPriceEnc_UpdateTables(
- CLenPriceEnc *p,
- unsigned numPosStates,
- const CLenEnc *enc,
- const CProbPrice *ProbPrices)
-{
- UInt32 b;
-
- {
- unsigned prob = enc->low[0];
- UInt32 a, c;
- unsigned posState;
- b = GET_PRICEa_1(prob);
- a = GET_PRICEa_0(prob);
- c = b + GET_PRICEa_0(enc->low[kLenNumLowSymbols]);
- for (posState = 0; posState < numPosStates; posState++)
- {
- UInt32 *prices = p->prices[posState];
- const CLzmaProb *probs = enc->low + (posState << (1 + kLenNumLowBits));
- SetPrices_3(probs, a, prices, ProbPrices);
- SetPrices_3(probs + kLenNumLowSymbols, c, prices + kLenNumLowSymbols, ProbPrices);
- }
- }
-
- /*
- {
- unsigned i;
- UInt32 b;
- a = GET_PRICEa_0(enc->low[0]);
- for (i = 0; i < kLenNumLowSymbols; i++)
- p->prices2[i] = a;
- a = GET_PRICEa_1(enc->low[0]);
- b = a + GET_PRICEa_0(enc->low[kLenNumLowSymbols]);
- for (i = kLenNumLowSymbols; i < kLenNumLowSymbols * 2; i++)
- p->prices2[i] = b;
- a += GET_PRICEa_1(enc->low[kLenNumLowSymbols]);
- }
- */
-
- // p->counter = numSymbols;
- // p->counter = 64;
-
- {
- unsigned i = p->tableSize;
-
- if (i > kLenNumLowSymbols * 2)
- {
- const CLzmaProb *probs = enc->high;
- UInt32 *prices = p->prices[0] + kLenNumLowSymbols * 2;
- i -= kLenNumLowSymbols * 2 - 1;
- i >>= 1;
- b += GET_PRICEa_1(enc->low[kLenNumLowSymbols]);
- do
- {
- /*
- p->prices2[i] = a +
- // RcTree_GetPrice(enc->high, kLenNumHighBits, i - kLenNumLowSymbols * 2, ProbPrices);
- LitEnc_GetPrice(probs, i - kLenNumLowSymbols * 2, ProbPrices);
- */
- // UInt32 price = a + RcTree_GetPrice(probs, kLenNumHighBits - 1, sym, ProbPrices);
- unsigned sym = --i + (1 << (kLenNumHighBits - 1));
- UInt32 price = b;
- do
- {
- unsigned bit = sym & 1;
- sym >>= 1;
- price += GET_PRICEa(probs[sym], bit);
- }
- while (sym >= 2);
-
- {
- unsigned prob = probs[(size_t)i + (1 << (kLenNumHighBits - 1))];
- prices[(size_t)i * 2 ] = price + GET_PRICEa_0(prob);
- prices[(size_t)i * 2 + 1] = price + GET_PRICEa_1(prob);
- }
- }
- while (i);
-
- {
- unsigned posState;
- size_t num = (p->tableSize - kLenNumLowSymbols * 2) * sizeof(p->prices[0][0]);
- for (posState = 1; posState < numPosStates; posState++)
- memcpy(p->prices[posState] + kLenNumLowSymbols * 2, p->prices[0] + kLenNumLowSymbols * 2, num);
- }
- }
- }
-}
-
-/*
- #ifdef SHOW_STAT
- g_STAT_OFFSET += num;
- printf("\n MovePos %u", num);
- #endif
-*/
-
-#define MOVE_POS(p, num) { \
- p->additionalOffset += (num); \
- p->matchFinder.Skip(p->matchFinderObj, (UInt32)(num)); }
-
-
-static unsigned ReadMatchDistances(CLzmaEnc *p, unsigned *numPairsRes)
-{
- unsigned numPairs;
-
- p->additionalOffset++;
- p->numAvail = p->matchFinder.GetNumAvailableBytes(p->matchFinderObj);
- numPairs = p->matchFinder.GetMatches(p->matchFinderObj, p->matches);
- *numPairsRes = numPairs;
-
- #ifdef SHOW_STAT
- printf("\n i = %u numPairs = %u ", g_STAT_OFFSET, numPairs / 2);
- g_STAT_OFFSET++;
- {
- unsigned i;
- for (i = 0; i < numPairs; i += 2)
- printf("%2u %6u | ", p->matches[i], p->matches[i + 1]);
- }
- #endif
-
- if (numPairs == 0)
- return 0;
- {
- unsigned len = p->matches[(size_t)numPairs - 2];
- if (len != p->numFastBytes)
- return len;
- {
- UInt32 numAvail = p->numAvail;
- if (numAvail > LZMA_MATCH_LEN_MAX)
- numAvail = LZMA_MATCH_LEN_MAX;
- {
- const Byte *p1 = p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - 1;
- const Byte *p2 = p1 + len;
- ptrdiff_t dif = (ptrdiff_t)-1 - p->matches[(size_t)numPairs - 1];
- const Byte *lim = p1 + numAvail;
- for (; p2 != lim && *p2 == p2[dif]; p2++)
- {}
- return (unsigned)(p2 - p1);
- }
- }
- }
-}
-
-#define MARK_LIT ((UInt32)(Int32)-1)
-
-#define MakeAs_Lit(p) { (p)->dist = MARK_LIT; (p)->extra = 0; }
-#define MakeAs_ShortRep(p) { (p)->dist = 0; (p)->extra = 0; }
-#define IsShortRep(p) ((p)->dist == 0)
-
-
-#define GetPrice_ShortRep(p, state, posState) \
- ( GET_PRICE_0(p->isRepG0[state]) + GET_PRICE_0(p->isRep0Long[state][posState]))
-
-#define GetPrice_Rep_0(p, state, posState) ( \
- GET_PRICE_1(p->isMatch[state][posState]) \
- + GET_PRICE_1(p->isRep0Long[state][posState])) \
- + GET_PRICE_1(p->isRep[state]) \
- + GET_PRICE_0(p->isRepG0[state])
-
-MY_FORCE_INLINE
-static UInt32 GetPrice_PureRep(const CLzmaEnc *p, unsigned repIndex, size_t state, size_t posState)
-{
- UInt32 price;
- UInt32 prob = p->isRepG0[state];
- if (repIndex == 0)
- {
- price = GET_PRICE_0(prob);
- price += GET_PRICE_1(p->isRep0Long[state][posState]);
- }
- else
- {
- price = GET_PRICE_1(prob);
- prob = p->isRepG1[state];
- if (repIndex == 1)
- price += GET_PRICE_0(prob);
- else
- {
- price += GET_PRICE_1(prob);
- price += GET_PRICE(p->isRepG2[state], repIndex - 2);
- }
- }
- return price;
-}
-
-
-static unsigned Backward(CLzmaEnc *p, unsigned cur)
-{
- unsigned wr = cur + 1;
- p->optEnd = wr;
-
- for (;;)
- {
- UInt32 dist = p->opt[cur].dist;
- unsigned len = (unsigned)p->opt[cur].len;
- unsigned extra = (unsigned)p->opt[cur].extra;
- cur -= len;
-
- if (extra)
- {
- wr--;
- p->opt[wr].len = (UInt32)len;
- cur -= extra;
- len = extra;
- if (extra == 1)
- {
- p->opt[wr].dist = dist;
- dist = MARK_LIT;
- }
- else
- {
- p->opt[wr].dist = 0;
- len--;
- wr--;
- p->opt[wr].dist = MARK_LIT;
- p->opt[wr].len = 1;
- }
- }
-
- if (cur == 0)
- {
- p->backRes = dist;
- p->optCur = wr;
- return len;
- }
-
- wr--;
- p->opt[wr].dist = dist;
- p->opt[wr].len = (UInt32)len;
- }
-}
-
-
-
-#define LIT_PROBS(pos, prevByte) \
- (p->litProbs + (UInt32)3 * (((((pos) << 8) + (prevByte)) & p->lpMask) << p->lc))
-
-
-static unsigned GetOptimum(CLzmaEnc *p, UInt32 position)
-{
- unsigned last, cur;
- UInt32 reps[LZMA_NUM_REPS];
- unsigned repLens[LZMA_NUM_REPS];
- UInt32 *matches;
-
- {
- UInt32 numAvail;
- unsigned numPairs, mainLen, repMaxIndex, i, posState;
- UInt32 matchPrice, repMatchPrice;
- const Byte *data;
- Byte curByte, matchByte;
-
- p->optCur = p->optEnd = 0;
-
- if (p->additionalOffset == 0)
- mainLen = ReadMatchDistances(p, &numPairs);
- else
- {
- mainLen = p->longestMatchLen;
- numPairs = p->numPairs;
- }
-
- numAvail = p->numAvail;
- if (numAvail < 2)
- {
- p->backRes = MARK_LIT;
- return 1;
- }
- if (numAvail > LZMA_MATCH_LEN_MAX)
- numAvail = LZMA_MATCH_LEN_MAX;
-
- data = p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - 1;
- repMaxIndex = 0;
-
- for (i = 0; i < LZMA_NUM_REPS; i++)
- {
- unsigned len;
- const Byte *data2;
- reps[i] = p->reps[i];
- data2 = data - reps[i];
- if (data[0] != data2[0] || data[1] != data2[1])
- {
- repLens[i] = 0;
- continue;
- }
- for (len = 2; len < numAvail && data[len] == data2[len]; len++)
- {}
- repLens[i] = len;
- if (len > repLens[repMaxIndex])
- repMaxIndex = i;
- }
-
- if (repLens[repMaxIndex] >= p->numFastBytes)
- {
- unsigned len;
- p->backRes = (UInt32)repMaxIndex;
- len = repLens[repMaxIndex];
- MOVE_POS(p, len - 1)
- return len;
- }
-
- matches = p->matches;
-
- if (mainLen >= p->numFastBytes)
- {
- p->backRes = matches[(size_t)numPairs - 1] + LZMA_NUM_REPS;
- MOVE_POS(p, mainLen - 1)
- return mainLen;
- }
-
- curByte = *data;
- matchByte = *(data - reps[0]);
-
- last = repLens[repMaxIndex];
- if (last <= mainLen)
- last = mainLen;
-
- if (last < 2 && curByte != matchByte)
- {
- p->backRes = MARK_LIT;
- return 1;
- }
-
- p->opt[0].state = (CState)p->state;
-
- posState = (position & p->pbMask);
-
- {
- const CLzmaProb *probs = LIT_PROBS(position, *(data - 1));
- p->opt[1].price = GET_PRICE_0(p->isMatch[p->state][posState]) +
- (!IsLitState(p->state) ?
- LitEnc_Matched_GetPrice(probs, curByte, matchByte, p->ProbPrices) :
- LitEnc_GetPrice(probs, curByte, p->ProbPrices));
- }
-
- MakeAs_Lit(&p->opt[1]);
-
- matchPrice = GET_PRICE_1(p->isMatch[p->state][posState]);
- repMatchPrice = matchPrice + GET_PRICE_1(p->isRep[p->state]);
-
- // 18.06
- if (matchByte == curByte && repLens[0] == 0)
- {
- UInt32 shortRepPrice = repMatchPrice + GetPrice_ShortRep(p, p->state, posState);
- if (shortRepPrice < p->opt[1].price)
- {
- p->opt[1].price = shortRepPrice;
- MakeAs_ShortRep(&p->opt[1]);
- }
- if (last < 2)
- {
- p->backRes = p->opt[1].dist;
- return 1;
- }
- }
-
- p->opt[1].len = 1;
-
- p->opt[0].reps[0] = reps[0];
- p->opt[0].reps[1] = reps[1];
- p->opt[0].reps[2] = reps[2];
- p->opt[0].reps[3] = reps[3];
-
- // ---------- REP ----------
-
- for (i = 0; i < LZMA_NUM_REPS; i++)
- {
- unsigned repLen = repLens[i];
- UInt32 price;
- if (repLen < 2)
- continue;
- price = repMatchPrice + GetPrice_PureRep(p, i, p->state, posState);
- do
- {
- UInt32 price2 = price + GET_PRICE_LEN(&p->repLenEnc, posState, repLen);
- COptimal *opt = &p->opt[repLen];
- if (price2 < opt->price)
- {
- opt->price = price2;
- opt->len = (UInt32)repLen;
- opt->dist = (UInt32)i;
- opt->extra = 0;
- }
- }
- while (--repLen >= 2);
- }
-
-
- // ---------- MATCH ----------
- {
- unsigned len = repLens[0] + 1;
- if (len <= mainLen)
- {
- unsigned offs = 0;
- UInt32 normalMatchPrice = matchPrice + GET_PRICE_0(p->isRep[p->state]);
-
- if (len < 2)
- len = 2;
- else
- while (len > matches[offs])
- offs += 2;
-
- for (; ; len++)
- {
- COptimal *opt;
- UInt32 dist = matches[(size_t)offs + 1];
- UInt32 price = normalMatchPrice + GET_PRICE_LEN(&p->lenEnc, posState, len);
- unsigned lenToPosState = GetLenToPosState(len);
-
- if (dist < kNumFullDistances)
- price += p->distancesPrices[lenToPosState][dist & (kNumFullDistances - 1)];
- else
- {
- unsigned slot;
- GetPosSlot2(dist, slot);
- price += p->alignPrices[dist & kAlignMask];
- price += p->posSlotPrices[lenToPosState][slot];
- }
-
- opt = &p->opt[len];
-
- if (price < opt->price)
- {
- opt->price = price;
- opt->len = (UInt32)len;
- opt->dist = dist + LZMA_NUM_REPS;
- opt->extra = 0;
- }
-
- if (len == matches[offs])
- {
- offs += 2;
- if (offs == numPairs)
- break;
- }
- }
- }
- }
-
-
- cur = 0;
-
- #ifdef SHOW_STAT2
- /* if (position >= 0) */
- {
- unsigned i;
- printf("\n pos = %4X", position);
- for (i = cur; i <= last; i++)
- printf("\nprice[%4X] = %u", position - cur + i, p->opt[i].price);
- }
- #endif
- }
-
-
-
- // ---------- Optimal Parsing ----------
-
- for (;;)
- {
- unsigned numAvail;
- UInt32 numAvailFull;
- unsigned newLen, numPairs, prev, state, posState, startLen;
- UInt32 litPrice, matchPrice, repMatchPrice;
- BoolInt nextIsLit;
- Byte curByte, matchByte;
- const Byte *data;
- COptimal *curOpt, *nextOpt;
-
- if (++cur == last)
- break;
-
- // 18.06
- if (cur >= kNumOpts - 64)
- {
- unsigned j, best;
- UInt32 price = p->opt[cur].price;
- best = cur;
- for (j = cur + 1; j <= last; j++)
- {
- UInt32 price2 = p->opt[j].price;
- if (price >= price2)
- {
- price = price2;
- best = j;
- }
- }
- {
- unsigned delta = best - cur;
- if (delta != 0)
- {
- MOVE_POS(p, delta);
- }
- }
- cur = best;
- break;
- }
-
- newLen = ReadMatchDistances(p, &numPairs);
-
- if (newLen >= p->numFastBytes)
- {
- p->numPairs = numPairs;
- p->longestMatchLen = newLen;
- break;
- }
-
- curOpt = &p->opt[cur];
-
- position++;
-
- // we need that check here, if skip_items in p->opt are possible
- /*
- if (curOpt->price >= kInfinityPrice)
- continue;
- */
-
- prev = cur - curOpt->len;
-
- if (curOpt->len == 1)
- {
- state = (unsigned)p->opt[prev].state;
- if (IsShortRep(curOpt))
- state = kShortRepNextStates[state];
- else
- state = kLiteralNextStates[state];
- }
- else
- {
- const COptimal *prevOpt;
- UInt32 b0;
- UInt32 dist = curOpt->dist;
-
- if (curOpt->extra)
- {
- prev -= (unsigned)curOpt->extra;
- state = kState_RepAfterLit;
- if (curOpt->extra == 1)
- state = (dist < LZMA_NUM_REPS ? kState_RepAfterLit : kState_MatchAfterLit);
- }
- else
- {
- state = (unsigned)p->opt[prev].state;
- if (dist < LZMA_NUM_REPS)
- state = kRepNextStates[state];
- else
- state = kMatchNextStates[state];
- }
-
- prevOpt = &p->opt[prev];
- b0 = prevOpt->reps[0];
-
- if (dist < LZMA_NUM_REPS)
- {
- if (dist == 0)
- {
- reps[0] = b0;
- reps[1] = prevOpt->reps[1];
- reps[2] = prevOpt->reps[2];
- reps[3] = prevOpt->reps[3];
- }
- else
- {
- reps[1] = b0;
- b0 = prevOpt->reps[1];
- if (dist == 1)
- {
- reps[0] = b0;
- reps[2] = prevOpt->reps[2];
- reps[3] = prevOpt->reps[3];
- }
- else
- {
- reps[2] = b0;
- reps[0] = prevOpt->reps[dist];
- reps[3] = prevOpt->reps[dist ^ 1];
- }
- }
- }
- else
- {
- reps[0] = (dist - LZMA_NUM_REPS + 1);
- reps[1] = b0;
- reps[2] = prevOpt->reps[1];
- reps[3] = prevOpt->reps[2];
- }
- }
-
- curOpt->state = (CState)state;
- curOpt->reps[0] = reps[0];
- curOpt->reps[1] = reps[1];
- curOpt->reps[2] = reps[2];
- curOpt->reps[3] = reps[3];
-
- data = p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - 1;
- curByte = *data;
- matchByte = *(data - reps[0]);
-
- posState = (position & p->pbMask);
-
- /*
- The order of Price checks:
- < LIT
- <= SHORT_REP
- < LIT : REP_0
- < REP [ : LIT : REP_0 ]
- < MATCH [ : LIT : REP_0 ]
- */
-
- {
- UInt32 curPrice = curOpt->price;
- unsigned prob = p->isMatch[state][posState];
- matchPrice = curPrice + GET_PRICE_1(prob);
- litPrice = curPrice + GET_PRICE_0(prob);
- }
-
- nextOpt = &p->opt[(size_t)cur + 1];
- nextIsLit = False;
-
- // here we can allow skip_items in p->opt, if we don't check (nextOpt->price < kInfinityPrice)
- // 18.new.06
- if ((nextOpt->price < kInfinityPrice
- // && !IsLitState(state)
- && matchByte == curByte)
- || litPrice > nextOpt->price
- )
- litPrice = 0;
- else
- {
- const CLzmaProb *probs = LIT_PROBS(position, *(data - 1));
- litPrice += (!IsLitState(state) ?
- LitEnc_Matched_GetPrice(probs, curByte, matchByte, p->ProbPrices) :
- LitEnc_GetPrice(probs, curByte, p->ProbPrices));
-
- if (litPrice < nextOpt->price)
- {
- nextOpt->price = litPrice;
- nextOpt->len = 1;
- MakeAs_Lit(nextOpt);
- nextIsLit = True;
- }
- }
-
- repMatchPrice = matchPrice + GET_PRICE_1(p->isRep[state]);
-
- numAvailFull = p->numAvail;
- {
- unsigned temp = kNumOpts - 1 - cur;
- if (numAvailFull > temp)
- numAvailFull = (UInt32)temp;
- }
-
- // 18.06
- // ---------- SHORT_REP ----------
- if (IsLitState(state)) // 18.new
- if (matchByte == curByte)
- if (repMatchPrice < nextOpt->price) // 18.new
- // if (numAvailFull < 2 || data[1] != *(data - reps[0] + 1))
- if (
- // nextOpt->price >= kInfinityPrice ||
- nextOpt->len < 2 // we can check nextOpt->len, if skip items are not allowed in p->opt
- || (nextOpt->dist != 0
- // && nextOpt->extra <= 1 // 17.old
- )
- )
- {
- UInt32 shortRepPrice = repMatchPrice + GetPrice_ShortRep(p, state, posState);
- // if (shortRepPrice <= nextOpt->price) // 17.old
- if (shortRepPrice < nextOpt->price) // 18.new
- {
- nextOpt->price = shortRepPrice;
- nextOpt->len = 1;
- MakeAs_ShortRep(nextOpt);
- nextIsLit = False;
- }
- }
-
- if (numAvailFull < 2)
- continue;
- numAvail = (numAvailFull <= p->numFastBytes ? numAvailFull : p->numFastBytes);
-
- // numAvail <= p->numFastBytes
-
- // ---------- LIT : REP_0 ----------
-
- if (!nextIsLit
- && litPrice != 0 // 18.new
- && matchByte != curByte
- && numAvailFull > 2)
- {
- const Byte *data2 = data - reps[0];
- if (data[1] == data2[1] && data[2] == data2[2])
- {
- unsigned len;
- unsigned limit = p->numFastBytes + 1;
- if (limit > numAvailFull)
- limit = numAvailFull;
- for (len = 3; len < limit && data[len] == data2[len]; len++)
- {}
-
- {
- unsigned state2 = kLiteralNextStates[state];
- unsigned posState2 = (position + 1) & p->pbMask;
- UInt32 price = litPrice + GetPrice_Rep_0(p, state2, posState2);
- {
- unsigned offset = cur + len;
-
- if (last < offset)
- last = offset;
-
- // do
- {
- UInt32 price2;
- COptimal *opt;
- len--;
- // price2 = price + GetPrice_Len_Rep_0(p, len, state2, posState2);
- price2 = price + GET_PRICE_LEN(&p->repLenEnc, posState2, len);
-
- opt = &p->opt[offset];
- // offset--;
- if (price2 < opt->price)
- {
- opt->price = price2;
- opt->len = (UInt32)len;
- opt->dist = 0;
- opt->extra = 1;
- }
- }
- // while (len >= 3);
- }
- }
- }
- }
-
- startLen = 2; /* speed optimization */
-
- {
- // ---------- REP ----------
- unsigned repIndex = 0; // 17.old
- // unsigned repIndex = IsLitState(state) ? 0 : 1; // 18.notused
- for (; repIndex < LZMA_NUM_REPS; repIndex++)
- {
- unsigned len;
- UInt32 price;
- const Byte *data2 = data - reps[repIndex];
- if (data[0] != data2[0] || data[1] != data2[1])
- continue;
-
- for (len = 2; len < numAvail && data[len] == data2[len]; len++)
- {}
-
- // if (len < startLen) continue; // 18.new: speed optimization
-
- {
- unsigned offset = cur + len;
- if (last < offset)
- last = offset;
- }
- {
- unsigned len2 = len;
- price = repMatchPrice + GetPrice_PureRep(p, repIndex, state, posState);
- do
- {
- UInt32 price2 = price + GET_PRICE_LEN(&p->repLenEnc, posState, len2);
- COptimal *opt = &p->opt[cur + len2];
- if (price2 < opt->price)
- {
- opt->price = price2;
- opt->len = (UInt32)len2;
- opt->dist = (UInt32)repIndex;
- opt->extra = 0;
- }
- }
- while (--len2 >= 2);
- }
-
- if (repIndex == 0) startLen = len + 1; // 17.old
- // startLen = len + 1; // 18.new
-
- /* if (_maxMode) */
- {
- // ---------- REP : LIT : REP_0 ----------
- // numFastBytes + 1 + numFastBytes
-
- unsigned len2 = len + 1;
- unsigned limit = len2 + p->numFastBytes;
- if (limit > numAvailFull)
- limit = numAvailFull;
-
- len2 += 2;
- if (len2 <= limit)
- if (data[len2 - 2] == data2[len2 - 2])
- if (data[len2 - 1] == data2[len2 - 1])
- {
- unsigned state2 = kRepNextStates[state];
- unsigned posState2 = (position + len) & p->pbMask;
- price += GET_PRICE_LEN(&p->repLenEnc, posState, len)
- + GET_PRICE_0(p->isMatch[state2][posState2])
- + LitEnc_Matched_GetPrice(LIT_PROBS(position + len, data[(size_t)len - 1]),
- data[len], data2[len], p->ProbPrices);
-
- // state2 = kLiteralNextStates[state2];
- state2 = kState_LitAfterRep;
- posState2 = (posState2 + 1) & p->pbMask;
-
-
- price += GetPrice_Rep_0(p, state2, posState2);
-
- for (; len2 < limit && data[len2] == data2[len2]; len2++)
- {}
-
- len2 -= len;
- // if (len2 >= 3)
- {
- {
- unsigned offset = cur + len + len2;
-
- if (last < offset)
- last = offset;
- // do
- {
- UInt32 price2;
- COptimal *opt;
- len2--;
- // price2 = price + GetPrice_Len_Rep_0(p, len2, state2, posState2);
- price2 = price + GET_PRICE_LEN(&p->repLenEnc, posState2, len2);
-
- opt = &p->opt[offset];
- // offset--;
- if (price2 < opt->price)
- {
- opt->price = price2;
- opt->len = (UInt32)len2;
- opt->extra = (CExtra)(len + 1);
- opt->dist = (UInt32)repIndex;
- }
- }
- // while (len2 >= 3);
- }
- }
- }
- }
- }
- }
-
-
- // ---------- MATCH ----------
- /* for (unsigned len = 2; len <= newLen; len++) */
- if (newLen > numAvail)
- {
- newLen = numAvail;
- for (numPairs = 0; newLen > matches[numPairs]; numPairs += 2);
- matches[numPairs] = (UInt32)newLen;
- numPairs += 2;
- }
-
- // startLen = 2; /* speed optimization */
-
- if (newLen >= startLen)
- {
- UInt32 normalMatchPrice = matchPrice + GET_PRICE_0(p->isRep[state]);
- UInt32 dist;
- unsigned offs, posSlot, len;
-
- {
- unsigned offset = cur + newLen;
- if (last < offset)
- last = offset;
- }
-
- offs = 0;
- while (startLen > matches[offs])
- offs += 2;
- dist = matches[(size_t)offs + 1];
-
- // if (dist >= kNumFullDistances)
- GetPosSlot2(dist, posSlot);
-
- for (len = /*2*/ startLen; ; len++)
- {
- UInt32 price = normalMatchPrice + GET_PRICE_LEN(&p->lenEnc, posState, len);
- {
- COptimal *opt;
- unsigned lenNorm = len - 2;
- lenNorm = GetLenToPosState2(lenNorm);
- if (dist < kNumFullDistances)
- price += p->distancesPrices[lenNorm][dist & (kNumFullDistances - 1)];
- else
- price += p->posSlotPrices[lenNorm][posSlot] + p->alignPrices[dist & kAlignMask];
-
- opt = &p->opt[cur + len];
- if (price < opt->price)
- {
- opt->price = price;
- opt->len = (UInt32)len;
- opt->dist = dist + LZMA_NUM_REPS;
- opt->extra = 0;
- }
- }
-
- if (len == matches[offs])
- {
- // if (p->_maxMode) {
- // MATCH : LIT : REP_0
-
- const Byte *data2 = data - dist - 1;
- unsigned len2 = len + 1;
- unsigned limit = len2 + p->numFastBytes;
- if (limit > numAvailFull)
- limit = numAvailFull;
-
- len2 += 2;
- if (len2 <= limit)
