From 4497d7fdaa9b0b4520416885c75742bf857bed66 Mon Sep 17 00:00:00 2001 From: Cacodemon345 Date: Tue, 19 Sep 2023 01:42:53 +0600 Subject: [PATCH] Update LZMA SDK to 23.01 As a bonus, add support for XZ-compressed files in ZIP --- libraries/lzma/C/7z.h | 12 +- libraries/lzma/C/7zAlloc.c | 89 + libraries/lzma/C/7zAlloc.h | 19 + libraries/lzma/C/7zArcIn.c | 405 +-- libraries/lzma/C/7zBuf.h | 6 +- libraries/lzma/C/7zBuf2.c | 52 + libraries/lzma/C/7zCrc.c | 130 +- libraries/lzma/C/7zCrc.h | 14 +- libraries/lzma/C/7zCrcOpt.c | 16 +- libraries/lzma/C/7zDec.c | 172 +- libraries/lzma/C/7zFile.c | 443 +++ libraries/lzma/C/7zFile.h | 92 + libraries/lzma/C/7zStream.c | 67 +- libraries/lzma/C/7zTypes.h | 274 +- libraries/lzma/C/7zVersion.h | 10 +- libraries/lzma/C/7zWindows.h | 101 + libraries/lzma/C/Aes.c | 393 +++ libraries/lzma/C/Aes.h | 60 + libraries/lzma/C/AesOpt.c | 840 +++++ libraries/lzma/C/Alloc.c | 535 +++ libraries/lzma/C/Alloc.h | 71 + libraries/lzma/C/Bcj2.c | 325 +- libraries/lzma/C/Bcj2.h | 270 +- libraries/lzma/C/Bcj2Enc.c | 506 +++ libraries/lzma/C/Bra.c | 528 ++- libraries/lzma/C/Bra.h | 117 +- libraries/lzma/C/Bra86.c | 221 +- libraries/lzma/C/BraIA64.c | 57 +- libraries/lzma/C/Compiler.h | 162 +- libraries/lzma/C/CpuArch.c | 847 +++-- libraries/lzma/C/CpuArch.h | 243 +- libraries/lzma/C/Delta.h | 6 +- libraries/lzma/C/DllSecur.c | 111 + libraries/lzma/C/DllSecur.h | 20 + libraries/lzma/C/LzFind.c | 519 +-- libraries/lzma/C/LzFind.h | 53 +- libraries/lzma/C/LzFindMt.c | 70 +- libraries/lzma/C/LzFindMt.h | 10 +- libraries/lzma/C/LzFindOpt.c | 14 +- libraries/lzma/C/LzHash.h | 8 +- libraries/lzma/C/Lzma2Dec.c | 12 +- libraries/lzma/C/Lzma2Dec.h | 15 +- libraries/lzma/C/Lzma2DecMt.c | 1095 +++++++ libraries/lzma/C/Lzma2DecMt.h | 81 + libraries/lzma/C/Lzma2Enc.c | 805 +++++ libraries/lzma/C/Lzma2Enc.h | 57 + libraries/lzma/C/Lzma86.h | 111 + libraries/lzma/C/Lzma86Dec.c | 53 + libraries/lzma/C/Lzma86Enc.c | 103 + libraries/lzma/C/LzmaDec.c | 190 +- libraries/lzma/C/LzmaDec.h | 17 +- libraries/lzma/C/LzmaEnc.c | 389 ++- libraries/lzma/C/LzmaEnc.h | 23 +- libraries/lzma/C/LzmaLib.c | 42 + libraries/lzma/C/LzmaLib.h | 138 + libraries/lzma/C/MtCoder.c | 571 ++++ libraries/lzma/C/MtCoder.h | 141 + libraries/lzma/C/MtDec.c | 1114 +++++++ libraries/lzma/C/MtDec.h | 202 ++ libraries/lzma/C/Ppmd.h | 12 +- libraries/lzma/C/Ppmd7.c | 186 +- libraries/lzma/C/Ppmd7.h | 10 +- libraries/lzma/C/Ppmd7Dec.c | 57 +- libraries/lzma/C/Ppmd7Enc.c | 338 ++ libraries/lzma/C/Precomp.h | 6 +- libraries/lzma/C/RotateDefs.h | 50 + libraries/lzma/C/Sha256.c | 516 +++ libraries/lzma/C/Sha256.h | 76 + libraries/lzma/C/Sha256Opt.c | 386 +++ libraries/lzma/C/Sort.c | 141 + libraries/lzma/C/Sort.h | 18 + libraries/lzma/C/SwapBytes.c | 800 +++++ libraries/lzma/C/SwapBytes.h | 17 + libraries/lzma/C/Threads.c | 86 +- libraries/lzma/C/Threads.h | 50 +- libraries/lzma/C/Xz.c | 90 + libraries/lzma/C/Xz.h | 535 +++ libraries/lzma/C/XzCrc64.c | 80 + libraries/lzma/C/XzCrc64.h | 26 + libraries/lzma/C/XzCrc64Opt.c | 61 + libraries/lzma/C/XzDec.c | 2875 +++++++++++++++++ libraries/lzma/C/XzEnc.c | 1362 ++++++++ libraries/lzma/C/XzEnc.h | 61 + libraries/lzma/C/XzIn.c | 340 ++ libraries/lzma/CMakeLists.txt | 26 + src/common/filesystem/include/fs_files.h | 1 + src/common/filesystem/source/file_7z.cpp | 2 +- src/common/filesystem/source/file_zip.cpp | 4 +- .../filesystem/source/files_decompress.cpp | 151 + 89 files changed, 19298 insertions(+), 2112 deletions(-) create mode 100644 libraries/lzma/C/7zAlloc.c create mode 100644 libraries/lzma/C/7zAlloc.h create mode 100644 libraries/lzma/C/7zBuf2.c create mode 100644 libraries/lzma/C/7zFile.c create mode 100644 libraries/lzma/C/7zFile.h create mode 100644 libraries/lzma/C/7zWindows.h create mode 100644 libraries/lzma/C/Aes.c create mode 100644 libraries/lzma/C/Aes.h create mode 100644 libraries/lzma/C/AesOpt.c create mode 100644 libraries/lzma/C/Alloc.c create mode 100644 libraries/lzma/C/Alloc.h create mode 100644 libraries/lzma/C/Bcj2Enc.c create mode 100644 libraries/lzma/C/DllSecur.c create mode 100644 libraries/lzma/C/DllSecur.h create mode 100644 libraries/lzma/C/Lzma2DecMt.c create mode 100644 libraries/lzma/C/Lzma2DecMt.h create mode 100644 libraries/lzma/C/Lzma2Enc.c create mode 100644 libraries/lzma/C/Lzma2Enc.h create mode 100644 libraries/lzma/C/Lzma86.h create mode 100644 libraries/lzma/C/Lzma86Dec.c create mode 100644 libraries/lzma/C/Lzma86Enc.c create mode 100644 libraries/lzma/C/LzmaLib.c create mode 100644 libraries/lzma/C/LzmaLib.h create mode 100644 libraries/lzma/C/MtCoder.c create mode 100644 libraries/lzma/C/MtCoder.h create mode 100644 libraries/lzma/C/MtDec.c create mode 100644 libraries/lzma/C/MtDec.h create mode 100644 libraries/lzma/C/Ppmd7Enc.c create mode 100644 libraries/lzma/C/RotateDefs.h create mode 100644 libraries/lzma/C/Sha256.c create mode 100644 libraries/lzma/C/Sha256.h create mode 100644 libraries/lzma/C/Sha256Opt.c create mode 100644 libraries/lzma/C/Sort.c create mode 100644 libraries/lzma/C/Sort.h create mode 100644 libraries/lzma/C/SwapBytes.c create mode 100644 libraries/lzma/C/SwapBytes.h create mode 100644 libraries/lzma/C/Xz.c create mode 100644 libraries/lzma/C/Xz.h create mode 100644 libraries/lzma/C/XzCrc64.c create mode 100644 libraries/lzma/C/XzCrc64.h create mode 100644 libraries/lzma/C/XzCrc64Opt.c create mode 100644 libraries/lzma/C/XzDec.c create mode 100644 libraries/lzma/C/XzEnc.c create mode 100644 libraries/lzma/C/XzEnc.h create mode 100644 libraries/lzma/C/XzIn.c diff --git a/libraries/lzma/C/7z.h b/libraries/lzma/C/7z.h index 304f75ffc..9e27c0152 100644 --- a/libraries/lzma/C/7z.h +++ b/libraries/lzma/C/7z.h @@ -1,8 +1,8 @@ /* 7z.h -- 7z interface -2018-07-02 : Igor Pavlov : Public domain */ +2023-04-02 : Igor Pavlov : Public domain */ -#ifndef __7Z_H -#define __7Z_H +#ifndef ZIP7_INC_7Z_H +#define ZIP7_INC_7Z_H #include "7zTypes.h" @@ -98,7 +98,7 @@ typedef struct UInt64 SzAr_GetFolderUnpackSize(const CSzAr *p, UInt32 folderIndex); SRes SzAr_DecodeFolder(const CSzAr *p, UInt32 folderIndex, - ILookInStream *stream, UInt64 startPos, + ILookInStreamPtr stream, UInt64 startPos, Byte *outBuffer, size_t outSize, ISzAllocPtr allocMain); @@ -174,7 +174,7 @@ UInt16 *SzArEx_GetFullNameUtf16_Back(const CSzArEx *p, size_t fileIndex, UInt16 SRes SzArEx_Extract( const CSzArEx *db, - ILookInStream *inStream, + ILookInStreamPtr inStream, UInt32 fileIndex, /* index of file */ UInt32 *blockIndex, /* index of solid block */ Byte **outBuffer, /* pointer to pointer to output buffer (allocated with allocMain) */ @@ -196,7 +196,7 @@ SZ_ERROR_INPUT_EOF SZ_ERROR_FAIL */ -SRes SzArEx_Open(CSzArEx *p, ILookInStream *inStream, +SRes SzArEx_Open(CSzArEx *p, ILookInStreamPtr inStream, ISzAllocPtr allocMain, ISzAllocPtr allocTemp); EXTERN_C_END diff --git a/libraries/lzma/C/7zAlloc.c b/libraries/lzma/C/7zAlloc.c new file mode 100644 index 000000000..2f0659af6 --- /dev/null +++ b/libraries/lzma/C/7zAlloc.c @@ -0,0 +1,89 @@ +/* 7zAlloc.c -- Allocation functions for 7z processing +2023-03-04 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include + +#include "7zAlloc.h" + +/* #define SZ_ALLOC_DEBUG */ +/* use SZ_ALLOC_DEBUG to debug alloc/free operations */ + +#ifdef SZ_ALLOC_DEBUG + +/* +#ifdef _WIN32 +#include "7zWindows.h" +#endif +*/ + +#include +static int g_allocCount = 0; +static int g_allocCountTemp = 0; + +static void Print_Alloc(const char *s, size_t size, int *counter) +{ + const unsigned size2 = (unsigned)size; + fprintf(stderr, "\n%s count = %10d : %10u bytes; ", s, *counter, size2); + (*counter)++; +} +static void Print_Free(const char *s, int *counter) +{ + (*counter)--; + fprintf(stderr, "\n%s count = %10d", s, *counter); +} +#endif + +void *SzAlloc(ISzAllocPtr p, size_t size) +{ + UNUSED_VAR(p) + if (size == 0) + return 0; + #ifdef SZ_ALLOC_DEBUG + Print_Alloc("Alloc", size, &g_allocCount); + #endif + return malloc(size); +} + +void SzFree(ISzAllocPtr p, void *address) +{ + UNUSED_VAR(p) + #ifdef SZ_ALLOC_DEBUG + if (address) + Print_Free("Free ", &g_allocCount); + #endif + free(address); +} + +void *SzAllocTemp(ISzAllocPtr p, size_t size) +{ + UNUSED_VAR(p) + if (size == 0) + return 0; + #ifdef SZ_ALLOC_DEBUG + Print_Alloc("Alloc_temp", size, &g_allocCountTemp); + /* + #ifdef _WIN32 + return HeapAlloc(GetProcessHeap(), 0, size); + #endif + */ + #endif + return malloc(size); +} + +void SzFreeTemp(ISzAllocPtr p, void *address) +{ + UNUSED_VAR(p) + #ifdef SZ_ALLOC_DEBUG + if (address) + Print_Free("Free_temp ", &g_allocCountTemp); + /* + #ifdef _WIN32 + HeapFree(GetProcessHeap(), 0, address); + return; + #endif + */ + #endif + free(address); +} diff --git a/libraries/lzma/C/7zAlloc.h b/libraries/lzma/C/7zAlloc.h new file mode 100644 index 000000000..b2b8b0cdd --- /dev/null +++ b/libraries/lzma/C/7zAlloc.h @@ -0,0 +1,19 @@ +/* 7zAlloc.h -- Allocation functions +2023-03-04 : Igor Pavlov : Public domain */ + +#ifndef ZIP7_INC_7Z_ALLOC_H +#define ZIP7_INC_7Z_ALLOC_H + +#include "7zTypes.h" + +EXTERN_C_BEGIN + +void *SzAlloc(ISzAllocPtr p, size_t size); +void SzFree(ISzAllocPtr p, void *address); + +void *SzAllocTemp(ISzAllocPtr p, size_t size); +void SzFreeTemp(ISzAllocPtr p, void *address); + +EXTERN_C_END + +#endif diff --git a/libraries/lzma/C/7zArcIn.c b/libraries/lzma/C/7zArcIn.c index 0d9dec41e..43fa7c214 100644 --- a/libraries/lzma/C/7zArcIn.c +++ b/libraries/lzma/C/7zArcIn.c @@ -1,5 +1,5 @@ /* 7zArcIn.c -- 7z Input functions -2021-02-09 : Igor Pavlov : Public domain */ +2023-05-11 : Igor Pavlov : Public domain */ #include "Precomp.h" @@ -10,10 +10,11 @@ #include "7zCrc.h" #include "CpuArch.h" -#define MY_ALLOC(T, p, size, alloc) { \ - if ((p = (T *)ISzAlloc_Alloc(alloc, (size) * sizeof(T))) == NULL) return SZ_ERROR_MEM; } +#define MY_ALLOC(T, p, size, alloc) \ + { if ((p = (T *)ISzAlloc_Alloc(alloc, (size) * sizeof(T))) == NULL) return SZ_ERROR_MEM; } -#define MY_ALLOC_ZE(T, p, size, alloc) { if ((size) == 0) p = NULL; else MY_ALLOC(T, p, size, alloc) } +#define MY_ALLOC_ZE(T, p, size, alloc) \ + { if ((size) == 0) p = NULL; else MY_ALLOC(T, p, size, alloc) } #define MY_ALLOC_AND_CPY(to, size, from, alloc) \ { MY_ALLOC(Byte, to, size, alloc); memcpy(to, from, size); } @@ -58,7 +59,7 @@ enum EIdEnum const Byte k7zSignature[k7zSignatureSize] = {'7', 'z', 0xBC, 0xAF, 0x27, 0x1C}; -#define SzBitUi32s_Init(p) { (p)->Defs = NULL; (p)->Vals = NULL; } +#define SzBitUi32s_INIT(p) { (p)->Defs = NULL; (p)->Vals = NULL; } static SRes SzBitUi32s_Alloc(CSzBitUi32s *p, size_t num, ISzAllocPtr alloc) { @@ -69,8 +70,8 @@ static SRes SzBitUi32s_Alloc(CSzBitUi32s *p, size_t num, ISzAllocPtr alloc) } else { - MY_ALLOC(Byte, p->Defs, (num + 7) >> 3, alloc); - MY_ALLOC(UInt32, p->Vals, num, alloc); + MY_ALLOC(Byte, p->Defs, (num + 7) >> 3, alloc) + MY_ALLOC(UInt32, p->Vals, num, alloc) } return SZ_OK; } @@ -81,7 +82,7 @@ static void SzBitUi32s_Free(CSzBitUi32s *p, ISzAllocPtr alloc) ISzAlloc_Free(alloc, p->Vals); p->Vals = NULL; } -#define SzBitUi64s_Init(p) { (p)->Defs = NULL; (p)->Vals = NULL; } +#define SzBitUi64s_INIT(p) { (p)->Defs = NULL; (p)->Vals = NULL; } static void SzBitUi64s_Free(CSzBitUi64s *p, ISzAllocPtr alloc) { @@ -96,7 +97,7 @@ static void SzAr_Init(CSzAr *p) p->NumFolders = 0; p->PackPositions = NULL; - SzBitUi32s_Init(&p->FolderCRCs); + SzBitUi32s_INIT(&p->FolderCRCs) p->FoCodersOffsets = NULL; p->FoStartPackStreamIndex = NULL; @@ -142,11 +143,11 @@ void SzArEx_Init(CSzArEx *p) p->FileNameOffsets = NULL; p->FileNames = NULL; - SzBitUi32s_Init(&p->CRCs); - SzBitUi32s_Init(&p->Attribs); - // SzBitUi32s_Init(&p->Parents); - SzBitUi64s_Init(&p->MTime); - SzBitUi64s_Init(&p->CTime); + SzBitUi32s_INIT(&p->CRCs) + SzBitUi32s_INIT(&p->Attribs) + // SzBitUi32s_INIT(&p->Parents) + SzBitUi64s_INIT(&p->MTime) + SzBitUi64s_INIT(&p->CTime) } void SzArEx_Free(CSzArEx *p, ISzAllocPtr alloc) @@ -180,11 +181,20 @@ static int TestSignatureCandidate(const Byte *testBytes) return 1; } -#define SzData_Clear(p) { (p)->Data = NULL; (p)->Size = 0; } +#define SzData_CLEAR(p) { (p)->Data = NULL; (p)->Size = 0; } + +#define SZ_READ_BYTE_SD_NOCHECK(_sd_, dest) \ + (_sd_)->Size--; dest = *(_sd_)->Data++; + +#define SZ_READ_BYTE_SD(_sd_, dest) \ + if ((_sd_)->Size == 0) return SZ_ERROR_ARCHIVE; \ + SZ_READ_BYTE_SD_NOCHECK(_sd_, dest) -#define SZ_READ_BYTE_SD(_sd_, dest) if ((_sd_)->Size == 0) return SZ_ERROR_ARCHIVE; (_sd_)->Size--; dest = *(_sd_)->Data++; #define SZ_READ_BYTE(dest) SZ_READ_BYTE_SD(sd, dest) -#define SZ_READ_BYTE_2(dest) if (sd.Size == 0) return SZ_ERROR_ARCHIVE; sd.Size--; dest = *sd.Data++; + +#define SZ_READ_BYTE_2(dest) \ + if (sd.Size == 0) return SZ_ERROR_ARCHIVE; \ + sd.Size--; dest = *sd.Data++; #define SKIP_DATA(sd, size) { sd->Size -= (size_t)(size); sd->Data += (size_t)(size); } #define SKIP_DATA2(sd, size) { sd.Size -= (size_t)(size); sd.Data += (size_t)(size); } @@ -192,25 +202,25 @@ static int TestSignatureCandidate(const Byte *testBytes) #define SZ_READ_32(dest) if (sd.Size < 4) return SZ_ERROR_ARCHIVE; \ dest = GetUi32(sd.Data); SKIP_DATA2(sd, 4); -static MY_NO_INLINE SRes ReadNumber(CSzData *sd, UInt64 *value) +static Z7_NO_INLINE SRes ReadNumber(CSzData *sd, UInt64 *value) { Byte firstByte, mask; unsigned i; UInt32 v; - SZ_READ_BYTE(firstByte); + SZ_READ_BYTE(firstByte) if ((firstByte & 0x80) == 0) { *value = firstByte; return SZ_OK; } - SZ_READ_BYTE(v); + SZ_READ_BYTE(v) if ((firstByte & 0x40) == 0) { *value = (((UInt32)firstByte & 0x3F) << 8) | v; return SZ_OK; } - SZ_READ_BYTE(mask); + SZ_READ_BYTE(mask) *value = v | ((UInt32)mask << 8); mask = 0x20; for (i = 2; i < 8; i++) @@ -218,11 +228,11 @@ static MY_NO_INLINE SRes ReadNumber(CSzData *sd, UInt64 *value) Byte b; if ((firstByte & mask) == 0) { - UInt64 highPart = (unsigned)firstByte & (unsigned)(mask - 1); + const UInt64 highPart = (unsigned)firstByte & (unsigned)(mask - 1); *value |= (highPart << (8 * i)); return SZ_OK; } - SZ_READ_BYTE(b); + SZ_READ_BYTE(b) *value |= ((UInt64)b << (8 * i)); mask >>= 1; } @@ -230,7 +240,7 @@ static MY_NO_INLINE SRes ReadNumber(CSzData *sd, UInt64 *value) } -static MY_NO_INLINE SRes SzReadNumber32(CSzData *sd, UInt32 *value) +static Z7_NO_INLINE SRes SzReadNumber32(CSzData *sd, UInt32 *value) { Byte firstByte; UInt64 value64; @@ -244,7 +254,7 @@ static MY_NO_INLINE SRes SzReadNumber32(CSzData *sd, UInt32 *value) sd->Size--; return SZ_OK; } - RINOK(ReadNumber(sd, &value64)); + RINOK(ReadNumber(sd, &value64)) if (value64 >= (UInt32)0x80000000 - 1) return SZ_ERROR_UNSUPPORTED; if (value64 >= ((UInt64)(1) << ((sizeof(size_t) - 1) * 8 + 4))) @@ -258,10 +268,10 @@ static MY_NO_INLINE SRes SzReadNumber32(CSzData *sd, UInt32 *value) static SRes SkipData(CSzData *sd) { UInt64 size; - RINOK(ReadNumber(sd, &size)); + RINOK(ReadNumber(sd, &size)) if (size > sd->Size) return SZ_ERROR_ARCHIVE; - SKIP_DATA(sd, size); + SKIP_DATA(sd, size) return SZ_OK; } @@ -270,22 +280,22 @@ static SRes WaitId(CSzData *sd, UInt32 id) for (;;) { UInt64 type; - RINOK(ReadID(sd, &type)); + RINOK(ReadID(sd, &type)) if (type == id) return SZ_OK; if (type == k7zIdEnd) return SZ_ERROR_ARCHIVE; - RINOK(SkipData(sd)); + RINOK(SkipData(sd)) } } static SRes RememberBitVector(CSzData *sd, UInt32 numItems, const Byte **v) { - UInt32 numBytes = (numItems + 7) >> 3; + const UInt32 numBytes = (numItems + 7) >> 3; if (numBytes > sd->Size) return SZ_ERROR_ARCHIVE; *v = sd->Data; - SKIP_DATA(sd, numBytes); + SKIP_DATA(sd, numBytes) return SZ_OK; } @@ -307,48 +317,48 @@ static UInt32 CountDefinedBits(const Byte *bits, UInt32 numItems) return sum; } -static MY_NO_INLINE SRes ReadBitVector(CSzData *sd, UInt32 numItems, Byte **v, ISzAllocPtr alloc) +static Z7_NO_INLINE SRes ReadBitVector(CSzData *sd, UInt32 numItems, Byte **v, ISzAllocPtr alloc) { Byte allAreDefined; Byte *v2; - UInt32 numBytes = (numItems + 7) >> 3; + const UInt32 numBytes = (numItems + 7) >> 3; *v = NULL; - SZ_READ_BYTE(allAreDefined); + SZ_READ_BYTE(allAreDefined) if (numBytes == 0) return SZ_OK; if (allAreDefined == 0) { if (numBytes > sd->Size) return SZ_ERROR_ARCHIVE; - MY_ALLOC_AND_CPY(*v, numBytes, sd->Data, alloc); - SKIP_DATA(sd, numBytes); + MY_ALLOC_AND_CPY(*v, numBytes, sd->Data, alloc) + SKIP_DATA(sd, numBytes) return SZ_OK; } - MY_ALLOC(Byte, *v, numBytes, alloc); + MY_ALLOC(Byte, *v, numBytes, alloc) v2 = *v; memset(v2, 0xFF, (size_t)numBytes); { - unsigned numBits = (unsigned)numItems & 7; + const unsigned numBits = (unsigned)numItems & 7; if (numBits != 0) v2[(size_t)numBytes - 1] = (Byte)((((UInt32)1 << numBits) - 1) << (8 - numBits)); } return SZ_OK; } -static MY_NO_INLINE SRes ReadUi32s(CSzData *sd2, UInt32 numItems, CSzBitUi32s *crcs, ISzAllocPtr alloc) +static Z7_NO_INLINE SRes ReadUi32s(CSzData *sd2, UInt32 numItems, CSzBitUi32s *crcs, ISzAllocPtr alloc) { UInt32 i; CSzData sd; UInt32 *vals; const Byte *defs; - MY_ALLOC_ZE(UInt32, crcs->Vals, numItems, alloc); + MY_ALLOC_ZE(UInt32, crcs->Vals, numItems, alloc) sd = *sd2; defs = crcs->Defs; vals = crcs->Vals; for (i = 0; i < numItems; i++) if (SzBitArray_Check(defs, i)) { - SZ_READ_32(vals[i]); + SZ_READ_32(vals[i]) } else vals[i] = 0; @@ -359,7 +369,7 @@ static MY_NO_INLINE SRes ReadUi32s(CSzData *sd2, UInt32 numItems, CSzBitUi32s *c static SRes ReadBitUi32s(CSzData *sd, UInt32 numItems, CSzBitUi32s *crcs, ISzAllocPtr alloc) { SzBitUi32s_Free(crcs, alloc); - RINOK(ReadBitVector(sd, numItems, &crcs->Defs, alloc)); + RINOK(ReadBitVector(sd, numItems, &crcs->Defs, alloc)) return ReadUi32s(sd, numItems, crcs, alloc); } @@ -367,36 +377,36 @@ static SRes SkipBitUi32s(CSzData *sd, UInt32 numItems) { Byte allAreDefined; UInt32 numDefined = numItems; - SZ_READ_BYTE(allAreDefined); + SZ_READ_BYTE(allAreDefined) if (!allAreDefined) { - size_t numBytes = (numItems + 7) >> 3; + const size_t numBytes = (numItems + 7) >> 3; if (numBytes > sd->Size) return SZ_ERROR_ARCHIVE; numDefined = CountDefinedBits(sd->Data, numItems); - SKIP_DATA(sd, numBytes); + SKIP_DATA(sd, numBytes) } if (numDefined > (sd->Size >> 2)) return SZ_ERROR_ARCHIVE; - SKIP_DATA(sd, (size_t)numDefined * 4); + SKIP_DATA(sd, (size_t)numDefined * 4) return SZ_OK; } static SRes ReadPackInfo(CSzAr *p, CSzData *sd, ISzAllocPtr alloc) { - RINOK(SzReadNumber32(sd, &p->NumPackStreams)); + RINOK(SzReadNumber32(sd, &p->NumPackStreams)) - RINOK(WaitId(sd, k7zIdSize)); - MY_ALLOC(UInt64, p->PackPositions, (size_t)p->NumPackStreams + 1, alloc); + RINOK(WaitId(sd, k7zIdSize)) + MY_ALLOC(UInt64, p->PackPositions, (size_t)p->NumPackStreams + 1, alloc) { UInt64 sum = 0; UInt32 i; - UInt32 numPackStreams = p->NumPackStreams; + const UInt32 numPackStreams = p->NumPackStreams; for (i = 0; i < numPackStreams; i++) { UInt64 packSize; p->PackPositions[i] = sum; - RINOK(ReadNumber(sd, &packSize)); + RINOK(ReadNumber(sd, &packSize)) sum += packSize; if (sum < packSize) return SZ_ERROR_ARCHIVE; @@ -407,16 +417,16 @@ static SRes ReadPackInfo(CSzAr *p, CSzData *sd, ISzAllocPtr alloc) for (;;) { UInt64 type; - RINOK(ReadID(sd, &type)); + RINOK(ReadID(sd, &type)) if (type == k7zIdEnd) return SZ_OK; if (type == k7zIdCRC) { /* CRC of packed streams is unused now */ - RINOK(SkipBitUi32s(sd, p->NumPackStreams)); + RINOK(SkipBitUi32s(sd, p->NumPackStreams)) continue; } - RINOK(SkipData(sd)); + RINOK(SkipData(sd)) } } @@ -442,7 +452,7 @@ SRes SzGetNextFolderItem(CSzFolder *f, CSzData *sd) f->NumPackStreams = 0; f->UnpackStream = 0; - RINOK(SzReadNumber32(sd, &numCoders)); + RINOK(SzReadNumber32(sd, &numCoders)) if (numCoders == 0 || numCoders > SZ_NUM_CODERS_IN_FOLDER_MAX) return SZ_ERROR_UNSUPPORTED; @@ -453,7 +463,7 @@ SRes SzGetNextFolderItem(CSzFolder *f, CSzData *sd) unsigned idSize, j; UInt64 id; - SZ_READ_BYTE(mainByte); + SZ_READ_BYTE(mainByte) if ((mainByte & 0xC0) != 0) return SZ_ERROR_UNSUPPORTED; @@ -481,12 +491,12 @@ SRes SzGetNextFolderItem(CSzFolder *f, CSzData *sd) { UInt32 numStreams; - RINOK(SzReadNumber32(sd, &numStreams)); + RINOK(SzReadNumber32(sd, &numStreams)) if (numStreams > k_NumCodersStreams_in_Folder_MAX) return SZ_ERROR_UNSUPPORTED; coder->NumStreams = (Byte)numStreams; - RINOK(SzReadNumber32(sd, &numStreams)); + RINOK(SzReadNumber32(sd, &numStreams)) if (numStreams != 1) return SZ_ERROR_UNSUPPORTED; } @@ -499,7 +509,7 @@ SRes SzGetNextFolderItem(CSzFolder *f, CSzData *sd) if ((mainByte & 0x20) != 0) { UInt32 propsSize = 0; - RINOK(SzReadNumber32(sd, &propsSize)); + RINOK(SzReadNumber32(sd, &propsSize)) if (propsSize > sd->Size) return SZ_ERROR_ARCHIVE; if (propsSize >= 0x80) @@ -549,12 +559,12 @@ SRes SzGetNextFolderItem(CSzFolder *f, CSzData *sd) { CSzBond *bp = f->Bonds + i; - RINOK(SzReadNumber32(sd, &bp->InIndex)); + RINOK(SzReadNumber32(sd, &bp->InIndex)) if (bp->InIndex >= numInStreams || streamUsed[bp->InIndex]) return SZ_ERROR_ARCHIVE; streamUsed[bp->InIndex] = True; - RINOK(SzReadNumber32(sd, &bp->OutIndex)); + RINOK(SzReadNumber32(sd, &bp->OutIndex)) if (bp->OutIndex >= numCoders || coderUsed[bp->OutIndex]) return SZ_ERROR_ARCHIVE; coderUsed[bp->OutIndex] = True; @@ -584,7 +594,7 @@ SRes SzGetNextFolderItem(CSzFolder *f, CSzData *sd) for (i = 0; i < numPackStreams; i++) { UInt32 index; - RINOK(SzReadNumber32(sd, &index)); + RINOK(SzReadNumber32(sd, &index)) if (index >= numInStreams || streamUsed[index]) return SZ_ERROR_ARCHIVE; streamUsed[index] = True; @@ -598,7 +608,7 @@ SRes SzGetNextFolderItem(CSzFolder *f, CSzData *sd) } -static MY_NO_INLINE SRes SkipNumbers(CSzData *sd2, UInt32 num) +static Z7_NO_INLINE SRes SkipNumbers(CSzData *sd2, UInt32 num) { CSzData sd; sd = *sd2; @@ -606,7 +616,7 @@ static MY_NO_INLINE SRes SkipNumbers(CSzData *sd2, UInt32 num) { Byte firstByte, mask; unsigned i; - SZ_READ_BYTE_2(firstByte); + SZ_READ_BYTE_2(firstByte) if ((firstByte & 0x80) == 0) continue; if ((firstByte & 0x40) == 0) @@ -622,7 +632,7 @@ static MY_NO_INLINE SRes SkipNumbers(CSzData *sd2, UInt32 num) mask >>= 1; if (i > sd.Size) return SZ_ERROR_ARCHIVE; - SKIP_DATA2(sd, i); + SKIP_DATA2(sd, i) } *sd2 = sd; return SZ_OK; @@ -645,30 +655,30 @@ static SRes ReadUnpackInfo(CSzAr *p, const Byte *startBufPtr; Byte external; - RINOK(WaitId(sd2, k7zIdFolder)); + RINOK(WaitId(sd2, k7zIdFolder)) - RINOK(SzReadNumber32(sd2, &numFolders)); + RINOK(SzReadNumber32(sd2, &numFolders)) if (numFolders > numFoldersMax) return SZ_ERROR_UNSUPPORTED; p->NumFolders = numFolders; - SZ_READ_BYTE_SD(sd2, external); + SZ_READ_BYTE_SD(sd2, external) if (external == 0) sd = *sd2; else { UInt32 index; - RINOK(SzReadNumber32(sd2, &index)); + RINOK(SzReadNumber32(sd2, &index)) if (index >= numTempBufs) return SZ_ERROR_ARCHIVE; sd.Data = tempBufs[index].data; sd.Size = tempBufs[index].size; } - MY_ALLOC(size_t, p->FoCodersOffsets, (size_t)numFolders + 1, alloc); - MY_ALLOC(UInt32, p->FoStartPackStreamIndex, (size_t)numFolders + 1, alloc); - MY_ALLOC(UInt32, p->FoToCoderUnpackSizes, (size_t)numFolders + 1, alloc); - MY_ALLOC_ZE(Byte, p->FoToMainUnpackSizeIndex, (size_t)numFolders, alloc); + MY_ALLOC(size_t, p->FoCodersOffsets, (size_t)numFolders + 1, alloc) + MY_ALLOC(UInt32, p->FoStartPackStreamIndex, (size_t)numFolders + 1, alloc) + MY_ALLOC(UInt32, p->FoToCoderUnpackSizes, (size_t)numFolders + 1, alloc) + MY_ALLOC_ZE(Byte, p->FoToMainUnpackSizeIndex, (size_t)numFolders, alloc) startBufPtr = sd.Data; @@ -681,7 +691,7 @@ static SRes ReadUnpackInfo(CSzAr *p, p->FoCodersOffsets[fo] = (size_t)(sd.Data - startBufPtr); - RINOK(SzReadNumber32(&sd, &numCoders)); + RINOK(SzReadNumber32(&sd, &numCoders)) if (numCoders == 0 || numCoders > k_Scan_NumCoders_MAX) return SZ_ERROR_UNSUPPORTED; @@ -691,7 +701,7 @@ static SRes ReadUnpackInfo(CSzAr *p, unsigned idSize; UInt32 coderInStreams; - SZ_READ_BYTE_2(mainByte); + SZ_READ_BYTE_2(mainByte) if ((mainByte & 0xC0) != 0) return SZ_ERROR_UNSUPPORTED; idSize = (mainByte & 0xF); @@ -699,15 +709,15 @@ static SRes ReadUnpackInfo(CSzAr *p, return SZ_ERROR_UNSUPPORTED; if (idSize > sd.Size) return SZ_ERROR_ARCHIVE; - SKIP_DATA2(sd, idSize); + SKIP_DATA2(sd, idSize) coderInStreams = 1; if ((mainByte & 0x10) != 0) { UInt32 coderOutStreams; - RINOK(SzReadNumber32(&sd, &coderInStreams)); - RINOK(SzReadNumber32(&sd, &coderOutStreams)); + RINOK(SzReadNumber32(&sd, &coderInStreams)) + RINOK(SzReadNumber32(&sd, &coderOutStreams)) if (coderInStreams > k_Scan_NumCodersStreams_in_Folder_MAX || coderOutStreams != 1) return SZ_ERROR_UNSUPPORTED; } @@ -717,10 +727,10 @@ static SRes ReadUnpackInfo(CSzAr *p, if ((mainByte & 0x20) != 0) { UInt32 propsSize; - RINOK(SzReadNumber32(&sd, &propsSize)); + RINOK(SzReadNumber32(&sd, &propsSize)) if (propsSize > sd.Size) return SZ_ERROR_ARCHIVE; - SKIP_DATA2(sd, propsSize); + SKIP_DATA2(sd, propsSize) } } @@ -734,7 +744,7 @@ static SRes ReadUnpackInfo(CSzAr *p, Byte coderUsed[k_Scan_NumCoders_MAX]; UInt32 i; - UInt32 numBonds = numCoders - 1; + const UInt32 numBonds = numCoders - 1; if (numInStreams < numBonds) return SZ_ERROR_ARCHIVE; @@ -750,12 +760,12 @@ static SRes ReadUnpackInfo(CSzAr *p, { UInt32 index; - RINOK(SzReadNumber32(&sd, &index)); + RINOK(SzReadNumber32(&sd, &index)) if (index >= numInStreams || streamUsed[index]) return SZ_ERROR_ARCHIVE; streamUsed[index] = True; - RINOK(SzReadNumber32(&sd, &index)); + RINOK(SzReadNumber32(&sd, &index)) if (index >= numCoders || coderUsed[index]) return SZ_ERROR_ARCHIVE; coderUsed[index] = True; @@ -767,7 +777,7 @@ static SRes ReadUnpackInfo(CSzAr *p, for (i = 0; i < numPackStreams; i++) { UInt32 index; - RINOK(SzReadNumber32(&sd, &index)); + RINOK(SzReadNumber32(&sd, &index)) if (index >= numInStreams || streamUsed[index]) return SZ_ERROR_ARCHIVE; streamUsed[index] = True; @@ -802,7 +812,7 @@ static SRes ReadUnpackInfo(CSzAr *p, const size_t dataSize = (size_t)(sd.Data - startBufPtr); p->FoStartPackStreamIndex[fo] = packStreamIndex; p->FoCodersOffsets[fo] = dataSize; - MY_ALLOC_ZE_AND_CPY(p->CodersData, dataSize, startBufPtr, alloc); + MY_ALLOC_ZE_AND_CPY(p->CodersData, dataSize, startBufPtr, alloc) } if (external != 0) @@ -812,21 +822,21 @@ static SRes ReadUnpackInfo(CSzAr *p, sd = *sd2; } - RINOK(WaitId(&sd, k7zIdCodersUnpackSize)); + RINOK(WaitId(&sd, k7zIdCodersUnpackSize)) - MY_ALLOC_ZE(UInt64, p->CoderUnpackSizes, (size_t)numCodersOutStreams, alloc); + MY_ALLOC_ZE(UInt64, p->CoderUnpackSizes, (size_t)numCodersOutStreams, alloc) { UInt32 i; for (i = 0; i < numCodersOutStreams; i++) { - RINOK(ReadNumber(&sd, p->CoderUnpackSizes + i)); + RINOK(ReadNumber(&sd, p->CoderUnpackSizes + i)) } } for (;;) { UInt64 type; - RINOK(ReadID(&sd, &type)); + RINOK(ReadID(&sd, &type)) if (type == k7zIdEnd) { *sd2 = sd; @@ -834,10 +844,10 @@ static SRes ReadUnpackInfo(CSzAr *p, } if (type == k7zIdCRC) { - RINOK(ReadBitUi32s(&sd, numFolders, &p->FolderCRCs, alloc)); + RINOK(ReadBitUi32s(&sd, numFolders, &p->FolderCRCs, alloc)) continue; } - RINOK(SkipData(&sd)); + RINOK(SkipData(&sd)) } } @@ -862,13 +872,13 @@ static SRes ReadSubStreamsInfo(CSzAr *p, CSzData *sd, CSubStreamInfo *ssi) { UInt64 type = 0; UInt32 numSubDigests = 0; - UInt32 numFolders = p->NumFolders; + const UInt32 numFolders = p->NumFolders; UInt32 numUnpackStreams = numFolders; UInt32 numUnpackSizesInData = 0; for (;;) { - RINOK(ReadID(sd, &type)); + RINOK(ReadID(sd, &type)) if (type == k7zIdNumUnpackStream) { UInt32 i; @@ -878,7 +888,7 @@ static SRes ReadSubStreamsInfo(CSzAr *p, CSzData *sd, CSubStreamInfo *ssi) for (i = 0; i < numFolders; i++) { UInt32 numStreams; - RINOK(SzReadNumber32(sd, &numStreams)); + RINOK(SzReadNumber32(sd, &numStreams)) if (numUnpackStreams > numUnpackStreams + numStreams) return SZ_ERROR_UNSUPPORTED; numUnpackStreams += numStreams; @@ -892,7 +902,7 @@ static SRes ReadSubStreamsInfo(CSzAr *p, CSzData *sd, CSubStreamInfo *ssi) } if (type == k7zIdCRC || type == k7zIdSize || type == k7zIdEnd) break; - RINOK(SkipData(sd)); + RINOK(SkipData(sd)) } if (!ssi->sdNumSubStreams.Data) @@ -908,9 +918,9 @@ static SRes ReadSubStreamsInfo(CSzAr *p, CSzData *sd, CSubStreamInfo *ssi) if (type == k7zIdSize) { ssi->sdSizes.Data = sd->Data; - RINOK(SkipNumbers(sd, numUnpackSizesInData)); + RINOK(SkipNumbers(sd, numUnpackSizesInData)) ssi->sdSizes.Size = (size_t)(sd->Data - ssi->sdSizes.Data); - RINOK(ReadID(sd, &type)); + RINOK(ReadID(sd, &type)) } for (;;) @@ -920,14 +930,14 @@ static SRes ReadSubStreamsInfo(CSzAr *p, CSzData *sd, CSubStreamInfo *ssi) if (type == k7zIdCRC) { ssi->sdCRCs.Data = sd->Data; - RINOK(SkipBitUi32s(sd, numSubDigests)); + RINOK(SkipBitUi32s(sd, numSubDigests)) ssi->sdCRCs.Size = (size_t)(sd->Data - ssi->sdCRCs.Data); } else { - RINOK(SkipData(sd)); + RINOK(SkipData(sd)) } - RINOK(ReadID(sd, &type)); + RINOK(ReadID(sd, &type)) } } @@ -940,31 +950,31 @@ static SRes SzReadStreamsInfo(CSzAr *p, { UInt64 type; - SzData_Clear(&ssi->sdSizes); - SzData_Clear(&ssi->sdCRCs); - SzData_Clear(&ssi->sdNumSubStreams); + SzData_CLEAR(&ssi->sdSizes) + SzData_CLEAR(&ssi->sdCRCs) + SzData_CLEAR(&ssi->sdNumSubStreams) *dataOffset = 0; - RINOK(ReadID(sd, &type)); + RINOK(ReadID(sd, &type)) if (type == k7zIdPackInfo) { - RINOK(ReadNumber(sd, dataOffset)); + RINOK(ReadNumber(sd, dataOffset)) if (*dataOffset > p->RangeLimit) return SZ_ERROR_ARCHIVE; - RINOK(ReadPackInfo(p, sd, alloc)); + RINOK(ReadPackInfo(p, sd, alloc)) if (p->PackPositions[p->NumPackStreams] > p->RangeLimit - *dataOffset) return SZ_ERROR_ARCHIVE; - RINOK(ReadID(sd, &type)); + RINOK(ReadID(sd, &type)) } if (type == k7zIdUnpackInfo) { - RINOK(ReadUnpackInfo(p, sd, numFoldersMax, tempBufs, numTempBufs, alloc)); - RINOK(ReadID(sd, &type)); + RINOK(ReadUnpackInfo(p, sd, numFoldersMax, tempBufs, numTempBufs, alloc)) + RINOK(ReadID(sd, &type)) } if (type == k7zIdSubStreamsInfo) { - RINOK(ReadSubStreamsInfo(p, sd, ssi)); - RINOK(ReadID(sd, &type)); + RINOK(ReadSubStreamsInfo(p, sd, ssi)) + RINOK(ReadID(sd, &type)) } else { @@ -976,7 +986,7 @@ static SRes SzReadStreamsInfo(CSzAr *p, } static SRes SzReadAndDecodePackedStreams( - ILookInStream *inStream, + ILookInStreamPtr inStream, CSzData *sd, CBuf *tempBufs, UInt32 numFoldersMax, @@ -988,7 +998,7 @@ static SRes SzReadAndDecodePackedStreams( UInt32 fo; CSubStreamInfo ssi; - RINOK(SzReadStreamsInfo(p, sd, numFoldersMax, NULL, 0, &dataStartPos, &ssi, allocTemp)); + RINOK(SzReadStreamsInfo(p, sd, numFoldersMax, NULL, 0, &dataStartPos, &ssi, allocTemp)) dataStartPos += baseOffset; if (p->NumFolders == 0) @@ -1000,7 +1010,7 @@ static SRes SzReadAndDecodePackedStreams( for (fo = 0; fo < p->NumFolders; fo++) { CBuf *tempBuf = tempBufs + fo; - UInt64 unpackSize = SzAr_GetFolderUnpackSize(p, fo); + const UInt64 unpackSize = SzAr_GetFolderUnpackSize(p, fo); if ((size_t)unpackSize != unpackSize) return SZ_ERROR_MEM; if (!Buf_Create(tempBuf, (size_t)unpackSize, allocTemp)) @@ -1010,8 +1020,8 @@ static SRes SzReadAndDecodePackedStreams( for (fo = 0; fo < p->NumFolders; fo++) { const CBuf *tempBuf = tempBufs + fo; - RINOK(LookInStream_SeekTo(inStream, dataStartPos)); - RINOK(SzAr_DecodeFolder(p, fo, inStream, dataStartPos, tempBuf->data, tempBuf->size, allocTemp)); + RINOK(LookInStream_SeekTo(inStream, dataStartPos)) + RINOK(SzAr_DecodeFolder(p, fo, inStream, dataStartPos, tempBuf->data, tempBuf->size, allocTemp)) } return SZ_OK; @@ -1046,7 +1056,7 @@ static SRes SzReadFileNames(const Byte *data, size_t size, UInt32 numFiles, size return (pos == size) ? SZ_OK : SZ_ERROR_ARCHIVE; } -static MY_NO_INLINE SRes ReadTime(CSzBitUi64s *p, UInt32 num, +static Z7_NO_INLINE SRes ReadTime(CSzBitUi64s *p, UInt32 num, CSzData *sd2, const CBuf *tempBufs, UInt32 numTempBufs, ISzAllocPtr alloc) @@ -1057,22 +1067,22 @@ static MY_NO_INLINE SRes ReadTime(CSzBitUi64s *p, UInt32 num, Byte *defs; Byte external; - RINOK(ReadBitVector(sd2, num, &p->Defs, alloc)); + RINOK(ReadBitVector(sd2, num, &p->Defs, alloc)) - SZ_READ_BYTE_SD(sd2, external); + SZ_READ_BYTE_SD(sd2, external) if (external == 0) sd = *sd2; else { UInt32 index; - RINOK(SzReadNumber32(sd2, &index)); + RINOK(SzReadNumber32(sd2, &index)) if (index >= numTempBufs) return SZ_ERROR_ARCHIVE; sd.Data = tempBufs[index].data; sd.Size = tempBufs[index].size; } - MY_ALLOC_ZE(CNtfsFileTime, p->Vals, num, alloc); + MY_ALLOC_ZE(CNtfsFileTime, p->Vals, num, alloc) vals = p->Vals; defs = p->Defs; for (i = 0; i < num; i++) @@ -1082,7 +1092,7 @@ static MY_NO_INLINE SRes ReadTime(CSzBitUi64s *p, UInt32 num, return SZ_ERROR_ARCHIVE; vals[i].Low = GetUi32(sd.Data); vals[i].High = GetUi32(sd.Data + 4); - SKIP_DATA2(sd, 8); + SKIP_DATA2(sd, 8) } else vals[i].High = vals[i].Low = 0; @@ -1100,7 +1110,7 @@ static MY_NO_INLINE SRes ReadTime(CSzBitUi64s *p, UInt32 num, static SRes SzReadHeader2( CSzArEx *p, /* allocMain */ CSzData *sd, - ILookInStream *inStream, + ILookInStreamPtr inStream, CBuf *tempBufs, UInt32 *numTempBufs, ISzAllocPtr allocMain, ISzAllocPtr allocTemp @@ -1111,26 +1121,26 @@ static SRes SzReadHeader2( { UInt64 type; - SzData_Clear(&ssi.sdSizes); - SzData_Clear(&ssi.sdCRCs); - SzData_Clear(&ssi.sdNumSubStreams); + SzData_CLEAR(&ssi.sdSizes) + SzData_CLEAR(&ssi.sdCRCs) + SzData_CLEAR(&ssi.sdNumSubStreams) ssi.NumSubDigests = 0; ssi.NumTotalSubStreams = 0; - RINOK(ReadID(sd, &type)); + RINOK(ReadID(sd, &type)) if (type == k7zIdArchiveProperties) { for (;;) { UInt64 type2; - RINOK(ReadID(sd, &type2)); + RINOK(ReadID(sd, &type2)) if (type2 == k7zIdEnd) break; - RINOK(SkipData(sd)); + RINOK(SkipData(sd)) } - RINOK(ReadID(sd, &type)); + RINOK(ReadID(sd, &type)) } if (type == k7zIdAdditionalStreamsInfo) @@ -1148,15 +1158,15 @@ static SRes SzReadHeader2( if (res != SZ_OK) return res; - RINOK(ReadID(sd, &type)); + RINOK(ReadID(sd, &type)) } if (type == k7zIdMainStreamsInfo) { RINOK(SzReadStreamsInfo(&p->db, sd, (UInt32)1 << 30, tempBufs, *numTempBufs, - &p->dataPos, &ssi, allocMain)); + &p->dataPos, &ssi, allocMain)) p->dataPos += p->startPosAfterHeader; - RINOK(ReadID(sd, &type)); + RINOK(ReadID(sd, &type)) } if (type == k7zIdEnd) @@ -1174,23 +1184,23 @@ static SRes SzReadHeader2( const Byte *emptyStreams = NULL; const Byte *emptyFiles = NULL; - RINOK(SzReadNumber32(sd, &numFiles)); + RINOK(SzReadNumber32(sd, &numFiles)) p->NumFiles = numFiles; for (;;) { UInt64 type; UInt64 size; - RINOK(ReadID(sd, &type)); + RINOK(ReadID(sd, &type)) if (type == k7zIdEnd) break; - RINOK(ReadNumber(sd, &size)); + RINOK(ReadNumber(sd, &size)) if (size > sd->Size) return SZ_ERROR_ARCHIVE; if (type >= ((UInt32)1 << 8)) { - SKIP_DATA(sd, size); + SKIP_DATA(sd, size) } else switch ((unsigned)type) { @@ -1200,7 +1210,7 @@ static SRes SzReadHeader2( const Byte *namesData; Byte external; - SZ_READ_BYTE(external); + SZ_READ_BYTE(external) if (external == 0) { namesSize = (size_t)size - 1; @@ -1209,7 +1219,7 @@ static SRes SzReadHeader2( else { UInt32 index; - RINOK(SzReadNumber32(sd, &index)); + RINOK(SzReadNumber32(sd, &index)) if (index >= *numTempBufs) return SZ_ERROR_ARCHIVE; namesData = (tempBufs)[index].data; @@ -1218,25 +1228,25 @@ static SRes SzReadHeader2( if ((namesSize & 1) != 0) return SZ_ERROR_ARCHIVE; - MY_ALLOC(size_t, p->FileNameOffsets, numFiles + 1, allocMain); - MY_ALLOC_ZE_AND_CPY(p->FileNames, namesSize, namesData, allocMain); + MY_ALLOC(size_t, p->FileNameOffsets, numFiles + 1, allocMain) + MY_ALLOC_ZE_AND_CPY(p->FileNames, namesSize, namesData, allocMain) RINOK(SzReadFileNames(p->FileNames, namesSize, numFiles, p->FileNameOffsets)) if (external == 0) { - SKIP_DATA(sd, namesSize); + SKIP_DATA(sd, namesSize) } break; } case k7zIdEmptyStream: { - RINOK(RememberBitVector(sd, numFiles, &emptyStreams)); + RINOK(RememberBitVector(sd, numFiles, &emptyStreams)) numEmptyStreams = CountDefinedBits(emptyStreams, numFiles); emptyFiles = NULL; break; } case k7zIdEmptyFile: { - RINOK(RememberBitVector(sd, numEmptyStreams, &emptyFiles)); + RINOK(RememberBitVector(sd, numEmptyStreams, &emptyFiles)) break; } case k7zIdWinAttrib: @@ -1245,22 +1255,22 @@ static SRes SzReadHeader2( CSzData sdSwitch; CSzData *sdPtr; SzBitUi32s_Free(&p->Attribs, allocMain); - RINOK(ReadBitVector(sd, numFiles, &p->Attribs.Defs, allocMain)); + RINOK(ReadBitVector(sd, numFiles, &p->Attribs.Defs, allocMain)) - SZ_READ_BYTE(external); + SZ_READ_BYTE(external) if (external == 0) sdPtr = sd; else { UInt32 index; - RINOK(SzReadNumber32(sd, &index)); + RINOK(SzReadNumber32(sd, &index)) if (index >= *numTempBufs) return SZ_ERROR_ARCHIVE; sdSwitch.Data = (tempBufs)[index].data; sdSwitch.Size = (tempBufs)[index].size; sdPtr = &sdSwitch; } - RINOK(ReadUi32s(sdPtr, numFiles, &p->Attribs, allocMain)); + RINOK(ReadUi32s(sdPtr, numFiles, &p->Attribs, allocMain)) break; } /* @@ -1273,11 +1283,11 @@ static SRes SzReadHeader2( break; } */ - case k7zIdMTime: RINOK(ReadTime(&p->MTime, numFiles, sd, tempBufs, *numTempBufs, allocMain)); break; - case k7zIdCTime: RINOK(ReadTime(&p->CTime, numFiles, sd, tempBufs, *numTempBufs, allocMain)); break; + case k7zIdMTime: RINOK(ReadTime(&p->MTime, numFiles, sd, tempBufs, *numTempBufs, allocMain)) break; + case k7zIdCTime: RINOK(ReadTime(&p->CTime, numFiles, sd, tempBufs, *numTempBufs, allocMain)) break; default: { - SKIP_DATA(sd, size); + SKIP_DATA(sd, size) } } } @@ -1288,10 +1298,10 @@ static SRes SzReadHeader2( for (;;) { UInt64 type; - RINOK(ReadID(sd, &type)); + RINOK(ReadID(sd, &type)) if (type == k7zIdEnd) break; - RINOK(SkipData(sd)); + RINOK(SkipData(sd)) } { @@ -1303,40 +1313,37 @@ static SRes SzReadHeader2( UInt64 unpackPos = 0; const Byte *digestsDefs = NULL; const Byte *digestsVals = NULL; - UInt32 digestsValsIndex = 0; - UInt32 digestIndex; - Byte allDigestsDefined = 0; + UInt32 digestIndex = 0; Byte isDirMask = 0; Byte crcMask = 0; Byte mask = 0x80; - MY_ALLOC(UInt32, p->FolderToFile, p->db.NumFolders + 1, allocMain); - MY_ALLOC_ZE(UInt32, p->FileToFolder, p->NumFiles, allocMain); - MY_ALLOC(UInt64, p->UnpackPositions, p->NumFiles + 1, allocMain); - MY_ALLOC_ZE(Byte, p->IsDirs, (p->NumFiles + 7) >> 3, allocMain); + MY_ALLOC(UInt32, p->FolderToFile, p->db.NumFolders + 1, allocMain) + MY_ALLOC_ZE(UInt32, p->FileToFolder, p->NumFiles, allocMain) + MY_ALLOC(UInt64, p->UnpackPositions, p->NumFiles + 1, allocMain) + MY_ALLOC_ZE(Byte, p->IsDirs, (p->NumFiles + 7) >> 3, allocMain) - RINOK(SzBitUi32s_Alloc(&p->CRCs, p->NumFiles, allocMain)); + RINOK(SzBitUi32s_Alloc(&p->CRCs, p->NumFiles, allocMain)) if (ssi.sdCRCs.Size != 0) { - SZ_READ_BYTE_SD(&ssi.sdCRCs, allDigestsDefined); + Byte allDigestsDefined = 0; + SZ_READ_BYTE_SD_NOCHECK(&ssi.sdCRCs, allDigestsDefined) if (allDigestsDefined) digestsVals = ssi.sdCRCs.Data; else { - size_t numBytes = (ssi.NumSubDigests + 7) >> 3; + const size_t numBytes = (ssi.NumSubDigests + 7) >> 3; digestsDefs = ssi.sdCRCs.Data; digestsVals = digestsDefs + numBytes; } } - digestIndex = 0; - for (i = 0; i < numFiles; i++, mask >>= 1) { if (mask == 0) { - UInt32 byteIndex = (i - 1) >> 3; + const UInt32 byteIndex = (i - 1) >> 3; p->IsDirs[byteIndex] = isDirMask; p->CRCs.Defs[byteIndex] = crcMask; isDirMask = 0; @@ -1374,18 +1381,17 @@ static SRes SzReadHeader2( numSubStreams = 1; if (ssi.sdNumSubStreams.Data) { - RINOK(SzReadNumber32(&ssi.sdNumSubStreams, &numSubStreams)); + RINOK(SzReadNumber32(&ssi.sdNumSubStreams, &numSubStreams)) } remSubStreams = numSubStreams; if (numSubStreams != 0) break; { - UInt64 folderUnpackSize = SzAr_GetFolderUnpackSize(&p->db, folderIndex); + const UInt64 folderUnpackSize = SzAr_GetFolderUnpackSize(&p->db, folderIndex); unpackPos += folderUnpackSize; if (unpackPos < folderUnpackSize) return SZ_ERROR_ARCHIVE; } - folderIndex++; } } @@ -1397,47 +1403,44 @@ static SRes SzReadHeader2( if (--remSubStreams == 0) { - UInt64 folderUnpackSize = SzAr_GetFolderUnpackSize(&p->db, folderIndex); - UInt64 startFolderUnpackPos = p->UnpackPositions[p->FolderToFile[folderIndex]]; + const UInt64 folderUnpackSize = SzAr_GetFolderUnpackSize(&p->db, folderIndex); + const UInt64 startFolderUnpackPos = p->UnpackPositions[p->FolderToFile[folderIndex]]; if (folderUnpackSize < unpackPos - startFolderUnpackPos) return SZ_ERROR_ARCHIVE; unpackPos = startFolderUnpackPos + folderUnpackSize; if (unpackPos < folderUnpackSize) return SZ_ERROR_ARCHIVE; - if (numSubStreams == 1 && SzBitWithVals_Check(&p->db.FolderCRCs, i)) + if (numSubStreams == 1 && SzBitWithVals_Check(&p->db.FolderCRCs, folderIndex)) { p->CRCs.Vals[i] = p->db.FolderCRCs.Vals[folderIndex]; crcMask |= mask; } - else if (allDigestsDefined || (digestsDefs && SzBitArray_Check(digestsDefs, digestIndex))) - { - p->CRCs.Vals[i] = GetUi32(digestsVals + (size_t)digestsValsIndex * 4); - digestsValsIndex++; - crcMask |= mask; - } - folderIndex++; } else { UInt64 v; - RINOK(ReadNumber(&ssi.sdSizes, &v)); + RINOK(ReadNumber(&ssi.sdSizes, &v)) unpackPos += v; if (unpackPos < v) return SZ_ERROR_ARCHIVE; - if (allDigestsDefined || (digestsDefs && SzBitArray_Check(digestsDefs, digestIndex))) + } + if ((crcMask & mask) == 0 && digestsVals) + { + if (!digestsDefs || SzBitArray_Check(digestsDefs, digestIndex)) { - p->CRCs.Vals[i] = GetUi32(digestsVals + (size_t)digestsValsIndex * 4); - digestsValsIndex++; + p->CRCs.Vals[i] = GetUi32(digestsVals); + digestsVals += 4; crcMask |= mask; } + digestIndex++; } } if (mask != 0x80) { - UInt32 byteIndex = (i - 1) >> 3; + const UInt32 byteIndex = (i - 1) >> 3; p->IsDirs[byteIndex] = isDirMask; p->CRCs.Defs[byteIndex] = crcMask; } @@ -1454,7 +1457,7 @@ static SRes SzReadHeader2( break; if (!ssi.sdNumSubStreams.Data) return SZ_ERROR_ARCHIVE; - RINOK(SzReadNumber32(&ssi.sdNumSubStreams, &numSubStreams)); + RINOK(SzReadNumber32(&ssi.sdNumSubStreams, &numSubStreams)) if (numSubStreams != 0) return SZ_ERROR_ARCHIVE; /* @@ -1479,7 +1482,7 @@ static SRes SzReadHeader2( static SRes SzReadHeader( CSzArEx *p, CSzData *sd, - ILookInStream *inStream, + ILookInStreamPtr inStream, ISzAllocPtr allocMain, ISzAllocPtr allocTemp) { @@ -1498,7 +1501,7 @@ static SRes SzReadHeader( for (i = 0; i < NUM_ADDITIONAL_STREAMS_MAX; i++) Buf_Free(tempBufs + i, allocTemp); - RINOK(res); + RINOK(res) if (sd->Size != 0) return SZ_ERROR_FAIL; @@ -1508,7 +1511,7 @@ static SRes SzReadHeader( static SRes SzArEx_Open2( CSzArEx *p, - ILookInStream *inStream, + ILookInStreamPtr inStream, ISzAllocPtr allocMain, ISzAllocPtr allocTemp) { @@ -1521,9 +1524,9 @@ static SRes SzArEx_Open2( SRes res; startArcPos = 0; - RINOK(ILookInStream_Seek(inStream, &startArcPos, SZ_SEEK_CUR)); + RINOK(ILookInStream_Seek(inStream, &startArcPos, SZ_SEEK_CUR)) - RINOK(LookInStream_Read2(inStream, header, k7zStartHeaderSize, SZ_ERROR_NO_ARCHIVE)); + RINOK(LookInStream_Read2(inStream, header, k7zStartHeaderSize, SZ_ERROR_NO_ARCHIVE)) if (!TestSignatureCandidate(header)) return SZ_ERROR_NO_ARCHIVE; @@ -1552,14 +1555,14 @@ static SRes SzArEx_Open2( { Int64 pos = 0; - RINOK(ILookInStream_Seek(inStream, &pos, SZ_SEEK_END)); + RINOK(ILookInStream_Seek(inStream, &pos, SZ_SEEK_END)) if ((UInt64)pos < (UInt64)startArcPos + nextHeaderOffset || (UInt64)pos < (UInt64)startArcPos + k7zStartHeaderSize + nextHeaderOffset || (UInt64)pos < (UInt64)startArcPos + k7zStartHeaderSize + nextHeaderOffset + nextHeaderSize) return SZ_ERROR_INPUT_EOF; } - RINOK(LookInStream_SeekTo(inStream, (UInt64)startArcPos + k7zStartHeaderSize + nextHeaderOffset)); + RINOK(LookInStream_SeekTo(inStream, (UInt64)startArcPos + k7zStartHeaderSize + nextHeaderOffset)) if (!Buf_Create(&buf, nextHeaderSizeT, allocTemp)) return SZ_ERROR_MEM; @@ -1634,10 +1637,10 @@ static SRes SzArEx_Open2( } -SRes SzArEx_Open(CSzArEx *p, ILookInStream *inStream, +SRes SzArEx_Open(CSzArEx *p, ILookInStreamPtr inStream, ISzAllocPtr allocMain, ISzAllocPtr allocTemp) { - SRes res = SzArEx_Open2(p, inStream, allocMain, allocTemp); + const SRes res = SzArEx_Open2(p, inStream, allocMain, allocTemp); if (res != SZ_OK) SzArEx_Free(p, allocMain); return res; @@ -1646,7 +1649,7 @@ SRes SzArEx_Open(CSzArEx *p, ILookInStream *inStream, SRes SzArEx_Extract( const CSzArEx *p, - ILookInStream *inStream, + ILookInStreamPtr inStream, UInt32 fileIndex, UInt32 *blockIndex, Byte **tempBuf, @@ -1656,7 +1659,7 @@ SRes SzArEx_Extract( ISzAllocPtr allocMain, ISzAllocPtr allocTemp) { - UInt32 folderIndex = p->FileToFolder[fileIndex]; + const UInt32 folderIndex = p->FileToFolder[fileIndex]; SRes res = SZ_OK; *offset = 0; @@ -1673,13 +1676,13 @@ SRes SzArEx_Extract( if (*tempBuf == NULL || *blockIndex != folderIndex) { - UInt64 unpackSizeSpec = SzAr_GetFolderUnpackSize(&p->db, folderIndex); + const UInt64 unpackSizeSpec = SzAr_GetFolderUnpackSize(&p->db, folderIndex); /* UInt64 unpackSizeSpec = p->UnpackPositions[p->FolderToFile[(size_t)folderIndex + 1]] - p->UnpackPositions[p->FolderToFile[folderIndex]]; */ - size_t unpackSize = (size_t)unpackSizeSpec; + const size_t unpackSize = (size_t)unpackSizeSpec; if (unpackSize != unpackSizeSpec) return SZ_ERROR_MEM; @@ -1707,7 +1710,7 @@ SRes SzArEx_Extract( if (res == SZ_OK) { - UInt64 unpackPos = p->UnpackPositions[fileIndex]; + const UInt64 unpackPos = p->UnpackPositions[fileIndex]; *offset = (size_t)(unpackPos - p->UnpackPositions[p->FolderToFile[folderIndex]]); *outSizeProcessed = (size_t)(p->UnpackPositions[(size_t)fileIndex + 1] - unpackPos); if (*offset + *outSizeProcessed > *outBufferSize) @@ -1723,8 +1726,8 @@ SRes SzArEx_Extract( size_t SzArEx_GetFileNameUtf16(const CSzArEx *p, size_t fileIndex, UInt16 *dest) { - size_t offs = p->FileNameOffsets[fileIndex]; - size_t len = p->FileNameOffsets[fileIndex + 1] - offs; + const size_t offs = p->FileNameOffsets[fileIndex]; + const size_t len = p->FileNameOffsets[fileIndex + 1] - offs; if (dest != 0) { size_t i; diff --git a/libraries/lzma/C/7zBuf.h b/libraries/lzma/C/7zBuf.h index 81d1b5b64..c0ba8a7b6 100644 --- a/libraries/lzma/C/7zBuf.h +++ b/libraries/lzma/C/7zBuf.h @@ -1,8 +1,8 @@ /* 7zBuf.h -- Byte Buffer -2017-04-03 : Igor Pavlov : Public domain */ +2023-03-04 : Igor Pavlov : Public domain */ -#ifndef __7Z_BUF_H -#define __7Z_BUF_H +#ifndef ZIP7_INC_7Z_BUF_H +#define ZIP7_INC_7Z_BUF_H #include "7zTypes.h" diff --git a/libraries/lzma/C/7zBuf2.c b/libraries/lzma/C/7zBuf2.c new file mode 100644 index 000000000..208347416 --- /dev/null +++ b/libraries/lzma/C/7zBuf2.c @@ -0,0 +1,52 @@ +/* 7zBuf2.c -- Byte Buffer +2017-04-03 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include + +#include "7zBuf.h" + +void DynBuf_Construct(CDynBuf *p) +{ + p->data = 0; + p->size = 0; + p->pos = 0; +} + +void DynBuf_SeekToBeg(CDynBuf *p) +{ + p->pos = 0; +} + +int DynBuf_Write(CDynBuf *p, const Byte *buf, size_t size, ISzAllocPtr alloc) +{ + if (size > p->size - p->pos) + { + size_t newSize = p->pos + size; + Byte *data; + newSize += newSize / 4; + data = (Byte *)ISzAlloc_Alloc(alloc, newSize); + if (!data) + return 0; + p->size = newSize; + if (p->pos != 0) + memcpy(data, p->data, p->pos); + ISzAlloc_Free(alloc, p->data); + p->data = data; + } + if (size != 0) + { + memcpy(p->data + p->pos, buf, size); + p->pos += size; + } + return 1; +} + +void DynBuf_Free(CDynBuf *p, ISzAllocPtr alloc) +{ + ISzAlloc_Free(alloc, p->data); + p->data = 0; + p->size = 0; + p->pos = 0; +} diff --git a/libraries/lzma/C/7zCrc.c b/libraries/lzma/C/7zCrc.c index f186324dd..c995a8be4 100644 --- a/libraries/lzma/C/7zCrc.c +++ b/libraries/lzma/C/7zCrc.c @@ -1,5 +1,5 @@ -/* 7zCrc.c -- CRC32 init -2021-04-01 : Igor Pavlov : Public domain */ +/* 7zCrc.c -- CRC32 calculation and init +2023-04-02 : Igor Pavlov : Public domain */ #include "Precomp.h" @@ -13,22 +13,20 @@ #else #define CRC_NUM_TABLES 9 - #define CRC_UINT32_SWAP(v) ((v >> 24) | ((v >> 8) & 0xFF00) | ((v << 8) & 0xFF0000) | (v << 24)) - - UInt32 MY_FAST_CALL CrcUpdateT1_BeT4(UInt32 v, const void *data, size_t size, const UInt32 *table); - UInt32 MY_FAST_CALL CrcUpdateT1_BeT8(UInt32 v, const void *data, size_t size, const UInt32 *table); + UInt32 Z7_FASTCALL CrcUpdateT1_BeT4(UInt32 v, const void *data, size_t size, const UInt32 *table); + UInt32 Z7_FASTCALL CrcUpdateT1_BeT8(UInt32 v, const void *data, size_t size, const UInt32 *table); #endif #ifndef MY_CPU_BE - UInt32 MY_FAST_CALL CrcUpdateT4(UInt32 v, const void *data, size_t size, const UInt32 *table); - UInt32 MY_FAST_CALL CrcUpdateT8(UInt32 v, const void *data, size_t size, const UInt32 *table); + UInt32 Z7_FASTCALL CrcUpdateT4(UInt32 v, const void *data, size_t size, const UInt32 *table); + UInt32 Z7_FASTCALL CrcUpdateT8(UInt32 v, const void *data, size_t size, const UInt32 *table); #endif -typedef UInt32 (MY_FAST_CALL *CRC_FUNC)(UInt32 v, const void *data, size_t size, const UInt32 *table); - +/* extern CRC_FUNC g_CrcUpdateT4; CRC_FUNC g_CrcUpdateT4; +*/ extern CRC_FUNC g_CrcUpdateT8; CRC_FUNC g_CrcUpdateT8; @@ -44,20 +42,22 @@ CRC_FUNC g_CrcUpdate; UInt32 g_CrcTable[256 * CRC_NUM_TABLES]; -UInt32 MY_FAST_CALL CrcUpdate(UInt32 v, const void *data, size_t size) +UInt32 Z7_FASTCALL CrcUpdate(UInt32 v, const void *data, size_t size) { return g_CrcUpdate(v, data, size, g_CrcTable); } -UInt32 MY_FAST_CALL CrcCalc(const void *data, size_t size) +UInt32 Z7_FASTCALL CrcCalc(const void *data, size_t size) { return g_CrcUpdate(CRC_INIT_VAL, data, size, g_CrcTable) ^ CRC_INIT_VAL; } +#if CRC_NUM_TABLES < 4 \ + || (CRC_NUM_TABLES == 4 && defined(MY_CPU_BE)) \ + || (!defined(MY_CPU_LE) && !defined(MY_CPU_BE)) #define CRC_UPDATE_BYTE_2(crc, b) (table[((crc) ^ (b)) & 0xFF] ^ ((crc) >> 8)) - -UInt32 MY_FAST_CALL CrcUpdateT1(UInt32 v, const void *data, size_t size, const UInt32 *table); -UInt32 MY_FAST_CALL CrcUpdateT1(UInt32 v, const void *data, size_t size, const UInt32 *table) +UInt32 Z7_FASTCALL CrcUpdateT1(UInt32 v, const void *data, size_t size, const UInt32 *table); +UInt32 Z7_FASTCALL CrcUpdateT1(UInt32 v, const void *data, size_t size, const UInt32 *table) { const Byte *p = (const Byte *)data; const Byte *pEnd = p + size; @@ -65,7 +65,7 @@ UInt32 MY_FAST_CALL CrcUpdateT1(UInt32 v, const void *data, size_t size, const U v = CRC_UPDATE_BYTE_2(v, *p); return v; } - +#endif /* ---------- hardware CRC ---------- */ @@ -78,16 +78,29 @@ UInt32 MY_FAST_CALL CrcUpdateT1(UInt32 v, const void *data, size_t size, const U #if defined(_MSC_VER) #if defined(MY_CPU_ARM64) #if (_MSC_VER >= 1910) + #ifndef __clang__ #define USE_ARM64_CRC + #include + #endif #endif #endif #elif (defined(__clang__) && (__clang_major__ >= 3)) \ || (defined(__GNUC__) && (__GNUC__ > 4)) #if !defined(__ARM_FEATURE_CRC32) #define __ARM_FEATURE_CRC32 1 - #if (!defined(__clang__) || (__clang_major__ > 3)) // fix these numbers + #if defined(__clang__) + #if defined(MY_CPU_ARM64) + #define ATTRIB_CRC __attribute__((__target__("crc"))) + #else + #define ATTRIB_CRC __attribute__((__target__("armv8-a,crc"))) + #endif + #else + #if defined(MY_CPU_ARM64) + #define ATTRIB_CRC __attribute__((__target__("+crc"))) + #else #define ATTRIB_CRC __attribute__((__target__("arch=armv8-a+crc"))) #endif + #endif #endif #if defined(__ARM_FEATURE_CRC32) #define USE_ARM64_CRC @@ -105,7 +118,7 @@ UInt32 MY_FAST_CALL CrcUpdateT1(UInt32 v, const void *data, size_t size, const U #pragma message("ARM64 CRC emulation") -MY_FORCE_INLINE +Z7_FORCE_INLINE UInt32 __crc32b(UInt32 v, UInt32 data) { const UInt32 *table = g_CrcTable; @@ -113,7 +126,7 @@ UInt32 __crc32b(UInt32 v, UInt32 data) return v; } -MY_FORCE_INLINE +Z7_FORCE_INLINE UInt32 __crc32w(UInt32 v, UInt32 data) { const UInt32 *table = g_CrcTable; @@ -124,7 +137,7 @@ UInt32 __crc32w(UInt32 v, UInt32 data) return v; } -MY_FORCE_INLINE +Z7_FORCE_INLINE UInt32 __crc32d(UInt32 v, UInt64 data) { const UInt32 *table = g_CrcTable; @@ -156,9 +169,9 @@ UInt32 __crc32d(UInt32 v, UInt64 data) // #pragma message("USE ARM HW CRC") ATTRIB_CRC -UInt32 MY_FAST_CALL CrcUpdateT0_32(UInt32 v, const void *data, size_t size, const UInt32 *table); +UInt32 Z7_FASTCALL CrcUpdateT0_32(UInt32 v, const void *data, size_t size, const UInt32 *table); ATTRIB_CRC -UInt32 MY_FAST_CALL CrcUpdateT0_32(UInt32 v, const void *data, size_t size, const UInt32 *table) +UInt32 Z7_FASTCALL CrcUpdateT0_32(UInt32 v, const void *data, size_t size, const UInt32 *table) { const Byte *p = (const Byte *)data; UNUSED_VAR(table); @@ -188,9 +201,9 @@ UInt32 MY_FAST_CALL CrcUpdateT0_32(UInt32 v, const void *data, size_t size, cons } ATTRIB_CRC -UInt32 MY_FAST_CALL CrcUpdateT0_64(UInt32 v, const void *data, size_t size, const UInt32 *table); +UInt32 Z7_FASTCALL CrcUpdateT0_64(UInt32 v, const void *data, size_t size, const UInt32 *table); ATTRIB_CRC -UInt32 MY_FAST_CALL CrcUpdateT0_64(UInt32 v, const void *data, size_t size, const UInt32 *table) +UInt32 Z7_FASTCALL CrcUpdateT0_64(UInt32 v, const void *data, size_t size, const UInt32 *table) { const Byte *p = (const Byte *)data; UNUSED_VAR(table); @@ -219,6 +232,9 @@ UInt32 MY_FAST_CALL CrcUpdateT0_64(UInt32 v, const void *data, size_t size, cons return v; } +#undef T0_32_UNROLL_BYTES +#undef T0_64_UNROLL_BYTES + #endif // defined(USE_ARM64_CRC) || defined(USE_CRC_EMU) #endif // MY_CPU_LE @@ -226,7 +242,7 @@ UInt32 MY_FAST_CALL CrcUpdateT0_64(UInt32 v, const void *data, size_t size, cons -void MY_FAST_CALL CrcGenerateTable() +void Z7_FASTCALL CrcGenerateTable(void) { UInt32 i; for (i = 0; i < 256; i++) @@ -239,64 +255,62 @@ void MY_FAST_CALL CrcGenerateTable() } for (i = 256; i < 256 * CRC_NUM_TABLES; i++) { - UInt32 r = g_CrcTable[(size_t)i - 256]; + const UInt32 r = g_CrcTable[(size_t)i - 256]; g_CrcTable[i] = g_CrcTable[r & 0xFF] ^ (r >> 8); } #if CRC_NUM_TABLES < 4 - - g_CrcUpdate = CrcUpdateT1; - - #else - - #ifdef MY_CPU_LE - - g_CrcUpdateT4 = CrcUpdateT4; - g_CrcUpdate = CrcUpdateT4; - - #if CRC_NUM_TABLES >= 8 + g_CrcUpdate = CrcUpdateT1; + #elif defined(MY_CPU_LE) + // g_CrcUpdateT4 = CrcUpdateT4; + #if CRC_NUM_TABLES < 8 + g_CrcUpdate = CrcUpdateT4; + #else // CRC_NUM_TABLES >= 8 g_CrcUpdateT8 = CrcUpdateT8; - + /* #ifdef MY_CPU_X86_OR_AMD64 if (!CPU_Is_InOrder()) #endif - g_CrcUpdate = CrcUpdateT8; + */ + g_CrcUpdate = CrcUpdateT8; #endif - #else { - #ifndef MY_CPU_BE + #ifndef MY_CPU_BE UInt32 k = 0x01020304; const Byte *p = (const Byte *)&k; if (p[0] == 4 && p[1] == 3) { - g_CrcUpdateT4 = CrcUpdateT4; - g_CrcUpdate = CrcUpdateT4; - #if CRC_NUM_TABLES >= 8 - g_CrcUpdateT8 = CrcUpdateT8; - g_CrcUpdate = CrcUpdateT8; + #if CRC_NUM_TABLES < 8 + // g_CrcUpdateT4 = CrcUpdateT4; + g_CrcUpdate = CrcUpdateT4; + #else // CRC_NUM_TABLES >= 8 + g_CrcUpdateT8 = CrcUpdateT8; + g_CrcUpdate = CrcUpdateT8; #endif } else if (p[0] != 1 || p[1] != 2) g_CrcUpdate = CrcUpdateT1; else - #endif + #endif // MY_CPU_BE { for (i = 256 * CRC_NUM_TABLES - 1; i >= 256; i--) { - UInt32 x = g_CrcTable[(size_t)i - 256]; - g_CrcTable[i] = CRC_UINT32_SWAP(x); + const UInt32 x = g_CrcTable[(size_t)i - 256]; + g_CrcTable[i] = Z7_BSWAP32(x); } - g_CrcUpdateT4 = CrcUpdateT1_BeT4; - g_CrcUpdate = CrcUpdateT1_BeT4; - #if CRC_NUM_TABLES >= 8 - g_CrcUpdateT8 = CrcUpdateT1_BeT8; - g_CrcUpdate = CrcUpdateT1_BeT8; + #if CRC_NUM_TABLES <= 4 + g_CrcUpdate = CrcUpdateT1; + #elif CRC_NUM_TABLES <= 8 + // g_CrcUpdateT4 = CrcUpdateT1_BeT4; + g_CrcUpdate = CrcUpdateT1_BeT4; + #else // CRC_NUM_TABLES > 8 + g_CrcUpdateT8 = CrcUpdateT1_BeT8; + g_CrcUpdate = CrcUpdateT1_BeT8; #endif } } - #endif - #endif + #endif // CRC_NUM_TABLES < 4 #ifdef MY_CPU_LE #ifdef USE_ARM64_CRC @@ -320,3 +334,7 @@ void MY_FAST_CALL CrcGenerateTable() #endif #endif } + +#undef kCrcPoly +#undef CRC64_NUM_TABLES +#undef CRC_UPDATE_BYTE_2 diff --git a/libraries/lzma/C/7zCrc.h b/libraries/lzma/C/7zCrc.h index 8fd579587..4afaeae4a 100644 --- a/libraries/lzma/C/7zCrc.h +++ b/libraries/lzma/C/7zCrc.h @@ -1,8 +1,8 @@ /* 7zCrc.h -- CRC32 calculation -2013-01-18 : Igor Pavlov : Public domain */ +2023-04-02 : Igor Pavlov : Public domain */ -#ifndef __7Z_CRC_H -#define __7Z_CRC_H +#ifndef ZIP7_INC_7Z_CRC_H +#define ZIP7_INC_7Z_CRC_H #include "7zTypes.h" @@ -11,14 +11,16 @@ EXTERN_C_BEGIN extern UInt32 g_CrcTable[]; /* Call CrcGenerateTable one time before other CRC functions */ -void MY_FAST_CALL CrcGenerateTable(void); +void Z7_FASTCALL CrcGenerateTable(void); #define CRC_INIT_VAL 0xFFFFFFFF #define CRC_GET_DIGEST(crc) ((crc) ^ CRC_INIT_VAL) #define CRC_UPDATE_BYTE(crc, b) (g_CrcTable[((crc) ^ (b)) & 0xFF] ^ ((crc) >> 8)) -UInt32 MY_FAST_CALL CrcUpdate(UInt32 crc, const void *data, size_t size); -UInt32 MY_FAST_CALL CrcCalc(const void *data, size_t size); +UInt32 Z7_FASTCALL CrcUpdate(UInt32 crc, const void *data, size_t size); +UInt32 Z7_FASTCALL CrcCalc(const void *data, size_t size); + +typedef UInt32 (Z7_FASTCALL *CRC_FUNC)(UInt32 v, const void *data, size_t size, const UInt32 *table); EXTERN_C_END diff --git a/libraries/lzma/C/7zCrcOpt.c b/libraries/lzma/C/7zCrcOpt.c index 69fad9ca2..9c649290d 100644 --- a/libraries/lzma/C/7zCrcOpt.c +++ b/libraries/lzma/C/7zCrcOpt.c @@ -1,5 +1,5 @@ /* 7zCrcOpt.c -- CRC32 calculation -2021-02-09 : Igor Pavlov : Public domain */ +2023-04-02 : Igor Pavlov : Public domain */ #include "Precomp.h" @@ -9,8 +9,8 @@ #define CRC_UPDATE_BYTE_2(crc, b) (table[((crc) ^ (b)) & 0xFF] ^ ((crc) >> 8)) -UInt32 MY_FAST_CALL CrcUpdateT4(UInt32 v, const void *data, size_t size, const UInt32 *table); -UInt32 MY_FAST_CALL CrcUpdateT4(UInt32 v, const void *data, size_t size, const UInt32 *table) +UInt32 Z7_FASTCALL CrcUpdateT4(UInt32 v, const void *data, size_t size, const UInt32 *table); +UInt32 Z7_FASTCALL CrcUpdateT4(UInt32 v, const void *data, size_t size, const UInt32 *table) { const Byte *p = (const Byte *)data; for (; size > 0 && ((unsigned)(ptrdiff_t)p & 3) != 0; size--, p++) @@ -29,8 +29,8 @@ UInt32 MY_FAST_CALL CrcUpdateT4(UInt32 v, const void *data, size_t size, const U return v; } -UInt32 MY_FAST_CALL CrcUpdateT8(UInt32 v, const void *data, size_t size, const UInt32 *table); -UInt32 MY_FAST_CALL CrcUpdateT8(UInt32 v, const void *data, size_t size, const UInt32 *table) +UInt32 Z7_FASTCALL CrcUpdateT8(UInt32 v, const void *data, size_t size, const UInt32 *table); +UInt32 Z7_FASTCALL CrcUpdateT8(UInt32 v, const void *data, size_t size, const UInt32 *table) { const Byte *p = (const Byte *)data; for (; size > 0 && ((unsigned)(ptrdiff_t)p & 7) != 0; size--, p++) @@ -61,11 +61,11 @@ UInt32 MY_FAST_CALL CrcUpdateT8(UInt32 v, const void *data, size_t size, const U #ifndef MY_CPU_LE -#define CRC_UINT32_SWAP(v) ((v >> 24) | ((v >> 8) & 0xFF00) | ((v << 8) & 0xFF0000) | (v << 24)) +#define CRC_UINT32_SWAP(v) Z7_BSWAP32(v) #define CRC_UPDATE_BYTE_2_BE(crc, b) (table[(((crc) >> 24) ^ (b))] ^ ((crc) << 8)) -UInt32 MY_FAST_CALL CrcUpdateT1_BeT4(UInt32 v, const void *data, size_t size, const UInt32 *table) +UInt32 Z7_FASTCALL CrcUpdateT1_BeT4(UInt32 v, const void *data, size_t size, const UInt32 *table) { const Byte *p = (const Byte *)data; table += 0x100; @@ -86,7 +86,7 @@ UInt32 MY_FAST_CALL CrcUpdateT1_BeT4(UInt32 v, const void *data, size_t size, co return CRC_UINT32_SWAP(v); } -UInt32 MY_FAST_CALL CrcUpdateT1_BeT8(UInt32 v, const void *data, size_t size, const UInt32 *table) +UInt32 Z7_FASTCALL CrcUpdateT1_BeT8(UInt32 v, const void *data, size_t size, const UInt32 *table) { const Byte *p = (const Byte *)data; table += 0x100; diff --git a/libraries/lzma/C/7zDec.c b/libraries/lzma/C/7zDec.c index fbfd016e1..96c60359a 100644 --- a/libraries/lzma/C/7zDec.c +++ b/libraries/lzma/C/7zDec.c @@ -1,11 +1,11 @@ /* 7zDec.c -- Decoding from 7z folder -2021-02-09 : Igor Pavlov : Public domain */ +2023-04-02 : Igor Pavlov : Public domain */ #include "Precomp.h" #include -/* #define _7ZIP_PPMD_SUPPPORT */ +/* #define Z7_PPMD_SUPPORT */ #include "7z.h" #include "7zCrc.h" @@ -16,27 +16,49 @@ #include "Delta.h" #include "LzmaDec.h" #include "Lzma2Dec.h" -#ifdef _7ZIP_PPMD_SUPPPORT +#ifdef Z7_PPMD_SUPPORT #include "Ppmd7.h" #endif #define k_Copy 0 -#ifndef _7Z_NO_METHOD_LZMA2 +#ifndef Z7_NO_METHOD_LZMA2 #define k_LZMA2 0x21 #endif #define k_LZMA 0x30101 #define k_BCJ2 0x303011B -#ifndef _7Z_NO_METHODS_FILTERS + +#if !defined(Z7_NO_METHODS_FILTERS) +#define Z7_USE_BRANCH_FILTER +#endif + +#if !defined(Z7_NO_METHODS_FILTERS) || \ + defined(Z7_USE_NATIVE_BRANCH_FILTER) && defined(MY_CPU_ARM64) +#define Z7_USE_FILTER_ARM64 +#ifndef Z7_USE_BRANCH_FILTER +#define Z7_USE_BRANCH_FILTER +#endif +#define k_ARM64 0xa +#endif + +#if !defined(Z7_NO_METHODS_FILTERS) || \ + defined(Z7_USE_NATIVE_BRANCH_FILTER) && defined(MY_CPU_ARMT) +#define Z7_USE_FILTER_ARMT +#ifndef Z7_USE_BRANCH_FILTER +#define Z7_USE_BRANCH_FILTER +#endif +#define k_ARMT 0x3030701 +#endif + +#ifndef Z7_NO_METHODS_FILTERS #define k_Delta 3 #define k_BCJ 0x3030103 #define k_PPC 0x3030205 #define k_IA64 0x3030401 #define k_ARM 0x3030501 -#define k_ARMT 0x3030701 #define k_SPARC 0x3030805 #endif -#ifdef _7ZIP_PPMD_SUPPPORT +#ifdef Z7_PPMD_SUPPORT #define k_PPMD 0x30401 @@ -49,12 +71,12 @@ typedef struct UInt64 processed; BoolInt extra; SRes res; - const ILookInStream *inStream; + ILookInStreamPtr inStream; } CByteInToLook; -static Byte ReadByte(const IByteIn *pp) +static Byte ReadByte(IByteInPtr pp) { - CByteInToLook *p = CONTAINER_FROM_VTBL(pp, CByteInToLook, vt); + Z7_CONTAINER_FROM_VTBL_TO_DECL_VAR_pp_vt_p(CByteInToLook) if (p->cur != p->end) return *p->cur++; if (p->res == SZ_OK) @@ -67,13 +89,13 @@ static Byte ReadByte(const IByteIn *pp) p->cur = p->begin; p->end = p->begin + size; if (size != 0) - return *p->cur++;; + return *p->cur++; } p->extra = True; return 0; } -static SRes SzDecodePpmd(const Byte *props, unsigned propsSize, UInt64 inSize, const ILookInStream *inStream, +static SRes SzDecodePpmd(const Byte *props, unsigned propsSize, UInt64 inSize, ILookInStreamPtr inStream, Byte *outBuffer, SizeT outSize, ISzAllocPtr allocMain) { CPpmd7 ppmd; @@ -138,14 +160,14 @@ static SRes SzDecodePpmd(const Byte *props, unsigned propsSize, UInt64 inSize, c #endif -static SRes SzDecodeLzma(const Byte *props, unsigned propsSize, UInt64 inSize, ILookInStream *inStream, +static SRes SzDecodeLzma(const Byte *props, unsigned propsSize, UInt64 inSize, ILookInStreamPtr inStream, Byte *outBuffer, SizeT outSize, ISzAllocPtr allocMain) { CLzmaDec state; SRes res = SZ_OK; - LzmaDec_Construct(&state); - RINOK(LzmaDec_AllocateProbs(&state, props, propsSize, allocMain)); + LzmaDec_CONSTRUCT(&state) + RINOK(LzmaDec_AllocateProbs(&state, props, propsSize, allocMain)) state.dic = outBuffer; state.dicBufSize = outSize; LzmaDec_Init(&state); @@ -196,18 +218,18 @@ static SRes SzDecodeLzma(const Byte *props, unsigned propsSize, UInt64 inSize, I } -#ifndef _7Z_NO_METHOD_LZMA2 +#ifndef Z7_NO_METHOD_LZMA2 -static SRes SzDecodeLzma2(const Byte *props, unsigned propsSize, UInt64 inSize, ILookInStream *inStream, +static SRes SzDecodeLzma2(const Byte *props, unsigned propsSize, UInt64 inSize, ILookInStreamPtr inStream, Byte *outBuffer, SizeT outSize, ISzAllocPtr allocMain) { CLzma2Dec state; SRes res = SZ_OK; - Lzma2Dec_Construct(&state); + Lzma2Dec_CONSTRUCT(&state) if (propsSize != 1) return SZ_ERROR_DATA; - RINOK(Lzma2Dec_AllocateProbs(&state, props[0], allocMain)); + RINOK(Lzma2Dec_AllocateProbs(&state, props[0], allocMain)) state.decoder.dic = outBuffer; state.decoder.dicBufSize = outSize; Lzma2Dec_Init(&state); @@ -257,7 +279,7 @@ static SRes SzDecodeLzma2(const Byte *props, unsigned propsSize, UInt64 inSize, #endif -static SRes SzDecodeCopy(UInt64 inSize, ILookInStream *inStream, Byte *outBuffer) +static SRes SzDecodeCopy(UInt64 inSize, ILookInStreamPtr inStream, Byte *outBuffer) { while (inSize > 0) { @@ -265,13 +287,13 @@ static SRes SzDecodeCopy(UInt64 inSize, ILookInStream *inStream, Byte *outBuffer size_t curSize = (1 << 18); if (curSize > inSize) curSize = (size_t)inSize; - RINOK(ILookInStream_Look(inStream, &inBuf, &curSize)); + RINOK(ILookInStream_Look(inStream, &inBuf, &curSize)) if (curSize == 0) return SZ_ERROR_INPUT_EOF; memcpy(outBuffer, inBuf, curSize); outBuffer += curSize; inSize -= curSize; - RINOK(ILookInStream_Skip(inStream, curSize)); + RINOK(ILookInStream_Skip(inStream, curSize)) } return SZ_OK; } @@ -282,12 +304,12 @@ static BoolInt IS_MAIN_METHOD(UInt32 m) { case k_Copy: case k_LZMA: - #ifndef _7Z_NO_METHOD_LZMA2 + #ifndef Z7_NO_METHOD_LZMA2 case k_LZMA2: - #endif - #ifdef _7ZIP_PPMD_SUPPPORT + #endif + #ifdef Z7_PPMD_SUPPORT case k_PPMD: - #endif + #endif return True; } return False; @@ -317,7 +339,7 @@ static SRes CheckSupportedFolder(const CSzFolder *f) } - #ifndef _7Z_NO_METHODS_FILTERS + #if defined(Z7_USE_BRANCH_FILTER) if (f->NumCoders == 2) { @@ -333,13 +355,20 @@ static SRes CheckSupportedFolder(const CSzFolder *f) return SZ_ERROR_UNSUPPORTED; switch ((UInt32)c->MethodID) { + #if !defined(Z7_NO_METHODS_FILTERS) case k_Delta: case k_BCJ: case k_PPC: case k_IA64: case k_SPARC: case k_ARM: + #endif + #ifdef Z7_USE_FILTER_ARM64 + case k_ARM64: + #endif + #ifdef Z7_USE_FILTER_ARMT case k_ARMT: + #endif break; default: return SZ_ERROR_UNSUPPORTED; @@ -372,15 +401,16 @@ static SRes CheckSupportedFolder(const CSzFolder *f) return SZ_ERROR_UNSUPPORTED; } -#ifndef _7Z_NO_METHODS_FILTERS -#define CASE_BRA_CONV(isa) case k_ ## isa: isa ## _Convert(outBuffer, outSize, 0, 0); break; -#endif + + + + static SRes SzFolder_Decode2(const CSzFolder *folder, const Byte *propsData, const UInt64 *unpackSizes, const UInt64 *packPositions, - ILookInStream *inStream, UInt64 startPos, + ILookInStreamPtr inStream, UInt64 startPos, Byte *outBuffer, SizeT outSize, ISzAllocPtr allocMain, Byte *tempBuf[]) { @@ -389,7 +419,7 @@ static SRes SzFolder_Decode2(const CSzFolder *folder, SizeT tempSize3 = 0; Byte *tempBuf3 = 0; - RINOK(CheckSupportedFolder(folder)); + RINOK(CheckSupportedFolder(folder)) for (ci = 0; ci < folder->NumCoders; ci++) { @@ -404,8 +434,8 @@ static SRes SzFolder_Decode2(const CSzFolder *folder, SizeT outSizeCur = outSize; if (folder->NumCoders == 4) { - UInt32 indices[] = { 3, 2, 0 }; - UInt64 unpackSize = unpackSizes[ci]; + const UInt32 indices[] = { 3, 2, 0 }; + const UInt64 unpackSize = unpackSizes[ci]; si = indices[ci]; if (ci < 2) { @@ -431,37 +461,37 @@ static SRes SzFolder_Decode2(const CSzFolder *folder, } offset = packPositions[si]; inSize = packPositions[(size_t)si + 1] - offset; - RINOK(LookInStream_SeekTo(inStream, startPos + offset)); + RINOK(LookInStream_SeekTo(inStream, startPos + offset)) if (coder->MethodID == k_Copy) { if (inSize != outSizeCur) /* check it */ return SZ_ERROR_DATA; - RINOK(SzDecodeCopy(inSize, inStream, outBufCur)); + RINOK(SzDecodeCopy(inSize, inStream, outBufCur)) } else if (coder->MethodID == k_LZMA) { - RINOK(SzDecodeLzma(propsData + coder->PropsOffset, coder->PropsSize, inSize, inStream, outBufCur, outSizeCur, allocMain)); + RINOK(SzDecodeLzma(propsData + coder->PropsOffset, coder->PropsSize, inSize, inStream, outBufCur, outSizeCur, allocMain)) } - #ifndef _7Z_NO_METHOD_LZMA2 + #ifndef Z7_NO_METHOD_LZMA2 else if (coder->MethodID == k_LZMA2) { - RINOK(SzDecodeLzma2(propsData + coder->PropsOffset, coder->PropsSize, inSize, inStream, outBufCur, outSizeCur, allocMain)); + RINOK(SzDecodeLzma2(propsData + coder->PropsOffset, coder->PropsSize, inSize, inStream, outBufCur, outSizeCur, allocMain)) } - #endif - #ifdef _7ZIP_PPMD_SUPPPORT + #endif + #ifdef Z7_PPMD_SUPPORT else if (coder->MethodID == k_PPMD) { - RINOK(SzDecodePpmd(propsData + coder->PropsOffset, coder->PropsSize, inSize, inStream, outBufCur, outSizeCur, allocMain)); + RINOK(SzDecodePpmd(propsData + coder->PropsOffset, coder->PropsSize, inSize, inStream, outBufCur, outSizeCur, allocMain)) } - #endif + #endif else return SZ_ERROR_UNSUPPORTED; } else if (coder->MethodID == k_BCJ2) { - UInt64 offset = packPositions[1]; - UInt64 s3Size = packPositions[2] - offset; + const UInt64 offset = packPositions[1]; + const UInt64 s3Size = packPositions[2] - offset; if (ci != 3) return SZ_ERROR_UNSUPPORTED; @@ -473,8 +503,8 @@ static SRes SzFolder_Decode2(const CSzFolder *folder, if (!tempBuf[2] && tempSizes[2] != 0) return SZ_ERROR_MEM; - RINOK(LookInStream_SeekTo(inStream, startPos + offset)); - RINOK(SzDecodeCopy(s3Size, inStream, tempBuf[2])); + RINOK(LookInStream_SeekTo(inStream, startPos + offset)) + RINOK(SzDecodeCopy(s3Size, inStream, tempBuf[2])) if ((tempSizes[0] & 3) != 0 || (tempSizes[1] & 3) != 0 || @@ -493,26 +523,22 @@ static SRes SzFolder_Decode2(const CSzFolder *folder, p.destLim = outBuffer + outSize; Bcj2Dec_Init(&p); - RINOK(Bcj2Dec_Decode(&p)); + RINOK(Bcj2Dec_Decode(&p)) { unsigned i; for (i = 0; i < 4; i++) if (p.bufs[i] != p.lims[i]) return SZ_ERROR_DATA; - - if (!Bcj2Dec_IsFinished(&p)) - return SZ_ERROR_DATA; - - if (p.dest != p.destLim - || p.state != BCJ2_STREAM_MAIN) + if (p.dest != p.destLim || !Bcj2Dec_IsMaybeFinished(&p)) return SZ_ERROR_DATA; } } } - #ifndef _7Z_NO_METHODS_FILTERS + #if defined(Z7_USE_BRANCH_FILTER) else if (ci == 1) { + #if !defined(Z7_NO_METHODS_FILTERS) if (coder->MethodID == k_Delta) { if (coder->PropsSize != 1) @@ -522,31 +548,53 @@ static SRes SzFolder_Decode2(const CSzFolder *folder, Delta_Init(state); Delta_Decode(state, (unsigned)(propsData[coder->PropsOffset]) + 1, outBuffer, outSize); } + continue; } - else + #endif + + #ifdef Z7_USE_FILTER_ARM64 + if (coder->MethodID == k_ARM64) + { + UInt32 pc = 0; + if (coder->PropsSize == 4) + pc = GetUi32(propsData + coder->PropsOffset); + else if (coder->PropsSize != 0) + return SZ_ERROR_UNSUPPORTED; + z7_BranchConv_ARM64_Dec(outBuffer, outSize, pc); + continue; + } + #endif + + #if !defined(Z7_NO_METHODS_FILTERS) || defined(Z7_USE_FILTER_ARMT) { if (coder->PropsSize != 0) return SZ_ERROR_UNSUPPORTED; + #define CASE_BRA_CONV(isa) case k_ ## isa: Z7_BRANCH_CONV_DEC(isa)(outBuffer, outSize, 0); break; // pc = 0; switch (coder->MethodID) { + #if !defined(Z7_NO_METHODS_FILTERS) case k_BCJ: { - UInt32 state; - x86_Convert_Init(state); - x86_Convert(outBuffer, outSize, 0, &state, 0); + UInt32 state = Z7_BRANCH_CONV_ST_X86_STATE_INIT_VAL; + z7_BranchConvSt_X86_Dec(outBuffer, outSize, 0, &state); // pc = 0 break; } CASE_BRA_CONV(PPC) CASE_BRA_CONV(IA64) CASE_BRA_CONV(SPARC) CASE_BRA_CONV(ARM) + #endif + #if !defined(Z7_NO_METHODS_FILTERS) || defined(Z7_USE_FILTER_ARMT) CASE_BRA_CONV(ARMT) + #endif default: return SZ_ERROR_UNSUPPORTED; } + continue; } - } - #endif + #endif + } // (c == 1) + #endif else return SZ_ERROR_UNSUPPORTED; } @@ -556,7 +604,7 @@ static SRes SzFolder_Decode2(const CSzFolder *folder, SRes SzAr_DecodeFolder(const CSzAr *p, UInt32 folderIndex, - ILookInStream *inStream, UInt64 startPos, + ILookInStreamPtr inStream, UInt64 startPos, Byte *outBuffer, size_t outSize, ISzAllocPtr allocMain) { diff --git a/libraries/lzma/C/7zFile.c b/libraries/lzma/C/7zFile.c new file mode 100644 index 000000000..ba5daa133 --- /dev/null +++ b/libraries/lzma/C/7zFile.c @@ -0,0 +1,443 @@ +/* 7zFile.c -- File IO +2023-04-02 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include "7zFile.h" + +#ifndef USE_WINDOWS_FILE + + #include + + #ifndef USE_FOPEN + #include + #include + #ifdef _WIN32 + #include + typedef int ssize_t; + typedef int off_t; + #else + #include + #endif + #endif + +#else + +/* + ReadFile and WriteFile functions in Windows have BUG: + If you Read or Write 64MB or more (probably min_failure_size = 64MB - 32KB + 1) + from/to Network file, it returns ERROR_NO_SYSTEM_RESOURCES + (Insufficient system resources exist to complete the requested service). + Probably in some version of Windows there are problems with other sizes: + for 32 MB (maybe also for 16 MB). + And message can be "Network connection was lost" +*/ + +#endif + +#define kChunkSizeMax (1 << 22) + +void File_Construct(CSzFile *p) +{ + #ifdef USE_WINDOWS_FILE + p->handle = INVALID_HANDLE_VALUE; + #elif defined(USE_FOPEN) + p->file = NULL; + #else + p->fd = -1; + #endif +} + +#if !defined(UNDER_CE) || !defined(USE_WINDOWS_FILE) + +static WRes File_Open(CSzFile *p, const char *name, int writeMode) +{ + #ifdef USE_WINDOWS_FILE + + p->handle = CreateFileA(name, + writeMode ? GENERIC_WRITE : GENERIC_READ, + FILE_SHARE_READ, NULL, + writeMode ? CREATE_ALWAYS : OPEN_EXISTING, + FILE_ATTRIBUTE_NORMAL, NULL); + return (p->handle != INVALID_HANDLE_VALUE) ? 0 : GetLastError(); + + #elif defined(USE_FOPEN) + + p->file = fopen(name, writeMode ? "wb+" : "rb"); + return (p->file != 0) ? 0 : + #ifdef UNDER_CE + 2; /* ENOENT */ + #else + errno; + #endif + + #else + + int flags = (writeMode ? (O_CREAT | O_EXCL | O_WRONLY) : O_RDONLY); + #ifdef O_BINARY + flags |= O_BINARY; + #endif + p->fd = open(name, flags, 0666); + return (p->fd != -1) ? 0 : errno; + + #endif +} + +WRes InFile_Open(CSzFile *p, const char *name) { return File_Open(p, name, 0); } + +WRes OutFile_Open(CSzFile *p, const char *name) +{ + #if defined(USE_WINDOWS_FILE) || defined(USE_FOPEN) + return File_Open(p, name, 1); + #else + p->fd = creat(name, 0666); + return (p->fd != -1) ? 0 : errno; + #endif +} + +#endif + + +#ifdef USE_WINDOWS_FILE +static WRes File_OpenW(CSzFile *p, const WCHAR *name, int writeMode) +{ + p->handle = CreateFileW(name, + writeMode ? GENERIC_WRITE : GENERIC_READ, + FILE_SHARE_READ, NULL, + writeMode ? CREATE_ALWAYS : OPEN_EXISTING, + FILE_ATTRIBUTE_NORMAL, NULL); + return (p->handle != INVALID_HANDLE_VALUE) ? 0 : GetLastError(); +} +WRes InFile_OpenW(CSzFile *p, const WCHAR *name) { return File_OpenW(p, name, 0); } +WRes OutFile_OpenW(CSzFile *p, const WCHAR *name) { return File_OpenW(p, name, 1); } +#endif + +WRes File_Close(CSzFile *p) +{ + #ifdef USE_WINDOWS_FILE + + if (p->handle != INVALID_HANDLE_VALUE) + { + if (!CloseHandle(p->handle)) + return GetLastError(); + p->handle = INVALID_HANDLE_VALUE; + } + + #elif defined(USE_FOPEN) + + if (p->file != NULL) + { + int res = fclose(p->file); + if (res != 0) + { + if (res == EOF) + return errno; + return res; + } + p->file = NULL; + } + + #else + + if (p->fd != -1) + { + if (close(p->fd) != 0) + return errno; + p->fd = -1; + } + + #endif + + return 0; +} + + +WRes File_Read(CSzFile *p, void *data, size_t *size) +{ + size_t originalSize = *size; + *size = 0; + if (originalSize == 0) + return 0; + + #ifdef USE_WINDOWS_FILE + + do + { + const DWORD curSize = (originalSize > kChunkSizeMax) ? kChunkSizeMax : (DWORD)originalSize; + DWORD processed = 0; + const BOOL res = ReadFile(p->handle, data, curSize, &processed, NULL); + data = (void *)((Byte *)data + processed); + originalSize -= processed; + *size += processed; + if (!res) + return GetLastError(); + // debug : we can break here for partial reading mode + if (processed == 0) + break; + } + while (originalSize > 0); + + #elif defined(USE_FOPEN) + + do + { + const size_t curSize = (originalSize > kChunkSizeMax) ? kChunkSizeMax : originalSize; + const size_t processed = fread(data, 1, curSize, p->file); + data = (void *)((Byte *)data + (size_t)processed); + originalSize -= processed; + *size += processed; + if (processed != curSize) + return ferror(p->file); + // debug : we can break here for partial reading mode + if (processed == 0) + break; + } + while (originalSize > 0); + + #else + + do + { + const size_t curSize = (originalSize > kChunkSizeMax) ? kChunkSizeMax : originalSize; + const ssize_t processed = read(p->fd, data, curSize); + if (processed == -1) + return errno; + if (processed == 0) + break; + data = (void *)((Byte *)data + (size_t)processed); + originalSize -= (size_t)processed; + *size += (size_t)processed; + // debug : we can break here for partial reading mode + // break; + } + while (originalSize > 0); + + #endif + + return 0; +} + + +WRes File_Write(CSzFile *p, const void *data, size_t *size) +{ + size_t originalSize = *size; + *size = 0; + if (originalSize == 0) + return 0; + + #ifdef USE_WINDOWS_FILE + + do + { + const DWORD curSize = (originalSize > kChunkSizeMax) ? kChunkSizeMax : (DWORD)originalSize; + DWORD processed = 0; + const BOOL res = WriteFile(p->handle, data, curSize, &processed, NULL); + data = (const void *)((const Byte *)data + processed); + originalSize -= processed; + *size += processed; + if (!res) + return GetLastError(); + if (processed == 0) + break; + } + while (originalSize > 0); + + #elif defined(USE_FOPEN) + + do + { + const size_t curSize = (originalSize > kChunkSizeMax) ? kChunkSizeMax : originalSize; + const size_t processed = fwrite(data, 1, curSize, p->file); + data = (void *)((Byte *)data + (size_t)processed); + originalSize -= processed; + *size += processed; + if (processed != curSize) + return ferror(p->file); + if (processed == 0) + break; + } + while (originalSize > 0); + + #else + + do + { + const size_t curSize = (originalSize > kChunkSizeMax) ? kChunkSizeMax : originalSize; + const ssize_t processed = write(p->fd, data, curSize); + if (processed == -1) + return errno; + if (processed == 0) + break; + data = (const void *)((const Byte *)data + (size_t)processed); + originalSize -= (size_t)processed; + *size += (size_t)processed; + } + while (originalSize > 0); + + #endif + + return 0; +} + + +WRes File_Seek(CSzFile *p, Int64 *pos, ESzSeek origin) +{ + #ifdef USE_WINDOWS_FILE + + DWORD moveMethod; + UInt32 low = (UInt32)*pos; + LONG high = (LONG)((UInt64)*pos >> 16 >> 16); /* for case when UInt64 is 32-bit only */ + // (int) to eliminate clang warning + switch ((int)origin) + { + case SZ_SEEK_SET: moveMethod = FILE_BEGIN; break; + case SZ_SEEK_CUR: moveMethod = FILE_CURRENT; break; + case SZ_SEEK_END: moveMethod = FILE_END; break; + default: return ERROR_INVALID_PARAMETER; + } + low = SetFilePointer(p->handle, (LONG)low, &high, moveMethod); + if (low == (UInt32)0xFFFFFFFF) + { + WRes res = GetLastError(); + if (res != NO_ERROR) + return res; + } + *pos = ((Int64)high << 32) | low; + return 0; + + #else + + int moveMethod; // = origin; + + switch ((int)origin) + { + case SZ_SEEK_SET: moveMethod = SEEK_SET; break; + case SZ_SEEK_CUR: moveMethod = SEEK_CUR; break; + case SZ_SEEK_END: moveMethod = SEEK_END; break; + default: return EINVAL; + } + + #if defined(USE_FOPEN) + { + int res = fseek(p->file, (long)*pos, moveMethod); + if (res == -1) + return errno; + *pos = ftell(p->file); + if (*pos == -1) + return errno; + return 0; + } + #else + { + off_t res = lseek(p->fd, (off_t)*pos, moveMethod); + if (res == -1) + return errno; + *pos = res; + return 0; + } + + #endif // USE_FOPEN + #endif // USE_WINDOWS_FILE +} + + +WRes File_GetLength(CSzFile *p, UInt64 *length) +{ + #ifdef USE_WINDOWS_FILE + + DWORD sizeHigh; + DWORD sizeLow = GetFileSize(p->handle, &sizeHigh); + if (sizeLow == 0xFFFFFFFF) + { + DWORD res = GetLastError(); + if (res != NO_ERROR) + return res; + } + *length = (((UInt64)sizeHigh) << 32) + sizeLow; + return 0; + + #elif defined(USE_FOPEN) + + long pos = ftell(p->file); + int res = fseek(p->file, 0, SEEK_END); + *length = ftell(p->file); + fseek(p->file, pos, SEEK_SET); + return res; + + #else + + off_t pos; + *length = 0; + pos = lseek(p->fd, 0, SEEK_CUR); + if (pos != -1) + { + const off_t len2 = lseek(p->fd, 0, SEEK_END); + const off_t res2 = lseek(p->fd, pos, SEEK_SET); + if (len2 != -1) + { + *length = (UInt64)len2; + if (res2 != -1) + return 0; + } + } + return errno; + + #endif +} + + +/* ---------- FileSeqInStream ---------- */ + +static SRes FileSeqInStream_Read(ISeqInStreamPtr pp, void *buf, size_t *size) +{ + Z7_CONTAINER_FROM_VTBL_TO_DECL_VAR_pp_vt_p(CFileSeqInStream) + const WRes wres = File_Read(&p->file, buf, size); + p->wres = wres; + return (wres == 0) ? SZ_OK : SZ_ERROR_READ; +} + +void FileSeqInStream_CreateVTable(CFileSeqInStream *p) +{ + p->vt.Read = FileSeqInStream_Read; +} + + +/* ---------- FileInStream ---------- */ + +static SRes FileInStream_Read(ISeekInStreamPtr pp, void *buf, size_t *size) +{ + Z7_CONTAINER_FROM_VTBL_TO_DECL_VAR_pp_vt_p(CFileInStream) + const WRes wres = File_Read(&p->file, buf, size); + p->wres = wres; + return (wres == 0) ? SZ_OK : SZ_ERROR_READ; +} + +static SRes FileInStream_Seek(ISeekInStreamPtr pp, Int64 *pos, ESzSeek origin) +{ + Z7_CONTAINER_FROM_VTBL_TO_DECL_VAR_pp_vt_p(CFileInStream) + const WRes wres = File_Seek(&p->file, pos, origin); + p->wres = wres; + return (wres == 0) ? SZ_OK : SZ_ERROR_READ; +} + +void FileInStream_CreateVTable(CFileInStream *p) +{ + p->vt.Read = FileInStream_Read; + p->vt.Seek = FileInStream_Seek; +} + + +/* ---------- FileOutStream ---------- */ + +static size_t FileOutStream_Write(ISeqOutStreamPtr pp, const void *data, size_t size) +{ + Z7_CONTAINER_FROM_VTBL_TO_DECL_VAR_pp_vt_p(CFileOutStream) + const WRes wres = File_Write(&p->file, data, &size); + p->wres = wres; + return size; +} + +void FileOutStream_CreateVTable(CFileOutStream *p) +{ + p->vt.Write = FileOutStream_Write; +} diff --git a/libraries/lzma/C/7zFile.h b/libraries/lzma/C/7zFile.h new file mode 100644 index 000000000..f5069cd9e --- /dev/null +++ b/libraries/lzma/C/7zFile.h @@ -0,0 +1,92 @@ +/* 7zFile.h -- File IO +2023-03-05 : Igor Pavlov : Public domain */ + +#ifndef ZIP7_INC_FILE_H +#define ZIP7_INC_FILE_H + +#ifdef _WIN32 +#define USE_WINDOWS_FILE +// #include +#endif + +#ifdef USE_WINDOWS_FILE +#include "7zWindows.h" + +#else +// note: USE_FOPEN mode is limited to 32-bit file size +// #define USE_FOPEN +// #include +#endif + +#include "7zTypes.h" + +EXTERN_C_BEGIN + +/* ---------- File ---------- */ + +typedef struct +{ + #ifdef USE_WINDOWS_FILE + HANDLE handle; + #elif defined(USE_FOPEN) + FILE *file; + #else + int fd; + #endif +} CSzFile; + +void File_Construct(CSzFile *p); +#if !defined(UNDER_CE) || !defined(USE_WINDOWS_FILE) +WRes InFile_Open(CSzFile *p, const char *name); +WRes OutFile_Open(CSzFile *p, const char *name); +#endif +#ifdef USE_WINDOWS_FILE +WRes InFile_OpenW(CSzFile *p, const WCHAR *name); +WRes OutFile_OpenW(CSzFile *p, const WCHAR *name); +#endif +WRes File_Close(CSzFile *p); + +/* reads max(*size, remain file's size) bytes */ +WRes File_Read(CSzFile *p, void *data, size_t *size); + +/* writes *size bytes */ +WRes File_Write(CSzFile *p, const void *data, size_t *size); + +WRes File_Seek(CSzFile *p, Int64 *pos, ESzSeek origin); +WRes File_GetLength(CSzFile *p, UInt64 *length); + + +/* ---------- FileInStream ---------- */ + +typedef struct +{ + ISeqInStream vt; + CSzFile file; + WRes wres; +} CFileSeqInStream; + +void FileSeqInStream_CreateVTable(CFileSeqInStream *p); + + +typedef struct +{ + ISeekInStream vt; + CSzFile file; + WRes wres; +} CFileInStream; + +void FileInStream_CreateVTable(CFileInStream *p); + + +typedef struct +{ + ISeqOutStream vt; + CSzFile file; + WRes wres; +} CFileOutStream; + +void FileOutStream_CreateVTable(CFileOutStream *p); + +EXTERN_C_END + +#endif diff --git a/libraries/lzma/C/7zStream.c b/libraries/lzma/C/7zStream.c index 28a14604f..74e75b65a 100644 --- a/libraries/lzma/C/7zStream.c +++ b/libraries/lzma/C/7zStream.c @@ -1,5 +1,5 @@ /* 7zStream.c -- 7z Stream functions -2021-02-09 : Igor Pavlov : Public domain */ +2023-04-02 : Igor Pavlov : Public domain */ #include "Precomp.h" @@ -7,12 +7,33 @@ #include "7zTypes.h" -SRes SeqInStream_Read2(const ISeqInStream *stream, void *buf, size_t size, SRes errorType) + +SRes SeqInStream_ReadMax(ISeqInStreamPtr stream, void *buf, size_t *processedSize) +{ + size_t size = *processedSize; + *processedSize = 0; + while (size != 0) + { + size_t cur = size; + const SRes res = ISeqInStream_Read(stream, buf, &cur); + *processedSize += cur; + buf = (void *)((Byte *)buf + cur); + size -= cur; + if (res != SZ_OK) + return res; + if (cur == 0) + return SZ_OK; + } + return SZ_OK; +} + +/* +SRes SeqInStream_Read2(ISeqInStreamPtr stream, void *buf, size_t size, SRes errorType) { while (size != 0) { size_t processed = size; - RINOK(ISeqInStream_Read(stream, buf, &processed)); + RINOK(ISeqInStream_Read(stream, buf, &processed)) if (processed == 0) return errorType; buf = (void *)((Byte *)buf + processed); @@ -21,42 +42,44 @@ SRes SeqInStream_Read2(const ISeqInStream *stream, void *buf, size_t size, SRes return SZ_OK; } -SRes SeqInStream_Read(const ISeqInStream *stream, void *buf, size_t size) +SRes SeqInStream_Read(ISeqInStreamPtr stream, void *buf, size_t size) { return SeqInStream_Read2(stream, buf, size, SZ_ERROR_INPUT_EOF); } +*/ -SRes SeqInStream_ReadByte(const ISeqInStream *stream, Byte *buf) + +SRes SeqInStream_ReadByte(ISeqInStreamPtr stream, Byte *buf) { size_t processed = 1; - RINOK(ISeqInStream_Read(stream, buf, &processed)); + RINOK(ISeqInStream_Read(stream, buf, &processed)) return (processed == 1) ? SZ_OK : SZ_ERROR_INPUT_EOF; } -SRes LookInStream_SeekTo(const ILookInStream *stream, UInt64 offset) +SRes LookInStream_SeekTo(ILookInStreamPtr stream, UInt64 offset) { Int64 t = (Int64)offset; return ILookInStream_Seek(stream, &t, SZ_SEEK_SET); } -SRes LookInStream_LookRead(const ILookInStream *stream, void *buf, size_t *size) +SRes LookInStream_LookRead(ILookInStreamPtr stream, void *buf, size_t *size) { const void *lookBuf; if (*size == 0) return SZ_OK; - RINOK(ILookInStream_Look(stream, &lookBuf, size)); + RINOK(ILookInStream_Look(stream, &lookBuf, size)) memcpy(buf, lookBuf, *size); return ILookInStream_Skip(stream, *size); } -SRes LookInStream_Read2(const ILookInStream *stream, void *buf, size_t size, SRes errorType) +SRes LookInStream_Read2(ILookInStreamPtr stream, void *buf, size_t size, SRes errorType) { while (size != 0) { size_t processed = size; - RINOK(ILookInStream_Read(stream, buf, &processed)); + RINOK(ILookInStream_Read(stream, buf, &processed)) if (processed == 0) return errorType; buf = (void *)((Byte *)buf + processed); @@ -65,16 +88,16 @@ SRes LookInStream_Read2(const ILookInStream *stream, void *buf, size_t size, SRe return SZ_OK; } -SRes LookInStream_Read(const ILookInStream *stream, void *buf, size_t size) +SRes LookInStream_Read(ILookInStreamPtr stream, void *buf, size_t size) { return LookInStream_Read2(stream, buf, size, SZ_ERROR_INPUT_EOF); } -#define GET_LookToRead2 CLookToRead2 *p = CONTAINER_FROM_VTBL(pp, CLookToRead2, vt); +#define GET_LookToRead2 Z7_CONTAINER_FROM_VTBL_TO_DECL_VAR_pp_vt_p(CLookToRead2) -static SRes LookToRead2_Look_Lookahead(const ILookInStream *pp, const void **buf, size_t *size) +static SRes LookToRead2_Look_Lookahead(ILookInStreamPtr pp, const void **buf, size_t *size) { SRes res = SZ_OK; GET_LookToRead2 @@ -93,7 +116,7 @@ static SRes LookToRead2_Look_Lookahead(const ILookInStream *pp, const void **buf return res; } -static SRes LookToRead2_Look_Exact(const ILookInStream *pp, const void **buf, size_t *size) +static SRes LookToRead2_Look_Exact(ILookInStreamPtr pp, const void **buf, size_t *size) { SRes res = SZ_OK; GET_LookToRead2 @@ -113,14 +136,14 @@ static SRes LookToRead2_Look_Exact(const ILookInStream *pp, const void **buf, si return res; } -static SRes LookToRead2_Skip(const ILookInStream *pp, size_t offset) +static SRes LookToRead2_Skip(ILookInStreamPtr pp, size_t offset) { GET_LookToRead2 p->pos += offset; return SZ_OK; } -static SRes LookToRead2_Read(const ILookInStream *pp, void *buf, size_t *size) +static SRes LookToRead2_Read(ILookInStreamPtr pp, void *buf, size_t *size) { GET_LookToRead2 size_t rem = p->size - p->pos; @@ -134,7 +157,7 @@ static SRes LookToRead2_Read(const ILookInStream *pp, void *buf, size_t *size) return SZ_OK; } -static SRes LookToRead2_Seek(const ILookInStream *pp, Int64 *pos, ESzSeek origin) +static SRes LookToRead2_Seek(ILookInStreamPtr pp, Int64 *pos, ESzSeek origin) { GET_LookToRead2 p->pos = p->size = 0; @@ -153,9 +176,9 @@ void LookToRead2_CreateVTable(CLookToRead2 *p, int lookahead) -static SRes SecToLook_Read(const ISeqInStream *pp, void *buf, size_t *size) +static SRes SecToLook_Read(ISeqInStreamPtr pp, void *buf, size_t *size) { - CSecToLook *p = CONTAINER_FROM_VTBL(pp, CSecToLook, vt); + Z7_CONTAINER_FROM_VTBL_TO_DECL_VAR_pp_vt_p(CSecToLook) return LookInStream_LookRead(p->realStream, buf, size); } @@ -164,9 +187,9 @@ void SecToLook_CreateVTable(CSecToLook *p) p->vt.Read = SecToLook_Read; } -static SRes SecToRead_Read(const ISeqInStream *pp, void *buf, size_t *size) +static SRes SecToRead_Read(ISeqInStreamPtr pp, void *buf, size_t *size) { - CSecToRead *p = CONTAINER_FROM_VTBL(pp, CSecToRead, vt); + Z7_CONTAINER_FROM_VTBL_TO_DECL_VAR_pp_vt_p(CSecToRead) return ILookInStream_Read(p->realStream, buf, size); } diff --git a/libraries/lzma/C/7zTypes.h b/libraries/lzma/C/7zTypes.h index fe4fde3ff..1fcb2473e 100644 --- a/libraries/lzma/C/7zTypes.h +++ b/libraries/lzma/C/7zTypes.h @@ -1,8 +1,8 @@ /* 7zTypes.h -- Basic types -2021-12-25 : Igor Pavlov : Public domain */ +2023-04-02 : Igor Pavlov : Public domain */ -#ifndef __7Z_TYPES_H -#define __7Z_TYPES_H +#ifndef ZIP7_7Z_TYPES_H +#define ZIP7_7Z_TYPES_H #ifdef _WIN32 /* #include */ @@ -52,6 +52,11 @@ typedef int SRes; #define MY_ALIGN(n) #endif #else + /* + // C11/C++11: + #include + #define MY_ALIGN(n) alignas(n) + */ #define MY_ALIGN(n) __attribute__ ((aligned(n))) #endif @@ -62,7 +67,7 @@ typedef int SRes; typedef unsigned WRes; #define MY_SRes_HRESULT_FROM_WRes(x) HRESULT_FROM_WIN32(x) -// #define MY_HRES_ERROR__INTERNAL_ERROR MY_SRes_HRESULT_FROM_WRes(ERROR_INTERNAL_ERROR) +// #define MY_HRES_ERROR_INTERNAL_ERROR MY_SRes_HRESULT_FROM_WRes(ERROR_INTERNAL_ERROR) #else // _WIN32 @@ -70,13 +75,13 @@ typedef unsigned WRes; typedef int WRes; // (FACILITY_ERRNO = 0x800) is 7zip's FACILITY constant to represent (errno) errors in HRESULT -#define MY__FACILITY_ERRNO 0x800 -#define MY__FACILITY_WIN32 7 -#define MY__FACILITY__WRes MY__FACILITY_ERRNO +#define MY_FACILITY_ERRNO 0x800 +#define MY_FACILITY_WIN32 7 +#define MY_FACILITY_WRes MY_FACILITY_ERRNO #define MY_HRESULT_FROM_errno_CONST_ERROR(x) ((HRESULT)( \ ( (HRESULT)(x) & 0x0000FFFF) \ - | (MY__FACILITY__WRes << 16) \ + | (MY_FACILITY_WRes << 16) \ | (HRESULT)0x80000000 )) #define MY_SRes_HRESULT_FROM_WRes(x) \ @@ -120,23 +125,19 @@ typedef int WRes; #define ERROR_INVALID_REPARSE_DATA ((HRESULT)0x80071128L) #define ERROR_REPARSE_TAG_INVALID ((HRESULT)0x80071129L) -// if (MY__FACILITY__WRes != FACILITY_WIN32), +// if (MY_FACILITY_WRes != FACILITY_WIN32), // we use FACILITY_WIN32 for COM errors: #define E_OUTOFMEMORY ((HRESULT)0x8007000EL) #define E_INVALIDARG ((HRESULT)0x80070057L) -#define MY__E_ERROR_NEGATIVE_SEEK ((HRESULT)0x80070083L) +#define MY_E_ERROR_NEGATIVE_SEEK ((HRESULT)0x80070083L) /* // we can use FACILITY_ERRNO for some COM errors, that have errno equivalents: #define E_OUTOFMEMORY MY_HRESULT_FROM_errno_CONST_ERROR(ENOMEM) #define E_INVALIDARG MY_HRESULT_FROM_errno_CONST_ERROR(EINVAL) -#define MY__E_ERROR_NEGATIVE_SEEK MY_HRESULT_FROM_errno_CONST_ERROR(EINVAL) +#define MY_E_ERROR_NEGATIVE_SEEK MY_HRESULT_FROM_errno_CONST_ERROR(EINVAL) */ -// gcc / clang : (sizeof(long) == sizeof(void*)) in 32/64 bits -typedef long INT_PTR; -typedef unsigned long UINT_PTR; - #define TEXT(quote) quote #define FILE_ATTRIBUTE_READONLY 0x0001 @@ -160,18 +161,18 @@ typedef unsigned long UINT_PTR; #ifndef RINOK -#define RINOK(x) { int __result__ = (x); if (__result__ != 0) return __result__; } +#define RINOK(x) { const int _result_ = (x); if (_result_ != 0) return _result_; } #endif #ifndef RINOK_WRes -#define RINOK_WRes(x) { WRes __result__ = (x); if (__result__ != 0) return __result__; } +#define RINOK_WRes(x) { const WRes _result_ = (x); if (_result_ != 0) return _result_; } #endif typedef unsigned char Byte; typedef short Int16; typedef unsigned short UInt16; -#ifdef _LZMA_UINT32_IS_ULONG +#ifdef Z7_DECL_Int32_AS_long typedef long Int32; typedef unsigned long UInt32; #else @@ -210,37 +211,51 @@ typedef size_t SIZE_T; #endif // _WIN32 -#define MY_HRES_ERROR__INTERNAL_ERROR ((HRESULT)0x8007054FL) +#define MY_HRES_ERROR_INTERNAL_ERROR ((HRESULT)0x8007054FL) -#ifdef _SZ_NO_INT_64 - -/* define _SZ_NO_INT_64, if your compiler doesn't support 64-bit integers. - NOTES: Some code will work incorrectly in that case! */ +#ifdef Z7_DECL_Int64_AS_long typedef long Int64; typedef unsigned long UInt64; #else -#if defined(_MSC_VER) || defined(__BORLANDC__) +#if (defined(_MSC_VER) || defined(__BORLANDC__)) && !defined(__clang__) typedef __int64 Int64; typedef unsigned __int64 UInt64; -#define UINT64_CONST(n) n +#else +#if defined(__clang__) || defined(__GNUC__) +#include +typedef int64_t Int64; +typedef uint64_t UInt64; #else typedef long long int Int64; typedef unsigned long long int UInt64; -#define UINT64_CONST(n) n ## ULL +// #define UINT64_CONST(n) n ## ULL +#endif #endif #endif -#ifdef _LZMA_NO_SYSTEM_SIZE_T -typedef UInt32 SizeT; +#define UINT64_CONST(n) n + + +#ifdef Z7_DECL_SizeT_AS_unsigned_int +typedef unsigned int SizeT; #else typedef size_t SizeT; #endif +/* +#if (defined(_MSC_VER) && _MSC_VER <= 1200) +typedef size_t MY_uintptr_t; +#else +#include +typedef uintptr_t MY_uintptr_t; +#endif +*/ + typedef int BoolInt; /* typedef BoolInt Bool; */ #define True 1 @@ -248,23 +263,23 @@ typedef int BoolInt; #ifdef _WIN32 -#define MY_STD_CALL __stdcall +#define Z7_STDCALL __stdcall #else -#define MY_STD_CALL +#define Z7_STDCALL #endif #ifdef _MSC_VER #if _MSC_VER >= 1300 -#define MY_NO_INLINE __declspec(noinline) +#define Z7_NO_INLINE __declspec(noinline) #else -#define MY_NO_INLINE +#define Z7_NO_INLINE #endif -#define MY_FORCE_INLINE __forceinline +#define Z7_FORCE_INLINE __forceinline -#define MY_CDECL __cdecl -#define MY_FAST_CALL __fastcall +#define Z7_CDECL __cdecl +#define Z7_FASTCALL __fastcall #else // _MSC_VER @@ -272,27 +287,25 @@ typedef int BoolInt; || (defined(__clang__) && (__clang_major__ >= 4)) \ || defined(__INTEL_COMPILER) \ || defined(__xlC__) -#define MY_NO_INLINE __attribute__((noinline)) -// #define MY_FORCE_INLINE __attribute__((always_inline)) inline +#define Z7_NO_INLINE __attribute__((noinline)) +#define Z7_FORCE_INLINE __attribute__((always_inline)) inline #else -#define MY_NO_INLINE +#define Z7_NO_INLINE +#define Z7_FORCE_INLINE #endif -#define MY_FORCE_INLINE - - -#define MY_CDECL +#define Z7_CDECL #if defined(_M_IX86) \ || defined(__i386__) -// #define MY_FAST_CALL __attribute__((fastcall)) -// #define MY_FAST_CALL __attribute__((cdecl)) -#define MY_FAST_CALL +// #define Z7_FASTCALL __attribute__((fastcall)) +// #define Z7_FASTCALL __attribute__((cdecl)) +#define Z7_FASTCALL #elif defined(MY_CPU_AMD64) -// #define MY_FAST_CALL __attribute__((ms_abi)) -#define MY_FAST_CALL +// #define Z7_FASTCALL __attribute__((ms_abi)) +#define Z7_FASTCALL #else -#define MY_FAST_CALL +#define Z7_FASTCALL #endif #endif // _MSC_VER @@ -300,41 +313,49 @@ typedef int BoolInt; /* The following interfaces use first parameter as pointer to structure */ -typedef struct IByteIn IByteIn; -struct IByteIn +// #define Z7_C_IFACE_CONST_QUAL +#define Z7_C_IFACE_CONST_QUAL const + +#define Z7_C_IFACE_DECL(a) \ + struct a ## _; \ + typedef Z7_C_IFACE_CONST_QUAL struct a ## _ * a ## Ptr; \ + typedef struct a ## _ a; \ + struct a ## _ + + +Z7_C_IFACE_DECL (IByteIn) { - Byte (*Read)(const IByteIn *p); /* reads one byte, returns 0 in case of EOF or error */ + Byte (*Read)(IByteInPtr p); /* reads one byte, returns 0 in case of EOF or error */ }; #define IByteIn_Read(p) (p)->Read(p) -typedef struct IByteOut IByteOut; -struct IByteOut +Z7_C_IFACE_DECL (IByteOut) { - void (*Write)(const IByteOut *p, Byte b); + void (*Write)(IByteOutPtr p, Byte b); }; #define IByteOut_Write(p, b) (p)->Write(p, b) -typedef struct ISeqInStream ISeqInStream; -struct ISeqInStream +Z7_C_IFACE_DECL (ISeqInStream) { - SRes (*Read)(const ISeqInStream *p, void *buf, size_t *size); + SRes (*Read)(ISeqInStreamPtr p, void *buf, size_t *size); /* if (input(*size) != 0 && output(*size) == 0) means end_of_stream. (output(*size) < input(*size)) is allowed */ }; #define ISeqInStream_Read(p, buf, size) (p)->Read(p, buf, size) +/* try to read as much as avail in stream and limited by (*processedSize) */ +SRes SeqInStream_ReadMax(ISeqInStreamPtr stream, void *buf, size_t *processedSize); /* it can return SZ_ERROR_INPUT_EOF */ -SRes SeqInStream_Read(const ISeqInStream *stream, void *buf, size_t size); -SRes SeqInStream_Read2(const ISeqInStream *stream, void *buf, size_t size, SRes errorType); -SRes SeqInStream_ReadByte(const ISeqInStream *stream, Byte *buf); +// SRes SeqInStream_Read(ISeqInStreamPtr stream, void *buf, size_t size); +// SRes SeqInStream_Read2(ISeqInStreamPtr stream, void *buf, size_t size, SRes errorType); +SRes SeqInStream_ReadByte(ISeqInStreamPtr stream, Byte *buf); -typedef struct ISeqOutStream ISeqOutStream; -struct ISeqOutStream +Z7_C_IFACE_DECL (ISeqOutStream) { - size_t (*Write)(const ISeqOutStream *p, const void *buf, size_t size); + size_t (*Write)(ISeqOutStreamPtr p, const void *buf, size_t size); /* Returns: result - the number of actually written bytes. (result < size) means error */ }; @@ -348,29 +369,26 @@ typedef enum } ESzSeek; -typedef struct ISeekInStream ISeekInStream; -struct ISeekInStream +Z7_C_IFACE_DECL (ISeekInStream) { - SRes (*Read)(const ISeekInStream *p, void *buf, size_t *size); /* same as ISeqInStream::Read */ - SRes (*Seek)(const ISeekInStream *p, Int64 *pos, ESzSeek origin); + SRes (*Read)(ISeekInStreamPtr p, void *buf, size_t *size); /* same as ISeqInStream::Read */ + SRes (*Seek)(ISeekInStreamPtr p, Int64 *pos, ESzSeek origin); }; #define ISeekInStream_Read(p, buf, size) (p)->Read(p, buf, size) #define ISeekInStream_Seek(p, pos, origin) (p)->Seek(p, pos, origin) -typedef struct ILookInStream ILookInStream; -struct ILookInStream +Z7_C_IFACE_DECL (ILookInStream) { - SRes (*Look)(const ILookInStream *p, const void **buf, size_t *size); + SRes (*Look)(ILookInStreamPtr p, const void **buf, size_t *size); /* if (input(*size) != 0 && output(*size) == 0) means end_of_stream. (output(*size) > input(*size)) is not allowed (output(*size) < input(*size)) is allowed */ - SRes (*Skip)(const ILookInStream *p, size_t offset); + SRes (*Skip)(ILookInStreamPtr p, size_t offset); /* offset must be <= output(*size) of Look */ - - SRes (*Read)(const ILookInStream *p, void *buf, size_t *size); + SRes (*Read)(ILookInStreamPtr p, void *buf, size_t *size); /* reads directly (without buffer). It's same as ISeqInStream::Read */ - SRes (*Seek)(const ILookInStream *p, Int64 *pos, ESzSeek origin); + SRes (*Seek)(ILookInStreamPtr p, Int64 *pos, ESzSeek origin); }; #define ILookInStream_Look(p, buf, size) (p)->Look(p, buf, size) @@ -379,19 +397,18 @@ struct ILookInStream #define ILookInStream_Seek(p, pos, origin) (p)->Seek(p, pos, origin) -SRes LookInStream_LookRead(const ILookInStream *stream, void *buf, size_t *size); -SRes LookInStream_SeekTo(const ILookInStream *stream, UInt64 offset); +SRes LookInStream_LookRead(ILookInStreamPtr stream, void *buf, size_t *size); +SRes LookInStream_SeekTo(ILookInStreamPtr stream, UInt64 offset); /* reads via ILookInStream::Read */ -SRes LookInStream_Read2(const ILookInStream *stream, void *buf, size_t size, SRes errorType); -SRes LookInStream_Read(const ILookInStream *stream, void *buf, size_t size); - +SRes LookInStream_Read2(ILookInStreamPtr stream, void *buf, size_t size, SRes errorType); +SRes LookInStream_Read(ILookInStreamPtr stream, void *buf, size_t size); typedef struct { ILookInStream vt; - const ISeekInStream *realStream; + ISeekInStreamPtr realStream; size_t pos; size_t size; /* it's data size */ @@ -403,13 +420,13 @@ typedef struct void LookToRead2_CreateVTable(CLookToRead2 *p, int lookahead); -#define LookToRead2_Init(p) { (p)->pos = (p)->size = 0; } +#define LookToRead2_INIT(p) { (p)->pos = (p)->size = 0; } typedef struct { ISeqInStream vt; - const ILookInStream *realStream; + ILookInStreamPtr realStream; } CSecToLook; void SecToLook_CreateVTable(CSecToLook *p); @@ -419,20 +436,19 @@ void SecToLook_CreateVTable(CSecToLook *p); typedef struct { ISeqInStream vt; - const ILookInStream *realStream; + ILookInStreamPtr realStream; } CSecToRead; void SecToRead_CreateVTable(CSecToRead *p); -typedef struct ICompressProgress ICompressProgress; - -struct ICompressProgress +Z7_C_IFACE_DECL (ICompressProgress) { - SRes (*Progress)(const ICompressProgress *p, UInt64 inSize, UInt64 outSize); + SRes (*Progress)(ICompressProgressPtr p, UInt64 inSize, UInt64 outSize); /* Returns: result. (result != SZ_OK) means break. Value (UInt64)(Int64)-1 for size means unknown value. */ }; + #define ICompressProgress_Progress(p, inSize, outSize) (p)->Progress(p, inSize, outSize) @@ -470,13 +486,13 @@ struct ISzAlloc -#ifndef MY_container_of +#ifndef Z7_container_of /* -#define MY_container_of(ptr, type, m) container_of(ptr, type, m) -#define MY_container_of(ptr, type, m) CONTAINING_RECORD(ptr, type, m) -#define MY_container_of(ptr, type, m) ((type *)((char *)(ptr) - offsetof(type, m))) -#define MY_container_of(ptr, type, m) (&((type *)0)->m == (ptr), ((type *)(((char *)(ptr)) - MY_offsetof(type, m)))) +#define Z7_container_of(ptr, type, m) container_of(ptr, type, m) +#define Z7_container_of(ptr, type, m) CONTAINING_RECORD(ptr, type, m) +#define Z7_container_of(ptr, type, m) ((type *)((char *)(ptr) - offsetof(type, m))) +#define Z7_container_of(ptr, type, m) (&((type *)0)->m == (ptr), ((type *)(((char *)(ptr)) - MY_offsetof(type, m)))) */ /* @@ -485,24 +501,64 @@ struct ISzAlloc GCC 4.8.1 : classes with non-public variable members" */ -#define MY_container_of(ptr, type, m) ((type *)(void *)((char *)(void *)(1 ? (ptr) : &((type *)0)->m) - MY_offsetof(type, m))) +#define Z7_container_of(ptr, type, m) \ + ((type *)(void *)((char *)(void *) \ + (1 ? (ptr) : &((type *)NULL)->m) - MY_offsetof(type, m))) + +#define Z7_container_of_CONST(ptr, type, m) \ + ((const type *)(const void *)((const char *)(const void *) \ + (1 ? (ptr) : &((type *)NULL)->m) - MY_offsetof(type, m))) + +/* +#define Z7_container_of_NON_CONST_FROM_CONST(ptr, type, m) \ + ((type *)(void *)(const void *)((const char *)(const void *) \ + (1 ? (ptr) : &((type *)NULL)->m) - MY_offsetof(type, m))) +*/ #endif -#define CONTAINER_FROM_VTBL_SIMPLE(ptr, type, m) ((type *)(void *)(ptr)) +#define Z7_CONTAINER_FROM_VTBL_SIMPLE(ptr, type, m) ((type *)(void *)(ptr)) +// #define Z7_CONTAINER_FROM_VTBL(ptr, type, m) Z7_CONTAINER_FROM_VTBL_SIMPLE(ptr, type, m) +#define Z7_CONTAINER_FROM_VTBL(ptr, type, m) Z7_container_of(ptr, type, m) +// #define Z7_CONTAINER_FROM_VTBL(ptr, type, m) Z7_container_of_NON_CONST_FROM_CONST(ptr, type, m) + +#define Z7_CONTAINER_FROM_VTBL_CONST(ptr, type, m) Z7_container_of_CONST(ptr, type, m) + +#define Z7_CONTAINER_FROM_VTBL_CLS(ptr, type, m) Z7_CONTAINER_FROM_VTBL_SIMPLE(ptr, type, m) /* -#define CONTAINER_FROM_VTBL(ptr, type, m) CONTAINER_FROM_VTBL_SIMPLE(ptr, type, m) +#define Z7_CONTAINER_FROM_VTBL_CLS(ptr, type, m) Z7_CONTAINER_FROM_VTBL(ptr, type, m) */ -#define CONTAINER_FROM_VTBL(ptr, type, m) MY_container_of(ptr, type, m) +#if defined (__clang__) || defined(__GNUC__) +#define Z7_DIAGNOSCTIC_IGNORE_BEGIN_CAST_QUAL \ + _Pragma("GCC diagnostic push") \ + _Pragma("GCC diagnostic ignored \"-Wcast-qual\"") +#define Z7_DIAGNOSCTIC_IGNORE_END_CAST_QUAL \ + _Pragma("GCC diagnostic pop") +#else +#define Z7_DIAGNOSCTIC_IGNORE_BEGIN_CAST_QUAL +#define Z7_DIAGNOSCTIC_IGNORE_END_CAST_QUAL +#endif -#define CONTAINER_FROM_VTBL_CLS(ptr, type, m) CONTAINER_FROM_VTBL_SIMPLE(ptr, type, m) -/* -#define CONTAINER_FROM_VTBL_CLS(ptr, type, m) CONTAINER_FROM_VTBL(ptr, type, m) -*/ +#define Z7_CONTAINER_FROM_VTBL_TO_DECL_VAR(ptr, type, m, p) \ + Z7_DIAGNOSCTIC_IGNORE_BEGIN_CAST_QUAL \ + type *p = Z7_CONTAINER_FROM_VTBL(ptr, type, m); \ + Z7_DIAGNOSCTIC_IGNORE_END_CAST_QUAL + +#define Z7_CONTAINER_FROM_VTBL_TO_DECL_VAR_pp_vt_p(type) \ + Z7_CONTAINER_FROM_VTBL_TO_DECL_VAR(pp, type, vt, p) -#define MY_memset_0_ARRAY(a) memset((a), 0, sizeof(a)) +// #define ZIP7_DECLARE_HANDLE(name) typedef void *name; +#define Z7_DECLARE_HANDLE(name) struct name##_dummy{int unused;}; typedef struct name##_dummy *name; + + +#define Z7_memset_0_ARRAY(a) memset((a), 0, sizeof(a)) + +#ifndef Z7_ARRAY_SIZE +#define Z7_ARRAY_SIZE(a) (sizeof(a) / sizeof((a)[0])) +#endif + #ifdef _WIN32 @@ -520,6 +576,22 @@ struct ISzAlloc #endif +#define k_PropVar_TimePrec_0 0 +#define k_PropVar_TimePrec_Unix 1 +#define k_PropVar_TimePrec_DOS 2 +#define k_PropVar_TimePrec_HighPrec 3 +#define k_PropVar_TimePrec_Base 16 +#define k_PropVar_TimePrec_100ns (k_PropVar_TimePrec_Base + 7) +#define k_PropVar_TimePrec_1ns (k_PropVar_TimePrec_Base + 9) + EXTERN_C_END #endif + +/* +#ifndef Z7_ST +#ifdef _7ZIP_ST +#define Z7_ST +#endif +#endif +*/ diff --git a/libraries/lzma/C/7zVersion.h b/libraries/lzma/C/7zVersion.h index e9363d37b..754923961 100644 --- a/libraries/lzma/C/7zVersion.h +++ b/libraries/lzma/C/7zVersion.h @@ -1,7 +1,7 @@ -#define MY_VER_MAJOR 21 -#define MY_VER_MINOR 07 +#define MY_VER_MAJOR 23 +#define MY_VER_MINOR 01 #define MY_VER_BUILD 0 -#define MY_VERSION_NUMBERS "21.07" +#define MY_VERSION_NUMBERS "23.01" #define MY_VERSION MY_VERSION_NUMBERS #ifdef MY_CPU_NAME @@ -10,12 +10,12 @@ #define MY_VERSION_CPU MY_VERSION #endif -#define MY_DATE "2021-12-26" +#define MY_DATE "2023-06-20" #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) 1999-2021 Igor Pavlov" +#define MY_COPYRIGHT_CR "Copyright (c) 1999-2023 Igor Pavlov" #ifdef USE_COPYRIGHT_CR #define MY_COPYRIGHT MY_COPYRIGHT_CR diff --git a/libraries/lzma/C/7zWindows.h b/libraries/lzma/C/7zWindows.h new file mode 100644 index 000000000..42c6db8bf --- /dev/null +++ b/libraries/lzma/C/7zWindows.h @@ -0,0 +1,101 @@ +/* 7zWindows.h -- StdAfx +2023-04-02 : Igor Pavlov : Public domain */ + +#ifndef ZIP7_INC_7Z_WINDOWS_H +#define ZIP7_INC_7Z_WINDOWS_H + +#ifdef _WIN32 + +#if defined(__clang__) +# pragma clang diagnostic push +#endif + +#if defined(_MSC_VER) + +#pragma warning(push) +#pragma warning(disable : 4668) // '_WIN32_WINNT' is not defined as a preprocessor macro, replacing with '0' for '#if/#elif' + +#if _MSC_VER == 1900 +// for old kit10 versions +// #pragma warning(disable : 4255) // winuser.h(13979): warning C4255: 'GetThreadDpiAwarenessContext': +#endif +// win10 Windows Kit: +#endif // _MSC_VER + +#if defined(_MSC_VER) && _MSC_VER <= 1200 && !defined(_WIN64) +// for msvc6 without sdk2003 +#define RPC_NO_WINDOWS_H +#endif + +#if defined(__MINGW32__) || defined(__MINGW64__) +// #if defined(__GNUC__) && !defined(__clang__) +#include +#else +#include +#endif +// #include +// #include + +// but if precompiled with clang-cl then we need +// #include +#if defined(_MSC_VER) +#pragma warning(pop) +#endif + +#if defined(__clang__) +# pragma clang diagnostic pop +#endif + +#if defined(_MSC_VER) && _MSC_VER <= 1200 && !defined(_WIN64) +#ifndef _W64 + +typedef long LONG_PTR, *PLONG_PTR; +typedef unsigned long ULONG_PTR, *PULONG_PTR; +typedef ULONG_PTR DWORD_PTR, *PDWORD_PTR; + +#define Z7_OLD_WIN_SDK +#endif // _W64 +#endif // _MSC_VER == 1200 + +#ifdef Z7_OLD_WIN_SDK + +#ifndef INVALID_FILE_ATTRIBUTES +#define INVALID_FILE_ATTRIBUTES ((DWORD)-1) +#endif +#ifndef INVALID_SET_FILE_POINTER +#define INVALID_SET_FILE_POINTER ((DWORD)-1) +#endif +#ifndef FILE_SPECIAL_ACCESS +#define FILE_SPECIAL_ACCESS (FILE_ANY_ACCESS) +#endif + +// ShlObj.h: +// #define BIF_NEWDIALOGSTYLE 0x0040 + +#pragma warning(disable : 4201) +// #pragma warning(disable : 4115) + +#undef VARIANT_TRUE +#define VARIANT_TRUE ((VARIANT_BOOL)-1) +#endif + +#endif // Z7_OLD_WIN_SDK + +#ifdef UNDER_CE +#undef VARIANT_TRUE +#define VARIANT_TRUE ((VARIANT_BOOL)-1) +#endif + + +#if defined(_MSC_VER) +#if _MSC_VER >= 1400 && _MSC_VER <= 1600 + // BaseTsd.h(148) : 'HandleToULong' : unreferenced inline function has been removed + // string.h + // #pragma warning(disable : 4514) +#endif +#endif + + +/* #include "7zTypes.h" */ + +#endif diff --git a/libraries/lzma/C/Aes.c b/libraries/lzma/C/Aes.c new file mode 100644 index 000000000..bcaafab11 --- /dev/null +++ b/libraries/lzma/C/Aes.c @@ -0,0 +1,393 @@ +/* Aes.c -- AES encryption / decryption +2023-04-02 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include "CpuArch.h" +#include "Aes.h" + +AES_CODE_FUNC g_AesCbc_Decode; +#ifndef Z7_SFX +AES_CODE_FUNC g_AesCbc_Encode; +AES_CODE_FUNC g_AesCtr_Code; +UInt32 g_Aes_SupportedFunctions_Flags; +#endif + +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}; + + +static UInt32 D[256 * 4]; +static Byte InvS[256]; + +#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)) + + +// #define Z7_SHOW_AES_STATUS + +#ifdef MY_CPU_X86_OR_AMD64 + #define USE_HW_AES +#elif defined(MY_CPU_ARM_OR_ARM64) && defined(MY_CPU_LE) + #if defined(__clang__) + #if (__clang_major__ >= 8) // fix that check + #define USE_HW_AES + #endif + #elif defined(__GNUC__) + #if (__GNUC__ >= 6) // fix that check + #define USE_HW_AES + #endif + #elif defined(_MSC_VER) + #if _MSC_VER >= 1910 + #define USE_HW_AES + #endif + #endif +#endif + +#ifdef USE_HW_AES +#ifdef Z7_SHOW_AES_STATUS +#include +#define PRF(x) x +#else +#define PRF(x) +#endif +#endif + + +void AesGenTables(void) +{ + unsigned i; + for (i = 0; i < 256; i++) + InvS[Sbox[i]] = (Byte)i; + + for (i = 0; i < 256; i++) + { + { + const UInt32 a1 = Sbox[i]; + const UInt32 a2 = xtime(a1); + const 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); + } + { + const UInt32 a1 = InvS[i]; + const UInt32 a2 = xtime(a1); + const UInt32 a4 = xtime(a2); + const UInt32 a8 = xtime(a4); + const UInt32 a9 = a8 ^ a1; + const UInt32 aB = a8 ^ a2 ^ a1; + const UInt32 aD = a8 ^ a4 ^ a1; + const 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); + } + } + + { + AES_CODE_FUNC d = AesCbc_Decode; + #ifndef Z7_SFX + AES_CODE_FUNC e = AesCbc_Encode; + AES_CODE_FUNC c = AesCtr_Code; + UInt32 flags = 0; + #endif + + #ifdef USE_HW_AES + if (CPU_IsSupported_AES()) + { + // #pragma message ("AES HW") + PRF(printf("\n===AES HW\n")); + d = AesCbc_Decode_HW; + + #ifndef Z7_SFX + e = AesCbc_Encode_HW; + c = AesCtr_Code_HW; + flags = k_Aes_SupportedFunctions_HW; + #endif + + #ifdef MY_CPU_X86_OR_AMD64 + if (CPU_IsSupported_VAES_AVX2()) + { + PRF(printf("\n===vaes avx2\n")); + d = AesCbc_Decode_HW_256; + #ifndef Z7_SFX + c = AesCtr_Code_HW_256; + flags |= k_Aes_SupportedFunctions_HW_256; + #endif + } + #endif + } + #endif + + g_AesCbc_Decode = d; + #ifndef Z7_SFX + g_AesCbc_Encode = e; + g_AesCtr_Code = c; + g_Aes_SupportedFunctions_Flags = flags; + #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 Z7_FASTCALL Aes_SetKey_Enc(UInt32 *w, const Byte *key, unsigned keySize) +{ + unsigned i, m; + const UInt32 *wLim; + UInt32 t; + UInt32 rcon = 1; + + keySize /= 4; + w[0] = ((UInt32)keySize / 2) + 3; + w += 4; + + for (i = 0; i < keySize; i++, key += 4) + w[i] = GetUi32(key); + + t = w[(size_t)keySize - 1]; + wLim = w + (size_t)keySize * 3 + 28; + m = 0; + do + { + if (m == 0) + { + t = Ui32(Sbox[gb1(t)] ^ rcon, Sbox[gb2(t)], Sbox[gb3(t)], Sbox[gb0(t)]); + rcon <<= 1; + if (rcon & 0x100) + rcon = 0x1b; + m = keySize; + } + else if (m == 4 && keySize > 6) + t = Ui32(Sbox[gb0(t)], Sbox[gb1(t)], Sbox[gb2(t)], Sbox[gb3(t)]); + m--; + t ^= w[0]; + w[keySize] = t; + } + while (++w != wLim); +} + +void Z7_FASTCALL 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 */ + +// Z7_FORCE_INLINE +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) +} + +Z7_FORCE_INLINE +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 Z7_FASTCALL 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 Z7_FASTCALL 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 Z7_FASTCALL 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) + { + const UInt32 t = temp[i]; + + #ifdef MY_CPU_LE_UNALIGN + *((UInt32 *)(void *)data) ^= t; + #else + data[0] = (Byte)(data[0] ^ (t & 0xFF)); + data[1] = (Byte)(data[1] ^ ((t >> 8) & 0xFF)); + data[2] = (Byte)(data[2] ^ ((t >> 16) & 0xFF)); + data[3] = (Byte)(data[3] ^ ((t >> 24))); + #endif + } + } +} + +#undef xtime +#undef Ui32 +#undef gb0 +#undef gb1 +#undef gb2 +#undef gb3 +#undef gb +#undef TT +#undef DD +#undef USE_HW_AES +#undef PRF diff --git a/libraries/lzma/C/Aes.h b/libraries/lzma/C/Aes.h new file mode 100644 index 000000000..7f0182ac0 --- /dev/null +++ b/libraries/lzma/C/Aes.h @@ -0,0 +1,60 @@ +/* Aes.h -- AES encryption / decryption +2023-04-02 : Igor Pavlov : Public domain */ + +#ifndef ZIP7_INC_AES_H +#define ZIP7_INC_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 (Z7_FASTCALL *AES_SET_KEY_FUNC)(UInt32 *aes, const Byte *key, unsigned keySize); +void Z7_FASTCALL Aes_SetKey_Enc(UInt32 *aes, const Byte *key, unsigned keySize); +void Z7_FASTCALL 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 (Z7_FASTCALL *AES_CODE_FUNC)(UInt32 *ivAes, Byte *data, size_t numBlocks); + +extern AES_CODE_FUNC g_AesCbc_Decode; +#ifndef Z7_SFX +extern AES_CODE_FUNC g_AesCbc_Encode; +extern AES_CODE_FUNC g_AesCtr_Code; +#define k_Aes_SupportedFunctions_HW (1 << 2) +#define k_Aes_SupportedFunctions_HW_256 (1 << 3) +extern UInt32 g_Aes_SupportedFunctions_Flags; +#endif + + +#define Z7_DECLARE_AES_CODE_FUNC(funcName) \ + void Z7_FASTCALL funcName(UInt32 *ivAes, Byte *data, size_t numBlocks); + +Z7_DECLARE_AES_CODE_FUNC (AesCbc_Encode) +Z7_DECLARE_AES_CODE_FUNC (AesCbc_Decode) +Z7_DECLARE_AES_CODE_FUNC (AesCtr_Code) + +Z7_DECLARE_AES_CODE_FUNC (AesCbc_Encode_HW) +Z7_DECLARE_AES_CODE_FUNC (AesCbc_Decode_HW) +Z7_DECLARE_AES_CODE_FUNC (AesCtr_Code_HW) + +Z7_DECLARE_AES_CODE_FUNC (AesCbc_Decode_HW_256) +Z7_DECLARE_AES_CODE_FUNC (AesCtr_Code_HW_256) + +EXTERN_C_END + +#endif diff --git a/libraries/lzma/C/AesOpt.c b/libraries/lzma/C/AesOpt.c new file mode 100644 index 000000000..cfa6413f4 --- /dev/null +++ b/libraries/lzma/C/AesOpt.c @@ -0,0 +1,840 @@ +/* AesOpt.c -- AES optimized code for x86 AES hardware instructions +2023-04-02 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include "Aes.h" +#include "CpuArch.h" + +#ifdef MY_CPU_X86_OR_AMD64 + + #if defined(__INTEL_COMPILER) + #if (__INTEL_COMPILER >= 1110) + #define USE_INTEL_AES + #if (__INTEL_COMPILER >= 1900) + #define USE_INTEL_VAES + #endif + #endif + #elif defined(__clang__) && (__clang_major__ > 3 || __clang_major__ == 3 && __clang_minor__ >= 8) \ + || defined(__GNUC__) && (__GNUC__ > 4 || __GNUC__ == 4 && __GNUC_MINOR__ >= 4) + #define USE_INTEL_AES + #if !defined(__AES__) + #define ATTRIB_AES __attribute__((__target__("aes"))) + #endif + #if defined(__clang__) && (__clang_major__ >= 8) \ + || defined(__GNUC__) && (__GNUC__ >= 8) + #define USE_INTEL_VAES + #if !defined(__AES__) || !defined(__VAES__) || !defined(__AVX__) || !defined(__AVX2__) + #define ATTRIB_VAES __attribute__((__target__("aes,vaes,avx,avx2"))) + #endif + #endif + #elif defined(_MSC_VER) + #if (_MSC_VER > 1500) || (_MSC_FULL_VER >= 150030729) + #define USE_INTEL_AES + #if (_MSC_VER >= 1910) + #define USE_INTEL_VAES + #endif + #endif + #endif + +#ifndef ATTRIB_AES + #define ATTRIB_AES +#endif +#ifndef ATTRIB_VAES + #define ATTRIB_VAES +#endif + + +#ifdef USE_INTEL_AES + +#include + +#ifndef USE_INTEL_VAES +#define AES_TYPE_keys UInt32 +#define AES_TYPE_data Byte +// #define AES_TYPE_keys __m128i +// #define AES_TYPE_data __m128i +#endif + +#define AES_FUNC_START(name) \ + void Z7_FASTCALL name(UInt32 *ivAes, Byte *data8, size_t numBlocks) + // void Z7_FASTCALL name(__m128i *p, __m128i *data, size_t numBlocks) + +#define AES_FUNC_START2(name) \ +AES_FUNC_START (name); \ +ATTRIB_AES \ +AES_FUNC_START (name) + +#define MM_OP(op, dest, src) dest = op(dest, src); +#define MM_OP_m(op, src) MM_OP(op, m, src) + +#define MM_XOR( dest, src) MM_OP(_mm_xor_si128, dest, src) +#define AVX_XOR(dest, src) MM_OP(_mm256_xor_si256, dest, src) + + +AES_FUNC_START2 (AesCbc_Encode_HW) +{ + __m128i *p = (__m128i *)(void *)ivAes; + __m128i *data = (__m128i *)(void *)data8; + __m128i m = *p; + const __m128i k0 = p[2]; + const __m128i k1 = p[3]; + const UInt32 numRounds2 = *(const UInt32 *)(p + 1) - 1; + for (; numBlocks != 0; numBlocks--, data++) + { + UInt32 r = numRounds2; + const __m128i *w = p + 4; + __m128i temp = *data; + MM_XOR (temp, k0) + MM_XOR (m, temp) + MM_OP_m (_mm_aesenc_si128, k1) + do + { + MM_OP_m (_mm_aesenc_si128, w[0]) + MM_OP_m (_mm_aesenc_si128, w[1]) + w += 2; + } + while (--r); + MM_OP_m (_mm_aesenclast_si128, w[0]) + *data = m; + } + *p = m; +} + + +#define WOP_1(op) +#define WOP_2(op) WOP_1 (op) op (m1, 1) +#define WOP_3(op) WOP_2 (op) op (m2, 2) +#define WOP_4(op) WOP_3 (op) op (m3, 3) +#ifdef MY_CPU_AMD64 +#define WOP_5(op) WOP_4 (op) op (m4, 4) +#define WOP_6(op) WOP_5 (op) op (m5, 5) +#define WOP_7(op) WOP_6 (op) op (m6, 6) +#define WOP_8(op) WOP_7 (op) op (m7, 7) +#endif +/* +#define WOP_9(op) WOP_8 (op) op (m8, 8); +#define WOP_10(op) WOP_9 (op) op (m9, 9); +#define WOP_11(op) WOP_10(op) op (m10, 10); +#define WOP_12(op) WOP_11(op) op (m11, 11); +#define WOP_13(op) WOP_12(op) op (m12, 12); +#define WOP_14(op) WOP_13(op) op (m13, 13); +*/ + +#ifdef MY_CPU_AMD64 + #define NUM_WAYS 8 + #define WOP_M1 WOP_8 +#else + #define NUM_WAYS 4 + #define WOP_M1 WOP_4 +#endif + +#define WOP(op) op (m0, 0) WOP_M1(op) + + +#define DECLARE_VAR(reg, ii) __m128i reg; +#define LOAD_data( reg, ii) reg = data[ii]; +#define STORE_data( reg, ii) data[ii] = reg; +#if (NUM_WAYS > 1) +#define XOR_data_M1(reg, ii) MM_XOR (reg, data[ii- 1]) +#endif + +#define AVX_DECLARE_VAR(reg, ii) __m256i reg; +#define AVX_LOAD_data( reg, ii) reg = ((const __m256i *)(const void *)data)[ii]; +#define AVX_STORE_data( reg, ii) ((__m256i *)(void *)data)[ii] = reg; +#define AVX_XOR_data_M1(reg, ii) AVX_XOR (reg, (((const __m256i *)(const void *)(data - 1))[ii])) + +#define MM_OP_key(op, reg) MM_OP(op, reg, key); + +#define AES_DEC( reg, ii) MM_OP_key (_mm_aesdec_si128, reg) +#define AES_DEC_LAST( reg, ii) MM_OP_key (_mm_aesdeclast_si128, reg) +#define AES_ENC( reg, ii) MM_OP_key (_mm_aesenc_si128, reg) +#define AES_ENC_LAST( reg, ii) MM_OP_key (_mm_aesenclast_si128, reg) +#define AES_XOR( reg, ii) MM_OP_key (_mm_xor_si128, reg) + + +#define AVX_AES_DEC( reg, ii) MM_OP_key (_mm256_aesdec_epi128, reg) +#define AVX_AES_DEC_LAST( reg, ii) MM_OP_key (_mm256_aesdeclast_epi128, reg) +#define AVX_AES_ENC( reg, ii) MM_OP_key (_mm256_aesenc_epi128, reg) +#define AVX_AES_ENC_LAST( reg, ii) MM_OP_key (_mm256_aesenclast_epi128, reg) +#define AVX_AES_XOR( reg, ii) MM_OP_key (_mm256_xor_si256, reg) + +#define CTR_START(reg, ii) MM_OP (_mm_add_epi64, ctr, one) reg = ctr; +#define CTR_END( reg, ii) MM_XOR (data[ii], reg) + +#define AVX_CTR_START(reg, ii) MM_OP (_mm256_add_epi64, ctr2, two) reg = _mm256_xor_si256(ctr2, key); +#define AVX_CTR_END( reg, ii) AVX_XOR (((__m256i *)(void *)data)[ii], reg) + +#define WOP_KEY(op, n) { \ + const __m128i key = w[n]; \ + WOP(op); } + +#define AVX_WOP_KEY(op, n) { \ + const __m256i key = w[n]; \ + WOP(op); } + + +#define WIDE_LOOP_START \ + dataEnd = data + numBlocks; \ + if (numBlocks >= NUM_WAYS) \ + { dataEnd -= NUM_WAYS; do { \ + + +#define WIDE_LOOP_END \ + data += NUM_WAYS; \ + } while (data <= dataEnd); \ + dataEnd += NUM_WAYS; } \ + + +#define SINGLE_LOOP \ + for (; data < dataEnd; data++) + + +#define NUM_AES_KEYS_MAX 15 + +#define WIDE_LOOP_START_AVX(OP) \ + dataEnd = data + numBlocks; \ + if (numBlocks >= NUM_WAYS * 2) \ + { __m256i keys[NUM_AES_KEYS_MAX]; \ + UInt32 ii; \ + OP \ + for (ii = 0; ii < numRounds; ii++) \ + keys[ii] = _mm256_broadcastsi128_si256(p[ii]); \ + dataEnd -= NUM_WAYS * 2; do { \ + + +#define WIDE_LOOP_END_AVX(OP) \ + data += NUM_WAYS * 2; \ + } while (data <= dataEnd); \ + dataEnd += NUM_WAYS * 2; \ + OP \ + _mm256_zeroupper(); \ + } \ + +/* MSVC for x86: If we don't call _mm256_zeroupper(), and -arch:IA32 is not specified, + MSVC still can insert vzeroupper instruction. */ + + +AES_FUNC_START2 (AesCbc_Decode_HW) +{ + __m128i *p = (__m128i *)(void *)ivAes; + __m128i *data = (__m128i *)(void *)data8; + __m128i iv = *p; + const __m128i *wStart = p + *(const UInt32 *)(p + 1) * 2 + 2 - 1; + const __m128i *dataEnd; + p += 2; + + WIDE_LOOP_START + { + const __m128i *w = wStart; + + WOP (DECLARE_VAR) + WOP (LOAD_data) + WOP_KEY (AES_XOR, 1) + + do + { + WOP_KEY (AES_DEC, 0) + w--; + } + while (w != p); + WOP_KEY (AES_DEC_LAST, 0) + + MM_XOR (m0, iv) + WOP_M1 (XOR_data_M1) + iv = data[NUM_WAYS - 1]; + WOP (STORE_data) + } + WIDE_LOOP_END + + SINGLE_LOOP + { + const __m128i *w = wStart - 1; + __m128i m = _mm_xor_si128 (w[2], *data); + do + { + MM_OP_m (_mm_aesdec_si128, w[1]) + MM_OP_m (_mm_aesdec_si128, w[0]) + w -= 2; + } + while (w != p); + MM_OP_m (_mm_aesdec_si128, w[1]) + MM_OP_m (_mm_aesdeclast_si128, w[0]) + + MM_XOR (m, iv) + iv = *data; + *data = m; + } + + p[-2] = iv; +} + + +AES_FUNC_START2 (AesCtr_Code_HW) +{ + __m128i *p = (__m128i *)(void *)ivAes; + __m128i *data = (__m128i *)(void *)data8; + __m128i ctr = *p; + UInt32 numRoundsMinus2 = *(const UInt32 *)(p + 1) * 2 - 1; + const __m128i *dataEnd; + __m128i one = _mm_cvtsi32_si128(1); + + p += 2; + + WIDE_LOOP_START + { + const __m128i *w = p; + UInt32 r = numRoundsMinus2; + WOP (DECLARE_VAR) + WOP (CTR_START) + WOP_KEY (AES_XOR, 0) + w += 1; + do + { + WOP_KEY (AES_ENC, 0) + w += 1; + } + while (--r); + WOP_KEY (AES_ENC_LAST, 0) + + WOP (CTR_END) + } + WIDE_LOOP_END + + SINGLE_LOOP + { + UInt32 numRounds2 = *(const UInt32 *)(p - 2 + 1) - 1; + const __m128i *w = p; + __m128i m; + MM_OP (_mm_add_epi64, ctr, one) + m = _mm_xor_si128 (ctr, p[0]); + w += 1; + do + { + MM_OP_m (_mm_aesenc_si128, w[0]) + MM_OP_m (_mm_aesenc_si128, w[1]) + w += 2; + } + while (--numRounds2); + MM_OP_m (_mm_aesenc_si128, w[0]) + MM_OP_m (_mm_aesenclast_si128, w[1]) + MM_XOR (*data, m) + } + + p[-2] = ctr; +} + + + +#ifdef USE_INTEL_VAES + +/* +GCC before 2013-Jun: + : + #ifdef __AVX__ + #include + #endif +GCC after 2013-Jun: + : + #include +CLANG 3.8+: +{ + : + #if !defined(_MSC_VER) || defined(__AVX__) + #include + #endif + + if (the compiler is clang for Windows and if global arch is not set for __AVX__) + [ if (defined(_MSC_VER) && !defined(__AVX__)) ] + { + doesn't include + and we have 2 ways to fix it: + 1) we can define required __AVX__ before + or + 2) we can include after + } +} + +If we include manually for GCC/CLANG, it's +required that must be included before . +*/ + +/* +#if defined(__clang__) && defined(_MSC_VER) +#define __AVX__ +#define __AVX2__ +#define __VAES__ +#endif +*/ + +#include +#if defined(__clang__) && defined(_MSC_VER) + #if !defined(__AVX__) + #include + #endif + #if !defined(__AVX2__) + #include + #endif + #if !defined(__VAES__) + #include + #endif +#endif // __clang__ && _MSC_VER + + +#define VAES_FUNC_START2(name) \ +AES_FUNC_START (name); \ +ATTRIB_VAES \ +AES_FUNC_START (name) + +VAES_FUNC_START2 (AesCbc_Decode_HW_256) +{ + __m128i *p = (__m128i *)(void *)ivAes; + __m128i *data = (__m128i *)(void *)data8; + __m128i iv = *p; + const __m128i *dataEnd; + UInt32 numRounds = *(const UInt32 *)(p + 1) * 2 + 1; + p += 2; + + WIDE_LOOP_START_AVX(;) + { + const __m256i *w = keys + numRounds - 2; + + WOP (AVX_DECLARE_VAR) + WOP (AVX_LOAD_data) + AVX_WOP_KEY (AVX_AES_XOR, 1) + + do + { + AVX_WOP_KEY (AVX_AES_DEC, 0) + w--; + } + while (w != keys); + AVX_WOP_KEY (AVX_AES_DEC_LAST, 0) + + AVX_XOR (m0, _mm256_setr_m128i(iv, data[0])) + WOP_M1 (AVX_XOR_data_M1) + iv = data[NUM_WAYS * 2 - 1]; + WOP (AVX_STORE_data) + } + WIDE_LOOP_END_AVX(;) + + SINGLE_LOOP + { + const __m128i *w = p + *(const UInt32 *)(p + 1 - 2) * 2 + 1 - 3; + __m128i m = _mm_xor_si128 (w[2], *data); + do + { + MM_OP_m (_mm_aesdec_si128, w[1]) + MM_OP_m (_mm_aesdec_si128, w[0]) + w -= 2; + } + while (w != p); + MM_OP_m (_mm_aesdec_si128, w[1]) + MM_OP_m (_mm_aesdeclast_si128, w[0]) + + MM_XOR (m, iv) + iv = *data; + *data = m; + } + + p[-2] = iv; +} + + +/* +SSE2: _mm_cvtsi32_si128 : movd +AVX: _mm256_setr_m128i : vinsertf128 +AVX2: _mm256_add_epi64 : vpaddq ymm, ymm, ymm + _mm256_extracti128_si256 : vextracti128 + _mm256_broadcastsi128_si256 : vbroadcasti128 +*/ + +#define AVX_CTR_LOOP_START \ + ctr2 = _mm256_setr_m128i(_mm_sub_epi64(ctr, one), ctr); \ + two = _mm256_setr_m128i(one, one); \ + two = _mm256_add_epi64(two, two); \ + +// two = _mm256_setr_epi64x(2, 0, 2, 0); + +#define AVX_CTR_LOOP_ENC \ + ctr = _mm256_extracti128_si256 (ctr2, 1); \ + +VAES_FUNC_START2 (AesCtr_Code_HW_256) +{ + __m128i *p = (__m128i *)(void *)ivAes; + __m128i *data = (__m128i *)(void *)data8; + __m128i ctr = *p; + UInt32 numRounds = *(const UInt32 *)(p + 1) * 2 + 1; + const __m128i *dataEnd; + __m128i one = _mm_cvtsi32_si128(1); + __m256i ctr2, two; + p += 2; + + WIDE_LOOP_START_AVX (AVX_CTR_LOOP_START) + { + const __m256i *w = keys; + UInt32 r = numRounds - 2; + WOP (AVX_DECLARE_VAR) + AVX_WOP_KEY (AVX_CTR_START, 0) + + w += 1; + do + { + AVX_WOP_KEY (AVX_AES_ENC, 0) + w += 1; + } + while (--r); + AVX_WOP_KEY (AVX_AES_ENC_LAST, 0) + + WOP (AVX_CTR_END) + } + WIDE_LOOP_END_AVX (AVX_CTR_LOOP_ENC) + + SINGLE_LOOP + { + UInt32 numRounds2 = *(const UInt32 *)(p - 2 + 1) - 1; + const __m128i *w = p; + __m128i m; + MM_OP (_mm_add_epi64, ctr, one) + m = _mm_xor_si128 (ctr, p[0]); + w += 1; + do + { + MM_OP_m (_mm_aesenc_si128, w[0]) + MM_OP_m (_mm_aesenc_si128, w[1]) + w += 2; + } + while (--numRounds2); + MM_OP_m (_mm_aesenc_si128, w[0]) + MM_OP_m (_mm_aesenclast_si128, w[1]) + MM_XOR (*data, m) + } + + p[-2] = ctr; +} + +#endif // USE_INTEL_VAES + +#else // USE_INTEL_AES + +/* no USE_INTEL_AES */ + +#pragma message("AES HW_SW stub was used") + +#define AES_TYPE_keys UInt32 +#define AES_TYPE_data Byte + +#define AES_FUNC_START(name) \ + void Z7_FASTCALL name(UInt32 *p, Byte *data, size_t numBlocks) \ + +#define AES_COMPAT_STUB(name) \ + AES_FUNC_START(name); \ + AES_FUNC_START(name ## _HW) \ + { name(p, data, numBlocks); } + +AES_COMPAT_STUB (AesCbc_Encode) +AES_COMPAT_STUB (AesCbc_Decode) +AES_COMPAT_STUB (AesCtr_Code) + +#endif // USE_INTEL_AES + + +#ifndef USE_INTEL_VAES + +#pragma message("VAES HW_SW stub was used") + +#define VAES_COMPAT_STUB(name) \ + void Z7_FASTCALL name ## _256(UInt32 *p, Byte *data, size_t numBlocks); \ + void Z7_FASTCALL name ## _256(UInt32 *p, Byte *data, size_t numBlocks) \ + { name((AES_TYPE_keys *)(void *)p, (AES_TYPE_data *)(void *)data, numBlocks); } + +VAES_COMPAT_STUB (AesCbc_Decode_HW) +VAES_COMPAT_STUB (AesCtr_Code_HW) + +#endif // ! USE_INTEL_VAES + + +#elif defined(MY_CPU_ARM_OR_ARM64) && defined(MY_CPU_LE) + + #if defined(__clang__) + #if (__clang_major__ >= 8) // fix that check + #define USE_HW_AES + #endif + #elif defined(__GNUC__) + #if (__GNUC__ >= 6) // fix that check + #define USE_HW_AES + #endif + #elif defined(_MSC_VER) + #if _MSC_VER >= 1910 + #define USE_HW_AES + #endif + #endif + +#ifdef USE_HW_AES + +// #pragma message("=== AES HW === ") + +#if defined(__clang__) || defined(__GNUC__) + #ifdef MY_CPU_ARM64 + #define ATTRIB_AES __attribute__((__target__("+crypto"))) + #else + #define ATTRIB_AES __attribute__((__target__("fpu=crypto-neon-fp-armv8"))) + #endif +#else + // _MSC_VER + // for arm32 + #define _ARM_USE_NEW_NEON_INTRINSICS +#endif + +#ifndef ATTRIB_AES + #define ATTRIB_AES +#endif + +#if defined(_MSC_VER) && defined(MY_CPU_ARM64) +#include +#else +#include +#endif + +typedef uint8x16_t v128; + +#define AES_FUNC_START(name) \ + void Z7_FASTCALL name(UInt32 *ivAes, Byte *data8, size_t numBlocks) + // void Z7_FASTCALL name(v128 *p, v128 *data, size_t numBlocks) + +#define AES_FUNC_START2(name) \ +AES_FUNC_START (name); \ +ATTRIB_AES \ +AES_FUNC_START (name) + +#define MM_OP(op, dest, src) dest = op(dest, src); +#define MM_OP_m(op, src) MM_OP(op, m, src) +#define MM_OP1_m(op) m = op(m); + +#define MM_XOR( dest, src) MM_OP(veorq_u8, dest, src) +#define MM_XOR_m( src) MM_XOR(m, src) + +#define AES_E_m(k) MM_OP_m (vaeseq_u8, k) +#define AES_E_MC_m(k) AES_E_m (k) MM_OP1_m(vaesmcq_u8) + + +AES_FUNC_START2 (AesCbc_Encode_HW) +{ + v128 *p = (v128*)(void*)ivAes; + v128 *data = (v128*)(void*)data8; + v128 m = *p; + const v128 k0 = p[2]; + const v128 k1 = p[3]; + const v128 k2 = p[4]; + const v128 k3 = p[5]; + const v128 k4 = p[6]; + const v128 k5 = p[7]; + const v128 k6 = p[8]; + const v128 k7 = p[9]; + const v128 k8 = p[10]; + const v128 k9 = p[11]; + const UInt32 numRounds2 = *(const UInt32 *)(p + 1); + const v128 *w = p + ((size_t)numRounds2 * 2); + const v128 k_z1 = w[1]; + const v128 k_z0 = w[2]; + for (; numBlocks != 0; numBlocks--, data++) + { + MM_XOR_m (*data); + AES_E_MC_m (k0) + AES_E_MC_m (k1) + AES_E_MC_m (k2) + AES_E_MC_m (k3) + AES_E_MC_m (k4) + AES_E_MC_m (k5) + AES_E_MC_m (k6) + AES_E_MC_m (k7) + AES_E_MC_m (k8) + if (numRounds2 >= 6) + { + AES_E_MC_m (k9) + AES_E_MC_m (p[12]) + if (numRounds2 != 6) + { + AES_E_MC_m (p[13]) + AES_E_MC_m (p[14]) + } + } + AES_E_m (k_z1) + MM_XOR_m (k_z0); + *data = m; + } + *p = m; +} + + +#define WOP_1(op) +#define WOP_2(op) WOP_1 (op) op (m1, 1) +#define WOP_3(op) WOP_2 (op) op (m2, 2) +#define WOP_4(op) WOP_3 (op) op (m3, 3) +#define WOP_5(op) WOP_4 (op) op (m4, 4) +#define WOP_6(op) WOP_5 (op) op (m5, 5) +#define WOP_7(op) WOP_6 (op) op (m6, 6) +#define WOP_8(op) WOP_7 (op) op (m7, 7) + + #define NUM_WAYS 8 + #define WOP_M1 WOP_8 + +#define WOP(op) op (m0, 0) WOP_M1(op) + +#define DECLARE_VAR(reg, ii) v128 reg; +#define LOAD_data( reg, ii) reg = data[ii]; +#define STORE_data( reg, ii) data[ii] = reg; +#if (NUM_WAYS > 1) +#define XOR_data_M1(reg, ii) MM_XOR (reg, data[ii- 1]) +#endif + +#define MM_OP_key(op, reg) MM_OP (op, reg, key) + +#define AES_D_m(k) MM_OP_m (vaesdq_u8, k) +#define AES_D_IMC_m(k) AES_D_m (k) MM_OP1_m (vaesimcq_u8) + +#define AES_XOR( reg, ii) MM_OP_key (veorq_u8, reg) +#define AES_D( reg, ii) MM_OP_key (vaesdq_u8, reg) +#define AES_E( reg, ii) MM_OP_key (vaeseq_u8, reg) + +#define AES_D_IMC( reg, ii) AES_D (reg, ii) reg = vaesimcq_u8(reg); +#define AES_E_MC( reg, ii) AES_E (reg, ii) reg = vaesmcq_u8(reg); + +#define CTR_START(reg, ii) MM_OP (vaddq_u64, ctr, one) reg = vreinterpretq_u8_u64(ctr); +#define CTR_END( reg, ii) MM_XOR (data[ii], reg) + +#define WOP_KEY(op, n) { \ + const v128 key = w[n]; \ + WOP(op) } + +#define WIDE_LOOP_START \ + dataEnd = data + numBlocks; \ + if (numBlocks >= NUM_WAYS) \ + { dataEnd -= NUM_WAYS; do { \ + +#define WIDE_LOOP_END \ + data += NUM_WAYS; \ + } while (data <= dataEnd); \ + dataEnd += NUM_WAYS; } \ + +#define SINGLE_LOOP \ + for (; data < dataEnd; data++) + + +AES_FUNC_START2 (AesCbc_Decode_HW) +{ + v128 *p = (v128*)(void*)ivAes; + v128 *data = (v128*)(void*)data8; + v128 iv = *p; + const v128 *wStart = p + ((size_t)*(const UInt32 *)(p + 1)) * 2; + const v128 *dataEnd; + p += 2; + + WIDE_LOOP_START + { + const v128 *w = wStart; + WOP (DECLARE_VAR) + WOP (LOAD_data) + WOP_KEY (AES_D_IMC, 2) + do + { + WOP_KEY (AES_D_IMC, 1) + WOP_KEY (AES_D_IMC, 0) + w -= 2; + } + while (w != p); + WOP_KEY (AES_D, 1) + WOP_KEY (AES_XOR, 0) + MM_XOR (m0, iv); + WOP_M1 (XOR_data_M1) + iv = data[NUM_WAYS - 1]; + WOP (STORE_data) + } + WIDE_LOOP_END + + SINGLE_LOOP + { + const v128 *w = wStart; + v128 m = *data; + AES_D_IMC_m (w[2]) + do + { + AES_D_IMC_m (w[1]); + AES_D_IMC_m (w[0]); + w -= 2; + } + while (w != p); + AES_D_m (w[1]); + MM_XOR_m (w[0]); + MM_XOR_m (iv); + iv = *data; + *data = m; + } + + p[-2] = iv; +} + + +AES_FUNC_START2 (AesCtr_Code_HW) +{ + v128 *p = (v128*)(void*)ivAes; + v128 *data = (v128*)(void*)data8; + uint64x2_t ctr = vreinterpretq_u64_u8(*p); + const v128 *wEnd = p + ((size_t)*(const UInt32 *)(p + 1)) * 2; + const v128 *dataEnd; + uint64x2_t one = vdupq_n_u64(0); + one = vsetq_lane_u64(1, one, 0); + p += 2; + + WIDE_LOOP_START + { + const v128 *w = p; + WOP (DECLARE_VAR) + WOP (CTR_START) + do + { + WOP_KEY (AES_E_MC, 0) + WOP_KEY (AES_E_MC, 1) + w += 2; + } + while (w != wEnd); + WOP_KEY (AES_E_MC, 0) + WOP_KEY (AES_E, 1) + WOP_KEY (AES_XOR, 2) + WOP (CTR_END) + } + WIDE_LOOP_END + + SINGLE_LOOP + { + const v128 *w = p; + v128 m; + CTR_START (m, 0); + do + { + AES_E_MC_m (w[0]); + AES_E_MC_m (w[1]); + w += 2; + } + while (w != wEnd); + AES_E_MC_m (w[0]) + AES_E_m (w[1]) + MM_XOR_m (w[2]) + CTR_END (m, 0) + } + + p[-2] = vreinterpretq_u8_u64(ctr); +} + +#endif // USE_HW_AES + +#endif // MY_CPU_ARM_OR_ARM64 + +#undef NUM_WAYS +#undef WOP_M1 +#undef WOP +#undef DECLARE_VAR +#undef LOAD_data +#undef STORE_data +#undef USE_INTEL_AES +#undef USE_HW_AES diff --git a/libraries/lzma/C/Alloc.c b/libraries/lzma/C/Alloc.c new file mode 100644 index 000000000..d841bf20a --- /dev/null +++ b/libraries/lzma/C/Alloc.c @@ -0,0 +1,535 @@ +/* Alloc.c -- Memory allocation functions +2023-04-02 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#ifdef _WIN32 +#include "7zWindows.h" +#endif +#include + +#include "Alloc.h" + +#ifdef _WIN32 +#ifdef Z7_LARGE_PAGES +#if defined(__clang__) || defined(__GNUC__) +typedef void (*Z7_voidFunction)(void); +#define MY_CAST_FUNC (Z7_voidFunction) +#elif defined(_MSC_VER) && _MSC_VER > 1920 +#define MY_CAST_FUNC (void *) +// #pragma warning(disable : 4191) // 'type cast': unsafe conversion from 'FARPROC' to 'void (__cdecl *)()' +#else +#define MY_CAST_FUNC +#endif +#endif // Z7_LARGE_PAGES +#endif // _WIN32 + +// #define SZ_ALLOC_DEBUG +/* #define SZ_ALLOC_DEBUG */ + +/* use SZ_ALLOC_DEBUG to debug alloc/free operations */ +#ifdef SZ_ALLOC_DEBUG + +#include +#include +static int g_allocCount = 0; +#ifdef _WIN32 +static int g_allocCountMid = 0; +static int g_allocCountBig = 0; +#endif + + +#define CONVERT_INT_TO_STR(charType, tempSize) \ + char temp[tempSize]; unsigned i = 0; \ + while (val >= 10) { temp[i++] = (char)('0' + (unsigned)(val % 10)); val /= 10; } \ + *s++ = (charType)('0' + (unsigned)val); \ + while (i != 0) { i--; *s++ = temp[i]; } \ + *s = 0; + +static void ConvertUInt64ToString(UInt64 val, char *s) +{ + CONVERT_INT_TO_STR(char, 24) +} + +#define GET_HEX_CHAR(t) ((char)(((t < 10) ? ('0' + t) : ('A' + (t - 10))))) + +static void ConvertUInt64ToHex(UInt64 val, char *s) +{ + UInt64 v = val; + unsigned i; + for (i = 1;; i++) + { + v >>= 4; + if (v == 0) + break; + } + s[i] = 0; + do + { + unsigned t = (unsigned)(val & 0xF); + val >>= 4; + s[--i] = GET_HEX_CHAR(t); + } + while (i); +} + +#define DEBUG_OUT_STREAM stderr + +static void Print(const char *s) +{ + fputs(s, DEBUG_OUT_STREAM); +} + +static void PrintAligned(const char *s, size_t align) +{ + size_t len = strlen(s); + for(;;) + { + fputc(' ', DEBUG_OUT_STREAM); + if (len >= align) + break; + ++len; + } + Print(s); +} + +static void PrintLn(void) +{ + Print("\n"); +} + +static void PrintHex(UInt64 v, size_t align) +{ + char s[32]; + ConvertUInt64ToHex(v, s); + PrintAligned(s, align); +} + +static void PrintDec(int v, size_t align) +{ + char s[32]; + ConvertUInt64ToString((unsigned)v, s); + PrintAligned(s, align); +} + +static void PrintAddr(void *p) +{ + PrintHex((UInt64)(size_t)(ptrdiff_t)p, 12); +} + + +#define PRINT_REALLOC(name, cnt, size, ptr) { \ + Print(name " "); \ + if (!ptr) PrintDec(cnt++, 10); \ + PrintHex(size, 10); \ + PrintAddr(ptr); \ + PrintLn(); } + +#define PRINT_ALLOC(name, cnt, size, ptr) { \ + Print(name " "); \ + PrintDec(cnt++, 10); \ + PrintHex(size, 