- if (data[len2 - 2] == data2[len2 - 2])
- if (data[len2 - 1] == data2[len2 - 1])
- {
- for (; len2 < limit && data[len2] == data2[len2]; len2++)
- {}
-
- len2 -= len;
-
- // if (len2 >= 3)
- {
- unsigned state2 = kMatchNextStates[state];
- unsigned posState2 = (position + len) & p->pbMask;
- unsigned offset;
- price += GET_PRICE_0(p->isMatch[state2][posState2]);
- price += LitEnc_Matched_GetPrice(LIT_PROBS(position + len, data[(size_t)len - 1]),
- data[len], data2[len], p->ProbPrices);
-
- // state2 = kLiteralNextStates[state2];
- state2 = kState_LitAfterMatch;
-
- posState2 = (posState2 + 1) & p->pbMask;
- price += GetPrice_Rep_0(p, state2, posState2);
-
- offset = cur + len + len2;
-
- if (last < offset)
- last = offset;
- // do
- {
- UInt32 price2;
- COptimal *opt;
- len2--;
- // price2 = price + GetPrice_Len_Rep_0(p, len2, state2, posState2);
- price2 = price + GET_PRICE_LEN(&p->repLenEnc, posState2, len2);
- opt = &p->opt[offset];
- // offset--;
- if (price2 < opt->price)
- {
- opt->price = price2;
- opt->len = (UInt32)len2;
- opt->extra = (CExtra)(len + 1);
- opt->dist = dist + LZMA_NUM_REPS;
- }
- }
- // while (len2 >= 3);
- }
-
- }
-
- offs += 2;
- if (offs == numPairs)
- break;
- dist = matches[(size_t)offs + 1];
- // if (dist >= kNumFullDistances)
- GetPosSlot2(dist, posSlot);
- }
- }
- }
- }
-
- do
- p->opt[last].price = kInfinityPrice;
- while (--last);
-
- return Backward(p, cur);
-}
-
-
-
-#define ChangePair(smallDist, bigDist) (((bigDist) >> 7) > (smallDist))
-
-
-
-static unsigned GetOptimumFast(CLzmaEnc *p)
-{
- UInt32 numAvail, mainDist;
- unsigned mainLen, numPairs, repIndex, repLen, i;
- const Byte *data;
-
- if (p->additionalOffset == 0)
- mainLen = ReadMatchDistances(p, &numPairs);
- else
- {
- mainLen = p->longestMatchLen;
- numPairs = p->numPairs;
- }
-
- numAvail = p->numAvail;
- p->backRes = MARK_LIT;
- if (numAvail < 2)
- return 1;
- // if (mainLen < 2 && p->state == 0) return 1; // 18.06.notused
- if (numAvail > LZMA_MATCH_LEN_MAX)
- numAvail = LZMA_MATCH_LEN_MAX;
- data = p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - 1;
- repLen = repIndex = 0;
-
- for (i = 0; i < LZMA_NUM_REPS; i++)
- {
- unsigned len;
- const Byte *data2 = data - p->reps[i];
- if (data[0] != data2[0] || data[1] != data2[1])
- continue;
- for (len = 2; len < numAvail && data[len] == data2[len]; len++)
- {}
- if (len >= p->numFastBytes)
- {
- p->backRes = (UInt32)i;
- MOVE_POS(p, len - 1)
- return len;
- }
- if (len > repLen)
- {
- repIndex = i;
- repLen = len;
- }
- }
-
- if (mainLen >= p->numFastBytes)
- {
- p->backRes = p->matches[(size_t)numPairs - 1] + LZMA_NUM_REPS;
- MOVE_POS(p, mainLen - 1)
- return mainLen;
- }
-
- mainDist = 0; /* for GCC */
-
- if (mainLen >= 2)
- {
- mainDist = p->matches[(size_t)numPairs - 1];
- while (numPairs > 2)
- {
- UInt32 dist2;
- if (mainLen != p->matches[(size_t)numPairs - 4] + 1)
- break;
- dist2 = p->matches[(size_t)numPairs - 3];
- if (!ChangePair(dist2, mainDist))
- break;
- numPairs -= 2;
- mainLen--;
- mainDist = dist2;
- }
- if (mainLen == 2 && mainDist >= 0x80)
- mainLen = 1;
- }
-
- if (repLen >= 2)
- if ( repLen + 1 >= mainLen
- || (repLen + 2 >= mainLen && mainDist >= (1 << 9))
- || (repLen + 3 >= mainLen && mainDist >= (1 << 15)))
- {
- p->backRes = (UInt32)repIndex;
- MOVE_POS(p, repLen - 1)
- return repLen;
- }
-
- if (mainLen < 2 || numAvail <= 2)
- return 1;
-
- {
- unsigned len1 = ReadMatchDistances(p, &p->numPairs);
- p->longestMatchLen = len1;
-
- if (len1 >= 2)
- {
- UInt32 newDist = p->matches[(size_t)p->numPairs - 1];
- if ( (len1 >= mainLen && newDist < mainDist)
- || (len1 == mainLen + 1 && !ChangePair(mainDist, newDist))
- || (len1 > mainLen + 1)
- || (len1 + 1 >= mainLen && mainLen >= 3 && ChangePair(newDist, mainDist)))
- return 1;
- }
- }
-
- data = p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - 1;
-
- for (i = 0; i < LZMA_NUM_REPS; i++)
- {
- unsigned len, limit;
- const Byte *data2 = data - p->reps[i];
- if (data[0] != data2[0] || data[1] != data2[1])
- continue;
- limit = mainLen - 1;
- for (len = 2;; len++)
- {
- if (len >= limit)
- return 1;
- if (data[len] != data2[len])
- break;
- }
- }
-
- p->backRes = mainDist + LZMA_NUM_REPS;
- if (mainLen != 2)
- {
- MOVE_POS(p, mainLen - 2)
- }
- return mainLen;
-}
-
-
-
-
-static void WriteEndMarker(CLzmaEnc *p, unsigned posState)
-{
- UInt32 range;
- range = p->rc.range;
- {
- UInt32 ttt, newBound;
- CLzmaProb *prob = &p->isMatch[p->state][posState];
- RC_BIT_PRE(&p->rc, prob)
- RC_BIT_1(&p->rc, prob)
- prob = &p->isRep[p->state];
- RC_BIT_PRE(&p->rc, prob)
- RC_BIT_0(&p->rc, prob)
- }
- p->state = kMatchNextStates[p->state];
-
- p->rc.range = range;
- LenEnc_Encode(&p->lenProbs, &p->rc, 0, posState);
- range = p->rc.range;
-
- {
- // RcTree_Encode_PosSlot(&p->rc, p->posSlotEncoder[0], (1 << kNumPosSlotBits) - 1);
- CLzmaProb *probs = p->posSlotEncoder[0];
- unsigned m = 1;
- do
- {
- UInt32 ttt, newBound;
- RC_BIT_PRE(p, probs + m)
- RC_BIT_1(&p->rc, probs + m);
- m = (m << 1) + 1;
- }
- while (m < (1 << kNumPosSlotBits));
- }
- {
- // RangeEnc_EncodeDirectBits(&p->rc, ((UInt32)1 << (30 - kNumAlignBits)) - 1, 30 - kNumAlignBits); UInt32 range = p->range;
- unsigned numBits = 30 - kNumAlignBits;
- do
- {
- range >>= 1;
- p->rc.low += range;
- RC_NORM(&p->rc)
- }
- while (--numBits);
- }
-
- {
- // RcTree_ReverseEncode(&p->rc, p->posAlignEncoder, kNumAlignBits, kAlignMask);
- CLzmaProb *probs = p->posAlignEncoder;
- unsigned m = 1;
- do
- {
- UInt32 ttt, newBound;
- RC_BIT_PRE(p, probs + m)
- RC_BIT_1(&p->rc, probs + m);
- m = (m << 1) + 1;
- }
- while (m < kAlignTableSize);
- }
- p->rc.range = range;
-}
-
-
-static SRes CheckErrors(CLzmaEnc *p)
-{
- if (p->result != SZ_OK)
- return p->result;
- if (p->rc.res != SZ_OK)
- p->result = SZ_ERROR_WRITE;
- if (p->matchFinderBase.result != SZ_OK)
- p->result = SZ_ERROR_READ;
- if (p->result != SZ_OK)
- p->finished = True;
- return p->result;
-}
-
-
-MY_NO_INLINE static SRes Flush(CLzmaEnc *p, UInt32 nowPos)
-{
- /* ReleaseMFStream(); */
- p->finished = True;
- if (p->writeEndMark)
- WriteEndMarker(p, nowPos & p->pbMask);
- RangeEnc_FlushData(&p->rc);
- RangeEnc_FlushStream(&p->rc);
- return CheckErrors(p);
-}
-
-
-MY_NO_INLINE static void FillAlignPrices(CLzmaEnc *p)
-{
- unsigned i;
- const CProbPrice *ProbPrices = p->ProbPrices;
- const CLzmaProb *probs = p->posAlignEncoder;
- // p->alignPriceCount = 0;
- for (i = 0; i < kAlignTableSize / 2; i++)
- {
- UInt32 price = 0;
- unsigned sym = i;
- unsigned m = 1;
- unsigned bit;
- UInt32 prob;
- bit = sym & 1; sym >>= 1; price += GET_PRICEa(probs[m], bit); m = (m << 1) + bit;
- bit = sym & 1; sym >>= 1; price += GET_PRICEa(probs[m], bit); m = (m << 1) + bit;
- bit = sym & 1; sym >>= 1; price += GET_PRICEa(probs[m], bit); m = (m << 1) + bit;
- prob = probs[m];
- p->alignPrices[i ] = price + GET_PRICEa_0(prob);
- p->alignPrices[i + 8] = price + GET_PRICEa_1(prob);
- // p->alignPrices[i] = RcTree_ReverseGetPrice(p->posAlignEncoder, kNumAlignBits, i, p->ProbPrices);
- }
-}
-
-
-MY_NO_INLINE static void FillDistancesPrices(CLzmaEnc *p)
-{
- // int y; for (y = 0; y < 100; y++) {
-
- UInt32 tempPrices[kNumFullDistances];
- unsigned i, lps;
-
- const CProbPrice *ProbPrices = p->ProbPrices;
- p->matchPriceCount = 0;
-
- for (i = kStartPosModelIndex / 2; i < kNumFullDistances / 2; i++)
- {
- unsigned posSlot = GetPosSlot1(i);
- unsigned footerBits = (posSlot >> 1) - 1;
- unsigned base = ((2 | (posSlot & 1)) << footerBits);
- const CLzmaProb *probs = p->posEncoders + (size_t)base * 2;
- // tempPrices[i] = RcTree_ReverseGetPrice(p->posEncoders + base, footerBits, i - base, p->ProbPrices);
- UInt32 price = 0;
- unsigned m = 1;
- unsigned sym = i;
- unsigned offset = (unsigned)1 << footerBits;
- base += i;
-
- if (footerBits)
- do
- {
- unsigned bit = sym & 1;
- sym >>= 1;
- price += GET_PRICEa(probs[m], bit);
- m = (m << 1) + bit;
- }
- while (--footerBits);
-
- {
- unsigned prob = probs[m];
- tempPrices[base ] = price + GET_PRICEa_0(prob);
- tempPrices[base + offset] = price + GET_PRICEa_1(prob);
- }
- }
-
- for (lps = 0; lps < kNumLenToPosStates; lps++)
- {
- unsigned slot;
- unsigned distTableSize2 = (p->distTableSize + 1) >> 1;
- UInt32 *posSlotPrices = p->posSlotPrices[lps];
- const CLzmaProb *probs = p->posSlotEncoder[lps];
-
- for (slot = 0; slot < distTableSize2; slot++)
- {
- // posSlotPrices[slot] = RcTree_GetPrice(encoder, kNumPosSlotBits, slot, p->ProbPrices);
- UInt32 price;
- unsigned bit;
- unsigned sym = slot + (1 << (kNumPosSlotBits - 1));
- unsigned prob;
- bit = sym & 1; sym >>= 1; price = GET_PRICEa(probs[sym], bit);
- bit = sym & 1; sym >>= 1; price += GET_PRICEa(probs[sym], bit);
- bit = sym & 1; sym >>= 1; price += GET_PRICEa(probs[sym], bit);
- bit = sym & 1; sym >>= 1; price += GET_PRICEa(probs[sym], bit);
- bit = sym & 1; sym >>= 1; price += GET_PRICEa(probs[sym], bit);
- prob = probs[(size_t)slot + (1 << (kNumPosSlotBits - 1))];
- posSlotPrices[(size_t)slot * 2 ] = price + GET_PRICEa_0(prob);
- posSlotPrices[(size_t)slot * 2 + 1] = price + GET_PRICEa_1(prob);
- }
-
- {
- UInt32 delta = ((UInt32)((kEndPosModelIndex / 2 - 1) - kNumAlignBits) << kNumBitPriceShiftBits);
- for (slot = kEndPosModelIndex / 2; slot < distTableSize2; slot++)
- {
- posSlotPrices[(size_t)slot * 2 ] += delta;
- posSlotPrices[(size_t)slot * 2 + 1] += delta;
- delta += ((UInt32)1 << kNumBitPriceShiftBits);
- }
- }
-
- {
- UInt32 *dp = p->distancesPrices[lps];
-
- dp[0] = posSlotPrices[0];
- dp[1] = posSlotPrices[1];
- dp[2] = posSlotPrices[2];
- dp[3] = posSlotPrices[3];
-
- for (i = 4; i < kNumFullDistances; i += 2)
- {
- UInt32 slotPrice = posSlotPrices[GetPosSlot1(i)];
- dp[i ] = slotPrice + tempPrices[i];
- dp[i + 1] = slotPrice + tempPrices[i + 1];
- }
- }
- }
- // }
-}
-
-
-
-void LzmaEnc_Construct(CLzmaEnc *p)
-{
- RangeEnc_Construct(&p->rc);
- MatchFinder_Construct(&p->matchFinderBase);
-
- #ifndef _7ZIP_ST
- MatchFinderMt_Construct(&p->matchFinderMt);
- p->matchFinderMt.MatchFinder = &p->matchFinderBase;
- #endif
-
- {
- CLzmaEncProps props;
- LzmaEncProps_Init(&props);
- LzmaEnc_SetProps(p, &props);
- }
-
- #ifndef LZMA_LOG_BSR
- LzmaEnc_FastPosInit(p->g_FastPos);
- #endif
-
- LzmaEnc_InitPriceTables(p->ProbPrices);
- p->litProbs = NULL;
- p->saveState.litProbs = NULL;
-
-}
-
-CLzmaEncHandle LzmaEnc_Create(ISzAllocPtr alloc)
-{
- void *p;
- p = ISzAlloc_Alloc(alloc, sizeof(CLzmaEnc));
- if (p)
- LzmaEnc_Construct((CLzmaEnc *)p);
- return p;
-}
-
-void LzmaEnc_FreeLits(CLzmaEnc *p, ISzAllocPtr alloc)
-{
- ISzAlloc_Free(alloc, p->litProbs);
- ISzAlloc_Free(alloc, p->saveState.litProbs);
- p->litProbs = NULL;
- p->saveState.litProbs = NULL;
-}
-
-void LzmaEnc_Destruct(CLzmaEnc *p, ISzAllocPtr alloc, ISzAllocPtr allocBig)
-{
- #ifndef _7ZIP_ST
- MatchFinderMt_Destruct(&p->matchFinderMt, allocBig);
- #endif
-
- MatchFinder_Free(&p->matchFinderBase, allocBig);
- LzmaEnc_FreeLits(p, alloc);
- RangeEnc_Free(&p->rc, alloc);
-}
-
-void LzmaEnc_Destroy(CLzmaEncHandle p, ISzAllocPtr alloc, ISzAllocPtr allocBig)
-{
- LzmaEnc_Destruct((CLzmaEnc *)p, alloc, allocBig);
- ISzAlloc_Free(alloc, p);
-}
-
-
-SRes LzmaEnc_CodeOneBlock(CLzmaEncHandle pp, UInt32 maxPackSize, UInt32 maxUnpackSize)
-{
- CLzmaEnc *p = (CLzmaEnc *) pp;
- UInt32 nowPos32, startPos32;
- if (p->needInit)
- {
- p->matchFinder.Init(p->matchFinderObj);
- p->needInit = 0;
- }
-
- if (p->finished)
- return p->result;
- RINOK(CheckErrors(p));
-
- nowPos32 = (UInt32)p->nowPos64;
- startPos32 = nowPos32;
-
- if (p->nowPos64 == 0)
- {
- unsigned numPairs;
- Byte curByte;
- if (p->matchFinder.GetNumAvailableBytes(p->matchFinderObj) == 0)
- return Flush(p, nowPos32);
- ReadMatchDistances(p, &numPairs);
- RangeEnc_EncodeBit_0(&p->rc, &p->isMatch[kState_Start][0]);
- // p->state = kLiteralNextStates[p->state];
- curByte = *(p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - p->additionalOffset);
- LitEnc_Encode(&p->rc, p->litProbs, curByte);
- p->additionalOffset--;
- nowPos32++;
- }
-
- if (p->matchFinder.GetNumAvailableBytes(p->matchFinderObj) != 0)
-
- for (;;)
- {
- UInt32 dist;
- unsigned len, posState;
- UInt32 range, ttt, newBound;
- CLzmaProb *probs;
-
- if (p->fastMode)
- len = GetOptimumFast(p);
- else
- {
- unsigned oci = p->optCur;
- if (p->optEnd == oci)
- len = GetOptimum(p, nowPos32);
- else
- {
- const COptimal *opt = &p->opt[oci];
- len = opt->len;
- p->backRes = opt->dist;
- p->optCur = oci + 1;
- }
- }
-
- posState = (unsigned)nowPos32 & p->pbMask;
- range = p->rc.range;
- probs = &p->isMatch[p->state][posState];
-
- RC_BIT_PRE(&p->rc, probs)
-
- dist = p->backRes;
-
- #ifdef SHOW_STAT2
- printf("\n pos = %6X, len = %3u pos = %6u", nowPos32, len, dist);
- #endif
-
- if (dist == MARK_LIT)
- {
- Byte curByte;
- const Byte *data;
- unsigned state;
-
- RC_BIT_0(&p->rc, probs);
- p->rc.range = range;
- data = p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - p->additionalOffset;
- probs = LIT_PROBS(nowPos32, *(data - 1));
- curByte = *data;
- state = p->state;
- p->state = kLiteralNextStates[state];
- if (IsLitState(state))
- LitEnc_Encode(&p->rc, probs, curByte);
- else
- LitEnc_EncodeMatched(&p->rc, probs, curByte, *(data - p->reps[0]));
- }
- else
- {
- RC_BIT_1(&p->rc, probs);
- probs = &p->isRep[p->state];
- RC_BIT_PRE(&p->rc, probs)
-
- if (dist < LZMA_NUM_REPS)
- {
- RC_BIT_1(&p->rc, probs);
- probs = &p->isRepG0[p->state];
- RC_BIT_PRE(&p->rc, probs)
- if (dist == 0)
- {
- RC_BIT_0(&p->rc, probs);
- probs = &p->isRep0Long[p->state][posState];
- RC_BIT_PRE(&p->rc, probs)
- if (len != 1)
- {
- RC_BIT_1_BASE(&p->rc, probs);
- }
- else
- {
- RC_BIT_0_BASE(&p->rc, probs);
- p->state = kShortRepNextStates[p->state];
- }
- }
- else
- {
- RC_BIT_1(&p->rc, probs);
- probs = &p->isRepG1[p->state];
- RC_BIT_PRE(&p->rc, probs)
- if (dist == 1)
- {
- RC_BIT_0_BASE(&p->rc, probs);
- dist = p->reps[1];
- }
- else
- {
- RC_BIT_1(&p->rc, probs);
- probs = &p->isRepG2[p->state];
- RC_BIT_PRE(&p->rc, probs)
- if (dist == 2)
- {
- RC_BIT_0_BASE(&p->rc, probs);
- dist = p->reps[2];
- }
- else
- {
- RC_BIT_1_BASE(&p->rc, probs);
- dist = p->reps[3];
- p->reps[3] = p->reps[2];
- }
- p->reps[2] = p->reps[1];
- }
- p->reps[1] = p->reps[0];
- p->reps[0] = dist;
- }
-
- RC_NORM(&p->rc)
-
- p->rc.range = range;
-
- if (len != 1)
- {
- LenEnc_Encode(&p->repLenProbs, &p->rc, len - LZMA_MATCH_LEN_MIN, posState);
- --p->repLenEncCounter;
- p->state = kRepNextStates[p->state];
- }
- }
- else
- {
- unsigned posSlot;
- RC_BIT_0(&p->rc, probs);
- p->rc.range = range;
- p->state = kMatchNextStates[p->state];
-
- LenEnc_Encode(&p->lenProbs, &p->rc, len - LZMA_MATCH_LEN_MIN, posState);
- // --p->lenEnc.counter;
-
- dist -= LZMA_NUM_REPS;
- p->reps[3] = p->reps[2];
- p->reps[2] = p->reps[1];
- p->reps[1] = p->reps[0];
- p->reps[0] = dist + 1;
-
- p->matchPriceCount++;
- GetPosSlot(dist, posSlot);
- // RcTree_Encode_PosSlot(&p->rc, p->posSlotEncoder[GetLenToPosState(len)], posSlot);
- {
- UInt32 sym = (UInt32)posSlot + (1 << kNumPosSlotBits);
- range = p->rc.range;
- probs = p->posSlotEncoder[GetLenToPosState(len)];
- do
- {
- CLzmaProb *prob = probs + (sym >> kNumPosSlotBits);
- UInt32 bit = (sym >> (kNumPosSlotBits - 1)) & 1;
- sym <<= 1;
- RC_BIT(&p->rc, prob, bit);
- }
- while (sym < (1 << kNumPosSlotBits * 2));
- p->rc.range = range;
- }
-
- if (dist >= kStartPosModelIndex)
- {
- unsigned footerBits = ((posSlot >> 1) - 1);
-
- if (dist < kNumFullDistances)
- {
- unsigned base = ((2 | (posSlot & 1)) << footerBits);
- RcTree_ReverseEncode(&p->rc, p->posEncoders + base, footerBits, (unsigned)(dist /* - base */));
- }
- else
- {
- UInt32 pos2 = (dist | 0xF) << (32 - footerBits);
- range = p->rc.range;
- // RangeEnc_EncodeDirectBits(&p->rc, posReduced >> kNumAlignBits, footerBits - kNumAlignBits);
- /*
- do
- {
- range >>= 1;
- p->rc.low += range & (0 - ((dist >> --footerBits) & 1));
- RC_NORM(&p->rc)
- }
- while (footerBits > kNumAlignBits);
- */
- do
- {
- range >>= 1;
- p->rc.low += range & (0 - (pos2 >> 31));
- pos2 += pos2;
- RC_NORM(&p->rc)
- }
- while (pos2 != 0xF0000000);
-
-
- // RcTree_ReverseEncode(&p->rc, p->posAlignEncoder, kNumAlignBits, posReduced & kAlignMask);
-
- {
- unsigned m = 1;
- unsigned bit;
- bit = dist & 1; dist >>= 1; RC_BIT(&p->rc, p->posAlignEncoder + m, bit); m = (m << 1) + bit;
- bit = dist & 1; dist >>= 1; RC_BIT(&p->rc, p->posAlignEncoder + m, bit); m = (m << 1) + bit;
- bit = dist & 1; dist >>= 1; RC_BIT(&p->rc, p->posAlignEncoder + m, bit); m = (m << 1) + bit;
- bit = dist & 1; RC_BIT(&p->rc, p->posAlignEncoder + m, bit);
- p->rc.range = range;
- // p->alignPriceCount++;
- }
- }
- }
- }
- }
-
- nowPos32 += (UInt32)len;
- p->additionalOffset -= len;
-
- if (p->additionalOffset == 0)
- {
- UInt32 processed;
-
- if (!p->fastMode)
- {
- /*
- if (p->alignPriceCount >= 16) // kAlignTableSize
- FillAlignPrices(p);
- if (p->matchPriceCount >= 128)
- FillDistancesPrices(p);
- if (p->lenEnc.counter <= 0)
- LenPriceEnc_UpdateTables(&p->lenEnc, 1 << p->pb, &p->lenProbs, p->ProbPrices);
- */
- if (p->matchPriceCount >= 64)
- {
- FillAlignPrices(p);
- // { int y; for (y = 0; y < 100; y++) {
- FillDistancesPrices(p);
- // }}
- LenPriceEnc_UpdateTables(&p->lenEnc, 1 << p->pb, &p->lenProbs, p->ProbPrices);
- }
- if (p->repLenEncCounter <= 0)
- {
- p->repLenEncCounter = REP_LEN_COUNT;
- LenPriceEnc_UpdateTables(&p->repLenEnc, 1 << p->pb, &p->repLenProbs, p->ProbPrices);
- }
- }
-
- if (p->matchFinder.GetNumAvailableBytes(p->matchFinderObj) == 0)
- break;
- processed = nowPos32 - startPos32;
-
- if (maxPackSize)
- {
- if (processed + kNumOpts + 300 >= maxUnpackSize
- || RangeEnc_GetProcessed_sizet(&p->rc) + kPackReserve >= maxPackSize)
- break;
- }
- else if (processed >= (1 << 17))
- {
- p->nowPos64 += nowPos32 - startPos32;
- return CheckErrors(p);
- }
- }
- }
-
- p->nowPos64 += nowPos32 - startPos32;
- return Flush(p, nowPos32);
-}
-
-
-
-#define kBigHashDicLimit ((UInt32)1 << 24)
-
-static SRes LzmaEnc_Alloc(CLzmaEnc *p, UInt32 keepWindowSize, ISzAllocPtr alloc, ISzAllocPtr allocBig)
-{
- UInt32 beforeSize = kNumOpts;
- if (!RangeEnc_Alloc(&p->rc, alloc))
- return SZ_ERROR_MEM;
-
- #ifndef _7ZIP_ST
- p->mtMode = (p->multiThread && !p->fastMode && (p->matchFinderBase.btMode != 0));
- #endif
-
- {
- unsigned lclp = p->lc + p->lp;
- if (!p->litProbs || !p->saveState.litProbs || p->lclp != lclp)
- {
- LzmaEnc_FreeLits(p, alloc);
- p->litProbs = (CLzmaProb *)ISzAlloc_Alloc(alloc, ((UInt32)0x300 << lclp) * sizeof(CLzmaProb));
- p->saveState.litProbs = (CLzmaProb *)ISzAlloc_Alloc(alloc, ((UInt32)0x300 << lclp) * sizeof(CLzmaProb));
- if (!p->litProbs || !p->saveState.litProbs)
- {
- LzmaEnc_FreeLits(p, alloc);
- return SZ_ERROR_MEM;
- }
- p->lclp = lclp;
- }
- }
-
- p->matchFinderBase.bigHash = (Byte)(p->dictSize > kBigHashDicLimit ? 1 : 0);
-
- if (beforeSize + p->dictSize < keepWindowSize)
- beforeSize = keepWindowSize - p->dictSize;
-
- #ifndef _7ZIP_ST
- if (p->mtMode)
- {
- RINOK(MatchFinderMt_Create(&p->matchFinderMt, p->dictSize, beforeSize, p->numFastBytes,
- LZMA_MATCH_LEN_MAX
- + 1 /* 18.04 */
- , allocBig));
- p->matchFinderObj = &p->matchFinderMt;
- p->matchFinderBase.bigHash = (Byte)(
- (p->dictSize > kBigHashDicLimit && p->matchFinderBase.hashMask >= 0xFFFFFF) ? 1 : 0);
- MatchFinderMt_CreateVTable(&p->matchFinderMt, &p->matchFinder);
- }
- else
- #endif
- {
- if (!MatchFinder_Create(&p->matchFinderBase, p->dictSize, beforeSize, p->numFastBytes, LZMA_MATCH_LEN_MAX, allocBig))
- return SZ_ERROR_MEM;
- p->matchFinderObj = &p->matchFinderBase;
- MatchFinder_CreateVTable(&p->matchFinderBase, &p->matchFinder);
- }
-
- return SZ_OK;
-}
-
-void LzmaEnc_Init(CLzmaEnc *p)
-{
- unsigned i;
- p->state = 0;
- p->reps[0] =
- p->reps[1] =
- p->reps[2] =
- p->reps[3] = 1;
-
- RangeEnc_Init(&p->rc);
-
- for (i = 0; i < (1 << kNumAlignBits); i++)