10); \ + PrintAddr(ptr); \ + PrintLn(); } + +#define PRINT_FREE(name, cnt, ptr) if (ptr) { \ + Print(name " "); \ + PrintDec(--cnt, 10); \ + PrintAddr(ptr); \ + PrintLn(); } + +#else + +#ifdef _WIN32 +#define PRINT_ALLOC(name, cnt, size, ptr) +#endif +#define PRINT_FREE(name, cnt, ptr) +#define Print(s) +#define PrintLn() +#define PrintHex(v, align) +#define PrintAddr(p) + +#endif + + +/* +by specification: + malloc(non_NULL, 0) : returns NULL or a unique pointer value that can later be successfully passed to free() + realloc(NULL, size) : the call is equivalent to malloc(size) + realloc(non_NULL, 0) : the call is equivalent to free(ptr) + +in main compilers: + malloc(0) : returns non_NULL + realloc(NULL, 0) : returns non_NULL + realloc(non_NULL, 0) : returns NULL +*/ + + +void *MyAlloc(size_t size) +{ + if (size == 0) + return NULL; + // PRINT_ALLOC("Alloc ", g_allocCount, size, NULL) + #ifdef SZ_ALLOC_DEBUG + { + void *p = malloc(size); + if (p) + { + PRINT_ALLOC("Alloc ", g_allocCount, size, p) + } + return p; + } + #else + return malloc(size); + #endif +} + +void MyFree(void *address) +{ + PRINT_FREE("Free ", g_allocCount, address) + + free(address); +} + +void *MyRealloc(void *address, size_t size) +{ + if (size == 0) + { + MyFree(address); + return NULL; + } + // PRINT_REALLOC("Realloc ", g_allocCount, size, address) + #ifdef SZ_ALLOC_DEBUG + { + void *p = realloc(address, size); + if (p) + { + PRINT_REALLOC("Realloc ", g_allocCount, size, address) + } + return p; + } + #else + return realloc(address, size); + #endif +} + + +#ifdef _WIN32 + +void *MidAlloc(size_t size) +{ + if (size == 0) + return NULL; + #ifdef SZ_ALLOC_DEBUG + { + void *p = VirtualAlloc(NULL, size, MEM_COMMIT, PAGE_READWRITE); + if (p) + { + PRINT_ALLOC("Alloc-Mid", g_allocCountMid, size, p) + } + return p; + } + #else + return VirtualAlloc(NULL, size, MEM_COMMIT, PAGE_READWRITE); + #endif +} + +void MidFree(void *address) +{ + PRINT_FREE("Free-Mid", g_allocCountMid, address) + + if (!address) + return; + VirtualFree(address, 0, MEM_RELEASE); +} + +#ifdef Z7_LARGE_PAGES + +#ifdef MEM_LARGE_PAGES + #define MY__MEM_LARGE_PAGES MEM_LARGE_PAGES +#else + #define MY__MEM_LARGE_PAGES 0x20000000 +#endif + +extern +SIZE_T g_LargePageSize; +SIZE_T g_LargePageSize = 0; +typedef SIZE_T (WINAPI *Func_GetLargePageMinimum)(VOID); + +void SetLargePageSize(void) +{ + #ifdef Z7_LARGE_PAGES + SIZE_T size; + const + Func_GetLargePageMinimum fn = + (Func_GetLargePageMinimum) MY_CAST_FUNC GetProcAddress(GetModuleHandle(TEXT("kernel32.dll")), + "GetLargePageMinimum"); + if (!fn) + return; + size = fn(); + if (size == 0 || (size & (size - 1)) != 0) + return; + g_LargePageSize = size; + #endif +} + +#endif // Z7_LARGE_PAGES + +void *BigAlloc(size_t size) +{ + if (size == 0) + return NULL; + + PRINT_ALLOC("Alloc-Big", g_allocCountBig, size, NULL) + + #ifdef Z7_LARGE_PAGES + { + SIZE_T ps = g_LargePageSize; + if (ps != 0 && ps <= (1 << 30) && size > (ps / 2)) + { + size_t size2; + ps--; + size2 = (size + ps) & ~ps; + if (size2 >= size) + { + void *p = VirtualAlloc(NULL, size2, MEM_COMMIT | MY__MEM_LARGE_PAGES, PAGE_READWRITE); + if (p) + { + PRINT_ALLOC("Alloc-BM ", g_allocCountMid, size2, p) + return p; + } + } + } + } + #endif + + return MidAlloc(size); +} + +void BigFree(void *address) +{ + PRINT_FREE("Free-Big", g_allocCountBig, address) + MidFree(address); +} + +#endif // _WIN32 + + +static void *SzAlloc(ISzAllocPtr p, size_t size) { UNUSED_VAR(p) return MyAlloc(size); } +static void SzFree(ISzAllocPtr p, void *address) { UNUSED_VAR(p) MyFree(address); } +const ISzAlloc g_Alloc = { SzAlloc, SzFree }; + +#ifdef _WIN32 +static void *SzMidAlloc(ISzAllocPtr p, size_t size) { UNUSED_VAR(p) return MidAlloc(size); } +static void SzMidFree(ISzAllocPtr p, void *address) { UNUSED_VAR(p) MidFree(address); } +static void *SzBigAlloc(ISzAllocPtr p, size_t size) { UNUSED_VAR(p) return BigAlloc(size); } +static void SzBigFree(ISzAllocPtr p, void *address) { UNUSED_VAR(p) BigFree(address); } +const ISzAlloc g_MidAlloc = { SzMidAlloc, SzMidFree }; +const ISzAlloc g_BigAlloc = { SzBigAlloc, SzBigFree }; +#endif + +/* + uintptr_t : C99 (optional) + : unsupported in VS6 +*/ + +#ifdef _WIN32 + typedef UINT_PTR UIntPtr; +#else + /* + typedef uintptr_t UIntPtr; + */ + typedef ptrdiff_t UIntPtr; +#endif + + +#define ADJUST_ALLOC_SIZE 0 +/* +#define ADJUST_ALLOC_SIZE (sizeof(void *) - 1) +*/ +/* + Use (ADJUST_ALLOC_SIZE = (sizeof(void *) - 1)), if + MyAlloc() can return address that is NOT multiple of sizeof(void *). +*/ + + +/* +#define MY_ALIGN_PTR_DOWN(p, align) ((void *)((char *)(p) - ((size_t)(UIntPtr)(p) & ((align) - 1)))) +*/ +#define MY_ALIGN_PTR_DOWN(p, align) ((void *)((((UIntPtr)(p)) & ~((UIntPtr)(align) - 1)))) + + +#if !defined(_WIN32) && defined(_POSIX_C_SOURCE) && (_POSIX_C_SOURCE >= 200112L) + #define USE_posix_memalign +#endif + +#ifndef USE_posix_memalign +#define MY_ALIGN_PTR_UP_PLUS(p, align) MY_ALIGN_PTR_DOWN(((char *)(p) + (align) + ADJUST_ALLOC_SIZE), align) +#endif + +/* + This posix_memalign() is for test purposes only. + We also need special Free() function instead of free(), + if this posix_memalign() is used. +*/ + +/* +static int posix_memalign(void **ptr, size_t align, size_t size) +{ + size_t newSize = size + align; + void *p; + void *pAligned; + *ptr = NULL; + if (newSize < size) + return 12; // ENOMEM + p = MyAlloc(newSize); + if (!p) + return 12; // ENOMEM + pAligned = MY_ALIGN_PTR_UP_PLUS(p, align); + ((void **)pAligned)[-1] = p; + *ptr = pAligned; + return 0; +} +*/ + +/* + ALLOC_ALIGN_SIZE >= sizeof(void *) + ALLOC_ALIGN_SIZE >= cache_line_size +*/ + +#define ALLOC_ALIGN_SIZE ((size_t)1 << 7) + +static void *SzAlignedAlloc(ISzAllocPtr pp, size_t size) +{ + #ifndef USE_posix_memalign + + void *p; + void *pAligned; + size_t newSize; + UNUSED_VAR(pp) + + /* also we can allocate additional dummy ALLOC_ALIGN_SIZE bytes after aligned + block to prevent cache line sharing with another allocated blocks */ + + newSize = size + ALLOC_ALIGN_SIZE * 1 + ADJUST_ALLOC_SIZE; + if (newSize < size) + return NULL; + + p = MyAlloc(newSize); + + if (!p) + return NULL; + pAligned = MY_ALIGN_PTR_UP_PLUS(p, ALLOC_ALIGN_SIZE); + + Print(" size="); PrintHex(size, 8); + Print(" a_size="); PrintHex(newSize, 8); + Print(" ptr="); PrintAddr(p); + Print(" a_ptr="); PrintAddr(pAligned); + PrintLn(); + + ((void **)pAligned)[-1] = p; + + return pAligned; + + #else + + void *p; + UNUSED_VAR(pp) + if (posix_memalign(&p, ALLOC_ALIGN_SIZE, size)) + return NULL; + + Print(" posix_memalign="); PrintAddr(p); + PrintLn(); + + return p; + + #endif +} + + +static void SzAlignedFree(ISzAllocPtr pp, void *address) +{ + UNUSED_VAR(pp) + #ifndef USE_posix_memalign + if (address) + MyFree(((void **)address)[-1]); + #else + free(address); + #endif +} + + +const ISzAlloc g_AlignedAlloc = { SzAlignedAlloc, SzAlignedFree }; + + + +#define MY_ALIGN_PTR_DOWN_1(p) MY_ALIGN_PTR_DOWN(p, sizeof(void *)) + +/* we align ptr to support cases where CAlignOffsetAlloc::offset is not multiply of sizeof(void *) */ +#define REAL_BLOCK_PTR_VAR(p) ((void **)MY_ALIGN_PTR_DOWN_1(p))[-1] +/* +#define REAL_BLOCK_PTR_VAR(p) ((void **)(p))[-1] +*/ + +static void *AlignOffsetAlloc_Alloc(ISzAllocPtr pp, size_t size) +{ + const CAlignOffsetAlloc *p = Z7_CONTAINER_FROM_VTBL_CONST(pp, CAlignOffsetAlloc, vt); + void *adr; + void *pAligned; + size_t newSize; + size_t extra; + size_t alignSize = (size_t)1 << p->numAlignBits; + + if (alignSize < sizeof(void *)) + alignSize = sizeof(void *); + + if (p->offset >= alignSize) + return NULL; + + /* also we can allocate additional dummy ALLOC_ALIGN_SIZE bytes after aligned + block to prevent cache line sharing with another allocated blocks */ + extra = p->offset & (sizeof(void *) - 1); + newSize = size + alignSize + extra + ADJUST_ALLOC_SIZE; + if (newSize < size) + return NULL; + + adr = ISzAlloc_Alloc(p->baseAlloc, newSize); + + if (!adr) + return NULL; + + pAligned = (char *)MY_ALIGN_PTR_DOWN((char *)adr + + alignSize - p->offset + extra + ADJUST_ALLOC_SIZE, alignSize) + p->offset; + + PrintLn(); + Print("- Aligned: "); + Print(" size="); PrintHex(size, 8); + Print(" a_size="); PrintHex(newSize, 8); + Print(" ptr="); PrintAddr(adr); + Print(" a_ptr="); PrintAddr(pAligned); + PrintLn(); + + REAL_BLOCK_PTR_VAR(pAligned) = adr; + + return pAligned; +} + + +static void AlignOffsetAlloc_Free(ISzAllocPtr pp, void *address) +{ + if (address) + { + const CAlignOffsetAlloc *p = Z7_CONTAINER_FROM_VTBL_CONST(pp, CAlignOffsetAlloc, vt); + PrintLn(); + Print("- Aligned Free: "); + PrintLn(); + ISzAlloc_Free(p->baseAlloc, REAL_BLOCK_PTR_VAR(address)); + } +} + + +void AlignOffsetAlloc_CreateVTable(CAlignOffsetAlloc *p) +{ + p->vt.Alloc = AlignOffsetAlloc_Alloc; + p->vt.Free = AlignOffsetAlloc_Free; +} diff --git a/libraries/lzma/C/Alloc.h b/libraries/lzma/C/Alloc.h new file mode 100644 index 000000000..fac5b62fb --- /dev/null +++ b/libraries/lzma/C/Alloc.h @@ -0,0 +1,71 @@ +/* Alloc.h -- Memory allocation functions +2023-03-04 : Igor Pavlov : Public domain */ + +#ifndef ZIP7_INC_ALLOC_H +#define ZIP7_INC_ALLOC_H + +#include "7zTypes.h" + +EXTERN_C_BEGIN + +/* + MyFree(NULL) : is allowed, as free(NULL) + MyAlloc(0) : returns NULL : but malloc(0) is allowed to return NULL or non_NULL + MyRealloc(NULL, 0) : returns NULL : but realloc(NULL, 0) is allowed to return NULL or non_NULL +MyRealloc() is similar to realloc() for the following cases: + MyRealloc(non_NULL, 0) : returns NULL and always calls MyFree(ptr) + MyRealloc(NULL, non_ZERO) : returns NULL, if allocation failed + MyRealloc(non_NULL, non_ZERO) : returns NULL, if reallocation failed +*/ + +void *MyAlloc(size_t size); +void MyFree(void *address); +void *MyRealloc(void *address, size_t size); + +#ifdef _WIN32 + +#ifdef Z7_LARGE_PAGES +void SetLargePageSize(void); +#endif + +void *MidAlloc(size_t size); +void MidFree(void *address); +void *BigAlloc(size_t size); +void BigFree(void *address); + +#else + +#define MidAlloc(size) MyAlloc(size) +#define MidFree(address) MyFree(address) +#define BigAlloc(size) MyAlloc(size) +#define BigFree(address) MyFree(address) + +#endif + +extern const ISzAlloc g_Alloc; + +#ifdef _WIN32 +extern const ISzAlloc g_BigAlloc; +extern const ISzAlloc g_MidAlloc; +#else +#define g_BigAlloc g_AlignedAlloc +#define g_MidAlloc g_AlignedAlloc +#endif + +extern const ISzAlloc g_AlignedAlloc; + + +typedef struct +{ + ISzAlloc vt; + ISzAllocPtr baseAlloc; + unsigned numAlignBits; /* ((1 << numAlignBits) >= sizeof(void *)) */ + size_t offset; /* (offset == (k * sizeof(void *)) && offset < (1 << numAlignBits) */ +} CAlignOffsetAlloc; + +void AlignOffsetAlloc_CreateVTable(CAlignOffsetAlloc *p); + + +EXTERN_C_END + +#endif diff --git a/libraries/lzma/C/Bcj2.c b/libraries/lzma/C/Bcj2.c index c7b956708..7cb57ad62 100644 --- a/libraries/lzma/C/Bcj2.c +++ b/libraries/lzma/C/Bcj2.c @@ -1,29 +1,24 @@ /* Bcj2.c -- BCJ2 Decoder (Converter for x86 code) -2021-02-09 : Igor Pavlov : Public domain */ +2023-03-01 : Igor Pavlov : Public domain */ #include "Precomp.h" #include "Bcj2.h" #include "CpuArch.h" -#define CProb UInt16 - #define kTopValue ((UInt32)1 << 24) -#define kNumModelBits 11 -#define kBitModelTotal (1 << kNumModelBits) +#define kNumBitModelTotalBits 11 +#define kBitModelTotal (1 << kNumBitModelTotalBits) #define kNumMoveBits 5 -#define _IF_BIT_0 ttt = *prob; bound = (p->range >> kNumModelBits) * ttt; if (p->code < bound) -#define _UPDATE_0 p->range = bound; *prob = (CProb)(ttt + ((kBitModelTotal - ttt) >> kNumMoveBits)); -#define _UPDATE_1 p->range -= bound; p->code -= bound; *prob = (CProb)(ttt - (ttt >> kNumMoveBits)); +// UInt32 bcj2_stats[256 + 2][2]; void Bcj2Dec_Init(CBcj2Dec *p) { unsigned i; - - p->state = BCJ2_DEC_STATE_OK; + p->state = BCJ2_STREAM_RC; // BCJ2_DEC_STATE_OK; p->ip = 0; - p->temp[3] = 0; + p->temp = 0; p->range = 0; p->code = 0; for (i = 0; i < sizeof(p->probs) / sizeof(p->probs[0]); i++) @@ -32,217 +27,248 @@ void Bcj2Dec_Init(CBcj2Dec *p) SRes Bcj2Dec_Decode(CBcj2Dec *p) { + UInt32 v = p->temp; + // const Byte *src; if (p->range <= 5) { - p->state = BCJ2_DEC_STATE_OK; + UInt32 code = p->code; + p->state = BCJ2_DEC_STATE_ERROR; /* for case if we return SZ_ERROR_DATA; */ for (; p->range != 5; p->range++) { - if (p->range == 1 && p->code != 0) + if (p->range == 1 && code != 0) return SZ_ERROR_DATA; - if (p->bufs[BCJ2_STREAM_RC] == p->lims[BCJ2_STREAM_RC]) { p->state = BCJ2_STREAM_RC; return SZ_OK; } - - p->code = (p->code << 8) | *(p->bufs[BCJ2_STREAM_RC])++; + code = (code << 8) | *(p->bufs[BCJ2_STREAM_RC])++; + p->code = code; } - - if (p->code == 0xFFFFFFFF) + if (code == 0xffffffff) return SZ_ERROR_DATA; - - p->range = 0xFFFFFFFF; + p->range = 0xffffffff; } - else if (p->state >= BCJ2_DEC_STATE_ORIG_0) + // else { - while (p->state <= BCJ2_DEC_STATE_ORIG_3) + unsigned state = p->state; + // we check BCJ2_IS_32BIT_STREAM() here instead of check in the main loop + if (BCJ2_IS_32BIT_STREAM(state)) + { + const Byte *cur = p->bufs[state]; + if (cur == p->lims[state]) + return SZ_OK; + p->bufs[state] = cur + 4; + { + const UInt32 ip = p->ip + 4; + v = GetBe32a(cur) - ip; + p->ip = ip; + } + state = BCJ2_DEC_STATE_ORIG_0; + } + if ((unsigned)(state - BCJ2_DEC_STATE_ORIG_0) < 4) { Byte *dest = p->dest; - if (dest == p->destLim) - return SZ_OK; - *dest = p->temp[(size_t)p->state - BCJ2_DEC_STATE_ORIG_0]; - p->state++; - p->dest = dest + 1; - } - } - - /* - if (BCJ2_IS_32BIT_STREAM(p->state)) - { - const Byte *cur = p->bufs[p->state]; - if (cur == p->lims[p->state]) - return SZ_OK; - p->bufs[p->state] = cur + 4; - - { - UInt32 val; - Byte *dest; - SizeT rem; - - p->ip += 4; - val = GetBe32(cur) - p->ip; - dest = p->dest; - rem = p->destLim - dest; - if (rem < 4) + for (;;) { - SizeT i; - SetUi32(p->temp, val); - for (i = 0; i < rem; i++) - dest[i] = p->temp[i]; - p->dest = dest + rem; - p->state = BCJ2_DEC_STATE_ORIG_0 + (unsigned)rem; - return SZ_OK; + if (dest == p->destLim) + { + p->state = state; + p->temp = v; + return SZ_OK; + } + *dest++ = (Byte)v; + p->dest = dest; + if (++state == BCJ2_DEC_STATE_ORIG_3 + 1) + break; + v >>= 8; } - SetUi32(dest, val); - p->temp[3] = (Byte)(val >> 24); - p->dest = dest + 4; - p->state = BCJ2_DEC_STATE_OK; } } - */ + // src = p->bufs[BCJ2_STREAM_MAIN]; for (;;) { + /* if (BCJ2_IS_32BIT_STREAM(p->state)) p->state = BCJ2_DEC_STATE_OK; else + */ { if (p->range < kTopValue) { if (p->bufs[BCJ2_STREAM_RC] == p->lims[BCJ2_STREAM_RC]) { p->state = BCJ2_STREAM_RC; + p->temp = v; return SZ_OK; } p->range <<= 8; p->code = (p->code << 8) | *(p->bufs[BCJ2_STREAM_RC])++; } - { const Byte *src = p->bufs[BCJ2_STREAM_MAIN]; const Byte *srcLim; - Byte *dest; - SizeT num = (SizeT)(p->lims[BCJ2_STREAM_MAIN] - src); - - if (num == 0) + Byte *dest = p->dest; { - p->state = BCJ2_STREAM_MAIN; - return SZ_OK; + const SizeT rem = (SizeT)(p->lims[BCJ2_STREAM_MAIN] - src); + SizeT num = (SizeT)(p->destLim - dest); + if (num >= rem) + num = rem; + #define NUM_ITERS 4 + #if (NUM_ITERS & (NUM_ITERS - 1)) == 0 + num &= ~((SizeT)NUM_ITERS - 1); // if (NUM_ITERS == (1 << x)) + #else + num -= num % NUM_ITERS; // if (NUM_ITERS != (1 << x)) + #endif + srcLim = src + num; } - - dest = p->dest; - if (num > (SizeT)(p->destLim - dest)) + + #define NUM_SHIFT_BITS 24 + #define ONE_ITER(indx) { \ + const unsigned b = src[indx]; \ + *dest++ = (Byte)b; \ + v = (v << NUM_SHIFT_BITS) | b; \ + if (((b + (0x100 - 0xe8)) & 0xfe) == 0) break; \ + if (((v - (((UInt32)0x0f << (NUM_SHIFT_BITS)) + 0x80)) & \ + ((((UInt32)1 << (4 + NUM_SHIFT_BITS)) - 0x1) << 4)) == 0) break; \ + /* ++dest */; /* v = b; */ } + + if (src != srcLim) + for (;;) { - num = (SizeT)(p->destLim - dest); - if (num == 0) + /* The dependency chain of 2-cycle for (v) calculation is not big problem here. + But we can remove dependency chain with v = b in the end of loop. */ + ONE_ITER(0) + #if (NUM_ITERS > 1) + ONE_ITER(1) + #if (NUM_ITERS > 2) + ONE_ITER(2) + #if (NUM_ITERS > 3) + ONE_ITER(3) + #if (NUM_ITERS > 4) + ONE_ITER(4) + #if (NUM_ITERS > 5) + ONE_ITER(5) + #if (NUM_ITERS > 6) + ONE_ITER(6) + #if (NUM_ITERS > 7) + ONE_ITER(7) + #endif + #endif + #endif + #endif + #endif + #endif + #endif + + src += NUM_ITERS; + if (src == srcLim) + break; + } + + if (src == srcLim) + #if (NUM_ITERS > 1) + for (;;) + #endif + { + #if (NUM_ITERS > 1) + if (src == p->lims[BCJ2_STREAM_MAIN] || dest == p->destLim) + #endif { - p->state = BCJ2_DEC_STATE_ORIG; + const SizeT num = (SizeT)(src - p->bufs[BCJ2_STREAM_MAIN]); + p->bufs[BCJ2_STREAM_MAIN] = src; + p->dest = dest; + p->ip += (UInt32)num; + /* state BCJ2_STREAM_MAIN has more priority than BCJ2_STATE_ORIG */ + p->state = + src == p->lims[BCJ2_STREAM_MAIN] ? + (unsigned)BCJ2_STREAM_MAIN : + (unsigned)BCJ2_DEC_STATE_ORIG; + p->temp = v; return SZ_OK; } + #if (NUM_ITERS > 1) + ONE_ITER(0) + src++; + #endif } - - srcLim = src + num; - if (p->temp[3] == 0x0F && (src[0] & 0xF0) == 0x80) - *dest = src[0]; - else for (;;) { - Byte b = *src; - *dest = b; - if (b != 0x0F) - { - if ((b & 0xFE) == 0xE8) - break; - dest++; - if (++src != srcLim) - continue; - break; - } - dest++; - if (++src == srcLim) - break; - if ((*src & 0xF0) != 0x80) - continue; - *dest = *src; - break; - } - - num = (SizeT)(src - p->bufs[BCJ2_STREAM_MAIN]); - - if (src == srcLim) - { - p->temp[3] = src[-1]; - p->bufs[BCJ2_STREAM_MAIN] = src; + const SizeT num = (SizeT)(dest - p->dest); + p->dest = dest; // p->dest += num; + p->bufs[BCJ2_STREAM_MAIN] += num; // = src; p->ip += (UInt32)num; - p->dest += num; - p->state = - p->bufs[BCJ2_STREAM_MAIN] == - p->lims[BCJ2_STREAM_MAIN] ? - (unsigned)BCJ2_STREAM_MAIN : - (unsigned)BCJ2_DEC_STATE_ORIG; - return SZ_OK; } - { UInt32 bound, ttt; - CProb *prob; - Byte b = src[0]; - Byte prev = (Byte)(num == 0 ? p->temp[3] : src[-1]); - - p->temp[3] = b; - p->bufs[BCJ2_STREAM_MAIN] = src + 1; - num++; - p->ip += (UInt32)num; - p->dest += num; - - prob = p->probs + (unsigned)(b == 0xE8 ? 2 + (unsigned)prev : (b == 0xE9 ? 1 : 0)); - - _IF_BIT_0 + CBcj2Prob *prob; // unsigned index; + /* + prob = p->probs + (unsigned)((Byte)v == 0xe8 ? + 2 + (Byte)(v >> 8) : + ((v >> 5) & 1)); // ((Byte)v < 0xe8 ? 0 : 1)); + */ { - _UPDATE_0 + const unsigned c = ((v + 0x17) >> 6) & 1; + prob = p->probs + (unsigned) + (((0 - c) & (Byte)(v >> NUM_SHIFT_BITS)) + c + ((v >> 5) & 1)); + // (Byte) + // 8x->0 : e9->1 : xxe8->xx+2 + // 8x->0x100 : e9->0x101 : xxe8->xx + // (((0x100 - (e & ~v)) & (0x100 | (v >> 8))) + (e & v)); + // (((0x101 + (~e | v)) & (0x100 | (v >> 8))) + (e & v)); + } + ttt = *prob; + bound = (p->range >> kNumBitModelTotalBits) * ttt; + if (p->code < bound) + { + // bcj2_stats[prob - p->probs][0]++; + p->range = bound; + *prob = (CBcj2Prob)(ttt + ((kBitModelTotal - ttt) >> kNumMoveBits)); continue; } - _UPDATE_1 - + { + // bcj2_stats[prob - p->probs][1]++; + p->range -= bound; + p->code -= bound; + *prob = (CBcj2Prob)(ttt - (ttt >> kNumMoveBits)); + } } } } - { - UInt32 val; - unsigned cj = (p->temp[3] == 0xE8) ? BCJ2_STREAM_CALL : BCJ2_STREAM_JUMP; + /* (v == 0xe8 ? 