- p->posAlignEncoder[i] = kProbInitValue;
-
- for (i = 0; i < kNumStates; i++)
- {
- unsigned j;
- for (j = 0; j < LZMA_NUM_PB_STATES_MAX; j++)
- {
- p->isMatch[i][j] = kProbInitValue;
- p->isRep0Long[i][j] = kProbInitValue;
- }
- p->isRep[i] = kProbInitValue;
- p->isRepG0[i] = kProbInitValue;
- p->isRepG1[i] = kProbInitValue;
- p->isRepG2[i] = kProbInitValue;
- }
-
- {
- for (i = 0; i < kNumLenToPosStates; i++)
- {
- CLzmaProb *probs = p->posSlotEncoder[i];
- unsigned j;
- for (j = 0; j < (1 << kNumPosSlotBits); j++)
- probs[j] = kProbInitValue;
- }
- }
- {
- for (i = 0; i < kNumFullDistances; i++)
- p->posEncoders[i] = kProbInitValue;
- }
-
- {
- UInt32 num = (UInt32)0x300 << (p->lp + p->lc);
- UInt32 k;
- CLzmaProb *probs = p->litProbs;
- for (k = 0; k < num; k++)
- probs[k] = kProbInitValue;
- }
-
-
- LenEnc_Init(&p->lenProbs);
- LenEnc_Init(&p->repLenProbs);
-
- p->optEnd = 0;
- p->optCur = 0;
-
- {
- for (i = 0; i < kNumOpts; i++)
- p->opt[i].price = kInfinityPrice;
- }
-
- p->additionalOffset = 0;
-
- p->pbMask = (1 << p->pb) - 1;
- p->lpMask = ((UInt32)0x100 << p->lp) - ((unsigned)0x100 >> p->lc);
-}
-
-
-void LzmaEnc_InitPrices(CLzmaEnc *p)
-{
- if (!p->fastMode)
- {
- FillDistancesPrices(p);
- FillAlignPrices(p);
- }
-
- p->lenEnc.tableSize =
- p->repLenEnc.tableSize =
- p->numFastBytes + 1 - LZMA_MATCH_LEN_MIN;
-
- p->repLenEncCounter = REP_LEN_COUNT;
-
- LenPriceEnc_UpdateTables(&p->lenEnc, 1 << p->pb, &p->lenProbs, p->ProbPrices);
- LenPriceEnc_UpdateTables(&p->repLenEnc, 1 << p->pb, &p->repLenProbs, p->ProbPrices);
-}
-
-static SRes LzmaEnc_AllocAndInit(CLzmaEnc *p, UInt32 keepWindowSize, ISzAllocPtr alloc, ISzAllocPtr allocBig)
-{
- unsigned i;
- for (i = kEndPosModelIndex / 2; i < kDicLogSizeMax; i++)
- if (p->dictSize <= ((UInt32)1 << i))
- break;
- p->distTableSize = i * 2;
-
- p->finished = False;
- p->result = SZ_OK;
- RINOK(LzmaEnc_Alloc(p, keepWindowSize, alloc, allocBig));
- LzmaEnc_Init(p);
- LzmaEnc_InitPrices(p);
- p->nowPos64 = 0;
- return SZ_OK;
-}
-
-SRes LzmaEnc_Prepare(CLzmaEncHandle pp, ISeqOutStream *outStream, ISeqInStream *inStream,
- ISzAllocPtr alloc, ISzAllocPtr allocBig)
-{
- CLzmaEnc *p = (CLzmaEnc *)pp;
- p->matchFinderBase.stream = inStream;
- p->needInit = 1;
- p->rc.outStream = outStream;
- return LzmaEnc_AllocAndInit(p, 0, alloc, allocBig);
-}
-
-SRes LzmaEnc_PrepareForLzma2(CLzmaEncHandle pp,
- ISeqInStream *inStream, UInt32 keepWindowSize,
- ISzAllocPtr alloc, ISzAllocPtr allocBig)
-{
- CLzmaEnc *p = (CLzmaEnc *)pp;
- p->matchFinderBase.stream = inStream;
- p->needInit = 1;
- return LzmaEnc_AllocAndInit(p, keepWindowSize, alloc, allocBig);
-}
-
-static void LzmaEnc_SetInputBuf(CLzmaEnc *p, const Byte *src, SizeT srcLen)
-{
- p->matchFinderBase.directInput = 1;
- p->matchFinderBase.bufferBase = (Byte *)src;
- p->matchFinderBase.directInputRem = srcLen;
-}
-
-SRes LzmaEnc_MemPrepare(CLzmaEncHandle pp, const Byte *src, SizeT srcLen,
- UInt32 keepWindowSize, ISzAllocPtr alloc, ISzAllocPtr allocBig)
-{
- CLzmaEnc *p = (CLzmaEnc *)pp;
- LzmaEnc_SetInputBuf(p, src, srcLen);
- p->needInit = 1;
-
- LzmaEnc_SetDataSize(pp, srcLen);
- return LzmaEnc_AllocAndInit(p, keepWindowSize, alloc, allocBig);
-}
-
-void LzmaEnc_Finish(CLzmaEncHandle pp)
-{
- #ifndef _7ZIP_ST
- CLzmaEnc *p = (CLzmaEnc *)pp;
- if (p->mtMode)
- MatchFinderMt_ReleaseStream(&p->matchFinderMt);
- #else
- UNUSED_VAR(pp);
- #endif
-}
-
-
-typedef struct
-{
- ISeqOutStream vt;
- Byte *data;
- SizeT rem;
- BoolInt overflow;
-} CLzmaEnc_SeqOutStreamBuf;
-
-static size_t SeqOutStreamBuf_Write(const ISeqOutStream *pp, const void *data, size_t size)
-{
- CLzmaEnc_SeqOutStreamBuf *p = CONTAINER_FROM_VTBL(pp, CLzmaEnc_SeqOutStreamBuf, vt);
- if (p->rem < size)
- {
- size = p->rem;
- p->overflow = True;
- }
- memcpy(p->data, data, size);
- p->rem -= size;
- p->data += size;
- return size;
-}
-
-
-UInt32 LzmaEnc_GetNumAvailableBytes(CLzmaEncHandle pp)
-{
- const CLzmaEnc *p = (CLzmaEnc *)pp;
- return p->matchFinder.GetNumAvailableBytes(p->matchFinderObj);
-}
-
-
-const Byte *LzmaEnc_GetCurBuf(CLzmaEncHandle pp)
-{
- const CLzmaEnc *p = (CLzmaEnc *)pp;
- return p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - p->additionalOffset;
-}
-
-
-SRes LzmaEnc_CodeOneMemBlock(CLzmaEncHandle pp, BoolInt reInit,
- Byte *dest, size_t *destLen, UInt32 desiredPackSize, UInt32 *unpackSize)
-{
- CLzmaEnc *p = (CLzmaEnc *)pp;
- UInt64 nowPos64;
- SRes res;
- CLzmaEnc_SeqOutStreamBuf outStream;
-
- outStream.vt.Write = SeqOutStreamBuf_Write;
- outStream.data = dest;
- outStream.rem = *destLen;
- outStream.overflow = False;
-
- p->writeEndMark = False;
- p->finished = False;
- p->result = SZ_OK;
-
- if (reInit)
- LzmaEnc_Init(p);
- LzmaEnc_InitPrices(p);
-
- nowPos64 = p->nowPos64;
- RangeEnc_Init(&p->rc);
- p->rc.outStream = &outStream.vt;
-
- if (desiredPackSize == 0)
- return SZ_ERROR_OUTPUT_EOF;
-
- res = LzmaEnc_CodeOneBlock(p, desiredPackSize, *unpackSize);
-
- *unpackSize = (UInt32)(p->nowPos64 - nowPos64);
- *destLen -= outStream.rem;
- if (outStream.overflow)
- return SZ_ERROR_OUTPUT_EOF;
-
- return res;
-}
-
-
-static SRes LzmaEnc_Encode2(CLzmaEnc *p, ICompressProgress *progress)
-{
- SRes res = SZ_OK;
-
- #ifndef _7ZIP_ST
- Byte allocaDummy[0x300];
- allocaDummy[0] = 0;
- allocaDummy[1] = allocaDummy[0];
- #endif
-
- for (;;)
- {
- res = LzmaEnc_CodeOneBlock(p, 0, 0);
- if (res != SZ_OK || p->finished)
- break;
- if (progress)
- {
- res = ICompressProgress_Progress(progress, p->nowPos64, RangeEnc_GetProcessed(&p->rc));
- if (res != SZ_OK)
- {
- res = SZ_ERROR_PROGRESS;
- break;
- }
- }
- }
-
- LzmaEnc_Finish(p);
-
- /*
- if (res == SZ_OK && !Inline_MatchFinder_IsFinishedOK(&p->matchFinderBase))
- res = SZ_ERROR_FAIL;
- }
- */
-
- return res;
-}
-
-
-SRes LzmaEnc_Encode(CLzmaEncHandle pp, ISeqOutStream *outStream, ISeqInStream *inStream, ICompressProgress *progress,
- ISzAllocPtr alloc, ISzAllocPtr allocBig)
-{
- RINOK(LzmaEnc_Prepare(pp, outStream, inStream, alloc, allocBig));
- return LzmaEnc_Encode2((CLzmaEnc *)pp, progress);
-}
-
-
-SRes LzmaEnc_WriteProperties(CLzmaEncHandle pp, Byte *props, SizeT *size)
-{
- CLzmaEnc *p = (CLzmaEnc *)pp;
- unsigned i;
- UInt32 dictSize = p->dictSize;
- if (*size < LZMA_PROPS_SIZE)
- return SZ_ERROR_PARAM;
- *size = LZMA_PROPS_SIZE;
- props[0] = (Byte)((p->pb * 5 + p->lp) * 9 + p->lc);
-
- if (dictSize >= ((UInt32)1 << 22))
- {
- UInt32 kDictMask = ((UInt32)1 << 20) - 1;
- if (dictSize < (UInt32)0xFFFFFFFF - kDictMask)
- dictSize = (dictSize + kDictMask) & ~kDictMask;
- }
- else for (i = 11; i <= 30; i++)
- {
- if (dictSize <= ((UInt32)2 << i)) { dictSize = (2 << i); break; }
- if (dictSize <= ((UInt32)3 << i)) { dictSize = (3 << i); break; }
- }
-
- for (i = 0; i < 4; i++)
- props[1 + i] = (Byte)(dictSize >> (8 * i));
- return SZ_OK;
-}
-
-
-unsigned LzmaEnc_IsWriteEndMark(CLzmaEncHandle pp)
-{
- return ((CLzmaEnc *)pp)->writeEndMark;
-}
-
-
-SRes LzmaEnc_MemEncode(CLzmaEncHandle pp, Byte *dest, SizeT *destLen, const Byte *src, SizeT srcLen,
- int writeEndMark, ICompressProgress *progress, ISzAllocPtr alloc, ISzAllocPtr allocBig)
-{
- SRes res;
- CLzmaEnc *p = (CLzmaEnc *)pp;
-
- CLzmaEnc_SeqOutStreamBuf outStream;
-
- outStream.vt.Write = SeqOutStreamBuf_Write;
- outStream.data = dest;
- outStream.rem = *destLen;
- outStream.overflow = False;
-
- p->writeEndMark = writeEndMark;
- p->rc.outStream = &outStream.vt;
-
- res = LzmaEnc_MemPrepare(pp, src, srcLen, 0, alloc, allocBig);
-
- if (res == SZ_OK)
- {
- res = LzmaEnc_Encode2(p, progress);
- if (res == SZ_OK && p->nowPos64 != srcLen)
- res = SZ_ERROR_FAIL;
- }
-
- *destLen -= outStream.rem;
- if (outStream.overflow)
- return SZ_ERROR_OUTPUT_EOF;
- return res;
-}
-
-
-SRes LzmaEncode(Byte *dest, SizeT *destLen, const Byte *src, SizeT srcLen,
- const CLzmaEncProps *props, Byte *propsEncoded, SizeT *propsSize, int writeEndMark,
- ICompressProgress *progress, ISzAllocPtr alloc, ISzAllocPtr allocBig)
-{
- CLzmaEnc *p = (CLzmaEnc *)LzmaEnc_Create(alloc);
- SRes res;
- if (!p)
- return SZ_ERROR_MEM;
-
- res = LzmaEnc_SetProps(p, props);
- if (res == SZ_OK)
- {
- res = LzmaEnc_WriteProperties(p, propsEncoded, propsSize);
- if (res == SZ_OK)
- res = LzmaEnc_MemEncode(p, dest, destLen, src, srcLen,
- writeEndMark, progress, alloc, allocBig);
- }
-
- LzmaEnc_Destroy(p, alloc, allocBig);
- return res;
-}
-
-BoolInt LzmaEnc_IsFinished(CLzmaEncHandle pp)
-{
- CLzmaEnc *p = (CLzmaEnc *)pp;
- return p->finished;
-}
-
\ No newline at end of file +/* LzmaEnc.c -- LZMA Encoder +2019-01-10: Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include <string.h> + +/* #define SHOW_STAT */ +/* #define SHOW_STAT2 */ + +#if defined(SHOW_STAT) || defined(SHOW_STAT2) +#include <stdio.h> +#endif + +#include "LzmaEnc.h" + +#include "LzFind.h" +#ifndef _7ZIP_ST +#include "LzFindMt.h" +#endif + +#ifdef SHOW_STAT +static unsigned g_STAT_OFFSET = 0; +#endif + +#define kLzmaMaxHistorySize ((UInt32)3 << 29) +/* #define kLzmaMaxHistorySize ((UInt32)7 << 29) */ + +#define kNumTopBits 24 +#define kTopValue ((UInt32)1 << kNumTopBits) + +#define kNumBitModelTotalBits 11 +#define kBitModelTotal (1 << kNumBitModelTotalBits) +#define kNumMoveBits 5 +#define kProbInitValue (kBitModelTotal >> 1) + +#define kNumMoveReducingBits 4 +#define kNumBitPriceShiftBits 4 +#define kBitPrice (1 << kNumBitPriceShiftBits) + +#define REP_LEN_COUNT 64 + +void LzmaEncProps_Init(CLzmaEncProps *p) +{ + p->level = 5; + p->dictSize = p->mc = 0; + p->reduceSize = (UInt64)(Int64)-1; + p->lc = p->lp = p->pb = p->algo = p->fb = p->btMode = p->numHashBytes = p->numThreads = -1; + p->writeEndMark = 0; +} + +void LzmaEncProps_Normalize(CLzmaEncProps *p) +{ + int level = p->level; + if (level < 0) level = 5; + p->level = level; + + if (p->dictSize == 0) p->dictSize = (level <= 5 ? (1 << (level * 2 + 14)) : (level <= 7 ? (1 << 25) : (1 << 26))); + if (p->dictSize > p->reduceSize) + { + unsigned i; + UInt32 reduceSize = (UInt32)p->reduceSize; + for (i = 11; i <= 30; i++) + { + if (reduceSize <= ((UInt32)2 << i)) { p->dictSize = ((UInt32)2 << i); break; } + if (reduceSize <= ((UInt32)3 << i)) { p->dictSize = ((UInt32)3 << i); break; } + } + } + + if (p->lc < 0) p->lc = 3; + if (p->lp < 0) p->lp = 0; + if (p->pb < 0) p->pb = 2; + + if (p->algo < 0) p->algo = (level < 5 ? 0 : 1); + if (p->fb < 0) p->fb = (level < 7 ? 32 : 64); + if (p->btMode < 0) p->btMode = (p->algo == 0 ? 0 : 1); + if (p->numHashBytes < 0) p->numHashBytes = 4; + if (p->mc == 0) p->mc = (16 + (p->fb >> 1)) >> (p->btMode ? 0 : 1); + + if (p->numThreads < 0) + p->numThreads = + #ifndef _7ZIP_ST + ((p->btMode && p->algo) ? 2 : 1); + #else + 1; + #endif +} + +UInt32 LzmaEncProps_GetDictSize(const CLzmaEncProps *props2) +{ + CLzmaEncProps props = *props2; + LzmaEncProps_Normalize(&props); + return props.dictSize; +} + +#if (_MSC_VER >= 1400) +/* BSR code is fast for some new CPUs */ +/* #define LZMA_LOG_BSR */ +#endif + +#ifdef LZMA_LOG_BSR + +#define kDicLogSizeMaxCompress 32 + +#define BSR2_RET(pos, res) { unsigned long zz; _BitScanReverse(&zz, (pos)); res = (zz + zz) + ((pos >> (zz - 1)) & 1); } + +static unsigned GetPosSlot1(UInt32 pos) +{ + unsigned res; + BSR2_RET(pos, res); + return res; +} +#define GetPosSlot2(pos, res) { BSR2_RET(pos, res); } +#define GetPosSlot(pos, res) { if (pos < 2) res = pos; else BSR2_RET(pos, res); } + +#else + +#define kNumLogBits (9 + sizeof(size_t) / 2) +/* #define kNumLogBits (11 + sizeof(size_t) / 8 * 3) */ + +#define kDicLogSizeMaxCompress ((kNumLogBits - 1) * 2 + 7) + +static void LzmaEnc_FastPosInit(Byte *g_FastPos) +{ + unsigned slot; + g_FastPos[0] = 0; + g_FastPos[1] = 1; + g_FastPos += 2; + + for (slot = 2; slot < kNumLogBits * 2; slot++) + { + size_t k = ((size_t)1 << ((slot >> 1) - 1)); + size_t j; + for (j = 0; j < k; j++) + g_FastPos[j] = (Byte)slot; + g_FastPos += k; + } +} + +/* we can use ((limit - pos) >> 31) only if (pos < ((UInt32)1 << 31)) */ +/* +#define BSR2_RET(pos, res) { unsigned zz = 6 + ((kNumLogBits - 1) & \ + (0 - (((((UInt32)1 << (kNumLogBits + 6)) - 1) - pos) >> 31))); \ + res = p->g_FastPos[pos >> zz] + (zz * 2); } +*/ + +/* +#define BSR2_RET(pos, res) { unsigned zz = 6 + ((kNumLogBits - 1) & \ + (0 - (((((UInt32)1 << (kNumLogBits)) - 1) - (pos >> 6)) >> 31))); \ + res = p->g_FastPos[pos >> zz] + (zz * 2); } +*/ + +#define BSR2_RET(pos, res) { unsigned zz = (pos < (1 << (kNumLogBits + 6))) ? 6 : 6 + kNumLogBits - 1; \ + res = p->g_FastPos[pos >> zz] + (zz * 2); } + +/* +#define BSR2_RET(pos, res) { res = (pos < (1 << (kNumLogBits + 6))) ? \ + p->g_FastPos[pos >> 6] + 12 : \ + p->g_FastPos[pos >> (6 + kNumLogBits - 1)] + (6 + (kNumLogBits - 1)) * 2; } +*/ + +#define GetPosSlot1(pos) p->g_FastPos[pos] +#define GetPosSlot2(pos, res) { BSR2_RET(pos, res); } +#define GetPosSlot(pos, res) { if (pos < kNumFullDistances) res = p->g_FastPos[pos & (kNumFullDistances - 1)]; else BSR2_RET(pos, res); } + +#endif + + +#define LZMA_NUM_REPS 4 + +typedef UInt16 CState; +typedef UInt16 CExtra; + +typedef struct +{ + UInt32 price; + CState state; + CExtra extra; + // 0 : normal + // 1 : LIT : MATCH + // > 1 : MATCH (extra-1) : LIT : REP0 (len) + UInt32 len; + UInt32 dist; + UInt32 reps[LZMA_NUM_REPS]; +} COptimal; + + +// 18.06 +#define kNumOpts (1 << 11) +#define kPackReserve (kNumOpts * 8) +// #define kNumOpts (1 << 12) +// #define kPackReserve (1 + kNumOpts * 2) + +#define kNumLenToPosStates 4 +#define kNumPosSlotBits 6 +#define kDicLogSizeMin 0 +#define kDicLogSizeMax 32 +#define kDistTableSizeMax (kDicLogSizeMax * 2) + +#define kNumAlignBits 4 +#define kAlignTableSize (1 << kNumAlignBits) +#define kAlignMask (kAlignTableSize - 1) + +#define kStartPosModelIndex 4 +#define kEndPosModelIndex 14 +#define kNumFullDistances (1 << (kEndPosModelIndex >> 1)) + +typedef +#ifdef _LZMA_PROB32 + UInt32 +#else + UInt16 +#endif + CLzmaProb; + +#define LZMA_PB_MAX 4 +#define LZMA_LC_MAX 8 +#define LZMA_LP_MAX 4 + +#define LZMA_NUM_PB_STATES_MAX (1 << LZMA_PB_MAX) + +#define kLenNumLowBits 3 +#define kLenNumLowSymbols (1 << kLenNumLowBits) +#define kLenNumHighBits 8 +#define kLenNumHighSymbols (1 << kLenNumHighBits) +#define kLenNumSymbolsTotal (kLenNumLowSymbols * 2 + kLenNumHighSymbols) + +#define LZMA_MATCH_LEN_MIN 2 +#define LZMA_MATCH_LEN_MAX (LZMA_MATCH_LEN_MIN + kLenNumSymbolsTotal - 1) + +#define kNumStates 12 + + +typedef struct +{ + CLzmaProb low[LZMA_NUM_PB_STATES_MAX << (kLenNumLowBits + 1)]; + CLzmaProb high[kLenNumHighSymbols]; +} CLenEnc; + + +typedef struct +{ + unsigned tableSize; + UInt32 prices[LZMA_NUM_PB_STATES_MAX][kLenNumSymbolsTotal]; + // UInt32 prices1[LZMA_NUM_PB_STATES_MAX][kLenNumLowSymbols * 2]; + // UInt32 prices2[kLenNumSymbolsTotal]; +} CLenPriceEnc; + +#define GET_PRICE_LEN(p, posState, len) \ + ((p)->prices[posState][(size_t)(len) - LZMA_MATCH_LEN_MIN]) + +/* +#define GET_PRICE_LEN(p, posState, len) \ + ((p)->prices2[(size_t)(len) - 2] + ((p)->prices1[posState][((len) - 2) & (kLenNumLowSymbols * 2 - 1)] & (((len) - 2 - kLenNumLowSymbols * 2) >> 9))) +*/ + +typedef struct +{ + UInt32 range; + unsigned cache; + UInt64 low; + UInt64 cacheSize; + Byte *buf; + Byte *bufLim; + Byte *bufBase; + ISeqOutStream *outStream; + UInt64 processed; + SRes res; +} CRangeEnc; + + +typedef struct +{ + CLzmaProb *litProbs; + + unsigned state; + UInt32 reps[LZMA_NUM_REPS]; + + CLzmaProb posAlignEncoder[1 << kNumAlignBits]; + CLzmaProb isRep[kNumStates]; + CLzmaProb isRepG0[kNumStates]; + CLzmaProb isRepG1[kNumStates]; + CLzmaProb isRepG2[kNumStates]; + CLzmaProb isMatch[kNumStates][LZMA_NUM_PB_STATES_MAX]; + CLzmaProb isRep0Long[kNumStates][LZMA_NUM_PB_STATES_MAX]; + + CLzmaProb posSlotEncoder[kNumLenToPosStates][1 << kNumPosSlotBits]; + CLzmaProb posEncoders[kNumFullDistances]; + + CLenEnc lenProbs; + CLenEnc repLenProbs; + +} CSaveState; + + +typedef UInt32 CProbPrice; + + +typedef struct +{ + void *matchFinderObj; + IMatchFinder matchFinder; + + unsigned optCur; + unsigned optEnd; + + unsigned longestMatchLen; + unsigned numPairs; + UInt32 numAvail; + + unsigned state; + unsigned numFastBytes; + unsigned additionalOffset; + UInt32 reps[LZMA_NUM_REPS]; + unsigned lpMask, pbMask; + CLzmaProb *litProbs; + CRangeEnc rc; + + UInt32 backRes; + + unsigned lc, lp, pb; + unsigned lclp; + + BoolInt fastMode; + BoolInt writeEndMark; + BoolInt finished; + BoolInt multiThread; + BoolInt needInit; + // BoolInt _maxMode; + + UInt64 nowPos64; + + unsigned matchPriceCount; + // unsigned alignPriceCount; + int repLenEncCounter; + + unsigned distTableSize; + + UInt32 dictSize; + SRes result; + + #ifndef _7ZIP_ST + BoolInt mtMode; + // begin of CMatchFinderMt is used in LZ thread + CMatchFinderMt matchFinderMt; + // end of CMatchFinderMt is used in BT and HASH threads + #endif + + CMatchFinder matchFinderBase; + + #ifndef _7ZIP_ST + Byte pad[128]; + #endif + + // LZ thread + CProbPrice ProbPrices[kBitModelTotal >> kNumMoveReducingBits]; + + UInt32 matches[LZMA_MATCH_LEN_MAX * 2 + 2 + 1]; + + UInt32 alignPrices[kAlignTableSize]; + UInt32 posSlotPrices[kNumLenToPosStates][kDistTableSizeMax]; + UInt32 distancesPrices[kNumLenToPosStates][kNumFullDistances]; + + CLzmaProb posAlignEncoder[1 << kNumAlignBits]; + CLzmaProb isRep[kNumStates]; + CLzmaProb isRepG0[kNumStates]; + CLzmaProb isRepG1[kNumStates]; + CLzmaProb isRepG2[kNumStates]; + CLzmaProb isMatch[kNumStates][LZMA_NUM_PB_STATES_MAX]; + CLzmaProb isRep0Long[kNumStates][LZMA_NUM_PB_STATES_MAX]; + CLzmaProb posSlotEncoder[kNumLenToPosStates][1 << kNumPosSlotBits]; + CLzmaProb posEncoders[kNumFullDistances]; + + CLenEnc lenProbs; + CLenEnc repLenProbs; + + #ifndef LZMA_LOG_BSR + Byte g_FastPos[1 << kNumLogBits]; + #endif + + CLenPriceEnc lenEnc; + CLenPriceEnc repLenEnc; + + COptimal opt[kNumOpts]; + + CSaveState saveState; + + #ifndef _7ZIP_ST + Byte pad2[128]; + #endif +} CLzmaEnc; + + + +#define COPY_ARR(dest, src, arr) memcpy(dest->arr, src->arr, sizeof(src->arr)); + +void LzmaEnc_SaveState(CLzmaEncHandle pp) +{ + CLzmaEnc *p = (CLzmaEnc *)pp; + CSaveState *dest = &p->saveState; + + dest->state = p->state; + + dest->lenProbs = p->lenProbs; + dest->repLenProbs = p->repLenProbs; + + COPY_ARR(dest, p, reps); + + COPY_ARR(dest, p, posAlignEncoder); + COPY_ARR(dest, p, isRep); + COPY_ARR(dest, p, isRepG0); + COPY_ARR(dest, p, isRepG1); + COPY_ARR(dest, p, isRepG2); + COPY_ARR(dest, p, isMatch); + COPY_ARR(dest, p, isRep0Long); + COPY_ARR(dest, p, posSlotEncoder); + COPY_ARR(dest, p, posEncoders); + + memcpy(dest->litProbs, p->litProbs, ((UInt32)0x300 << p->lclp) * sizeof(CLzmaProb)); +} + + +void LzmaEnc_RestoreState(CLzmaEncHandle pp) +{ + CLzmaEnc *dest = (CLzmaEnc *)pp; + const CSaveState *p = &dest->saveState; + + dest->state = p->state; + + dest->lenProbs = p->lenProbs; + dest->repLenProbs = p->repLenProbs; + + COPY_ARR(dest, p, reps); + + COPY_ARR(dest, p, posAlignEncoder); + COPY_ARR(dest, p, isRep); + COPY_ARR(dest, p, isRepG0); + COPY_ARR(dest, p, isRepG1); + COPY_ARR(dest, p, isRepG2); + COPY_ARR(dest, p, isMatch); + COPY_ARR(dest, p, isRep0Long); + COPY_ARR(dest, p, posSlotEncoder); + COPY_ARR(dest, p, posEncoders); + + memcpy(dest->litProbs, p->litProbs, ((UInt32)0x300 << dest->lclp) * sizeof(CLzmaProb)); +} + + + +SRes LzmaEnc_SetProps(CLzmaEncHandle pp, const CLzmaEncProps *props2) +{ + CLzmaEnc *p = (CLzmaEnc *)pp; + CLzmaEncProps props = *props2; + LzmaEncProps_Normalize(&props); + + if (props.lc > LZMA_LC_MAX + || props.lp > LZMA_LP_MAX + || props.pb > LZMA_PB_MAX + || props.dictSize > ((UInt64)1 << kDicLogSizeMaxCompress) + || props.dictSize > kLzmaMaxHistorySize) + return SZ_ERROR_PARAM; + + p->dictSize = props.dictSize; + { + unsigned fb = props.fb; + if (fb < 5) + fb = 5; + if (fb > LZMA_MATCH_LEN_MAX) + fb = LZMA_MATCH_LEN_MAX; + p->numFastBytes = fb; + } + p->lc = props.lc; + p->lp = props.lp; + p->pb = props.pb; + p->fastMode = (props.algo == 0); + // p->_maxMode = True; + p->matchFinderBase.btMode = (Byte)(props.btMode ? 1 : 0); + { + unsigned numHashBytes = 4; + if (props.btMode) + { + if (props.numHashBytes < 2) + numHashBytes = 2; + else if (props.numHashBytes < 4) + numHashBytes = props.numHashBytes; + } + p->matchFinderBase.numHashBytes = numHashBytes; + } + + p->matchFinderBase.cutValue = props.mc; + + p->writeEndMark = props.writeEndMark; + + #ifndef _7ZIP_ST + /* + if (newMultiThread != _multiThread) + { + ReleaseMatchFinder(); + _multiThread = newMultiThread; + } + */ + p->multiThread = (props.numThreads > 1); + #endif + + return SZ_OK; +} + + +void LzmaEnc_SetDataSize(CLzmaEncHandle pp, UInt64 expectedDataSiize) +{ + CLzmaEnc *p = (CLzmaEnc *)pp; + p->matchFinderBase.expectedDataSize = expectedDataSiize; +} + + +#define kState_Start 0 +#define kState_LitAfterMatch 4 +#define kState_LitAfterRep 5 +#define kState_MatchAfterLit 7 +#define kState_RepAfterLit 8 + +static const Byte kLiteralNextStates[kNumStates] = {0, 0, 0, 0, 1, 2, 3, 4, 5, 6, 4, 5}; +static const Byte kMatchNextStates[kNumStates] = {7, 7, 7, 7, 7, 7, 7, 10, 10, 10, 10, 10}; +static const Byte kRepNextStates[kNumStates] = {8, 8, 8, 8, 8, 8, 8, 11, 11, 11, 11, 11}; +static const Byte kShortRepNextStates[kNumStates]= {9, 9, 9, 9, 9, 9, 9, 11, 11, 11, 11, 11}; + +#define IsLitState(s) ((s) < 7) +#define GetLenToPosState2(len) (((len) < kNumLenToPosStates - 1) ? (len) : kNumLenToPosStates - 1) +#define GetLenToPosState(len) (((len) < kNumLenToPosStates + 1) ? (len) - 2 : kNumLenToPosStates - 1) + +#define kInfinityPrice (1 << 30) + +static void RangeEnc_Construct(CRangeEnc *p) +{ + p->outStream = NULL; + p->bufBase = NULL; +} + +#define RangeEnc_GetProcessed(p) ((p)->processed + ((p)->buf - (p)->bufBase) + (p)->cacheSize) +#define RangeEnc_GetProcessed_sizet(p) ((size_t)(p)->processed + ((p)->buf - (p)->bufBase) + (size_t)(p)->cacheSize) + +#define RC_BUF_SIZE (1 << 16) + +static int RangeEnc_Alloc(CRangeEnc *p, ISzAllocPtr alloc) +{ + if (!p->bufBase) + { + p->bufBase = (Byte *)ISzAlloc_Alloc(alloc, RC_BUF_SIZE); + if (!p->bufBase) + return 0; + p->bufLim = p->bufBase + RC_BUF_SIZE; + } + return 1; +} + +static void RangeEnc_Free(CRangeEnc *p, ISzAllocPtr alloc) +{ + ISzAlloc_Free(alloc, p->bufBase); + p->bufBase = 0; +} + +static void RangeEnc_Init(CRangeEnc *p) +{ + /* Stream.Init(); */ + p->range = 0xFFFFFFFF; + p->cache = 0; + p->low = 0; + p->cacheSize = 0; + + p->buf = p->bufBase; + + p->processed = 0; + p->res = SZ_OK; +} + +MY_NO_INLINE static void RangeEnc_FlushStream(CRangeEnc *p) +{ + size_t num; + if (p->res != SZ_OK) + return; + num = p->buf - p->bufBase; + if (num != ISeqOutStream_Write(p->outStream, p->bufBase, num)) + p->res = SZ_ERROR_WRITE; + p->processed += num; + p->buf = p->bufBase; +} + +MY_NO_INLINE static void MY_FAST_CALL RangeEnc_ShiftLow(CRangeEnc *p) +{ + UInt32 low = (UInt32)p->low; + unsigned high = (unsigned)(p->low >> 32); + p->low = (UInt32)(low << 8); + if (low < (UInt32)0xFF000000 || high != 0) + { + { + Byte *buf = p->buf; + *buf++ = (Byte)(p->cache + high); + p->cache = (unsigned)(low >> 24); + p->buf = buf; + if (buf == p->bufLim) + RangeEnc_FlushStream(p); + if (p->cacheSize == 0) + return; + } + high += 0xFF; + for (;;) + { + Byte *buf = p->buf; + *buf++ = (Byte)(high); + p->buf = buf; + if (buf == p->bufLim) + RangeEnc_FlushStream(p); + if (--p->cacheSize == 0) + return; + } + } + p->cacheSize++; +} + +static void RangeEnc_FlushData(CRangeEnc *p) +{ + int i; + for (i = 0; i < 5; i++) + RangeEnc_ShiftLow(p); +} + +#define RC_NORM(p) if (range < kTopValue) { range <<= 8; RangeEnc_ShiftLow(p); } + +#define RC_BIT_PRE(p, prob) \ + ttt = *(prob); \ + newBound = (range >> kNumBitModelTotalBits) * ttt; + +// #define _LZMA_ENC_USE_BRANCH + +#ifdef _LZMA_ENC_USE_BRANCH + +#define RC_BIT(p, prob, bit) { \ + RC_BIT_PRE(p, prob) \ + if (bit == 0) { range = newBound; ttt += (kBitModelTotal - ttt) >> kNumMoveBits; } \ + else { (p)->low += newBound; range -= newBound; ttt -= ttt >> kNumMoveBits; } \ + *(prob) = (CLzmaProb)ttt; \ + RC_NORM(p) \ + } + +#else + +#define RC_BIT(p, prob, bit) { \ + UInt32 mask; \ + RC_BIT_PRE(p, prob) \ + mask = 0 - (UInt32)bit; \ + range &= mask; \ + mask &= newBound; \ + range -= mask; \ + (p)->low += mask; \ + mask = (UInt32)bit - 1; \ + range += newBound & mask; \ + mask &= (kBitModelTotal - ((1 << kNumMoveBits) - 1)); \ + mask += ((1 << kNumMoveBits) - 1); \ + ttt += (Int32)(mask - ttt) >> kNumMoveBits; \ + *(prob) = (CLzmaProb)ttt; \ + RC_NORM(p) \ + } + +#endif + + + + +#define RC_BIT_0_BASE(p, prob) \ + range = newBound; *(prob) = (CLzmaProb)(ttt + ((kBitModelTotal - ttt) >> kNumMoveBits)); + +#define RC_BIT_1_BASE(p, prob) \ + range -= newBound; (p)->low += newBound; *(prob) = (CLzmaProb)(ttt - (ttt >> kNumMoveBits)); \ + +#define RC_BIT_0(p, prob) \ + RC_BIT_0_BASE(p, prob) \ + RC_NORM(p) + +#define RC_BIT_1(p, prob) \ + RC_BIT_1_BASE(p, prob) \ + RC_NORM(p) + +static void RangeEnc_EncodeBit_0(CRangeEnc *p, CLzmaProb *prob) +{ + UInt32 range, ttt, newBound; + range = p->range; + RC_BIT_PRE(p, prob) + RC_BIT_0(p, prob) + p->range = range; +} + +static void LitEnc_Encode(CRangeEnc *p, CLzmaProb *probs, UInt32 sym) +{ + UInt32 range = p->range; + sym |= 0x100; + do + { + UInt32 ttt, newBound; + // RangeEnc_EncodeBit(p, probs + (sym >> 8), (sym >> 7) & 1); + CLzmaProb *prob = probs + (sym >> 8); + UInt32 bit = (sym >> 7) & 1; + sym <<= 1; + RC_BIT(p, prob, bit); + } + while (sym < 0x10000); + p->range = range; +} + +static void LitEnc_EncodeMatched(CRangeEnc *p, CLzmaProb *probs, UInt32 sym, UInt32 matchByte) +{ + UInt32 range = p->range; + UInt32 offs = 0x100; + sym |= 0x100; + do + { + UInt32 ttt, newBound; + CLzmaProb *prob; + UInt32 bit; + matchByte <<= 1; + // RangeEnc_EncodeBit(p, probs + (offs + (matchByte & offs) + (sym >> 8)), (sym >> 7) & 1); + prob = probs + (offs + (matchByte & offs) + (sym >> 8)); + bit = (sym >> 7) & 1; + sym <<= 1; + offs &= ~(matchByte ^ sym); + RC_BIT(p, prob, bit); + } + while (sym < 0x10000); + p->range = range; +} + + + +static void LzmaEnc_InitPriceTables(CProbPrice *ProbPrices) +{ + UInt32 i; + for (i = 0; i < (kBitModelTotal >> kNumMoveReducingBits); i++) + { + const unsigned kCyclesBits = kNumBitPriceShiftBits; + UInt32 w = (i << kNumMoveReducingBits) + (1 << (kNumMoveReducingBits - 1)); + unsigned bitCount = 0; + unsigned j; + for (j = 0; j < kCyclesBits; j++) + { + w = w * w; + bitCount <<= 1; + while (w >= ((UInt32)1 << 16)) + { + w >>= 1; + bitCount++; + } + } + ProbPrices[i] = (CProbPrice)((kNumBitModelTotalBits << kCyclesBits) - 15 - bitCount); + // printf("\n%3d: %5d", i, ProbPrices[i]); + } +} + + +#define GET_PRICE(prob, bit) \ + p->ProbPrices[((prob) ^ (unsigned)(((-(int)(bit))) & (kBitModelTotal - 1))) >> kNumMoveReducingBits]; + +#define GET_PRICEa(prob, bit) \ + ProbPrices[((prob) ^ (unsigned)((-((int)(bit))) & (kBitModelTotal - 1))) >> kNumMoveReducingBits]; + +#define GET_PRICE_0(prob) p->ProbPrices[(prob) >> kNumMoveReducingBits] +#define GET_PRICE_1(prob) p->ProbPrices[((prob) ^ (kBitModelTotal - 1)) >> kNumMoveReducingBits] + +#define GET_PRICEa_0(prob) ProbPrices[(prob) >> kNumMoveReducingBits] +#define GET_PRICEa_1(prob) ProbPrices[((prob) ^ (kBitModelTotal - 1)) >> kNumMoveReducingBits] + + +static UInt32 LitEnc_GetPrice(const CLzmaProb *probs, UInt32 sym, const CProbPrice *ProbPrices) +{ + UInt32 price = 0; + sym |= 0x100; + do + { + unsigned bit = sym & 1; + sym >>= 1; + price += GET_PRICEa(probs[sym], bit); + } + while (sym >= 2); + return price; +} + + +static UInt32 LitEnc_Matched_GetPrice(const CLzmaProb *probs, UInt32 sym, UInt32 matchByte, const CProbPrice *ProbPrices) +{ + UInt32 price = 0; + UInt32 offs = 0x100; + sym |= 0x100; + do + { + matchByte <<= 1; + price += GET_PRICEa(probs[offs + (matchByte & offs) + (sym >> 8)], (sym >> 7) & 1); + sym <<= 1; + offs &= ~(matchByte ^ sym); + } + while (sym < 0x10000); + return price; +} + + +static void RcTree_ReverseEncode(CRangeEnc *rc, CLzmaProb *probs, unsigned numBits, unsigned sym) +{ + UInt32 range = rc->range; + unsigned m = 1; + do + { + UInt32 ttt, newBound; + unsigned bit = sym & 1; + // RangeEnc_EncodeBit(rc, probs + m, bit); + sym >>= 1; + RC_BIT(rc, probs + m, bit); + m = (m << 1) | bit; + } + while (--numBits); + rc->range = range; +} + + + +static void LenEnc_Init(CLenEnc *p) +{ + unsigned i; + for (i = 0; i < (LZMA_NUM_PB_STATES_MAX << (kLenNumLowBits + 1)); i++) + p->low[i] = kProbInitValue; + for (i = 0; i < kLenNumHighSymbols; i++) + p->high[i] = kProbInitValue; +} + +static void LenEnc_Encode(CLenEnc *p, CRangeEnc *rc, unsigned sym, unsigned posState) +{ + UInt32 range, ttt, newBound; + CLzmaProb *probs = p->low; + range = rc->range; + RC_BIT_PRE(rc, probs); + if (sym >= kLenNumLowSymbols) + { + RC_BIT_1(rc, probs); + probs += kLenNumLowSymbols; + RC_BIT_PRE(rc, probs); + if (sym >= kLenNumLowSymbols * 2) + { + RC_BIT_1(rc, probs); + rc->range = range; + // RcTree_Encode(rc, p->high, kLenNumHighBits, sym - kLenNumLowSymbols * 2); + LitEnc_Encode(rc, p->high, sym - kLenNumLowSymbols * 2); + return; + } + sym -= kLenNumLowSymbols; + } + + // RcTree_Encode(rc, probs + (posState << kLenNumLowBits), kLenNumLowBits, sym); + { + unsigned m; + unsigned bit; + RC_BIT_0(rc, probs); + probs += (posState << (1 + kLenNumLowBits)); + bit = (sym >> 2) ; RC_BIT(rc, probs + 1, bit); m = (1 << 1) + bit; + bit = (sym >> 1) & 1; RC_BIT(rc, probs + m, bit); m = (m << 1) + bit; + bit = sym & 1; RC_BIT(rc, probs + m, bit); + rc->range = range; + } +} + +static void SetPrices_3(const CLzmaProb *probs, UInt32 startPrice, UInt32 *prices, const CProbPrice *ProbPrices) +{ + unsigned i; + for (i = 0; i < 8; i += 2) + { + UInt32 price = startPrice; + UInt32 prob; + price += GET_PRICEa(probs[1 ], (i >> 2)); + price += GET_PRICEa(probs[2 + (i >> 2)], (i >> 1) & 1); + prob = probs[4 + (i >> 1)]; + prices[i ] = price + GET_PRICEa_0(prob); + prices[i + 1] = price + GET_PRICEa_1(prob); + } +} + + +MY_NO_INLINE static void MY_FAST_CALL LenPriceEnc_UpdateTables( + CLenPriceEnc *p, + unsigned numPosStates, + const CLenEnc *enc, + const CProbPrice *ProbPrices) +{ + UInt32 b; + + { + unsigned prob = enc->low[0]; + UInt32 a, c; + unsigned posState; + b = GET_PRICEa_1(prob); + a = GET_PRICEa_0(prob); + c = b + GET_PRICEa_0(enc->low[kLenNumLowSymbols]); + for (posState = 0; posState < numPosStates; posState++) + { + UInt32 *prices = p->prices[posState]; + const CLzmaProb *probs = enc->low + (posState << (1 + kLenNumLowBits)); + SetPrices_3(probs, a, prices, ProbPrices); + SetPrices_3(probs + kLenNumLowSymbols, c, prices + kLenNumLowSymbols, ProbPrices); + } + } + + /* + { + unsigned i; + UInt32 b; + a = GET_PRICEa_0(enc->low[0]); + for (i = 0; i < kLenNumLowSymbols; i++) + p->prices2[i] = a; + a = GET_PRICEa_1(enc->low[0]); + b = a + GET_PRICEa_0(enc->low[kLenNumLowSymbols]); + for (i = kLenNumLowSymbols; i < kLenNumLowSymbols * 2; i++) + p->prices2[i] = b; + a += GET_PRICEa_1(enc->low[kLenNumLowSymbols]); + } + */ + + // p->counter = numSymbols; + // p->counter = 64; + + { + unsigned i = p->tableSize; + + if (i > kLenNumLowSymbols * 2) + { + const CLzmaProb *probs = enc->high; + UInt32 *prices = p->prices[0] + kLenNumLowSymbols * 2; + i -= kLenNumLowSymbols * 2 - 1; + i >>= 1; + b += GET_PRICEa_1(enc->low[kLenNumLowSymbols]); + do + { + /* + p->prices2[i] = a + + // RcTree_GetPrice(enc->high, kLenNumHighBits, i - kLenNumLowSymbols * 2, ProbPrices); + LitEnc_GetPrice(probs, i - kLenNumLowSymbols * 2, ProbPrices); + */ + // UInt32 price = a + RcTree_GetPrice(probs, kLenNumHighBits - 1, sym, ProbPrices); + unsigned sym = --i + (1 << (kLenNumHighBits - 1)); + UInt32 price = b; + do + { + unsigned bit = sym & 1; + sym >>= 1; + price += GET_PRICEa(probs[sym], bit); + } + while (sym >= 2); + + { + unsigned prob = probs[(size_t)i + (1 << (kLenNumHighBits - 1))]; + prices[(size_t)i * 2 ] = price + GET_PRICEa_0(prob); + prices[(size_t)i * 2 + 1] = price + GET_PRICEa_1(prob); + } + } + while (i); + + { + unsigned posState; + size_t num = (p->tableSize - kLenNumLowSymbols * 2) * sizeof(p->prices[0][0]); + for (posState = 1; posState < numPosStates; posState++) + memcpy(p->prices[posState] + kLenNumLowSymbols * 2, p->prices[0] + kLenNumLowSymbols * 2, num); + } + } + } +} + +/* + #ifdef SHOW_STAT + g_STAT_OFFSET += num; + printf("\n MovePos %u", num); + #endif +*/ + +#define MOVE_POS(p, num) { \ + p->additionalOffset += (num); \ + p->matchFinder.Skip(p->matchFinderObj, (UInt32)(num)); } + + +static unsigned ReadMatchDistances(CLzmaEnc *p, unsigned *numPairsRes) +{ + unsigned numPairs; + + p->additionalOffset++; + p->numAvail = p->matchFinder.GetNumAvailableBytes(p->matchFinderObj); + numPairs = p->matchFinder.GetMatches(p->matchFinderObj, p->matches); + *numPairsRes = numPairs; + + #ifdef SHOW_STAT + printf("\n i = %u numPairs = %u ", g_STAT_OFFSET, numPairs / 2); + g_STAT_OFFSET++; + { + unsigned i; + for (i = 0; i < numPairs; i += 2) + printf("%2u %6u | ", p->matches[i], p->matches[i + 1]); + } + #endif + + if (numPairs == 0) + return 0; + { + unsigned len = p->matches[(size_t)numPairs - 2]; + if (len != p->numFastBytes) + return len; + { + UInt32 numAvail = p->numAvail; + if (numAvail > LZMA_MATCH_LEN_MAX) + numAvail = LZMA_MATCH_LEN_MAX; + { + const Byte *p1 = p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - 1; + const Byte *p2 = p1 + len; + ptrdiff_t dif = (ptrdiff_t)-1 - p->matches[(size_t)numPairs - 1]; + const Byte *lim = p1 + numAvail; + for (; p2 != lim && *p2 == p2[dif]; p2++) + {} + return (unsigned)(p2 - p1); + } + } + } +} + +#define MARK_LIT ((UInt32)(Int32)-1) + +#define MakeAs_Lit(p) { (p)->dist = MARK_LIT; (p)->extra = 0; } +#define MakeAs_ShortRep(p) { (p)->dist = 0; (p)->extra = 0; } +#define IsShortRep(p) ((p)->dist == 0) + + +#define GetPrice_ShortRep(p, state, posState) \ + ( GET_PRICE_0(p->isRepG0[state]) + GET_PRICE_0(p->isRep0Long[state][posState])) + +#define GetPrice_Rep_0(p, state, posState) ( \ + GET_PRICE_1(p->isMatch[state][posState]) \ + + GET_PRICE_1(p->isRep0Long[state][posState])) \ + + GET_PRICE_1(p->isRep[state]) \ + + GET_PRICE_0(p->isRepG0[state]) + +MY_FORCE_INLINE +static UInt32 GetPrice_PureRep(const CLzmaEnc *p, unsigned repIndex, size_t state, size_t posState) +{ + UInt32 price; + UInt32 prob = p->isRepG0[state]; + if (repIndex == 0) + { + price = GET_PRICE_0(prob); + price += GET_PRICE_1(p->isRep0Long[state][posState]); + } + else + { + price = GET_PRICE_1(prob); + prob = p->isRepG1[state]; + if (repIndex == 1) + price += GET_PRICE_0(prob); + else + { + price += GET_PRICE_1(prob); + price += GET_PRICE(p->isRepG2[state], repIndex - 2); + } + } + return price; +} + + +static unsigned Backward(CLzmaEnc *p, unsigned cur) +{ + unsigned wr = cur + 1; + p->optEnd = wr; + + for (;;) + { + UInt32 dist = p->opt[cur].dist; + unsigned len = (unsigned)p->opt[cur].len; + unsigned extra = (unsigned)p->opt[cur].extra; + cur -= len; + + if (extra) + { + wr--; + p->opt[wr].len = (UInt32)len; + cur -= extra; + len = extra; + if (extra == 1) + { + p->opt[wr].dist = dist; + dist = MARK_LIT; + } + else + { + p->opt[wr].dist = 0; + len--; + wr--; + p->opt[wr].dist = MARK_LIT; + p->opt[wr].len = 1; + } + } + + if (cur == 0) + { + p->backRes = dist; + p->optCur = wr; + return len; + } + + wr--; + p->opt[wr].dist = dist; + p->opt[wr].len = (UInt32)len; + } +} + + + +#define LIT_PROBS(pos, prevByte) \ + (p->litProbs + (UInt32)3 * (((((pos) << 8) + (prevByte)) & p->lpMask) << p->lc)) + + +static unsigned GetOptimum(CLzmaEnc *p, UInt32 position) +{ + unsigned last, cur; + UInt32 reps[LZMA_NUM_REPS]; + unsigned repLens[LZMA_NUM_REPS]; + UInt32 *matches; + + { + UInt32 numAvail; + unsigned numPairs, mainLen, repMaxIndex, i, posState; + UInt32 matchPrice, repMatchPrice; + const Byte *data; + Byte curByte, matchByte; + + p->optCur = p->optEnd = 0; + + if (p->additionalOffset == 0) + mainLen = ReadMatchDistances(p, &numPairs); + else + { + mainLen = p->longestMatchLen; + numPairs = p->numPairs; + } + + numAvail = p->numAvail; + if (numAvail < 2) + { + p->backRes = MARK_LIT; + return 1; + } + if (numAvail > LZMA_MATCH_LEN_MAX) + numAvail = LZMA_MATCH_LEN_MAX; + + data = p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - 1; + repMaxIndex = 0; + + for (i = 0; i < LZMA_NUM_REPS; i++) + { + unsigned len; + const Byte *data2; + reps[i] = p->reps[i]; + data2 = data - reps[i]; + if (data[0] != data2[0] || data[1] != data2[1]) + { + repLens[i] = 0; + continue; + } + for (len = 2; len < numAvail && data[len] == data2[len]; len++) + {} + repLens[i] = len; + if (len > repLens[repMaxIndex]) + repMaxIndex = i; + } + + if (repLens[repMaxIndex] >= p->numFastBytes) + { + unsigned len; + p->backRes = (UInt32)repMaxIndex; + len = repLens[repMaxIndex]; + MOVE_POS(p, len - 1) + return len; + } + + matches = p->matches; + + if (mainLen >= p->numFastBytes) + { + p->backRes = matches[(size_t)numPairs - 1] + LZMA_NUM_REPS; + MOVE_POS(p, mainLen - 1) + return mainLen; + } + + curByte = *data; + matchByte = *(data - reps[0]); + + last = repLens[repMaxIndex]; + if (last <= mainLen) + last = mainLen; + + if (last < 2 && curByte != matchByte) + { + p->backRes = MARK_LIT; + return 1; + } + + p->opt[0].state = (CState)p->state; + + posState = (position & p->pbMask); + + { + const CLzmaProb *probs = LIT_PROBS(position, *(data - 1)); + p->opt[1].price = GET_PRICE_0(p->isMatch[p->state][posState]) + + (!IsLitState(p->state) ? + LitEnc_Matched_GetPrice(probs, curByte, matchByte, p->ProbPrices) : + LitEnc_GetPrice(probs, curByte, p->ProbPrices)); + } + + MakeAs_Lit(&p->opt[1]); + + matchPrice = GET_PRICE_1(p->isMatch[p->state][posState]); + repMatchPrice = matchPrice + GET_PRICE_1(p->isRep[p->state]); + + // 18.06 + if (matchByte == curByte && repLens[0] == 0) + { + UInt32 shortRepPrice = repMatchPrice + GetPrice_ShortRep(p, p->state, posState); + if (shortRepPrice < p->opt[1].price) + { + p->opt[1].price = shortRepPrice; + MakeAs_ShortRep(&p->opt[1]); + } + if (last < 2) + { + p->backRes = p->opt[1].dist; + return 1; + } + } + + p->opt[1].len = 1; + + p->opt[0].reps[0] = reps[0]; + p->opt[0].reps[1] = reps[1]; + p->opt[0].reps[2] = reps[2]; + p->opt[0].reps[3] = reps[3]; + + // ---------- REP ---------- + + for (i = 0; i < LZMA_NUM_REPS; i++) + { + unsigned repLen = repLens[i]; + UInt32 price; + if (repLen < 2) + continue; + price = repMatchPrice + GetPrice_PureRep(p, i, p->state, posState); + do + { + UInt32 price2 = price + GET_PRICE_LEN(&p->repLenEnc, posState, repLen); + COptimal *opt = &p->opt[repLen]; + if (price2 < opt->price) + { + opt->price = price2; + opt->len = (UInt32)repLen; + opt->dist = (UInt32)i; + opt->extra = 0; + } + } + while (--repLen >= 2); + } + + + // ---------- MATCH ---------- + { + unsigned len = repLens[0] + 1; + if (len <= mainLen) + { + unsigned offs = 0; + UInt32 normalMatchPrice = matchPrice + GET_PRICE_0(p->isRep[p->state]); + + if (len < 2) + len = 2; + else + while (len > matches[offs]) + offs += 2; + + for (; ; len++) + { + COptimal *opt; + UInt32 dist = matches[(size_t)offs + 1]; + UInt32 price = normalMatchPrice + GET_PRICE_LEN(&p->lenEnc, posState, len); + unsigned lenToPosState = GetLenToPosState(len); + + if (dist < kNumFullDistances) + price += p->distancesPrices[lenToPosState][dist & (kNumFullDistances - 1)]; + else + { + unsigned slot; + GetPosSlot2(dist, slot); + price += p->alignPrices[dist & kAlignMask]; + price += p->posSlotPrices[lenToPosState][slot]; + } + + opt = &p->opt[len]; + + if (price < opt->price) + { + opt->price = price; + opt->len = (UInt32)len; + opt->dist = dist + LZMA_NUM_REPS; + opt->extra = 0; + } + + if (len == matches[offs]) + { + offs += 2; + if (offs == numPairs) + break; + } + } + } + } + + + cur = 0; + + #ifdef SHOW_STAT2 + /* if (position >= 0) */ + { + unsigned i; + printf("\n pos = %4X", position); + for (i = cur; i <= last; i++) + printf("\nprice[%4X] = %u", position - cur + i, p->opt[i].price); + } + #endif + } + + + + // ---------- Optimal Parsing ---------- + + for (;;) + { + unsigned numAvail; + UInt32 numAvailFull; + unsigned newLen, numPairs, prev, state, posState, startLen; + UInt32 litPrice, matchPrice, repMatchPrice; + BoolInt nextIsLit; + Byte curByte, matchByte; + const Byte *data; + COptimal *curOpt, *nextOpt; + + if (++cur == last) + break; + + // 18.06 + if (cur >= kNumOpts - 64) + { + unsigned j, best; + UInt32 price = p->opt[cur].price; + best = cur; + for (j = cur + 1; j <= last; j++) + { + UInt32 price2 = p->opt[j].price; + if (price >= price2) + { + price = price2; + best = j; + } + } + { + unsigned delta = best - cur; + if (delta != 0) + { + MOVE_POS(p, delta); + } + } + cur = best; + break; + } + + newLen = ReadMatchDistances(p, &numPairs); + + if (newLen >= p->numFastBytes) + { + p->numPairs = numPairs; + p->longestMatchLen = newLen; + break; + } + + curOpt = &p->opt[cur]; + + position++; + + // we need that check here, if skip_items in p->opt are possible + /* + if (curOpt->price >= kInfinityPrice) + continue; + */ + + prev = cur - curOpt->len; + + if (curOpt->len == 1) + { + state = (unsigned)p->opt[prev].state; + if (IsShortRep(curOpt)) + state = kShortRepNextStates[state]; + else + state = kLiteralNextStates[state]; + } + else + { + const COptimal *prevOpt; + UInt32 b0; + UInt32 dist = curOpt->dist; + + if (curOpt->extra) + { + prev -= (unsigned)curOpt->extra; + state = kState_RepAfterLit; + if (curOpt->extra == 1) + state = (dist < LZMA_NUM_REPS ? kState_RepAfterLit : kState_MatchAfterLit); + } + else + { + state = (unsigned)p->opt[prev].state; + if (dist < LZMA_NUM_REPS) + state = kRepNextStates[state]; + else + state = kMatchNextStates[state]; + } + + prevOpt = &p->opt[prev]; + b0 = prevOpt->reps[0]; + + if (dist < LZMA_NUM_REPS) + { + if (dist == 0) + { + reps[0] = b0; + reps[1] = prevOpt->reps[1]; + reps[2] = prevOpt->reps[2]; + reps[3] = prevOpt->reps[3]; + } + else + { + reps[1] = b0; + b0 = prevOpt->reps[1]; + if (dist == 1) + { + reps[0] = b0; + reps[2] = prevOpt->reps[2]; + reps[3] = prevOpt->reps[3]; + } + else + { + reps[2] = b0; + reps[0] = prevOpt->reps[dist]; + reps[3] = prevOpt->reps[dist ^ 1]; + } + } + } + else + { + reps[0] = (dist - LZMA_NUM_REPS + 1); + reps[1] = b0; + reps[2] = prevOpt->reps[1]; + reps[3] = prevOpt->reps[2]; + } + } + + curOpt->state = (CState)state; + curOpt->reps[0] = reps[0]; + curOpt->reps[1] = reps[1]; + curOpt->reps[2] = reps[2]; + curOpt->reps[3] = reps[3]; + + data = p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - 1; + curByte = *data; + matchByte = *(data - reps[0]); + + posState = (position & p->pbMask); + + /* + The order of Price checks: + < LIT + <= SHORT_REP + < LIT : REP_0 + < REP [ : LIT : REP_0 ] + < MATCH [ : LIT : REP_0 ] + */ + + { + UInt32 curPrice = curOpt->price; + unsigned prob = p->isMatch[state][posState]; + matchPrice = curPrice + GET_PRICE_1(prob); + litPrice = curPrice + GET_PRICE_0(prob); + } + + nextOpt = &p->opt[(size_t)cur + 1]; + nextIsLit = False; + + // here we can allow skip_items in p->opt, if we don't check (nextOpt->price < kInfinityPrice) + // 18.new.06 + if ((nextOpt->price < kInfinityPrice + // && !IsLitState(state) + && matchByte == curByte) + || litPrice > nextOpt->price + ) + litPrice = 0; + else + { + const CLzmaProb *probs = LIT_PROBS(position, *(data - 1)); + litPrice += (!IsLitState(state) ? + LitEnc_Matched_GetPrice(probs, curByte, matchByte, p->ProbPrices) : + LitEnc_GetPrice(probs, curByte, p->ProbPrices)); + + if (litPrice < nextOpt->price) + { + nextOpt->price = litPrice; + nextOpt->len = 1; + MakeAs_Lit(nextOpt); + nextIsLit = True; + } + } + + repMatchPrice = matchPrice + GET_PRICE_1(p->isRep[state]); + + numAvailFull = p->numAvail; + { + unsigned temp = kNumOpts - 1 - cur; + if (numAvailFull > temp) + numAvailFull = (UInt32)temp; + } + + // 18.06 + // ---------- SHORT_REP ---------- + if (IsLitState(state)) // 18.new + if (matchByte == curByte) + if (repMatchPrice < nextOpt->price) // 18.new + // if (numAvailFull < 2 || data[1] != *(data - reps[0] + 1)) + if ( + // nextOpt->price >= kInfinityPrice || + nextOpt->len < 2 // we can check nextOpt->len, if skip items are not allowed in p->opt + || (nextOpt->dist != 0 + // && nextOpt->extra <= 1 // 17.old + ) + ) + { + UInt32 shortRepPrice = repMatchPrice + GetPrice_ShortRep(p, state, posState); + // if (shortRepPrice <= nextOpt->price) // 17.old + if (shortRepPrice < nextOpt->price) // 18.new + { + nextOpt->price = shortRepPrice; + nextOpt->len = 1; + MakeAs_ShortRep(nextOpt); + nextIsLit = False; + } + } + + if (numAvailFull < 2) + continue; + numAvail = (numAvailFull <= p->numFastBytes ? numAvailFull : p->numFastBytes); + + // numAvail <= p->numFastBytes + + // ---------- LIT : REP_0 ---------- + + if (!nextIsLit + && litPrice != 0 // 18.new + && matchByte != curByte + && numAvailFull > 2) + { + const Byte *data2 = data - reps[0]; + if (data[1] == data2[1] && data[2] == data2[2]) + { + unsigned len; + unsigned limit = p->numFastBytes + 1; + if (limit > numAvailFull) + limit = numAvailFull; + for (len = 3; len < limit && data[len] == data2[len]; len++) + {} + + { + unsigned state2 = kLiteralNextStates[state]; + unsigned posState2 = (position + 1) & p->pbMask; + UInt32 price = litPrice + GetPrice_Rep_0(p, state2, posState2); + { + unsigned offset = cur + len; + + if (last < offset) + last = offset; + + // do + { + UInt32 price2; + COptimal *opt; + len--; + // price2 = price + GetPrice_Len_Rep_0(p, len, state2, posState2); + price2 = price + GET_PRICE_LEN(&p->repLenEnc, posState2, len); + + opt = &p->opt[offset]; + // offset--; + if (price2 < opt->price) + { + opt->price = price2; + opt->len = (UInt32)len; + opt->dist = 0; + opt->extra = 1; + } + } + // while (len >= 3); + } + } + } + } + + startLen = 2; /* speed optimization */ + + { + // ---------- REP ---------- + unsigned repIndex = 0; // 17.old + // unsigned repIndex = IsLitState(state) ? 0 : 1; // 18.notused + for (; repIndex < LZMA_NUM_REPS; repIndex++) + { + unsigned len; + UInt32 price; + const Byte *data2 = data - reps[repIndex]; + if (data[0] != data2[0] || data[1] != data2[1]) + continue; + + for (len = 2; len < numAvail && data[len] == data2[len]; len++) + {} + + // if (len < startLen) continue; // 18.new: speed optimization + + { + unsigned offset = cur + len; + if (last < offset) + last = offset; + } + { + unsigned len2 = len; + price = repMatchPrice + GetPrice_PureRep(p, repIndex, state, posState); + do + { + UInt32 price2 = price + GET_PRICE_LEN(&p->repLenEnc, posState, len2); + COptimal *opt = &p->opt[cur + len2]; + if (price2 < opt->price) + { + opt->price = price2; + opt->len = (UInt32)len2; + opt->dist = (UInt32)repIndex; + opt->extra = 0; + } + } + while (--len2 >= 2); + } + + if (repIndex == 0) startLen = len + 1; // 17.old + // startLen = len + 1; // 18.new + + /* if (_maxMode) */ + { + // ---------- REP : LIT : REP_0 ---------- + // numFastBytes + 1 + numFastBytes + + unsigned len2 = len + 1; + unsigned limit = len2 + p->numFastBytes; + if (limit > numAvailFull) + limit = numAvailFull; + + len2 += 2; + if (len2 <= limit) + if (data[len2 - 2] == data2[len2 - 2]) + if (data[len2 - 1] == data2[len2 - 1]) + { + unsigned state2 = kRepNextStates[state]; + unsigned posState2 = (position + len) & p->pbMask; + price += GET_PRICE_LEN(&p->repLenEnc, posState, len) + + GET_PRICE_0(p->isMatch[state2][posState2]) + + LitEnc_Matched_GetPrice(LIT_PROBS(position + len, data[(size_t)len - 1]), + data[len], data2[len], p->ProbPrices); + + // state2 = kLiteralNextStates[state2]; + state2 = kState_LitAfterRep; + posState2 = (posState2 + 1) & p->pbMask; + + + price += GetPrice_Rep_0(p, state2, posState2); + + for (; len2 < limit && data[len2] == data2[len2]; len2++) + {} + + len2 -= len; + // if (len2 >= 3) + { + { + unsigned offset = cur + len + len2; + + if (last < offset) + last = offset; + // do + { + UInt32 price2; + COptimal *opt; + len2--; + // price2 = price + GetPrice_Len_Rep_0(p, len2, state2, posState2); + price2 = price + GET_PRICE_LEN(&p->repLenEnc, posState2, len2); + + opt = &p->opt[offset]; + // offset--; + if (price2 < opt->price) + { + opt->price = price2; + opt->len = (UInt32)len2; + opt->extra = (CExtra)(len + 1); + opt->dist = (UInt32)repIndex; + } + } + // while (len2 >= 3); + } + } + } + } + } + } + + + // ---------- MATCH ---------- + /* for (unsigned len = 2; len <= newLen; len++) */ + if (newLen > numAvail) + { + newLen = numAvail; + for (numPairs = 0; newLen > matches[numPairs]; numPairs += 2); + matches[numPairs] = (UInt32)newLen; + numPairs += 2; + } + + // startLen = 2; /* speed optimization */ + + if (newLen >= startLen) + { + UInt32 normalMatchPrice = matchPrice + GET_PRICE_0(p->isRep[state]); + UInt32 dist; + unsigned offs, posSlot, len; + + { + unsigned offset = cur + newLen; + if (last < offset) + last = offset; + } + + offs = 0; + while (startLen > matches[offs]) + offs += 2; + dist = matches[(size_t)offs + 1]; + + // if (dist >= kNumFullDistances) + GetPosSlot2(dist, posSlot); + + for (len = /*2*/ startLen; ; len++) + { + UInt32 price = normalMatchPrice + GET_PRICE_LEN(&p->lenEnc, posState, len); + { + COptimal *opt; + unsigned lenNorm = len - 2; + lenNorm = GetLenToPosState2(lenNorm); + if (dist < kNumFullDistances) + price += p->distancesPrices[lenNorm][dist & (kNumFullDistances - 1)]; + else + price += p->posSlotPrices[lenNorm][posSlot] + p->alignPrices[dist & kAlignMask]; + + opt = &p->opt[cur + len]; + if (price < opt->price) + { + opt->price = price; + opt->len = (UInt32)len; + opt->dist = dist + LZMA_NUM_REPS; + opt->extra = 0; + } + } + + if (len == matches[offs]) + { + // if (p->_maxMode) { + // MATCH : LIT : REP_0 + + const Byte *data2 = data - dist - 1; + unsigned len2 = len + 1; + unsigned limit = len2 + p->numFastBytes; + if (limit > numAvailFull) + limit = numAvailFull; + + len2 += 2; + if (len2 <= limit) + if (data[len2 - 2] == data2[len2 - 2]) + if (data[len2 - 1] == data2[len2 - 1]) + { + for (; len2 < limit && data[len2] == data2[len2]; len2++) + {} + + len2 -= len; + + // if (len2 >= 3) + { + unsigned state2 = kMatchNextStates[state]; + unsigned posState2 = (position + len) & p->pbMask; + unsigned offset; + price += GET_PRICE_0(p->isMatch[state2][posState2]); + price += LitEnc_Matched_GetPrice(LIT_PROBS(position + len, data[(size_t)len - 1]), + data[len], data2[len], p->ProbPrices); + + // state2 = kLiteralNextStates[state2]; + state2 = kState_LitAfterMatch; + + posState2 = (posState2 + 1) & p->pbMask; + price += GetPrice_Rep_0(p, state2, posState2); + + offset = cur + len + len2; + + if (last < offset) + last = offset; + // do + { + UInt32 price2; + COptimal *opt; + len2--; + // price2 = price + GetPrice_Len_Rep_0(p, len2, state2, posState2); + price2 = price + GET_PRICE_LEN(&p->repLenEnc, posState2, len2); + opt = &p->opt[offset]; + // offset--; + if (price2 < opt->price) + { + opt->price = price2; + opt->len = (UInt32)len2; + opt->extra = (CExtra)(len + 1); + opt->dist = dist + LZMA_NUM_REPS; + } + } + // while (len2 >= 3); + } + + } + + offs += 2; + if (offs == numPairs) + break; + dist = matches[(size_t)offs + 1]; + // if (dist >= kNumFullDistances) + GetPosSlot2(dist, posSlot); + } + } + } + } + + do + p->opt[last].price = kInfinityPrice; + while (--last); + + return Backward(p, cur); +} + + + +#define ChangePair(smallDist, bigDist) (((bigDist) >> 7) > (smallDist)) + + + +static unsigned GetOptimumFast(CLzmaEnc *p) +{ + UInt32 numAvail, mainDist; + unsigned mainLen, numPairs, repIndex, repLen, i; + const Byte *data; + + if (p->additionalOffset == 0) + mainLen = ReadMatchDistances(p, &numPairs); + else + { + mainLen = p->longestMatchLen; + numPairs = p->numPairs; + } + + numAvail = p->numAvail; + p->backRes = MARK_LIT; + if (numAvail < 2) + return 1; + // if (mainLen < 2 && p->state == 0) return 1; // 18.06.notused + if (numAvail > LZMA_MATCH_LEN_MAX) + numAvail = LZMA_MATCH_LEN_MAX; + data = p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - 1; + repLen = repIndex = 0; + + for (i = 0; i < LZMA_NUM_REPS; i++) + { + unsigned len; + const Byte *data2 = data - p->reps[i]; + if (data[0] != data2[0] || data[1] != data2[1]) + continue; + for (len = 2; len < numAvail && data[len] == data2[len]; len++) + {} + if (len >= p->numFastBytes) + { + p->backRes = (UInt32)i; + MOVE_POS(p, len - 1) + return len; + } + if (len > repLen) + { + repIndex = i; + repLen = len; + } + } + + if (mainLen >= p->numFastBytes) + { + p->backRes = p->matches[(size_t)numPairs - 1] + LZMA_NUM_REPS; + MOVE_POS(p, mainLen - 1) + return mainLen; + } + + mainDist = 0; /* for GCC */ + + if (mainLen >= 2) + { + mainDist = p->matches[(size_t)numPairs - 1]; + while (numPairs > 2) + { + UInt32 dist2; + if (mainLen != p->matches[(size_t)numPairs - 4] + 1) + break; + dist2 = p->matches[(size_t)numPairs - 3]; + if (!ChangePair(dist2, mainDist)) + break; + numPairs -= 2; + mainLen--; + mainDist = dist2; + } + if (mainLen == 2 && mainDist >= 0x80) + mainLen = 1; + } + + if (repLen >= 2) + if ( repLen + 1 >= mainLen + || (repLen + 2 >= mainLen && mainDist >= (1 << 9)) + || (repLen + 3 >= mainLen && mainDist >= (1 << 15))) + { + p->backRes = (UInt32)repIndex; + MOVE_POS(p, repLen - 1) + return repLen; + } + + if (mainLen < 2 || numAvail <= 2) + return 1; + + { + unsigned len1 = ReadMatchDistances(p, &p->numPairs); + p->longestMatchLen = len1; + + if (len1 >= 2) + { + UInt32 newDist = p->matches[(size_t)p->numPairs - 1]; + if ( (len1 >= mainLen && newDist < mainDist) + || (len1 == mainLen + 1 && !ChangePair(mainDist, newDist)) + || (len1 > mainLen + 1) + || (len1 + 1 >= mainLen && mainLen >= 3 && ChangePair(newDist, mainDist))) + return 1; + } + } + + data = p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - 1; + + for (i = 0; i < LZMA_NUM_REPS; i++) + { + unsigned len, limit; + const Byte *data2 = data - p->reps[i]; + if (data[0] != data2[0] || data[1] != data2[1]) + continue; + limit = mainLen - 1; + for (len = 2;; len++) + { + if (len >= limit) + return 1; + if (data[len] != data2[len]) + break; + } + } + + p->backRes = mainDist + LZMA_NUM_REPS; + if (mainLen != 2) + { + MOVE_POS(p, mainLen - 2) + } + return mainLen; +} + + + + +static void WriteEndMarker(CLzmaEnc *p, unsigned posState) +{ + UInt32 range; + range = p->rc.range; + { + UInt32 ttt, newBound; + CLzmaProb *prob = &p->isMatch[p->state][posState]; + RC_BIT_PRE(&p->rc, prob) + RC_BIT_1(&p->rc, prob) + prob = &p->isRep[p->state]; + RC_BIT_PRE(&p->rc, prob) + RC_BIT_0(&p->rc, prob) + } + p->state = kMatchNextStates[p->state]; + + p->rc.range = range; + LenEnc_Encode(&p->lenProbs, &p->rc, 0, posState); + range = p->rc.range; + + { + // RcTree_Encode_PosSlot(&p->rc, p->posSlotEncoder[0], (1 << kNumPosSlotBits) - 1); + CLzmaProb *probs = p->posSlotEncoder[0]; + unsigned m = 1; + do + { + UInt32 ttt, newBound; + RC_BIT_PRE(p, probs + m) + RC_BIT_1(&p->rc, probs + m); + m = (m << 1) + 1; + } + while (m < (1 << kNumPosSlotBits)); + } + { + // RangeEnc_EncodeDirectBits(&p->rc, ((UInt32)1 << (30 - kNumAlignBits)) - 1, 30 - kNumAlignBits); UInt32 range = p->range; + unsigned numBits = 30 - kNumAlignBits; + do + { + range >>= 1; + p->rc.low += range; + RC_NORM(&p->rc) + } + while (--numBits); + } + + { + // RcTree_ReverseEncode(&p->rc, p->posAlignEncoder, kNumAlignBits, kAlignMask); + CLzmaProb *probs = p->posAlignEncoder; + unsigned m = 1; + do + { + UInt32 ttt, newBound; + RC_BIT_PRE(p, probs + m) + RC_BIT_1(&p->rc, probs + m); + m = (m << 1) + 1; + } + while (m < kAlignTableSize); + } + p->rc.range = range; +} + + +static SRes CheckErrors(CLzmaEnc *p) +{ + if (p->result != SZ_OK) + return p->result; + if (p->rc.res != SZ_OK) + p->result = SZ_ERROR_WRITE; + if (p->matchFinderBase.result != SZ_OK) + p->result = SZ_ERROR_READ; + if (p->result != SZ_OK) + p->finished = True; + return p->result; +} + + +MY_NO_INLINE static SRes Flush(CLzmaEnc *p, UInt32 nowPos) +{ + /* ReleaseMFStream(); */ + p->finished = True; + if (p->writeEndMark) + WriteEndMarker(p, nowPos & p->pbMask); + RangeEnc_FlushData(&p->rc); + RangeEnc_FlushStream(&p->rc); + return CheckErrors(p); +} + + +MY_NO_INLINE static void FillAlignPrices(CLzmaEnc *p) +{ + unsigned i; + const CProbPrice *ProbPrices = p->ProbPrices; + const CLzmaProb *probs = p->posAlignEncoder; + // p->alignPriceCount = 0; + for (i = 0; i < kAlignTableSize / 2; i++) + { + UInt32 price = 0; + unsigned sym = i; + unsigned m = 1; + unsigned bit; + UInt32 prob; + bit = sym & 1; sym >>= 1; price += GET_PRICEa(probs[m], bit); m = (m << 1) + bit; + bit = sym & 1; sym >>= 1; price += GET_PRICEa(probs[m], bit); m = (m << 1) + bit; + bit = sym & 1; sym >>= 1; price += GET_PRICEa(probs[m], bit); m = (m << 1) + bit; + prob = probs[m]; + p->alignPrices[i ] = price + GET_PRICEa_0(prob); + p->alignPrices[i + 8] = price + GET_PRICEa_1(prob); + // p->alignPrices[i] = RcTree_ReverseGetPrice(p->posAlignEncoder, kNumAlignBits, i, p->ProbPrices); + } +} + + +MY_NO_INLINE static void FillDistancesPrices(CLzmaEnc *p) +{ + // int y; for (y = 0; y < 100; y++) { + + UInt32 tempPrices[kNumFullDistances]; + unsigned i, lps; + + const CProbPrice *ProbPrices = p->ProbPrices; + p->matchPriceCount = 0; + + for (i = kStartPosModelIndex / 2; i < kNumFullDistances / 2; i++) + { + unsigned posSlot = GetPosSlot1(i); + unsigned footerBits = (posSlot >> 1) - 1; + unsigned base = ((2 | (posSlot & 1)) << footerBits); + const CLzmaProb *probs = p->posEncoders + (size_t)base * 2; + // tempPrices[i] = RcTree_ReverseGetPrice(p->posEncoders + base, footerBits, i - base, p->ProbPrices); + UInt32 price = 0; + unsigned m = 1; + unsigned sym = i; + unsigned offset = (unsigned)1 << footerBits; + base += i; + + if (footerBits) + do + { + unsigned bit = sym & 1; + sym >>= 1; + price += GET_PRICEa(probs[m], bit); + m = (m << 1) + bit; + } + while (--footerBits); + + { + unsigned prob = probs[m]; + tempPrices[base ] = price + GET_PRICEa_0(prob); + tempPrices[base + offset] = price + GET_PRICEa_1(prob); + } + } + + for (lps = 0; lps < kNumLenToPosStates; lps++) + { + unsigned slot; + unsigned distTableSize2 = (p->distTableSize + 1) >> 1; + UInt32 *posSlotPrices = p->posSlotPrices[lps]; + const CLzmaProb *probs = p->posSlotEncoder[lps]; + + for (slot = 0; slot < distTableSize2; slot++) + { + // posSlotPrices[slot] = RcTree_GetPrice(encoder, kNumPosSlotBits, slot, p->ProbPrices); + UInt32 price; + unsigned bit; + unsigned sym = slot + (1 << (kNumPosSlotBits - 1)); + unsigned prob; + bit = sym & 1; sym >>= 1; price = GET_PRICEa(probs[sym], bit); + bit = sym & 1; sym >>= 1; price += GET_PRICEa(probs[sym], bit); + bit = sym & 1; sym >>= 1; price += GET_PRICEa(probs[sym], bit); + bit = sym & 1; sym >>= 1; price += GET_PRICEa(probs[sym], bit); + bit = sym & 1; sym >>= 1; price += GET_PRICEa(probs[sym], bit); + prob = probs[(size_t)slot + (1 << (kNumPosSlotBits - 1))]; + posSlotPrices[(size_t)slot * 2 ] = price + GET_PRICEa_0(prob); + posSlotPrices[(size_t)slot * 2 + 1] = price + GET_PRICEa_1(prob); + } + + { + UInt32 delta = ((UInt32)((kEndPosModelIndex / 2 - 1) - kNumAlignBits) << kNumBitPriceShiftBits); + for (slot = kEndPosModelIndex / 2; slot < distTableSize2; slot++) + { + posSlotPrices[(size_t)slot * 2 ] += delta; + posSlotPrices[(size_t)slot * 2 + 1] += delta; + delta += ((UInt32)1 << kNumBitPriceShiftBits); + } + } + + { + UInt32 *dp = p->distancesPrices[lps]; + + dp[0] = posSlotPrices[0]; + dp[1] = posSlotPrices[1]; + dp[2] = posSlotPrices[2]; + dp[3] = posSlotPrices[3]; + + for (i = 4; i < kNumFullDistances; i += 2) + { + UInt32 slotPrice = posSlotPrices[GetPosSlot1(i)]; + dp[i ] = slotPrice + tempPrices[i]; + dp[i + 1] = slotPrice + tempPrices[i + 1]; + } + } + } + // } +} + + + +void LzmaEnc_Construct(CLzmaEnc *p) +{ + RangeEnc_Construct(&p->rc); + MatchFinder_Construct(&p->matchFinderBase); + + #ifndef _7ZIP_ST + MatchFinderMt_Construct(&p->matchFinderMt); + p->matchFinderMt.MatchFinder = &p->matchFinderBase; + #endif + + { + CLzmaEncProps props; + LzmaEncProps_Init(&props); + LzmaEnc_SetProps(p, &props); + } + + #ifndef LZMA_LOG_BSR + LzmaEnc_FastPosInit(p->g_FastPos); + #endif + + LzmaEnc_InitPriceTables(p->ProbPrices); + p->litProbs = NULL; + p->saveState.litProbs = NULL; + +} + +CLzmaEncHandle LzmaEnc_Create(ISzAllocPtr alloc) +{ + void *p; + p = ISzAlloc_Alloc(alloc, sizeof(CLzmaEnc)); + if (p) + LzmaEnc_Construct((CLzmaEnc *)p); + return p; +} + +void LzmaEnc_FreeLits(CLzmaEnc *p, ISzAllocPtr alloc) +{ + ISzAlloc_Free(alloc, p->litProbs); + ISzAlloc_Free(alloc, p->saveState.litProbs); + p->litProbs = NULL; + p->saveState.litProbs = NULL; +} + +void LzmaEnc_Destruct(CLzmaEnc *p, ISzAllocPtr alloc, ISzAllocPtr allocBig) +{ + #ifndef _7ZIP_ST + MatchFinderMt_Destruct(&p->matchFinderMt, allocBig); + #endif + + MatchFinder_Free(&p->matchFinderBase, allocBig); + LzmaEnc_FreeLits(p, alloc); + RangeEnc_Free(&p->rc, alloc); +} + +void LzmaEnc_Destroy(CLzmaEncHandle p, ISzAllocPtr alloc, ISzAllocPtr allocBig) +{ + LzmaEnc_Destruct((CLzmaEnc *)p, alloc, allocBig); + ISzAlloc_Free(alloc, p); +} + + +SRes LzmaEnc_CodeOneBlock(CLzmaEncHandle pp, UInt32 maxPackSize, UInt32 maxUnpackSize) +{ + CLzmaEnc *p = (CLzmaEnc *) pp; + UInt32 nowPos32, startPos32; + if (p->needInit) + { + p->matchFinder.Init(p->matchFinderObj); + p->needInit = 0; + } + + if (p->finished) + return p->result; + RINOK(CheckErrors(p)); + + nowPos32 = (UInt32)p->nowPos64; + startPos32 = nowPos32; + + if (p->nowPos64 == 0) + { + unsigned numPairs; + Byte curByte; + if (p->matchFinder.GetNumAvailableBytes(p->matchFinderObj) == 0) + return Flush(p, nowPos32); + ReadMatchDistances(p, &numPairs); + RangeEnc_EncodeBit_0(&p->rc, &p->isMatch[kState_Start][0]); + // p->state = kLiteralNextStates[p->state]; + curByte = *(p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - p->additionalOffset); + LitEnc_Encode(&p->rc, p->litProbs, curByte); + p->additionalOffset--; + nowPos32++; + } + + if (p->matchFinder.GetNumAvailableBytes(p->matchFinderObj) != 0) + + for (;;) + { + UInt32 dist; + unsigned len, posState; + UInt32 range, ttt, newBound; + CLzmaProb *probs; + + if (p->fastMode) + len = GetOptimumFast(p); + else + { + unsigned oci = p->optCur; + if (p->optEnd == oci) + len = GetOptimum(p, nowPos32); + else + { + const COptimal *opt = &p->opt[oci]; + len = opt->len; + p->backRes = opt->dist; + p->optCur = oci + 1; + } + } + + posState = (unsigned)nowPos32 & p->pbMask; + range = p->rc.range; + probs = &p->isMatch[p->state][posState]; + + RC_BIT_PRE(&p->rc, probs) + + dist = p->backRes; + + #ifdef SHOW_STAT2 + printf("\n pos = %6X, len = %3u pos = %6u", nowPos32, len, dist); + #endif + + if (dist == MARK_LIT) + { + Byte curByte; + const Byte *data; + unsigned state; + + RC_BIT_0(&p->rc, probs); + p->rc.range = range; + data = p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - p->additionalOffset; + probs = LIT_PROBS(nowPos32, *(data - 1)); + curByte = *data; + state = p->state; + p->state = kLiteralNextStates[state]; + if (IsLitState(state)) + LitEnc_Encode(&p->rc, probs, curByte); + else + LitEnc_EncodeMatched(&p->rc, probs, curByte, *(data - p->reps[0])); + } + else + { + RC_BIT_1(&p->rc, probs); + probs = &p->isRep[p->state]; + RC_BIT_PRE(&p->rc, probs) + + if (dist < LZMA_NUM_REPS) + { + RC_BIT_1(&p->rc, probs); + probs = &p->isRepG0[p->state]; + RC_BIT_PRE(&p->rc, probs) + if (dist == 0) + { + RC_BIT_0(&p->rc, probs); + probs = &p->isRep0Long[p->state][posState]; + RC_BIT_PRE(&p->rc, probs) + if (len != 1) + { + RC_BIT_1_BASE(&p->rc, probs); + } + else + { + RC_BIT_0_BASE(&p->rc, probs); + p->state = kShortRepNextStates[p->state]; + } + } + else + { + RC_BIT_1(&p->rc, probs); + probs = &p->isRepG1[p->state]; + RC_BIT_PRE(&p->rc, probs) + if (dist == 1) + { + RC_BIT_0_BASE(&p->rc, probs); + dist = p->reps[1]; + } + else + { + RC_BIT_1(&p->rc, probs); + probs = &p->isRepG2[p->state]; + RC_BIT_PRE(&p->rc, probs) + if (dist == 2) + { + RC_BIT_0_BASE(&p->rc, probs); + dist = p->reps[2]; + } + else + { + RC_BIT_1_BASE(&p->rc, probs); + dist = p->reps[3]; + p->reps[3] = p->reps[2]; + } + p->reps[2] = p->reps[1]; + } + p->reps[1] = p->reps[0]; + p->reps[0] = dist; + } + + RC_NORM(&p->rc) + + p->rc.range = range; + + if (len != 1) + { + LenEnc_Encode(&p->repLenProbs, &p->rc, len - LZMA_MATCH_LEN_MIN, posState); + --p->repLenEncCounter; + p->state = kRepNextStates[p->state]; + } + } + else + { + unsigned posSlot; + RC_BIT_0(&p->rc, probs); + p->rc.range = range; + p->state = kMatchNextStates[p->state]; + + LenEnc_Encode(&p->lenProbs, &p->rc, len - LZMA_MATCH_LEN_MIN, posState); + // --p->lenEnc.counter; + + dist -= LZMA_NUM_REPS; + p->reps[3] = p->reps[2]; + p->reps[2] = p->reps[1]; + p->reps[1] = p->reps[0]; + p->reps[0] = dist + 1; + + p->matchPriceCount++; + GetPosSlot(dist, posSlot); + // RcTree_Encode_PosSlot(&p->rc, p->posSlotEncoder[GetLenToPosState(len)], posSlot); + { + UInt32 sym = (UInt32)posSlot + (1 << kNumPosSlotBits); + range = p->rc.range; + probs = p->posSlotEncoder[GetLenToPosState(len)]; + do + { + CLzmaProb *prob = probs + (sym >> kNumPosSlotBits); + UInt32 bit = (sym >> (kNumPosSlotBits - 1)) & 1; + sym <<= 1; + RC_BIT(&p->rc, prob, bit); + } + while (sym < (1 << kNumPosSlotBits * 2)); + p->rc.range = range; + } + + if (dist >= kStartPosModelIndex) + { + unsigned footerBits = ((posSlot >> 1) - 1); + + if (dist < kNumFullDistances) + { + unsigned base = ((2 | (posSlot & 1)) << footerBits); + RcTree_ReverseEncode(&p->rc, p->posEncoders + base, footerBits, (unsigned)(dist /* - base */)); + } + else + { + UInt32 pos2 = (dist | 0xF) << (32 - footerBits); + range = p->rc.range; + // RangeEnc_EncodeDirectBits(&p->rc, posReduced >> kNumAlignBits, footerBits - kNumAlignBits); + /* + do + { + range >>= 1; + p->rc.low += range & (0 - ((dist >> --footerBits) & 1)); + RC_NORM(&p->rc) + } + while (footerBits > kNumAlignBits); + */ + do + { + range >>= 1; + p->rc.low += range & (0 - (pos2 >> 31)); + pos2 += pos2; + RC_NORM(&p->rc) + } + while (pos2 != 0xF0000000); + + + // RcTree_ReverseEncode(&p->rc, p->posAlignEncoder, kNumAlignBits, posReduced & kAlignMask); + + { + unsigned m = 1; + unsigned bit; + bit = dist & 1; dist >>= 1; RC_BIT(&p->rc, p->posAlignEncoder + m, bit); m = (m << 1) + bit; + bit = dist & 1; dist >>= 1; RC_BIT(&p->rc, p->posAlignEncoder + m, bit); m = (m << 1) + bit; + bit = dist & 1; dist >>= 1; RC_BIT(&p->rc, p->posAlignEncoder + m, bit); m = (m << 1) + bit; + bit = dist & 1; RC_BIT(&p->rc, p->posAlignEncoder + m, bit); + p->rc.range = range; + // p->alignPriceCount++; + } + } + } + } + } + + nowPos32 += (UInt32)len; + p->additionalOffset -= len; + + if (p->additionalOffset == 0) + { + UInt32 processed; + + if (!p->fastMode) + { + /* + if (p->alignPriceCount >= 16) // kAlignTableSize + FillAlignPrices(p); + if (p->matchPriceCount >= 128) + FillDistancesPrices(p); + if (p->lenEnc.counter <= 0) + LenPriceEnc_UpdateTables(&p->lenEnc, 1 << p->pb, &p->lenProbs, p->ProbPrices); + */ + if (p->matchPriceCount >= 64) + { + FillAlignPrices(p); + // { int y; for (y = 0; y < 100; y++) { + FillDistancesPrices(p); + // }} + LenPriceEnc_UpdateTables(&p->lenEnc, 1 << p->pb, &p->lenProbs, p->ProbPrices); + } + if (p->repLenEncCounter <= 0) + { + p->repLenEncCounter = REP_LEN_COUNT; + LenPriceEnc_UpdateTables(&p->repLenEnc, 1 << p->pb, &p->repLenProbs, p->ProbPrices); + } + } + + if (p->matchFinder.GetNumAvailableBytes(p->matchFinderObj) == 0) + break; + processed = nowPos32 - startPos32; + + if (maxPackSize) + { + if (processed + kNumOpts + 300 >= maxUnpackSize + || RangeEnc_GetProcessed_sizet(&p->rc) + kPackReserve >= maxPackSize) + break; + } + else if (processed >= (1 << 17)) + { + p->nowPos64 += nowPos32 - startPos32; + return CheckErrors(p); + } + } + } + + p->nowPos64 += nowPos32 - startPos32; + return Flush(p, nowPos32); +} + + + +#define kBigHashDicLimit ((UInt32)1 << 24) + +static SRes LzmaEnc_Alloc(CLzmaEnc *p, UInt32 keepWindowSize, ISzAllocPtr alloc, ISzAllocPtr allocBig) +{ + UInt32 beforeSize = kNumOpts; + if (!RangeEnc_Alloc(&p->rc, alloc)) + return SZ_ERROR_MEM; + + #ifndef _7ZIP_ST + p->mtMode = (p->multiThread && !p->fastMode && (p->matchFinderBase.btMode != 0)); + #endif + + { + unsigned lclp = p->lc + p->lp; + if (!p->litProbs || !p->saveState.litProbs || p->lclp != lclp) + { + LzmaEnc_FreeLits(p, alloc); + p->litProbs = (CLzmaProb *)ISzAlloc_Alloc(alloc, ((UInt32)0x300 << lclp) * sizeof(CLzmaProb)); + p->saveState.litProbs = (CLzmaProb *)ISzAlloc_Alloc(alloc, ((UInt32)0x300 << lclp) * sizeof(CLzmaProb)); + if (!p->litProbs || !p->saveState.litProbs) + { + LzmaEnc_FreeLits(p, alloc); + return SZ_ERROR_MEM; + } + p->lclp = lclp; + } + } + + p->matchFinderBase.bigHash = (Byte)(p->dictSize > kBigHashDicLimit ? 1 : 0); + + if (beforeSize + p->dictSize < keepWindowSize) + beforeSize = keepWindowSize - p->dictSize; + + #ifndef _7ZIP_ST + if (p->mtMode) + { + RINOK(MatchFinderMt_Create(&p->matchFinderMt, p->dictSize, beforeSize, p->numFastBytes, + LZMA_MATCH_LEN_MAX + + 1 /* 18.04 */ + , allocBig)); + p->matchFinderObj = &p->matchFinderMt; + p->matchFinderBase.bigHash = (Byte)( + (p->dictSize > kBigHashDicLimit && p->matchFinderBase.hashMask >= 0xFFFFFF) ? 1 : 0); + MatchFinderMt_CreateVTable(&p->matchFinderMt, &p->matchFinder); + } + else + #endif + { + if (!MatchFinder_Create(&p->matchFinderBase, p->dictSize, beforeSize, p->numFastBytes, LZMA_MATCH_LEN_MAX, allocBig)) + return SZ_ERROR_MEM; + p->matchFinderObj = &p->matchFinderBase; + MatchFinder_CreateVTable(&p->matchFinderBase, &p->matchFinder); + } + + return SZ_OK; +} + +void LzmaEnc_Init(CLzmaEnc *p) +{ + unsigned i; + p->state = 0; + p->reps[0] = + p->reps[1] = + p->reps[2] = + p->reps[3] = 1; + + RangeEnc_Init(&p->rc); + + for (i = 0; i < (1 << kNumAlignBits); i++) + p->posAlignEncoder[i] = kProbInitValue; + + for (i = 0; i < kNumStates; i++) + { + unsigned j; + for (j = 0; j < LZMA_NUM_PB_STATES_MAX; j++) + { + p->isMatch[i][j] = kProbInitValue; + p->isRep0Long[i][j] = kProbInitValue; + } + p->isRep[i] = kProbInitValue; + p->isRepG0[i] = kProbInitValue; + p->isRepG1[i] = kProbInitValue; + p->isRepG2[i] = kProbInitValue; + } + + { + for (i = 0; i < kNumLenToPosStates; i++) + { + CLzmaProb *probs = p->posSlotEncoder[i]; + unsigned j; + for (j = 0; j < (1 << kNumPosSlotBits); j++) + probs[j] = kProbInitValue; + } + } + { + for (i = 0; i < kNumFullDistances; i++) + p->posEncoders[i] = kProbInitValue; + } + + { + UInt32 num = (UInt32)0x300 << (p->lp + p->lc); + UInt32 k; + CLzmaProb *probs = p->litProbs; + for (k = 0; k < num; k++) + probs[k] = kProbInitValue; + } + + + LenEnc_Init(&p->lenProbs); + LenEnc_Init(&p->repLenProbs); + + p->optEnd = 0; + p->optCur = 0; + + { + for (i = 0; i < kNumOpts; i++) + p->opt[i].price = kInfinityPrice; + } + + p->additionalOffset = 0; + + p->pbMask = (1 << p->pb) - 1; + p->lpMask = ((UInt32)0x100 << p->lp) - ((unsigned)0x100 >> p->lc); +} + + +void LzmaEnc_InitPrices(CLzmaEnc *p) +{ + if (!p->fastMode) + { + FillDistancesPrices(p); + FillAlignPrices(p); + } + + p->lenEnc.tableSize = + p->repLenEnc.tableSize = + p->numFastBytes + 1 - LZMA_MATCH_LEN_MIN; + + p->repLenEncCounter = REP_LEN_COUNT; + + LenPriceEnc_UpdateTables(&p->lenEnc, 1 << p->pb, &p->lenProbs, p->ProbPrices); + LenPriceEnc_UpdateTables(&p->repLenEnc, 1 << p->pb, &p->repLenProbs, p->ProbPrices); +} + +static SRes LzmaEnc_AllocAndInit(CLzmaEnc *p, UInt32 keepWindowSize, ISzAllocPtr alloc, ISzAllocPtr allocBig) +{ + unsigned i; + for (i = kEndPosModelIndex / 2; i < kDicLogSizeMax; i++) + if (p->dictSize <= ((UInt32)1 << i)) + break; + p->distTableSize = i * 2; + + p->finished = False; + p->result = SZ_OK; + RINOK(LzmaEnc_Alloc(p, keepWindowSize, alloc, allocBig)); + LzmaEnc_Init(p); + LzmaEnc_InitPrices(p); + p->nowPos64 = 0; + return SZ_OK; +} + +SRes LzmaEnc_Prepare(CLzmaEncHandle pp, ISeqOutStream *outStream, ISeqInStream *inStream, + ISzAllocPtr alloc, ISzAllocPtr allocBig) +{ + CLzmaEnc *p = (CLzmaEnc *)pp; + p->matchFinderBase.stream = inStream; + p->needInit = 1; + p->rc.outStream = outStream; + return LzmaEnc_AllocAndInit(p, 0, alloc, allocBig); +} + +SRes LzmaEnc_PrepareForLzma2(CLzmaEncHandle pp, + ISeqInStream *inStream, UInt32 keepWindowSize, + ISzAllocPtr alloc, ISzAllocPtr allocBig) +{ + CLzmaEnc *p = (CLzmaEnc *)pp; + p->matchFinderBase.stream = inStream; + p->needInit = 1; + return LzmaEnc_AllocAndInit(p, keepWindowSize, alloc, allocBig); +} + +static void LzmaEnc_SetInputBuf(CLzmaEnc *p, const Byte *src, SizeT srcLen) +{ + p->matchFinderBase.directInput = 1; + p->matchFinderBase.bufferBase = (Byte *)src; + p->matchFinderBase.directInputRem = srcLen; +} + +SRes LzmaEnc_MemPrepare(CLzmaEncHandle pp, const Byte *src, SizeT srcLen, + UInt32 keepWindowSize, ISzAllocPtr alloc, ISzAllocPtr allocBig) +{ + CLzmaEnc *p = (CLzmaEnc *)pp; + LzmaEnc_SetInputBuf(p, src, srcLen); + p->needInit = 1; + + LzmaEnc_SetDataSize(pp, srcLen); + return LzmaEnc_AllocAndInit(p, keepWindowSize, alloc, allocBig); +} + +void LzmaEnc_Finish(CLzmaEncHandle pp) +{ + #ifndef _7ZIP_ST + CLzmaEnc *p = (CLzmaEnc *)pp; + if (p->mtMode) + MatchFinderMt_ReleaseStream(&p->matchFinderMt); + #else + UNUSED_VAR(pp); + #endif +} + + +typedef struct +{ + ISeqOutStream vt; + Byte *data; + SizeT rem; + BoolInt overflow; +} CLzmaEnc_SeqOutStreamBuf; + +static size_t SeqOutStreamBuf_Write(const ISeqOutStream *pp, const void *data, size_t size) +{ + CLzmaEnc_SeqOutStreamBuf *p = CONTAINER_FROM_VTBL(pp, CLzmaEnc_SeqOutStreamBuf, vt); + if (p->rem < size) + { + size = p->rem; + p->overflow = True; + } + memcpy(p->data, data, size); + p->rem -= size; + p->data += size; + return size; +} + + +UInt32 LzmaEnc_GetNumAvailableBytes(CLzmaEncHandle pp) +{ + const CLzmaEnc *p = (CLzmaEnc *)pp; + return p->matchFinder.GetNumAvailableBytes(p->matchFinderObj); +} + + +const Byte *LzmaEnc_GetCurBuf(CLzmaEncHandle pp) +{ + const CLzmaEnc *p = (CLzmaEnc *)pp; + return p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - p->additionalOffset; +} + + +SRes LzmaEnc_CodeOneMemBlock(CLzmaEncHandle pp, BoolInt reInit, + Byte *dest, size_t *destLen, UInt32 desiredPackSize, UInt32 *unpackSize) +{ + CLzmaEnc *p = (CLzmaEnc *)pp; + UInt64 nowPos64; + SRes res; + CLzmaEnc_SeqOutStreamBuf outStream; + + outStream.vt.Write = SeqOutStreamBuf_Write; + outStream.data = dest; + outStream.rem = *destLen; + outStream.overflow = False; + + p->writeEndMark = False; + p->finished = False; + p->result = SZ_OK; + + if (reInit) + LzmaEnc_Init(p); + LzmaEnc_InitPrices(p); + + nowPos64 = p->nowPos64; + RangeEnc_Init(&p->rc); + p->rc.outStream = &outStream.vt; + + if (desiredPackSize == 0) + return SZ_ERROR_OUTPUT_EOF; + + res = LzmaEnc_CodeOneBlock(p, desiredPackSize, *unpackSize); + + *unpackSize = (UInt32)(p->nowPos64 - nowPos64); + *destLen -= outStream.rem; + if (outStream.overflow) + return SZ_ERROR_OUTPUT_EOF; + + return res; +} + + +static SRes LzmaEnc_Encode2(CLzmaEnc *p, ICompressProgress *progress) +{ + SRes res = SZ_OK; + + #ifndef _7ZIP_ST + Byte allocaDummy[0x300]; + allocaDummy[0] = 0; + allocaDummy[1] = allocaDummy[0]; + #endif + + for (;;) + { + res = LzmaEnc_CodeOneBlock(p, 0, 0); + if (res != SZ_OK || p->finished) + break; + if (progress) + { + res = ICompressProgress_Progress(progress, p->nowPos64, RangeEnc_GetProcessed(&p->rc)); + if (res != SZ_OK) + { + res = SZ_ERROR_PROGRESS; + break; + } + } + } + + LzmaEnc_Finish(p); + + /* + if (res == SZ_OK && !Inline_MatchFinder_IsFinishedOK(&p->matchFinderBase)) + res = SZ_ERROR_FAIL; + } + */ + + return res; +} + + +SRes LzmaEnc_Encode(CLzmaEncHandle pp, ISeqOutStream *outStream, ISeqInStream *inStream, ICompressProgress *progress, + ISzAllocPtr alloc, ISzAllocPtr allocBig) +{ + RINOK(LzmaEnc_Prepare(pp, outStream, inStream, alloc, allocBig)); + return LzmaEnc_Encode2((CLzmaEnc *)pp, progress); +} + + +SRes LzmaEnc_WriteProperties(CLzmaEncHandle pp, Byte *props, SizeT *size) +{ + CLzmaEnc *p = (CLzmaEnc *)pp; + unsigned i; + UInt32 dictSize = p->dictSize; + if (*size < LZMA_PROPS_SIZE) + return SZ_ERROR_PARAM; + *size = LZMA_PROPS_SIZE; + props[0] = (Byte)((p->pb * 5 + p->lp) * 9 + p->lc); + + if (dictSize >= ((UInt32)1 << 22)) + { + UInt32 kDictMask = ((UInt32)1 << 20) - 1; + if (dictSize < (UInt32)0xFFFFFFFF - kDictMask) + dictSize = (dictSize + kDictMask) & ~kDictMask; + } + else for (i = 11; i <= 30; i++) + { + if (dictSize <= ((UInt32)2 << i)) { dictSize = (2 << i); break; } + if (dictSize <= ((UInt32)3 << i)) { dictSize = (3 << i); break; } + } + + for (i = 0; i < 4; i++) + props[1 + i] = (Byte)(dictSize >> (8 * i)); + return SZ_OK; +} + + +unsigned LzmaEnc_IsWriteEndMark(CLzmaEncHandle pp) +{ + return ((CLzmaEnc *)pp)->writeEndMark; +} + + +SRes LzmaEnc_MemEncode(CLzmaEncHandle pp, Byte *dest, SizeT *destLen, const Byte *src, SizeT srcLen, + int writeEndMark, ICompressProgress *progress, ISzAllocPtr alloc, ISzAllocPtr allocBig) +{ + SRes res; + CLzmaEnc *p = (CLzmaEnc *)pp; + + CLzmaEnc_SeqOutStreamBuf outStream; + + outStream.vt.Write = SeqOutStreamBuf_Write; + outStream.data = dest; + outStream.rem = *destLen; + outStream.overflow = False; + + p->writeEndMark = writeEndMark; + p->rc.outStream = &outStream.vt; + + res = LzmaEnc_MemPrepare(pp, src, srcLen, 0, alloc, allocBig); + + if (res == SZ_OK) + { + res = LzmaEnc_Encode2(p, progress); + if (res == SZ_OK && p->nowPos64 != srcLen) + res = SZ_ERROR_FAIL; + } + + *destLen -= outStream.rem; + if (outStream.overflow) + return SZ_ERROR_OUTPUT_EOF; + return res; +} + + +SRes LzmaEncode(Byte *dest, SizeT *destLen, const Byte *src, SizeT srcLen, + const CLzmaEncProps *props, Byte *propsEncoded, SizeT *propsSize, int writeEndMark, + ICompressProgress *progress, ISzAllocPtr alloc, ISzAllocPtr allocBig) +{ + CLzmaEnc *p = (CLzmaEnc *)LzmaEnc_Create(alloc); + SRes res; + if (!p) + return SZ_ERROR_MEM; + + res = LzmaEnc_SetProps(p, props); + if (res == SZ_OK) + { + res = LzmaEnc_WriteProperties(p, propsEncoded, propsSize); + if (res == SZ_OK) + res = LzmaEnc_MemEncode(p, dest, destLen, src, srcLen, + writeEndMark, progress, alloc, allocBig); + } + + LzmaEnc_Destroy(p, alloc, allocBig); + return res; +} + +BoolInt LzmaEnc_IsFinished(CLzmaEncHandle pp) +{ + CLzmaEnc *p = (CLzmaEnc *)pp; + return p->finished; +} + diff --git a/contrib/libs/lzmasdk/LzmaEnc.h b/contrib/libs/lzmasdk/LzmaEnc.h index b7d9dd2055..24b3919ce8 100644 --- a/contrib/libs/lzmasdk/LzmaEnc.h +++ b/contrib/libs/lzmasdk/LzmaEnc.h @@ -1,83 +1,83 @@ -/* LzmaEnc.h -- LZMA Encoder
-2017-07-27 : Igor Pavlov : Public domain */
-
-#ifndef __LZMA_ENC_H
-#define __LZMA_ENC_H
-
-#include "7zTypes.h"
-
-EXTERN_C_BEGIN
-
-#define LZMA_PROPS_SIZE 5
-
-typedef struct _CLzmaEncProps
-{
- int level; /* 0 <= level <= 9 */
- UInt32 dictSize; /* (1 << 12) <= dictSize <= (1 << 27) for 32-bit version
- (1 << 12) <= dictSize <= (3 << 29) for 64-bit version
- default = (1 << 24) */
- int lc; /* 0 <= lc <= 8, default = 3 */
- int lp; /* 0 <= lp <= 4, default = 0 */
- int pb; /* 0 <= pb <= 4, default = 2 */
- int algo; /* 0 - fast, 1 - normal, default = 1 */
- int fb; /* 5 <= fb <= 273, default = 32 */
- int btMode; /* 0 - hashChain Mode, 1 - binTree mode - normal, default = 1 */
- int numHashBytes; /* 2, 3 or 4, default = 4 */
- UInt32 mc; /* 1 <= mc <= (1 << 30), default = 32 */
- unsigned writeEndMark; /* 0 - do not write EOPM, 1 - write EOPM, default = 0 */
- int numThreads; /* 1 or 2, default = 2 */
-
- UInt64 reduceSize; /* estimated size of data that will be compressed. default = (UInt64)(Int64)-1.
- Encoder uses this value to reduce dictionary size */
-} CLzmaEncProps;
-
-void LzmaEncProps_Init(CLzmaEncProps *p);
-void LzmaEncProps_Normalize(CLzmaEncProps *p);
-UInt32 LzmaEncProps_GetDictSize(const CLzmaEncProps *props2);
-
-
-/* ---------- CLzmaEncHandle Interface ---------- */
-
-/* LzmaEnc* functions can return the following exit codes:
-SRes:
- SZ_OK - OK
- SZ_ERROR_MEM - Memory allocation error
- SZ_ERROR_PARAM - Incorrect paramater in props
- SZ_ERROR_WRITE - ISeqOutStream write callback error
- SZ_ERROR_OUTPUT_EOF - output buffer overflow - version with (Byte *) output
- SZ_ERROR_PROGRESS - some break from progress callback
- SZ_ERROR_THREAD - error in multithreading functions (only for Mt version)
-*/
-
-typedef void * CLzmaEncHandle;
-
-CLzmaEncHandle LzmaEnc_Create(ISzAllocPtr alloc);
-void LzmaEnc_Destroy(CLzmaEncHandle p, ISzAllocPtr alloc, ISzAllocPtr allocBig);
-
-SRes LzmaEnc_SetProps(CLzmaEncHandle p, const CLzmaEncProps *props);
-void LzmaEnc_SetDataSize(CLzmaEncHandle p, UInt64 expectedDataSiize);
-SRes LzmaEnc_WriteProperties(CLzmaEncHandle p, Byte *properties, SizeT *size);
-unsigned LzmaEnc_IsWriteEndMark(CLzmaEncHandle p);
-
-SRes LzmaEnc_Encode(CLzmaEncHandle p, ISeqOutStream *outStream, ISeqInStream *inStream,
- ICompressProgress *progress, ISzAllocPtr alloc, ISzAllocPtr allocBig);
-SRes LzmaEnc_MemEncode(CLzmaEncHandle p, Byte *dest, SizeT *destLen, const Byte *src, SizeT srcLen,
- int writeEndMark, ICompressProgress *progress, ISzAllocPtr alloc, ISzAllocPtr allocBig);
-
-
-/* ---------- One Call Interface ---------- */
-
-SRes LzmaEncode(Byte *dest, SizeT *destLen, const Byte *src, SizeT srcLen,
- const CLzmaEncProps *props, Byte *propsEncoded, SizeT *propsSize, int writeEndMark,
- ICompressProgress *progress, ISzAllocPtr alloc, ISzAllocPtr allocBig);
-
-EXTERN_C_END
-
-/* ---------- Streaming Interface ---------- */
-
-SRes LzmaEnc_Prepare(CLzmaEncHandle pp, ISeqOutStream *outStream, ISeqInStream *inStream, ISzAllocPtr alloc, ISzAllocPtr allocBig);
-SRes LzmaEnc_CodeOneBlock(CLzmaEncHandle pp, UInt32 maxPackSize, UInt32 maxUnpackSize);
-BoolInt LzmaEnc_IsFinished(CLzmaEncHandle pp);
-void LzmaEnc_Finish(CLzmaEncHandle pp);
-
-#endif
+/* LzmaEnc.h -- LZMA Encoder +2017-07-27 : Igor Pavlov : Public domain */ + +#ifndef __LZMA_ENC_H +#define __LZMA_ENC_H + +#include "7zTypes.h" + +EXTERN_C_BEGIN + +#define LZMA_PROPS_SIZE 5 + +typedef struct _CLzmaEncProps +{ + int level; /* 0 <= level <= 9 */ + UInt32 dictSize; /* (1 << 12) <= dictSize <= (1 << 27) for 32-bit version + (1 << 12) <= dictSize <= (3 << 29) for 64-bit version + default = (1 << 24) */ + int lc; /* 0 <= lc <= 8, default = 3 */ + int lp; /* 0 <= lp <= 4, default = 0 */ + int pb; /* 0 <= pb <= 4, default = 2 */ + int algo; /* 0 - fast, 1 - normal, default = 1 */ + int fb; /* 5 <= fb <= 273, default = 32 */ + int btMode; /* 0 - hashChain Mode, 1 - binTree mode - normal, default = 1 */ + int numHashBytes; /* 2, 3 or 4, default = 4 */ + UInt32 mc; /* 1 <= mc <= (1 << 30), default = 32 */ + unsigned writeEndMark; /* 0 - do not write EOPM, 1 - write EOPM, default = 0 */ + int numThreads; /* 1 or 2, default = 2 */ + + UInt64 reduceSize; /* estimated size of data that will be compressed. default = (UInt64)(Int64)-1. + Encoder uses this value to reduce dictionary size */ +} CLzmaEncProps; + +void LzmaEncProps_Init(CLzmaEncProps *p); +void LzmaEncProps_Normalize(CLzmaEncProps *p); +UInt32 LzmaEncProps_GetDictSize(const CLzmaEncProps *props2); + + +/* ---------- CLzmaEncHandle Interface ---------- */ + +/* LzmaEnc* functions can return the following exit codes: +SRes: + SZ_OK - OK + SZ_ERROR_MEM - Memory allocation error + SZ_ERROR_PARAM - Incorrect paramater in props + SZ_ERROR_WRITE - ISeqOutStream write callback error + SZ_ERROR_OUTPUT_EOF - output buffer overflow - version with (Byte *) output + SZ_ERROR_PROGRESS - some break from progress callback + SZ_ERROR_THREAD - error in multithreading functions (only for Mt version) +*/ + +typedef void * CLzmaEncHandle; + +CLzmaEncHandle LzmaEnc_Create(ISzAllocPtr alloc); +void LzmaEnc_Destroy(CLzmaEncHandle p, ISzAllocPtr alloc, ISzAllocPtr allocBig); + +SRes LzmaEnc_SetProps(CLzmaEncHandle p, const CLzmaEncProps *props); +void LzmaEnc_SetDataSize(CLzmaEncHandle p, UInt64 expectedDataSiize); +SRes LzmaEnc_WriteProperties(CLzmaEncHandle p, Byte *properties, SizeT *size); +unsigned LzmaEnc_IsWriteEndMark(CLzmaEncHandle p); + +SRes LzmaEnc_Encode(CLzmaEncHandle p, ISeqOutStream *outStream, ISeqInStream *inStream, + ICompressProgress *progress, ISzAllocPtr alloc, ISzAllocPtr allocBig); +SRes LzmaEnc_MemEncode(CLzmaEncHandle p, Byte *dest, SizeT *destLen, const Byte *src, SizeT srcLen, + int writeEndMark, ICompressProgress *progress, ISzAllocPtr alloc, ISzAllocPtr allocBig); + + +/* ---------- One Call Interface ---------- */ + +SRes LzmaEncode(Byte *dest, SizeT *destLen, const Byte *src, SizeT srcLen, + const CLzmaEncProps *props, Byte *propsEncoded, SizeT *propsSize, int writeEndMark, + ICompressProgress *progress, ISzAllocPtr alloc, ISzAllocPtr allocBig); + +EXTERN_C_END + +/* ---------- Streaming Interface ---------- */ + +SRes LzmaEnc_Prepare(CLzmaEncHandle pp, ISeqOutStream *outStream, ISeqInStream *inStream, ISzAllocPtr alloc, ISzAllocPtr allocBig); +SRes LzmaEnc_CodeOneBlock(CLzmaEncHandle pp, UInt32 maxPackSize, UInt32 maxUnpackSize); +BoolInt LzmaEnc_IsFinished(CLzmaEncHandle pp); +void LzmaEnc_Finish(CLzmaEncHandle pp); + +#endif diff --git a/contrib/libs/lzmasdk/MtCoder.h b/contrib/libs/lzmasdk/MtCoder.h index 603329d367..20e8cc16c6 100644 --- a/contrib/libs/lzmasdk/MtCoder.h +++ b/contrib/libs/lzmasdk/MtCoder.h @@ -1,141 +1,141 @@ -/* MtCoder.h -- Multi-thread Coder
-2018-07-04 : Igor Pavlov : Public domain */
-
-#ifndef __MT_CODER_H
-#define __MT_CODER_H
-
-#include "MtDec.h"
-
-EXTERN_C_BEGIN
-
-/*
- if ( defined MTCODER__USE_WRITE_THREAD) : main thread writes all data blocks to output stream
- if (not defined MTCODER__USE_WRITE_THREAD) : any coder thread can write data blocks to output stream
-*/
-/* #define MTCODER__USE_WRITE_THREAD */
-
-#ifndef _7ZIP_ST
- #define MTCODER__GET_NUM_BLOCKS_FROM_THREADS(numThreads) ((numThreads) + (numThreads) / 8 + 1)
- #define MTCODER__THREADS_MAX 64
- #define MTCODER__BLOCKS_MAX (MTCODER__GET_NUM_BLOCKS_FROM_THREADS(MTCODER__THREADS_MAX) + 3)
-#else
- #define MTCODER__THREADS_MAX 1
- #define MTCODER__BLOCKS_MAX 1
-#endif
-
-
-#ifndef _7ZIP_ST
-
-
-typedef struct
-{
- ICompressProgress vt;
- CMtProgress *mtProgress;
- UInt64 inSize;
- UInt64 outSize;
-} CMtProgressThunk;
-
-void MtProgressThunk_CreateVTable(CMtProgressThunk *p);
-
-#define MtProgressThunk_Init(p) { (p)->inSize = 0; (p)->outSize = 0; }
-
-
-struct _CMtCoder;
-
-
-typedef struct
-{
- struct _CMtCoder *mtCoder;
- unsigned index;
- int stop;
- Byte *inBuf;
-
- CAutoResetEvent startEvent;
- CThread thread;
-} CMtCoderThread;
-
-
-typedef struct
-{
- SRes (*Code)(void *p, unsigned coderIndex, unsigned outBufIndex,
- const Byte *src, size_t srcSize, int finished);
- SRes (*Write)(void *p, unsigned outBufIndex);
-} IMtCoderCallback2;
-
-
-typedef struct
-{
- SRes res;
- unsigned bufIndex;
- BoolInt finished;
-} CMtCoderBlock;
-
-
-typedef struct _CMtCoder
-{
- /* input variables */
-
- size_t blockSize; /* size of input block */
- unsigned numThreadsMax;
- UInt64 expectedDataSize;
-
- ISeqInStream *inStream;
- const Byte *inData;
- size_t inDataSize;
-
- ICompressProgress *progress;
- ISzAllocPtr allocBig;
-
- IMtCoderCallback2 *mtCallback;
- void *mtCallbackObject;
-
-
- /* internal variables */
-
- size_t allocatedBufsSize;
-
- CAutoResetEvent readEvent;
- CSemaphore blocksSemaphore;
-
- BoolInt stopReading;
- SRes readRes;
-
- #ifdef MTCODER__USE_WRITE_THREAD
- CAutoResetEvent writeEvents[MTCODER__BLOCKS_MAX];
- #else
- CAutoResetEvent finishedEvent;
- SRes writeRes;
- unsigned writeIndex;
- Byte ReadyBlocks[MTCODER__BLOCKS_MAX];
- LONG numFinishedThreads;
- #endif
-
- unsigned numStartedThreadsLimit;
- unsigned numStartedThreads;
-
- unsigned numBlocksMax;
- unsigned blockIndex;
- UInt64 readProcessed;
-
- CCriticalSection cs;
-
- unsigned freeBlockHead;
- unsigned freeBlockList[MTCODER__BLOCKS_MAX];
-
- CMtProgress mtProgress;
- CMtCoderBlock blocks[MTCODER__BLOCKS_MAX];
- CMtCoderThread threads[MTCODER__THREADS_MAX];
-} CMtCoder;
-
-
-void MtCoder_Construct(CMtCoder *p);
-void MtCoder_Destruct(CMtCoder *p);
-SRes MtCoder_Code(CMtCoder *p);
-
-
-#endif
-
-
-EXTERN_C_END
-
-#endif
+/* MtCoder.h -- Multi-thread Coder +2018-07-04 : Igor Pavlov : Public domain */ + +#ifndef __MT_CODER_H +#define __MT_CODER_H + +#include "MtDec.h" + +EXTERN_C_BEGIN + +/* + if ( defined MTCODER__USE_WRITE_THREAD) : main thread writes all data blocks to output stream + if (not defined MTCODER__USE_WRITE_THREAD) : any coder thread can write data blocks to output stream +*/ +/* #define MTCODER__USE_WRITE_THREAD */ + +#ifndef _7ZIP_ST + #define MTCODER__GET_NUM_BLOCKS_FROM_THREADS(numThreads) ((numThreads) + (numThreads) / 8 + 1) + #define MTCODER__THREADS_MAX 64 + #define MTCODER__BLOCKS_MAX (MTCODER__GET_NUM_BLOCKS_FROM_THREADS(MTCODER__THREADS_MAX) + 3) +#else + #define MTCODER__THREADS_MAX 1 + #define MTCODER__BLOCKS_MAX 1 +#endif + + +#ifndef _7ZIP_ST + + +typedef struct +{ + ICompressProgress vt; + CMtProgress *mtProgress; + UInt64 inSize; + UInt64 outSize; +} CMtProgressThunk; + +void MtProgressThunk_CreateVTable(CMtProgressThunk *p); + +#define MtProgressThunk_Init(p) { (p)->inSize = 0; (p)->outSize = 0; } + + +struct _CMtCoder; + + +typedef struct +{ + struct _CMtCoder *mtCoder; + unsigned index; + int stop; + Byte *inBuf; + + CAutoResetEvent startEvent; + CThread thread; +} CMtCoderThread; + + +typedef struct +{ + SRes (*Code)(void *p, unsigned coderIndex, unsigned outBufIndex, + const Byte *src, size_t srcSize, int finished); + SRes (*Write)(void *p, unsigned outBufIndex); +} IMtCoderCallback2; + + +typedef struct +{ + SRes res; + unsigned bufIndex; + BoolInt finished; +} CMtCoderBlock; + + +typedef struct _CMtCoder +{ + /* input variables */ + + size_t blockSize; /* size of input block */ + unsigned numThreadsMax; + UInt64 expectedDataSize; + + ISeqInStream *inStream; + const Byte *inData; + size_t inDataSize; + + ICompressProgress *progress; + ISzAllocPtr allocBig; + + IMtCoderCallback2 *mtCallback; + void *mtCallbackObject; + + + /* internal variables */ + + size_t allocatedBufsSize; + + CAutoResetEvent readEvent; + CSemaphore blocksSemaphore; + + BoolInt stopReading; + SRes readRes; + + #ifdef MTCODER__USE_WRITE_THREAD + CAutoResetEvent writeEvents[MTCODER__BLOCKS_MAX]; + #else + CAutoResetEvent finishedEvent; + SRes writeRes; + unsigned writeIndex; + Byte ReadyBlocks[MTCODER__BLOCKS_MAX]; + LONG numFinishedThreads; + #endif + + unsigned numStartedThreadsLimit; + unsigned numStartedThreads; + + unsigned numBlocksMax; + unsigned blockIndex; + UInt64 readProcessed; + + CCriticalSection cs; + + unsigned freeBlockHead; + unsigned freeBlockList[MTCODER__BLOCKS_MAX]; + + CMtProgress mtProgress; + CMtCoderBlock blocks[MTCODER__BLOCKS_MAX]; + CMtCoderThread threads[MTCODER__THREADS_MAX]; +} CMtCoder; + + +void MtCoder_Construct(CMtCoder *p); +void MtCoder_Destruct(CMtCoder *p); +SRes MtCoder_Code(CMtCoder *p); + + +#endif + + +EXTERN_C_END + +#endif diff --git a/contrib/libs/lzmasdk/MtDec.h b/contrib/libs/lzmasdk/MtDec.h index 9864cc8741..78154c8b67 100644 --- a/contrib/libs/lzmasdk/MtDec.h +++ b/contrib/libs/lzmasdk/MtDec.h @@ -1,201 +1,201 @@ -/* MtDec.h -- Multi-thread Decoder
-2018-07-04 : Igor Pavlov : Public domain */
-
-#ifndef __MT_DEC_H
-#define __MT_DEC_H
-
-#include "7zTypes.h"
-
-#ifndef _7ZIP_ST
-#include "Threads.h"
-#endif
-
-EXTERN_C_BEGIN
-
-#ifndef _7ZIP_ST
-
-#ifndef _7ZIP_ST
- #define MTDEC__THREADS_MAX 32
-#else
- #define MTDEC__THREADS_MAX 1
-#endif
-
-
-typedef struct
-{
- ICompressProgress *progress;
- SRes res;
- UInt64 totalInSize;
- UInt64 totalOutSize;
- CCriticalSection cs;
-} CMtProgress;
-
-void MtProgress_Init(CMtProgress *p, ICompressProgress *progress);
-SRes MtProgress_Progress_ST(CMtProgress *p);
-SRes MtProgress_ProgressAdd(CMtProgress *p, UInt64 inSize, UInt64 outSize);
-SRes MtProgress_GetError(CMtProgress *p);
-void MtProgress_SetError(CMtProgress *p, SRes res);
-
-struct _CMtDec;
-
-typedef struct
-{
- struct _CMtDec *mtDec;
- unsigned index;
- void *inBuf;
-
- size_t inDataSize_Start; // size of input data in start block
- UInt64 inDataSize; // total size of input data in all blocks
-
- CThread thread;
- CAutoResetEvent canRead;
- CAutoResetEvent canWrite;
- void *allocaPtr;
-} CMtDecThread;
-
-void MtDecThread_FreeInBufs(CMtDecThread *t);
-
-
-typedef enum
-{
- MTDEC_PARSE_CONTINUE, // continue this block with more input data
- MTDEC_PARSE_OVERFLOW, // MT buffers overflow, need switch to single-thread
- MTDEC_PARSE_NEW, // new block
- MTDEC_PARSE_END // end of block threading. But we still can return to threading after Write(&needContinue)
-} EMtDecParseState;
-
-typedef struct
-{
- // in
- int startCall;
- const Byte *src;
- size_t srcSize;
- // in : (srcSize == 0) is allowed
- // out : it's allowed to return less that actually was used ?
- int srcFinished;
-
- // out
- EMtDecParseState state;
- BoolInt canCreateNewThread;
- UInt64 outPos; // check it (size_t)
-} CMtDecCallbackInfo;
-
-
-typedef struct
-{
- void (*Parse)(void *p, unsigned coderIndex, CMtDecCallbackInfo *ci);
-
- // PreCode() and Code():
- // (SRes_return_result != SZ_OK) means stop decoding, no need another blocks
- SRes (*PreCode)(void *p, unsigned coderIndex);
- SRes (*Code)(void *p, unsigned coderIndex,
- const Byte *src, size_t srcSize, int srcFinished,
- UInt64 *inCodePos, UInt64 *outCodePos, int *stop);
- // stop - means stop another Code calls
-
-
- /* Write() must be called, if Parse() was called
- set (needWrite) if
- {
- && (was not interrupted by progress)
- && (was not interrupted in previous block)
- }
-
- out:
- if (*needContinue), decoder still need to continue decoding with new iteration,
- even after MTDEC_PARSE_END
- if (*canRecode), we didn't flush current block data, so we still can decode current block later.
- */
- SRes (*Write)(void *p, unsigned coderIndex,
- BoolInt needWriteToStream,
- const Byte *src, size_t srcSize,
- // int srcFinished,
- BoolInt *needContinue,
- BoolInt *canRecode);
-} IMtDecCallback;
-
-
-
-typedef struct _CMtDec
-{
- /* input variables */
-
- size_t inBufSize; /* size of input block */
- unsigned numThreadsMax;
- // size_t inBlockMax;
- unsigned numThreadsMax_2;
-
- ISeqInStream *inStream;
- // const Byte *inData;
- // size_t inDataSize;
-
- ICompressProgress *progress;
- ISzAllocPtr alloc;
-
- IMtDecCallback *mtCallback;
- void *mtCallbackObject;
-
-
- /* internal variables */
-
- size_t allocatedBufsSize;
-
- BoolInt exitThread;
- WRes exitThreadWRes;
-
- UInt64 blockIndex;
- BoolInt isAllocError;
- BoolInt overflow;
- SRes threadingErrorSRes;
-
- BoolInt needContinue;
-
- // CAutoResetEvent finishedEvent;
-
- SRes readRes;
- SRes codeRes;
-
- BoolInt wasInterrupted;
-
- unsigned numStartedThreads_Limit;
- unsigned numStartedThreads;
-
- Byte *crossBlock;
- size_t crossStart;
- size_t crossEnd;
- UInt64 readProcessed;
- BoolInt readWasFinished;
- UInt64 inProcessed;
-
- unsigned filledThreadStart;
- unsigned numFilledThreads;
-
- #ifndef _7ZIP_ST
- BoolInt needInterrupt;
- UInt64 interruptIndex;
- CMtProgress mtProgress;
- CMtDecThread threads[MTDEC__THREADS_MAX];
- #endif
-} CMtDec;
-
-
-void MtDec_Construct(CMtDec *p);
-void MtDec_Destruct(CMtDec *p);
-
-/*
-MtDec_Code() returns:
- SZ_OK - in most cases
- MY_SRes_HRESULT_FROM_WRes(WRes_error) - in case of unexpected error in threading function
-*/
-
-SRes MtDec_Code(CMtDec *p);
-Byte *MtDec_GetCrossBuff(CMtDec *p);
-
-int MtDec_PrepareRead(CMtDec *p);
-const Byte *MtDec_Read(CMtDec *p, size_t *inLim);
-
-#endif
-
-EXTERN_C_END
-
-#endif
+/* MtDec.h -- Multi-thread Decoder +2018-07-04 : Igor Pavlov : Public domain */ + +#ifndef __MT_DEC_H +#define __MT_DEC_H + +#include "7zTypes.h" + +#ifndef _7ZIP_ST +#include "Threads.h" +#endif + +EXTERN_C_BEGIN + +#ifndef _7ZIP_ST + +#ifndef _7ZIP_ST + #define MTDEC__THREADS_MAX 32 +#else + #define MTDEC__THREADS_MAX 1 +#endif + + +typedef struct +{ + ICompressProgress *progress; + SRes res; + UInt64 totalInSize; + UInt64 totalOutSize; + CCriticalSection cs; +} CMtProgress; + +void MtProgress_Init(CMtProgress *p, ICompressProgress *progress); +SRes MtProgress_Progress_ST(CMtProgress *p); +SRes MtProgress_ProgressAdd(CMtProgress *p, UInt64 inSize, UInt64 outSize); +SRes MtProgress_GetError(CMtProgress *p); +void MtProgress_SetError(CMtProgress *p, SRes res); + +struct _CMtDec; + +typedef struct +{ + struct _CMtDec *mtDec; + unsigned index; + void *inBuf; + + size_t inDataSize_Start; // size of input data in start block + UInt64 inDataSize; // total size of input data in all blocks + + CThread thread; + CAutoResetEvent canRead; + CAutoResetEvent canWrite; + void *allocaPtr; +} CMtDecThread; + +void MtDecThread_FreeInBufs(CMtDecThread *t); + + +typedef enum +{ + MTDEC_PARSE_CONTINUE, // continue this block with more input data + MTDEC_PARSE_OVERFLOW, // MT buffers overflow, need switch to single-thread + MTDEC_PARSE_NEW, // new block + MTDEC_PARSE_END // end of block threading. But we still can return to threading after Write(&needContinue) +} EMtDecParseState; + +typedef struct +{ + // in + int startCall; + const Byte *src; + size_t srcSize; + // in : (srcSize == 0) is allowed + // out : it's allowed to return less that actually was used ? + int srcFinished; + + // out + EMtDecParseState state; + BoolInt canCreateNewThread; + UInt64 outPos; // check it (size_t) +} CMtDecCallbackInfo; + + +typedef struct +{ + void (*Parse)(void *p, unsigned coderIndex, CMtDecCallbackInfo *ci); + + // PreCode() and Code(): + // (SRes_return_result != SZ_OK) means stop decoding, no need another blocks + SRes (*PreCode)(void *p, unsigned coderIndex); + SRes (*Code)(void *p, unsigned coderIndex, + const Byte *src, size_t srcSize, int srcFinished, + UInt64 *inCodePos, UInt64 *outCodePos, int *stop); + // stop - means stop another Code calls + + + /* Write() must be called, if Parse() was called + set (needWrite) if + { + && (was not interrupted by progress) + && (was not interrupted in previous block) + } + + out: + if (*needContinue), decoder still need to continue decoding with new iteration, + even after MTDEC_PARSE_END + if (*canRecode), we didn't flush current block data, so we still can decode current block later. + */ + SRes (*Write)(void *p, unsigned coderIndex, + BoolInt needWriteToStream, + const Byte *src, size_t srcSize, + // int srcFinished, + BoolInt *needContinue, + BoolInt *canRecode); +} IMtDecCallback; + + + +typedef struct _CMtDec +{ + /* input variables */ + + size_t inBufSize; /* size of input block */ + unsigned numThreadsMax; + // size_t inBlockMax; + unsigned numThreadsMax_2; + + ISeqInStream *inStream; + // const Byte *inData; + // size_t inDataSize; + + ICompressProgress *progress; + ISzAllocPtr alloc; + + IMtDecCallback *mtCallback; + void *mtCallbackObject; + + + /* internal variables */ + + size_t allocatedBufsSize; + + BoolInt exitThread; + WRes exitThreadWRes; + + UInt64 blockIndex; + BoolInt isAllocError; + BoolInt overflow; + SRes threadingErrorSRes; + + BoolInt needContinue; + + // CAutoResetEvent finishedEvent; + + SRes readRes; + SRes codeRes; + + BoolInt wasInterrupted; + + unsigned numStartedThreads_Limit; + unsigned numStartedThreads; + + Byte *crossBlock; + size_t crossStart; + size_t crossEnd; + UInt64 readProcessed; + BoolInt readWasFinished; + UInt64 inProcessed; + + unsigned filledThreadStart; + unsigned numFilledThreads; + + #ifndef _7ZIP_ST + BoolInt needInterrupt; + UInt64 interruptIndex; + CMtProgress mtProgress; + CMtDecThread threads[MTDEC__THREADS_MAX]; + #endif +} CMtDec; + + +void MtDec_Construct(CMtDec *p); +void MtDec_Destruct(CMtDec *p); + +/* +MtDec_Code() returns: + SZ_OK - in most cases + MY_SRes_HRESULT_FROM_WRes(WRes_error) - in case of unexpected error in threading function +*/ + +SRes MtDec_Code(CMtDec *p); +Byte *MtDec_GetCrossBuff(CMtDec *p); + +int MtDec_PrepareRead(CMtDec *p); +const Byte *MtDec_Read(CMtDec *p, size_t *inLim); + +#endif + +EXTERN_C_END + +#endif diff --git a/contrib/libs/lzmasdk/RotateDefs.h b/contrib/libs/lzmasdk/RotateDefs.h index 6c790e791e..129199f232 100644 --- a/contrib/libs/lzmasdk/RotateDefs.h +++ b/contrib/libs/lzmasdk/RotateDefs.h @@ -1,30 +1,30 @@ -/* RotateDefs.h -- Rotate functions
-2015-03-25 : Igor Pavlov : Public domain */
-
-#ifndef __ROTATE_DEFS_H
-#define __ROTATE_DEFS_H
-
-#ifdef _MSC_VER
-
-#include <stdlib.h>
-
-/* don't use _rotl with MINGW. It can insert slow call to function. */
-
-/* #if (_MSC_VER >= 1200) */
-#pragma intrinsic(_rotl)
-#pragma intrinsic(_rotr)
-/* #endif */
-
-#define rotlFixed(x, n) _rotl((x), (n))
-#define rotrFixed(x, n) _rotr((x), (n))
-
-#else
-
-/* new compilers can translate these macros to fast commands. */
-
-#define rotlFixed(x, n) (((x) << (n)) | ((x) >> (32 - (n))))
-#define rotrFixed(x, n) (((x) >> (n)) | ((x) << (32 - (n))))
-
-#endif
-
-#endif
+/* RotateDefs.h -- Rotate functions +2015-03-25 : Igor Pavlov : Public domain */ + +#ifndef __ROTATE_DEFS_H +#define __ROTATE_DEFS_H + +#ifdef _MSC_VER + +#include <stdlib.h> + +/* don't use _rotl with MINGW. It can insert slow call to function. */ + +/* #if (_MSC_VER >= 1200) */ +#pragma intrinsic(_rotl) +#pragma intrinsic(_rotr) +/* #endif */ + +#define rotlFixed(x, n) _rotl((x), (n)) +#define rotrFixed(x, n) _rotr((x), (n)) + +#else + +/* new compilers can translate these macros to fast commands. */ + +#define rotlFixed(x, n) (((x) << (n)) | ((x) >> (32 - (n)))) +#define rotrFixed(x, n) (((x) >> (n)) | ((x) << (32 - (n)))) + +#endif + +#endif diff --git a/contrib/libs/lzmasdk/Sha256.c b/contrib/libs/lzmasdk/Sha256.c index 90994e5abb..56daa25d5f 100644 --- a/contrib/libs/lzmasdk/Sha256.c +++ b/contrib/libs/lzmasdk/Sha256.c @@ -1,248 +1,248 @@ -/* Crypto/Sha256.c -- SHA-256 Hash
-2017-04-03 : Igor Pavlov : Public domain
-This code is based on public domain code from Wei Dai's Crypto++ library. */
-
-#include "Precomp.h"
-
-#include <string.h>
-
-#include "CpuArch.h"
-#include "RotateDefs.h"
-#include "Sha256.h"
-
-/* define it for speed optimization */
-#ifndef _SFX
-#define _SHA256_UNROLL
-#define _SHA256_UNROLL2
-#endif
-
-/* #define _SHA256_UNROLL2 */
-
-void Sha256_Init(CSha256 *p)
-{
- p->state[0] = 0x6a09e667;
- p->state[1] = 0xbb67ae85;
- p->state[2] = 0x3c6ef372;
- p->state[3] = 0xa54ff53a;
- p->state[4] = 0x510e527f;
- p->state[5] = 0x9b05688c;
- p->state[6] = 0x1f83d9ab;
- p->state[7] = 0x5be0cd19;
- p->count = 0;
-}
-
-#define S0(x) (rotrFixed(x, 2) ^ rotrFixed(x,13) ^ rotrFixed(x, 22))
-#define S1(x) (rotrFixed(x, 6) ^ rotrFixed(x,11) ^ rotrFixed(x, 25))
-#define s0(x) (rotrFixed(x, 7) ^ rotrFixed(x,18) ^ (x >> 3))
-#define s1(x) (rotrFixed(x,17) ^ rotrFixed(x,19) ^ (x >> 10))
-
-#define blk0(i) (W[i])
-#define blk2(i) (W[i] += s1(W[((i)-2)&15]) + W[((i)-7)&15] + s0(W[((i)-15)&15]))
-
-#define Ch(x,y,z) (z^(x&(y^z)))
-#define Maj(x,y,z) ((x&y)|(z&(x|y)))
-
-#ifdef _SHA256_UNROLL2
-
-#define R(a,b,c,d,e,f,g,h, i) \
- h += S1(e) + Ch(e,f,g) + K[(i)+(size_t)(j)] + (j ? blk2(i) : blk0(i)); \
- d += h; \
- h += S0(a) + Maj(a, b, c)
-
-#define RX_8(i) \
- R(a,b,c,d,e,f,g,h, i); \
- R(h,a,b,c,d,e,f,g, i+1); \
- R(g,h,a,b,c,d,e,f, i+2); \
- R(f,g,h,a,b,c,d,e, i+3); \
- R(e,f,g,h,a,b,c,d, i+4); \
- R(d,e,f,g,h,a,b,c, i+5); \
- R(c,d,e,f,g,h,a,b, i+6); \
- R(b,c,d,e,f,g,h,a, i+7)
-
-#define RX_16 RX_8(0); RX_8(8);
-
-#else
-
-#define a(i) T[(0-(i))&7]
-#define b(i) T[(1-(i))&7]
-#define c(i) T[(2-(i))&7]
-#define d(i) T[(3-(i))&7]
-#define e(i) T[(4-(i))&7]
-#define f(i) T[(5-(i))&7]
-#define g(i) T[(6-(i))&7]
-#define h(i) T[(7-(i))&7]
-
-#define R(i) \
- h(i) += S1(e(i)) + Ch(e(i),f(i),g(i)) + K[(i)+(size_t)(j)] + (j ? blk2(i) : blk0(i)); \
- d(i) += h(i); \
- h(i) += S0(a(i)) + Maj(a(i), b(i), c(i)) \
-
-#ifdef _SHA256_UNROLL
-
-#define RX_8(i) R(i+0); R(i+1); R(i+2); R(i+3); R(i+4); R(i+5); R(i+6); R(i+7);
-#define RX_16 RX_8(0); RX_8(8);
-
-#else
-
-#define RX_16 unsigned i; for (i = 0; i < 16; i++) { R(i); }
-
-#endif
-
-#endif
-
-static const UInt32 K[64] = {
- 0x428a2f98, 0x71374491, 0xb5c0fbcf, 0xe9b5dba5,
- 0x3956c25b, 0x59f111f1, 0x923f82a4, 0xab1c5ed5,
- 0xd807aa98, 0x12835b01, 0x243185be, 0x550c7dc3,
- 0x72be5d74, 0x80deb1fe, 0x9bdc06a7, 0xc19bf174,
- 0xe49b69c1, 0xefbe4786, 0x0fc19dc6, 0x240ca1cc,
- 0x2de92c6f, 0x4a7484aa, 0x5cb0a9dc, 0x76f988da,
- 0x983e5152, 0xa831c66d, 0xb00327c8, 0xbf597fc7,
- 0xc6e00bf3, 0xd5a79147, 0x06ca6351, 0x14292967,
- 0x27b70a85, 0x2e1b2138, 0x4d2c6dfc, 0x53380d13,
- 0x650a7354, 0x766a0abb, 0x81c2c92e, 0x92722c85,
- 0xa2bfe8a1, 0xa81a664b, 0xc24b8b70, 0xc76c51a3,
- 0xd192e819, 0xd6990624, 0xf40e3585, 0x106aa070,
- 0x19a4c116, 0x1e376c08, 0x2748774c, 0x34b0bcb5,
- 0x391c0cb3, 0x4ed8aa4a, 0x5b9cca4f, 0x682e6ff3,
- 0x748f82ee, 0x78a5636f, 0x84c87814, 0x8cc70208,
- 0x90befffa, 0xa4506ceb, 0xbef9a3f7, 0xc67178f2
-};
-
-static void Sha256_WriteByteBlock(CSha256 *p)
-{
- UInt32 W[16];
- unsigned j;
- UInt32 *state;
-
- #ifdef _SHA256_UNROLL2
- UInt32 a,b,c,d,e,f,g,h;
- #else
- UInt32 T[8];
- #endif
-
- for (j = 0; j < 16; j += 4)
- {
- const Byte *ccc = p->buffer + j * 4;
- W[j ] = GetBe32(ccc);
- W[j + 1] = GetBe32(ccc + 4);
- W[j + 2] = GetBe32(ccc + 8);
- W[j + 3] = GetBe32(ccc + 12);
- }
-
- state = p->state;
-
- #ifdef _SHA256_UNROLL2
- a = state[0];
- b = state[1];
- c = state[2];
- d = state[3];
- e = state[4];
- f = state[5];
- g = state[6];
- h = state[7];
- #else
- for (j = 0; j < 8; j++)
- T[j] = state[j];
- #endif
-
- for (j = 0; j < 64; j += 16)
- {
- RX_16
- }
-
- #ifdef _SHA256_UNROLL2
- state[0] += a;
- state[1] += b;
- state[2] += c;
- state[3] += d;
- state[4] += e;
- state[5] += f;
- state[6] += g;
- state[7] += h;
- #else
- for (j = 0; j < 8; j++)
- state[j] += T[j];
- #endif
-
- /* Wipe variables */
- /* memset(W, 0, sizeof(W)); */
- /* memset(T, 0, sizeof(T)); */
-}
-
-#undef S0
-#undef S1
-#undef s0
-#undef s1
-
-void Sha256_Update(CSha256 *p, const Byte *data, size_t size)
-{
- if (size == 0)
- return;
-
- {
- unsigned pos = (unsigned)p->count & 0x3F;
- unsigned num;
-
- p->count += size;
-
- num = 64 - pos;
- if (num > size)
- {
- memcpy(p->buffer + pos, data, size);
- return;
- }
-
- size -= num;
- memcpy(p->buffer + pos, data, num);
- data += num;
- }
-
- for (;;)
- {
- Sha256_WriteByteBlock(p);
- if (size < 64)
- break;
- size -= 64;
- memcpy(p->buffer, data, 64);
- data += 64;
- }
-
- if (size != 0)
- memcpy(p->buffer, data, size);
-}
-
-void Sha256_Final(CSha256 *p, Byte *digest)
-{
- unsigned pos = (unsigned)p->count & 0x3F;
- unsigned i;
-
- p->buffer[pos++] = 0x80;
-
- while (pos != (64 - 8))
- {
- pos &= 0x3F;
- if (pos == 0)
- Sha256_WriteByteBlock(p);
- p->buffer[pos++] = 0;
- }
-
- {
- UInt64 numBits = (p->count << 3);
- SetBe32(p->buffer + 64 - 8, (UInt32)(numBits >> 32));
- SetBe32(p->buffer + 64 - 4, (UInt32)(numBits));
- }
-
- Sha256_WriteByteBlock(p);
-
- for (i = 0; i < 8; i += 2)
- {
- UInt32 v0 = p->state[i];
- UInt32 v1 = p->state[i + 1];
- SetBe32(digest , v0);
- SetBe32(digest + 4, v1);
- digest += 8;
- }
-
- Sha256_Init(p);
-}
+/* Crypto/Sha256.c -- SHA-256 Hash +2017-04-03 : Igor Pavlov : Public domain +This code is based on public domain code from Wei Dai's Crypto++ library. */ + +#include "Precomp.h" + +#include <string.h> + +#include "CpuArch.h" +#include "RotateDefs.h" +#include "Sha256.h" + +/* define it for speed optimization */ +#ifndef _SFX +#define _SHA256_UNROLL +#define _SHA256_UNROLL2 +#endif + +/* #define _SHA256_UNROLL2 */ + +void Sha256_Init(CSha256 *p) +{ + p->state[0] = 0x6a09e667; + p->state[1] = 0xbb67ae85; + p->state[2] = 0x3c6ef372; + p->state[3] = 0xa54ff53a; + p->state[4] = 0x510e527f; + p->state[5] = 0x9b05688c; + p->state[6] = 0x1f83d9ab; + p->state[7] = 0x5be0cd19; + p->count = 0; +} + +#define S0(x) (rotrFixed(x, 2) ^ rotrFixed(x,13) ^ rotrFixed(x, 22)) +#define S1(x) (rotrFixed(x, 6) ^ rotrFixed(x,11) ^ rotrFixed(x, 25)) +#define s0(x) (rotrFixed(x, 7) ^ rotrFixed(x,18) ^ (x >> 3)) +#define s1(x) (rotrFixed(x,17) ^ rotrFixed(x,19) ^ (x >> 10)) + +#define blk0(i) (W[i]) +#define blk2(i) (W[i] += s1(W[((i)-2)&15]) + W[((i)-7)&15] + s0(W[((i)-15)&15])) + +#define Ch(x,y,z) (z^(x&(y^z))) +#define Maj(x,y,z) ((x&y)|(z&(x|y))) + +#ifdef _SHA256_UNROLL2 + +#define R(a,b,c,d,e,f,g,h, i) \ + h += S1(e) + Ch(e,f,g) + K[(i)+(size_t)(j)] + (j ? blk2(i) : blk0(i)); \ + d += h; \ + h += S0(a) + Maj(a, b, c) + +#define RX_8(i) \ + R(a,b,c,d,e,f,g,h, i); \ + R(h,a,b,c,d,e,f,g, i+1); \ + R(g,h,a,b,c,d,e,f, i+2); \ + R(f,g,h,a,b,c,d,e, i+3); \ + R(e,f,g,h,a,b,c,d, i+4); \ + R(d,e,f,g,h,a,b,c, i+5); \ + R(c,d,e,f,g,h,a,b, i+6); \ + R(b,c,d,e,f,g,h,a, i+7) + +#define RX_16 RX_8(0); RX_8(8); + +#else + +#define a(i) T[(0-(i))&7] +#define b(i) T[(1-(i))&7] +#define c(i) T[(2-(i))&7] +#define d(i) T[(3-(i))&7] +#define e(i) T[(4-(i))&7] +#define f(i) T[(5-(i))&7] +#define g(i) T[(6-(i))&7] +#define h(i) T[(7-(i))&7] + +#define R(i) \ + h(i) += S1(e(i)) + Ch(e(i),f(i),g(i)) + K[(i)+(size_t)(j)] + (j ? blk2(i) : blk0(i)); \ + d(i) += h(i); \ + h(i) += S0(a(i)) + Maj(a(i), b(i), c(i)) \ + +#ifdef _SHA256_UNROLL + +#define RX_8(i) R(i+0); R(i+1); R(i+2); R(i+3); R(i+4); R(i+5); R(i+6); R(i+7); +#define RX_16 RX_8(0); RX_8(8); + +#else + +#define RX_16 unsigned i; for (i = 0; i < 16; i++) { R(i); } + +#endif + +#endif + +static const UInt32 K[64] = { + 0x428a2f98, 0x71374491, 0xb5c0fbcf, 0xe9b5dba5, + 0x3956c25b, 0x59f111f1, 0x923f82a4, 0xab1c5ed5, + 0xd807aa98, 0x12835b01, 0x243185be, 0x550c7dc3, + 0x72be5d74, 0x80deb1fe, 0x9bdc06a7, 0xc19bf174, + 0xe49b69c1, 0xefbe4786, 0x0fc19dc6, 0x240ca1cc, + 0x2de92c6f, 0x4a7484aa, 0x5cb0a9dc, 0x76f988da, + 0x983e5152, 0xa831c66d, 0xb00327c8, 0xbf597fc7, + 0xc6e00bf3, 0xd5a79147, 0x06ca6351, 0x14292967, + 0x27b70a85, 0x2e1b2138, 0x4d2c6dfc, 0x53380d13, + 0x650a7354, 0x766a0abb, 0x81c2c92e, 0x92722c85, + 0xa2bfe8a1, 0xa81a664b, 0xc24b8b70, 0xc76c51a3, + 0xd192e819, 0xd6990624, 0xf40e3585, 0x106aa070, + 0x19a4c116, 0x1e376c08, 0x2748774c, 0x34b0bcb5, + 0x391c0cb3, 0x4ed8aa4a, 0x5b9cca4f, 0x682e6ff3, + 0x748f82ee, 0x78a5636f, 0x84c87814, 0x8cc70208, + 0x90befffa, 0xa4506ceb, 0xbef9a3f7, 0xc67178f2 +}; + +static void Sha256_WriteByteBlock(CSha256 *p) +{ + UInt32 W[16]; + unsigned j; + UInt32 *state; + + #ifdef _SHA256_UNROLL2 + UInt32 a,b,c,d,e,f,g,h; + #else + UInt32 T[8]; + #endif + + for (j = 0; j < 16; j += 4) + { + const Byte *ccc = p->buffer + j * 4; + W[j ] = GetBe32(ccc); + W[j + 1] = GetBe32(ccc + 4); + W[j + 2] = GetBe32(ccc + 8); + W[j + 3] = GetBe32(ccc + 12); + } + + state = p->state; + + #ifdef _SHA256_UNROLL2 + a = state[0]; + b = state[1]; + c = state[2]; + d = state[3]; + e = state[4]; + f = state[5]; + g = state[6]; + h = state[7]; + #else + for (j = 0; j < 8; j++) + T[j] = state[j]; + #endif + + for (j = 0; j < 64; j += 16) + { + RX_16 + } + + #ifdef _SHA256_UNROLL2 + state[0] += a; + state[1] += b; + state[2] += c; + state[3] += d; + state[4] += e; + state[5] += f; + state[6] += g; + state[7] += h; + #else + for (j = 0; j < 8; j++) + state[j] += T[j]; + #endif + + /* Wipe variables */ + /* memset(W, 0, sizeof(W)); */ + /* memset(T, 0, sizeof(T)); */ +} + +#undef S0 +#undef S1 +#undef s0 +#undef s1 + +void Sha256_Update(CSha256 *p, const Byte *data, size_t size) +{ + if (size == 0) + return; + + { + unsigned pos = (unsigned)p->count & 0x3F; + unsigned num; + + p->count += size; + + num = 64 - pos; + if (num > size) + { + memcpy(p->buffer + pos, data, size); + return; + } + + size -= num; + memcpy(p->buffer + pos, data, num); + data += num; + } + + for (;;) + { + Sha256_WriteByteBlock(p); + if (size < 64) + break; + size -= 64; + memcpy(p->buffer, data, 64); + data += 64; + } + + if (size != 0) + memcpy(p->buffer, data, size); +} + +void Sha256_Final(CSha256 *p, Byte *digest) +{ + unsigned pos = (unsigned)p->count & 0x3F; + unsigned i; + + p->buffer[pos++] = 0x80; + + while (pos != (64 - 8)) + { + pos &= 0x3F; + if (pos == 0) + Sha256_WriteByteBlock(p); + p->buffer[pos++] = 0; + } + + { + UInt64 numBits = (p->count << 3); + SetBe32(p->buffer + 64 - 8, (UInt32)(numBits >> 32)); + SetBe32(p->buffer + 64 - 4, (UInt32)(numBits)); + } + + Sha256_WriteByteBlock(p); + + for (i = 0; i < 8; i += 2) + { + UInt32 v0 = p->state[i]; + UInt32 v1 = p->state[i + 1]; + SetBe32(digest , v0); + SetBe32(digest + 4, v1); + digest += 8; + } + + Sha256_Init(p); +} diff --git a/contrib/libs/lzmasdk/Sha256.h b/contrib/libs/lzmasdk/Sha256.h index 7f17ccf9c9..b8ee495207 100644 --- a/contrib/libs/lzmasdk/Sha256.h +++ b/contrib/libs/lzmasdk/Sha256.h @@ -1,26 +1,26 @@ -/* Sha256.h -- SHA-256 Hash
-2013-01-18 : Igor Pavlov : Public domain */
-
-#ifndef __CRYPTO_SHA256_H
-#define __CRYPTO_SHA256_H
-
-#include "7zTypes.h"
-
-EXTERN_C_BEGIN
-
-#define SHA256_DIGEST_SIZE 32
-
-typedef struct
-{
- UInt32 state[8];
- UInt64 count;
- Byte buffer[64];
-} CSha256;
-
-void Sha256_Init(CSha256 *p);
-void Sha256_Update(CSha256 *p, const Byte *data, size_t size);
-void Sha256_Final(CSha256 *p, Byte *digest);
-
-EXTERN_C_END
-
-#endif
+/* Sha256.h -- SHA-256 Hash +2013-01-18 : Igor Pavlov : Public domain */ + +#ifndef __CRYPTO_SHA256_H +#define __CRYPTO_SHA256_H + +#include "7zTypes.h" + +EXTERN_C_BEGIN + +#define SHA256_DIGEST_SIZE 32 + +typedef struct +{ + UInt32 state[8]; + UInt64 count; + Byte buffer[64]; +} CSha256; + +void Sha256_Init(CSha256 *p); +void Sha256_Update(CSha256 *p, const Byte *data, size_t size); +void Sha256_Final(CSha256 *p, Byte *digest); + +EXTERN_C_END + +#endif diff --git a/contrib/libs/lzmasdk/Threads.h b/contrib/libs/lzmasdk/Threads.h index f913241aea..17a1a5a550 100644 --- a/contrib/libs/lzmasdk/Threads.h +++ b/contrib/libs/lzmasdk/Threads.h @@ -1,68 +1,68 @@ -/* Threads.h -- multithreading library
-2017-06-18 : Igor Pavlov : Public domain */
-
-#ifndef __7Z_THREADS_H
-#define __7Z_THREADS_H
-
-#ifdef _WIN32
-#include <windows.h>
-#endif
-
-#include "7zTypes.h"
-
-EXTERN_C_BEGIN
-
-WRes HandlePtr_Close(HANDLE *h);
-WRes Handle_WaitObject(HANDLE h);
-
-typedef HANDLE CThread;
-#define Thread_Construct(p) *(p) = NULL
-#define Thread_WasCreated(p) (*(p) != NULL)
-#define Thread_Close(p) HandlePtr_Close(p)
-#define Thread_Wait(p) Handle_WaitObject(*(p))
-
-typedef
-#ifdef UNDER_CE
- DWORD
-#else
- unsigned
-#endif
- THREAD_FUNC_RET_TYPE;
-
-#define THREAD_FUNC_CALL_TYPE MY_STD_CALL
-#define THREAD_FUNC_DECL THREAD_FUNC_RET_TYPE THREAD_FUNC_CALL_TYPE
-typedef THREAD_FUNC_RET_TYPE (THREAD_FUNC_CALL_TYPE * THREAD_FUNC_TYPE)(void *);
-WRes Thread_Create(CThread *p, THREAD_FUNC_TYPE func, LPVOID param);
-
-typedef HANDLE CEvent;
-typedef CEvent CAutoResetEvent;
-typedef CEvent CManualResetEvent;
-#define Event_Construct(p) *(p) = NULL
-#define Event_IsCreated(p) (*(p) != NULL)
-#define Event_Close(p) HandlePtr_Close(p)
-#define Event_Wait(p) Handle_WaitObject(*(p))
-WRes Event_Set(CEvent *p);
-WRes Event_Reset(CEvent *p);
-WRes ManualResetEvent_Create(CManualResetEvent *p, int signaled);
-WRes ManualResetEvent_CreateNotSignaled(CManualResetEvent *p);
-WRes AutoResetEvent_Create(CAutoResetEvent *p, int signaled);
-WRes AutoResetEvent_CreateNotSignaled(CAutoResetEvent *p);
-
-typedef HANDLE CSemaphore;
-#define Semaphore_Construct(p) *(p) = NULL
-#define Semaphore_IsCreated(p) (*(p) != NULL)
-#define Semaphore_Close(p) HandlePtr_Close(p)
-#define Semaphore_Wait(p) Handle_WaitObject(*(p))
-WRes Semaphore_Create(CSemaphore *p, UInt32 initCount, UInt32 maxCount);
-WRes Semaphore_ReleaseN(CSemaphore *p, UInt32 num);
-WRes Semaphore_Release1(CSemaphore *p);
-
-typedef CRITICAL_SECTION CCriticalSection;
-WRes CriticalSection_Init(CCriticalSection *p);
-#define CriticalSection_Delete(p) DeleteCriticalSection(p)
-#define CriticalSection_Enter(p) EnterCriticalSection(p)
-#define CriticalSection_Leave(p) LeaveCriticalSection(p)
-
-EXTERN_C_END
-
-#endif
+/* Threads.h -- multithreading library +2017-06-18 : Igor Pavlov : Public domain */ + +#ifndef __7Z_THREADS_H +#define __7Z_THREADS_H + +#ifdef _WIN32 +#include <windows.h> +#endif + +#include "7zTypes.h" + +EXTERN_C_BEGIN + +WRes HandlePtr_Close(HANDLE *h); +WRes Handle_WaitObject(HANDLE h); + +typedef HANDLE CThread; +#define Thread_Construct(p) *(p) = NULL +#define Thread_WasCreated(p) (*(p) != NULL) +#define Thread_Close(p) HandlePtr_Close(p) +#define Thread_Wait(p) Handle_WaitObject(*(p)) + +typedef +#ifdef UNDER_CE + DWORD +#else + unsigned +#endif + THREAD_FUNC_RET_TYPE; + +#define THREAD_FUNC_CALL_TYPE MY_STD_CALL +#define THREAD_FUNC_DECL THREAD_FUNC_RET_TYPE THREAD_FUNC_CALL_TYPE +typedef THREAD_FUNC_RET_TYPE (THREAD_FUNC_CALL_TYPE * THREAD_FUNC_TYPE)(void *); +WRes Thread_Create(CThread *p, THREAD_FUNC_TYPE func, LPVOID param); + +typedef HANDLE CEvent; +typedef CEvent CAutoResetEvent; +typedef CEvent CManualResetEvent; +#define Event_Construct(p) *(p) = NULL +#define Event_IsCreated(p) (*(p) != NULL) +#define Event_Close(p) HandlePtr_Close(p) +#define Event_Wait(p) Handle_WaitObject(*(p)) +WRes Event_Set(CEvent *p); +WRes Event_Reset(CEvent *p); +WRes ManualResetEvent_Create(CManualResetEvent *p, int signaled); +WRes ManualResetEvent_CreateNotSignaled(CManualResetEvent *p); +WRes AutoResetEvent_Create(CAutoResetEvent *p, int signaled); +WRes AutoResetEvent_CreateNotSignaled(CAutoResetEvent *p); + +typedef HANDLE CSemaphore; +#define Semaphore_Construct(p) *(p) = NULL +#define Semaphore_IsCreated(p) (*(p) != NULL) +#define Semaphore_Close(p) HandlePtr_Close(p) +#define Semaphore_Wait(p) Handle_WaitObject(*(p)) +WRes Semaphore_Create(CSemaphore *p, UInt32 initCount, UInt32 maxCount); +WRes Semaphore_ReleaseN(CSemaphore *p, UInt32 num); +WRes Semaphore_Release1(CSemaphore *p); + +typedef CRITICAL_SECTION CCriticalSection; +WRes CriticalSection_Init(CCriticalSection *p); +#define CriticalSection_Delete(p) DeleteCriticalSection(p) +#define CriticalSection_Enter(p) EnterCriticalSection(p) +#define CriticalSection_Leave(p) LeaveCriticalSection(p) + +EXTERN_C_END + +#endif diff --git a/contrib/libs/lzmasdk/ya.make b/contrib/libs/lzmasdk/ya.make index db0a55788d..fd283a8089 100644 --- a/contrib/libs/lzmasdk/ya.make +++ b/contrib/libs/lzmasdk/ya.make @@ -18,20 +18,20 @@ NO_UTIL() SRCS( 7zStream.c - Aes.c - AesOpt.c + Aes.c + AesOpt.c Alloc.c - Bra.c - Bra86.c - BraIA64.c - CpuArch.c - LzFind.c - Lzma2Dec.c - Lzma2Enc.c + Bra.c + Bra86.c + BraIA64.c + CpuArch.c + LzFind.c + Lzma2Dec.c + Lzma2Enc.c LzmaDec.c LzmaEnc.c LzmaLib.c - Sha256.c + Sha256.c ) END() |