0 : 1) uses setcc instruction with additional zero register usage in x64 MSVC. */ + // const unsigned cj = ((Byte)v == 0xe8) ? BCJ2_STREAM_CALL : BCJ2_STREAM_JUMP; + const unsigned cj = (((v + 0x57) >> 6) & 1) + BCJ2_STREAM_CALL; const Byte *cur = p->bufs[cj]; Byte *dest; SizeT rem; - if (cur == p->lims[cj]) { p->state = cj; break; } - - val = GetBe32(cur); + v = GetBe32a(cur); p->bufs[cj] = cur + 4; - - p->ip += 4; - val -= p->ip; + { + const UInt32 ip = p->ip + 4; + v -= ip; + p->ip = ip; + } dest = p->dest; rem = (SizeT)(p->destLim - dest); - if (rem < 4) { - p->temp[0] = (Byte)val; if (rem > 0) dest[0] = (Byte)val; val >>= 8; - p->temp[1] = (Byte)val; if (rem > 1) dest[1] = (Byte)val; val >>= 8; - p->temp[2] = (Byte)val; if (rem > 2) dest[2] = (Byte)val; val >>= 8; - p->temp[3] = (Byte)val; + if ((unsigned)rem > 0) { dest[0] = (Byte)v; v >>= 8; + if ((unsigned)rem > 1) { dest[1] = (Byte)v; v >>= 8; + if ((unsigned)rem > 2) { dest[2] = (Byte)v; v >>= 8; }}} + p->temp = v; p->dest = dest + rem; p->state = BCJ2_DEC_STATE_ORIG_0 + (unsigned)rem; break; } - - SetUi32(dest, val); - p->temp[3] = (Byte)(val >> 24); + SetUi32(dest, v) + v >>= 24; p->dest = dest + 4; } } @@ -252,6 +278,13 @@ SRes Bcj2Dec_Decode(CBcj2Dec *p) p->range <<= 8; p->code = (p->code << 8) | *(p->bufs[BCJ2_STREAM_RC])++; } - return SZ_OK; } + +#undef NUM_ITERS +#undef ONE_ITER +#undef NUM_SHIFT_BITS +#undef kTopValue +#undef kNumBitModelTotalBits +#undef kBitModelTotal +#undef kNumMoveBits diff --git a/libraries/lzma/C/Bcj2.h b/libraries/lzma/C/Bcj2.h index 8824080ac..4575545b6 100644 --- a/libraries/lzma/C/Bcj2.h +++ b/libraries/lzma/C/Bcj2.h @@ -1,8 +1,8 @@ -/* Bcj2.h -- BCJ2 Converter for x86 code -2014-11-10 : Igor Pavlov : Public domain */ +/* Bcj2.h -- BCJ2 converter for x86 code (Branch CALL/JUMP variant2) +2023-03-02 : Igor Pavlov : Public domain */ -#ifndef __BCJ2_H -#define __BCJ2_H +#ifndef ZIP7_INC_BCJ2_H +#define ZIP7_INC_BCJ2_H #include "7zTypes.h" @@ -26,37 +26,68 @@ enum BCJ2_DEC_STATE_ORIG_3, BCJ2_DEC_STATE_ORIG, - BCJ2_DEC_STATE_OK + BCJ2_DEC_STATE_ERROR /* after detected data error */ }; enum { BCJ2_ENC_STATE_ORIG = BCJ2_NUM_STREAMS, - BCJ2_ENC_STATE_OK + BCJ2_ENC_STATE_FINISHED /* it's state after fully encoded stream */ }; -#define BCJ2_IS_32BIT_STREAM(s) ((s) == BCJ2_STREAM_CALL || (s) == BCJ2_STREAM_JUMP) +/* #define BCJ2_IS_32BIT_STREAM(s) ((s) == BCJ2_STREAM_CALL || (s) == BCJ2_STREAM_JUMP) */ +#define BCJ2_IS_32BIT_STREAM(s) ((unsigned)((unsigned)(s) - (unsigned)BCJ2_STREAM_CALL) < 2) /* CBcj2Dec / CBcj2Enc bufs sizes: BUF_SIZE(n) = lims[n] - bufs[n] -bufs sizes for BCJ2_STREAM_CALL and BCJ2_STREAM_JUMP must be mutliply of 4: +bufs sizes for BCJ2_STREAM_CALL and BCJ2_STREAM_JUMP must be multiply of 4: (BUF_SIZE(BCJ2_STREAM_CALL) & 3) == 0 (BUF_SIZE(BCJ2_STREAM_JUMP) & 3) == 0 */ +// typedef UInt32 CBcj2Prob; +typedef UInt16 CBcj2Prob; + +/* +BCJ2 encoder / decoder internal requirements: + - If last bytes of stream contain marker (e8/e8/0f8x), then + there is also encoded symbol (0 : no conversion) in RC stream. + - One case of overlapped instructions is supported, + if last byte of converted instruction is (0f) and next byte is (8x): + marker [xx xx xx 0f] 8x + then the pair (0f 8x) is treated as marker. +*/ + +/* ---------- BCJ2 Decoder ---------- */ + /* CBcj2Dec: -dest is allowed to overlap with bufs[BCJ2_STREAM_MAIN], with the following conditions: +(dest) is allowed to overlap with bufs[BCJ2_STREAM_MAIN], with the following conditions: bufs[BCJ2_STREAM_MAIN] >= dest && - bufs[BCJ2_STREAM_MAIN] - dest >= tempReserv + + bufs[BCJ2_STREAM_MAIN] - dest >= BUF_SIZE(BCJ2_STREAM_CALL) + BUF_SIZE(BCJ2_STREAM_JUMP) - tempReserv = 0 : for first call of Bcj2Dec_Decode - tempReserv = 4 : for any other calls of Bcj2Dec_Decode - overlap with offset = 1 is not allowed + reserve = bufs[BCJ2_STREAM_MAIN] - dest - + ( BUF_SIZE(BCJ2_STREAM_CALL) + + BUF_SIZE(BCJ2_STREAM_JUMP) ) + and additional conditions: + if (it's first call of Bcj2Dec_Decode() after Bcj2Dec_Init()) + { + (reserve != 1) : if (ver < v23.00) + } + else // if there are more than one calls of Bcj2Dec_Decode() after Bcj2Dec_Init()) + { + (reserve >= 6) : if (ver < v23.00) + (reserve >= 4) : if (ver >= v23.00) + We need that (reserve) because after first call of Bcj2Dec_Decode(), + CBcj2Dec::temp can contain up to 4 bytes for writing to (dest). + } + (reserve == 0) is allowed, if we decode full stream via single call of Bcj2Dec_Decode(). + (reserve == 0) also is allowed in case of multi-call, if we use fixed buffers, + and (reserve) is calculated from full (final) sizes of all streams before first call. */ typedef struct @@ -68,22 +99,66 @@ typedef struct unsigned state; /* BCJ2_STREAM_MAIN has more priority than BCJ2_STATE_ORIG */ - UInt32 ip; - Byte temp[4]; + UInt32 ip; /* property of starting base for decoding */ + UInt32 temp; /* Byte temp[4]; */ UInt32 range; UInt32 code; - UInt16 probs[2 + 256]; + CBcj2Prob probs[2 + 256]; } CBcj2Dec; + +/* Note: + Bcj2Dec_Init() sets (CBcj2Dec::ip = 0) + if (ip != 0) property is required, the caller must set CBcj2Dec::ip after Bcj2Dec_Init() +*/ void Bcj2Dec_Init(CBcj2Dec *p); -/* Returns: SZ_OK or SZ_ERROR_DATA */ + +/* Bcj2Dec_Decode(): + returns: + SZ_OK + SZ_ERROR_DATA : if data in 5 starting bytes of BCJ2_STREAM_RC stream are not correct +*/ SRes Bcj2Dec_Decode(CBcj2Dec *p); -#define Bcj2Dec_IsFinished(_p_) ((_p_)->code == 0) +/* To check that decoding was finished you can compare + sizes of processed streams with sizes known from another sources. + You must do at least one mandatory check from the two following options: + - the check for size of processed output (ORIG) stream. + - the check for size of processed input (MAIN) stream. + additional optional checks: + - the checks for processed sizes of all input streams (MAIN, CALL, JUMP, RC) + - the checks Bcj2Dec_IsMaybeFinished*() + also before actual decoding you can check that the + following condition is met for stream sizes: + ( size(ORIG) == size(MAIN) + size(CALL) + size(JUMP) ) +*/ + +/* (state == BCJ2_STREAM_MAIN) means that decoder is ready for + additional input data in BCJ2_STREAM_MAIN stream. + Note that (state == BCJ2_STREAM_MAIN) is allowed for non-finished decoding. +*/ +#define Bcj2Dec_IsMaybeFinished_state_MAIN(_p_) ((_p_)->state == BCJ2_STREAM_MAIN) + +/* if the stream decoding was finished correctly, then range decoder + part of CBcj2Dec also was finished, and then (CBcj2Dec::code == 0). + Note that (CBcj2Dec::code == 0) is allowed for non-finished decoding. +*/ +#define Bcj2Dec_IsMaybeFinished_code(_p_) ((_p_)->code == 0) + +/* use Bcj2Dec_IsMaybeFinished() only as additional check + after at least one mandatory check from the two following options: + - the check for size of processed output (ORIG) stream. + - the check for size of processed input (MAIN) stream. +*/ +#define Bcj2Dec_IsMaybeFinished(_p_) ( \ + Bcj2Dec_IsMaybeFinished_state_MAIN(_p_) && \ + Bcj2Dec_IsMaybeFinished_code(_p_)) +/* ---------- BCJ2 Encoder ---------- */ + typedef enum { BCJ2_ENC_FINISH_MODE_CONTINUE, @@ -91,6 +166,91 @@ typedef enum BCJ2_ENC_FINISH_MODE_END_STREAM } EBcj2Enc_FinishMode; +/* + BCJ2_ENC_FINISH_MODE_CONTINUE: + process non finished encoding. + It notifies the encoder that additional further calls + can provide more input data (src) than provided by current call. + In that case the CBcj2Enc encoder still can move (src) pointer + up to (srcLim), but CBcj2Enc encoder can store some of the last + processed bytes (up to 4 bytes) from src to internal CBcj2Enc::temp[] buffer. + at return: + (CBcj2Enc::src will point to position that includes + processed data and data copied to (temp[]) buffer) + That data from (temp[]) buffer will be used in further calls. + + BCJ2_ENC_FINISH_MODE_END_BLOCK: + finish encoding of current block (ended at srcLim) without RC flushing. + at return: if (CBcj2Enc::state == BCJ2_ENC_STATE_ORIG) && + CBcj2Enc::src == CBcj2Enc::srcLim) + : it shows that block encoding was finished. And the encoder is + ready for new (src) data or for stream finish operation. + finished block means + { + CBcj2Enc has completed block encoding up to (srcLim). + (1 + 4 bytes) or (2 + 4 bytes) CALL/JUMP cortages will + not cross block boundary at (srcLim). + temporary CBcj2Enc buffer for (ORIG) src data is empty. + 3 output uncompressed streams (MAIN, CALL, JUMP) were flushed. + RC stream was not flushed. And RC stream will cross block boundary. + } + Note: some possible implementation of BCJ2 encoder could + write branch marker (e8/e8/0f8x) in one call of Bcj2Enc_Encode(), + and it could calculate symbol for RC in another call of Bcj2Enc_Encode(). + BCJ2 encoder uses ip/fileIp/fileSize/relatLimit values to calculate RC symbol. + And these CBcj2Enc variables can have different values in different Bcj2Enc_Encode() calls. + So caller must finish each block with BCJ2_ENC_FINISH_MODE_END_BLOCK + to ensure that RC symbol is calculated and written in proper block. + + BCJ2_ENC_FINISH_MODE_END_STREAM + finish encoding of stream (ended at srcLim) fully including RC flushing. + at return: if (CBcj2Enc::state == BCJ2_ENC_STATE_FINISHED) + : it shows that stream encoding was finished fully, + and all output streams were flushed fully. + also Bcj2Enc_IsFinished() can be called. +*/ + + +/* + 32-bit relative offset in JUMP/CALL commands is + - (mod 4 GiB) for 32-bit x86 code + - signed Int32 for 64-bit x86-64 code + BCJ2 encoder also does internal relative to absolute address conversions. + And there are 2 possible ways to do it: + before v23: we used 32-bit variables and (mod 4 GiB) conversion + since v23: we use 64-bit variables and (signed Int32 offset) conversion. + The absolute address condition for conversion in v23: + ((UInt64)((Int64)ip64 - (Int64)fileIp64 + 5 + (Int32)offset) < (UInt64)fileSize64) + note that if (fileSize64 > 2 GiB). there is difference between + old (mod 4 GiB) way (v22) and new (signed Int32 offset) way (v23). + And new (v23) way is more suitable to encode 64-bit x86-64 code for (fileSize64 > 2 GiB) cases. +*/ + +/* +// for old (v22) way for conversion: +typedef UInt32 CBcj2Enc_ip_unsigned; +typedef Int32 CBcj2Enc_ip_signed; +#define BCJ2_ENC_FileSize_MAX ((UInt32)1 << 31) +*/ +typedef UInt64 CBcj2Enc_ip_unsigned; +typedef Int64 CBcj2Enc_ip_signed; + +/* maximum size of file that can be used for conversion condition */ +#define BCJ2_ENC_FileSize_MAX ((CBcj2Enc_ip_unsigned)0 - 2) + +/* default value of fileSize64_minus1 variable that means + that absolute address limitation will not be used */ +#define BCJ2_ENC_FileSizeField_UNLIMITED ((CBcj2Enc_ip_unsigned)0 - 1) + +/* calculate value that later can be set to CBcj2Enc::fileSize64_minus1 */ +#define BCJ2_ENC_GET_FileSizeField_VAL_FROM_FileSize(fileSize) \ + ((CBcj2Enc_ip_unsigned)(fileSize) - 1) + +/* set CBcj2Enc::fileSize64_minus1 variable from size of file */ +#define Bcj2Enc_SET_FileSize(p, fileSize) \ + (p)->fileSize64_minus1 = BCJ2_ENC_GET_FileSizeField_VAL_FROM_FileSize(fileSize); + + typedef struct { Byte *bufs[BCJ2_NUM_STREAMS]; @@ -101,45 +261,71 @@ typedef struct unsigned state; EBcj2Enc_FinishMode finishMode; - Byte prevByte; + Byte context; + Byte flushRem; + Byte isFlushState; Byte cache; UInt32 range; UInt64 low; UInt64 cacheSize; + + // UInt32 context; // for marker version, it can include marker flag. - UInt32 ip; - - /* 32-bit ralative offset in JUMP/CALL commands is - - (mod 4 GB) in 32-bit mode - - signed Int32 in 64-bit mode - We use (mod 4 GB) check for fileSize. - Use fileSize up to 2 GB, if you want to support 32-bit and 64-bit code conversion. */ - UInt32 fileIp; - UInt32 fileSize; /* (fileSize <= ((UInt32)1 << 31)), 0 means no_limit */ - UInt32 relatLimit; /* (relatLimit <= ((UInt32)1 << 31)), 0 means desable_conversion */ + /* (ip64) and (fileIp64) correspond to virtual source stream position + that doesn't include data in temp[] */ + CBcj2Enc_ip_unsigned ip64; /* current (ip) position */ + CBcj2Enc_ip_unsigned fileIp64; /* start (ip) position of current file */ + CBcj2Enc_ip_unsigned fileSize64_minus1; /* size of current file (for conversion limitation) */ + UInt32 relatLimit; /* (relatLimit <= ((UInt32)1 << 31)) : 0 means disable_conversion */ + // UInt32 relatExcludeBits; UInt32 tempTarget; - unsigned tempPos; - Byte temp[4 * 2]; - - unsigned flushPos; - - UInt16 probs[2 + 256]; + unsigned tempPos; /* the number of bytes that were copied to temp[] buffer + (tempPos <= 4) outside of Bcj2Enc_Encode() */ + // Byte temp[4]; // for marker version + Byte temp[8]; + CBcj2Prob probs[2 + 256]; } CBcj2Enc; void Bcj2Enc_Init(CBcj2Enc *p); + + +/* +Bcj2Enc_Encode(): at exit: + p->State < BCJ2_NUM_STREAMS : we need more buffer space for output stream + (bufs[p->State] == lims[p->State]) + p->State == BCJ2_ENC_STATE_ORIG : we need more data in input src stream + (src == srcLim) + p->State == BCJ2_ENC_STATE_FINISHED : after fully encoded stream +*/ void Bcj2Enc_Encode(CBcj2Enc *p); -#define Bcj2Enc_Get_InputData_Size(p) ((SizeT)((p)->srcLim - (p)->src) + (p)->tempPos) -#define Bcj2Enc_IsFinished(p) ((p)->flushPos == 5) +/* Bcj2Enc encoder can look ahead for up 4 bytes of source stream. + CBcj2Enc::tempPos : is the number of bytes that were copied from input stream to temp[] buffer. + (CBcj2Enc::src) after Bcj2Enc_Encode() is starting position after + fully processed data and after data copied to temp buffer. + So if the caller needs to get real number of fully processed input + bytes (without look ahead data in temp buffer), + the caller must subtruct (CBcj2Enc::tempPos) value from processed size + value that is calculated based on current (CBcj2Enc::src): + cur_processed_pos = Calc_Big_Processed_Pos(enc.src)) - + Bcj2Enc_Get_AvailInputSize_in_Temp(&enc); +*/ +/* get the size of input data that was stored in temp[] buffer: */ +#define Bcj2Enc_Get_AvailInputSize_in_Temp(p) ((p)->tempPos) +#define Bcj2Enc_IsFinished(p) ((p)->flushRem == 0) -#define BCJ2_RELAT_LIMIT_NUM_BITS 26 -#define BCJ2_RELAT_LIMIT ((UInt32)1 << BCJ2_RELAT_LIMIT_NUM_BITS) - -/* limit for CBcj2Enc::fileSize variable */ -#define BCJ2_FileSize_MAX ((UInt32)1 << 31) +/* Note : the decoder supports overlapping of marker (0f 80). + But we can eliminate such overlapping cases by setting + the limit for relative offset conversion as + CBcj2Enc::relatLimit <= (0x0f << 24) == (240 MiB) +*/ +/* default value for CBcj2Enc::relatLimit */ +#define BCJ2_ENC_RELAT_LIMIT_DEFAULT ((UInt32)0x0f << 24) +#define BCJ2_ENC_RELAT_LIMIT_MAX ((UInt32)1 << 31) +// #define BCJ2_RELAT_EXCLUDE_NUM_BITS 5 EXTERN_C_END diff --git a/libraries/lzma/C/Bcj2Enc.c b/libraries/lzma/C/Bcj2Enc.c new file mode 100644 index 000000000..79460bb15 --- /dev/null +++ b/libraries/lzma/C/Bcj2Enc.c @@ -0,0 +1,506 @@ +/* Bcj2Enc.c -- BCJ2 Encoder converter for x86 code (Branch CALL/JUMP variant2) +2023-04-02 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +/* #define SHOW_STAT */ +#ifdef SHOW_STAT +#include +#define PRF2(s) printf("%s ip=%8x tempPos=%d src= %8x\n", s, (unsigned)p->ip64, p->tempPos, (unsigned)(p->srcLim - p->src)); +#else +#define PRF2(s) +#endif + +#include "Bcj2.h" +#include "CpuArch.h" + +#define kTopValue ((UInt32)1 << 24) +#define kNumBitModelTotalBits 11 +#define kBitModelTotal (1 << kNumBitModelTotalBits) +#define kNumMoveBits 5 + +void Bcj2Enc_Init(CBcj2Enc *p) +{ + unsigned i; + p->state = BCJ2_ENC_STATE_ORIG; + p->finishMode = BCJ2_ENC_FINISH_MODE_CONTINUE; + p->context = 0; + p->flushRem = 5; + p->isFlushState = 0; + p->cache = 0; + p->range = 0xffffffff; + p->low = 0; + p->cacheSize = 1; + p->ip64 = 0; + p->fileIp64 = 0; + p->fileSize64_minus1 = BCJ2_ENC_FileSizeField_UNLIMITED; + p->relatLimit = BCJ2_ENC_RELAT_LIMIT_DEFAULT; + // p->relatExcludeBits = 0; + p->tempPos = 0; + for (i = 0; i < sizeof(p->probs) / sizeof(p->probs[0]); i++) + p->probs[i] = kBitModelTotal >> 1; +} + +// Z7_NO_INLINE +Z7_FORCE_INLINE +static BoolInt Bcj2_RangeEnc_ShiftLow(CBcj2Enc *p) +{ + const UInt32 low = (UInt32)p->low; + const unsigned high = (unsigned) + #if defined(Z7_MSC_VER_ORIGINAL) \ + && defined(MY_CPU_X86) \ + && defined(MY_CPU_LE) \ + && !defined(MY_CPU_64BIT) + // we try to rid of __aullshr() call in MSVS-x86 + (((const UInt32 *)&p->low)[1]); // [1] : for little-endian only + #else + (p->low >> 32); + #endif + if (low < (UInt32)0xff000000 || high != 0) + { + Byte *buf = p->bufs[BCJ2_STREAM_RC]; + do + { + if (buf == p->lims[BCJ2_STREAM_RC]) + { + p->state = BCJ2_STREAM_RC; + p->bufs[BCJ2_STREAM_RC] = buf; + return True; + } + *buf++ = (Byte)(p->cache + high); + p->cache = 0xff; + } + while (--p->cacheSize); + p->bufs[BCJ2_STREAM_RC] = buf; + p->cache = (Byte)(low >> 24); + } + p->cacheSize++; + p->low = low << 8; + return False; +} + + +/* +We can use 2 alternative versions of code: +1) non-marker version: + Byte CBcj2Enc::context + Byte temp[8]; + Last byte of marker (e8/e9/[0f]8x) can be written to temp[] buffer. + Encoder writes last byte of marker (e8/e9/[0f]8x) to dest, only in conjunction + with writing branch symbol to range coder in same Bcj2Enc_Encode_2() call. + +2) marker version: + UInt32 CBcj2Enc::context + Byte CBcj2Enc::temp[4]; + MARKER_FLAG in CBcj2Enc::context shows that CBcj2Enc::context contains finded marker. + it's allowed that + one call of Bcj2Enc_Encode_2() writes last byte of marker (e8/e9/[0f]8x) to dest, + and another call of Bcj2Enc_Encode_2() does offset conversion. + So different values of (fileIp) and (fileSize) are possible + in these different Bcj2Enc_Encode_2() calls. + +Also marker version requires additional if((v & MARKER_FLAG) == 0) check in main loop. +So we use non-marker version. +*/ + +/* + Corner cases with overlap in multi-block. + before v23: there was one corner case, where converted instruction + could start in one sub-stream and finish in next sub-stream. + If multi-block (solid) encoding is used, + and BCJ2_ENC_FINISH_MODE_END_BLOCK is used for each sub-stream. + and (0f) is last byte of previous sub-stream + and (8x) is first byte of current sub-stream + then (0f 8x) pair is treated as marker by BCJ2 encoder and decoder. + BCJ2 encoder can converts 32-bit offset for that (0f 8x) cortage, + if that offset meets limit requirements. + If encoder allows 32-bit offset conversion for such overlap case, + then the data in 3 uncompressed BCJ2 streams for some sub-stream + can depend from data of previous sub-stream. + That corner case is not big problem, and it's rare case. + Since v23.00 we do additional check to prevent conversions in such overlap cases. +*/ + +/* + Bcj2Enc_Encode_2() output variables at exit: + { + if (Bcj2Enc_Encode_2() exits with (p->state == BCJ2_ENC_STATE_ORIG)) + { + it means that encoder needs more input data. + if (p->srcLim == p->src) at exit, then + { + (p->finishMode != BCJ2_ENC_FINISH_MODE_END_STREAM) + all input data were read and processed, and we are ready for + new input data. + } + else + { + (p->srcLim != p->src) + (p->finishMode == BCJ2_ENC_FINISH_MODE_CONTINUE) + The encoder have found e8/e9/0f_8x marker, + and p->src points to last byte of that marker, + Bcj2Enc_Encode_2() needs more input data to get totally + 5 bytes (last byte of marker and 32-bit branch offset) + as continuous array starting from p->src. + (p->srcLim - p->src < 5) requirement is met after exit. + So non-processed resedue from p->src to p->srcLim is always less than 5 bytes. + } + } + } +*/ + +Z7_NO_INLINE +static void Bcj2Enc_Encode_2(CBcj2Enc *p) +{ + if (!p->isFlushState) + { + const Byte *src; + UInt32 v; + { + const unsigned state = p->state; + if (BCJ2_IS_32BIT_STREAM(state)) + { + Byte *cur = p->bufs[state]; + if (cur == p->lims[state]) + return; + SetBe32a(cur, p->tempTarget) + p->bufs[state] = cur + 4; + } + } + p->state = BCJ2_ENC_STATE_ORIG; // for main reason of exit + src = p->src; + v = p->context; + + // #define WRITE_CONTEXT p->context = v; // for marker version + #define WRITE_CONTEXT p->context = (Byte)v; + #define WRITE_CONTEXT_AND_SRC p->src = src; WRITE_CONTEXT + + for (;;) + { + // const Byte *src; + // UInt32 v; + CBcj2Enc_ip_unsigned ip; + if (p->range < kTopValue) + { + // to reduce register pressure and code size: we save and restore local variables. + WRITE_CONTEXT_AND_SRC + if (Bcj2_RangeEnc_ShiftLow(p)) + return; + p->range <<= 8; + src = p->src; + v = p->context; + } + // src = p->src; + // #define MARKER_FLAG ((UInt32)1 << 17) + // if ((v & MARKER_FLAG) == 0) // for marker version + { + const Byte *srcLim; + Byte *dest = p->bufs[BCJ2_STREAM_MAIN]; + { + const SizeT remSrc = (SizeT)(p->srcLim - src); + SizeT rem = (SizeT)(p->lims[BCJ2_STREAM_MAIN] - dest); + if (rem >= remSrc) + rem = remSrc; + srcLim = src + rem; + } + /* p->context contains context of previous byte: + bits [0 : 7] : src[-1], if (src) was changed in this call + bits [8 : 31] : are undefined for non-marker version + */ + // v = p->context; + #define NUM_SHIFT_BITS 24 + #define CONV_FLAG ((UInt32)1 << 16) + #define ONE_ITER { \ + b = src[0]; \ + *dest++ = (Byte)b; \ + v = (v << NUM_SHIFT_BITS) | b; \ + if (((b + (0x100 - 0xe8)) & 0xfe) == 0) break; \ + if (((v - (((UInt32)0x0f << (NUM_SHIFT_BITS)) + 0x80)) & \ + ((((UInt32)1 << (4 + NUM_SHIFT_BITS)) - 0x1) << 4)) == 0) break; \ + src++; if (src == srcLim) { break; } } + + if (src != srcLim) + for (;;) + { + /* clang can generate ineffective code with setne instead of two jcc instructions. + we can use 2 iterations and external (unsigned b) to avoid that ineffective code genaration. */ + unsigned b; + ONE_ITER + ONE_ITER + } + + ip = p->ip64 + (CBcj2Enc_ip_unsigned)(SizeT)(dest - p->bufs[BCJ2_STREAM_MAIN]); + p->bufs[BCJ2_STREAM_MAIN] = dest; + p->ip64 = ip; + + if (src == srcLim) + { + WRITE_CONTEXT_AND_SRC + if (src != p->srcLim) + { + p->state = BCJ2_STREAM_MAIN; + return; + } + /* (p->src == p->srcLim) + (p->state == BCJ2_ENC_STATE_ORIG) */ + if (p->finishMode != BCJ2_ENC_FINISH_MODE_END_STREAM) + return; + /* (p->finishMode == BCJ2_ENC_FINISH_MODE_END_STREAM */ + // (p->flushRem == 5); + p->isFlushState = 1; + break; + } + src++; + // p->src = src; + } + // ip = p->ip; // for marker version + /* marker was found */ + /* (v) contains marker that was found: + bits [NUM_SHIFT_BITS : NUM_SHIFT_BITS + 7] + : value of src[-2] : xx/xx/0f + bits [0 : 7] : value of src[-1] : e8/e9/8x + */ + { + { + #if NUM_SHIFT_BITS != 24 + v &= ~(UInt32)CONV_FLAG; + #endif + // UInt32 relat = 0; + if ((SizeT)(p->srcLim - src) >= 4) + { + /* + if (relat != 0 || (Byte)v != 0xe8) + BoolInt isBigOffset = True; + */ + const UInt32 relat = GetUi32(src); + /* + #define EXCLUDE_FLAG ((UInt32)1 << 4) + #define NEED_CONVERT(rel) ((((rel) + EXCLUDE_FLAG) & (0 - EXCLUDE_FLAG * 2)) != 0) + if (p->relatExcludeBits != 0) + { + const UInt32 flag = (UInt32)1 << (p->relatExcludeBits - 1); + isBigOffset = (((relat + flag) & (0 - flag * 2)) != 0); + } + // isBigOffset = False; // for debug + */ + ip -= p->fileIp64; + // Use the following if check, if (ip) is 64-bit: + if (ip > (((v + 0x20) >> 5) & 1)) // 23.00 : we eliminate milti-block overlap for (Of 80) and (e8/e9) + if ((CBcj2Enc_ip_unsigned)((CBcj2Enc_ip_signed)ip + 4 + (Int32)relat) <= p->fileSize64_minus1) + if (((UInt32)(relat + p->relatLimit) >> 1) < p->relatLimit) + v |= CONV_FLAG; + } + else if (p->finishMode == BCJ2_ENC_FINISH_MODE_CONTINUE) + { + // (p->srcLim - src < 4) + // /* + // for non-marker version + p->ip64--; // p->ip = ip - 1; + p->bufs[BCJ2_STREAM_MAIN]--; + src--; + v >>= NUM_SHIFT_BITS; + // (0 < p->srcLim - p->src <= 4) + // */ + // v |= MARKER_FLAG; // for marker version + /* (p->state == BCJ2_ENC_STATE_ORIG) */ + WRITE_CONTEXT_AND_SRC + return; + } + { + const unsigned c = ((v + 0x17) >> 6) & 1; + CBcj2Prob *prob = p->probs + (unsigned) + (((0 - c) & (Byte)(v >> NUM_SHIFT_BITS)) + c + ((v >> 5) & 1)); + /* + ((Byte)v == 0xe8 ? 2 + ((Byte)(v >> 8)) : + ((Byte)v < 0xe8 ? 0 : 1)); // ((v >> 5) & 1)); + */ + const unsigned ttt = *prob; + const UInt32 bound = (p->range >> kNumBitModelTotalBits) * ttt; + if ((v & CONV_FLAG) == 0) + { + // static int yyy = 0; yyy++; printf("\n!needConvert = %d\n", yyy); + // v = (Byte)v; // for marker version + p->range = bound; + *prob = (CBcj2Prob)(ttt + ((kBitModelTotal - ttt) >> kNumMoveBits)); + // WRITE_CONTEXT_AND_SRC + continue; + } + p->low += bound; + p->range -= bound; + *prob = (CBcj2Prob)(ttt - (ttt >> kNumMoveBits)); + } + // p->context = src[3]; + { + // const unsigned cj = ((Byte)v == 0xe8 ? BCJ2_STREAM_CALL : BCJ2_STREAM_JUMP); + const unsigned cj = (((v + 0x57) >> 6) & 1) + BCJ2_STREAM_CALL; + ip = p->ip64; + v = GetUi32(src); // relat + ip += 4; + p->ip64 = ip; + src += 4; + // p->src = src; + { + const UInt32 absol = (UInt32)ip + v; + Byte *cur = p->bufs[cj]; + v >>= 24; + // WRITE_CONTEXT + if (cur == p->lims[cj]) + { + p->state = cj; + p->tempTarget = absol; + WRITE_CONTEXT_AND_SRC + return; + } + SetBe32a(cur, absol) + p->bufs[cj] = cur + 4; + } + } + } + } + } // end of loop + } + + for (; p->flushRem != 0; p->flushRem--) + if (Bcj2_RangeEnc_ShiftLow(p)) + return; + p->state = BCJ2_ENC_STATE_FINISHED; +} + + +/* +BCJ2 encoder needs look ahead for up to 4 bytes in (src) buffer. +So base function Bcj2Enc_Encode_2() + in BCJ2_ENC_FINISH_MODE_CONTINUE mode can return with + (p->state == BCJ2_ENC_STATE_ORIG && p->src < p->srcLim) +Bcj2Enc_Encode() solves that look ahead problem by using p->temp[] buffer. + so if (p->state == BCJ2_ENC_STATE_ORIG) after Bcj2Enc_Encode(), + then (p->src == p->srcLim). + And the caller's code is simpler with Bcj2Enc_Encode(). +*/ + +Z7_NO_INLINE +void Bcj2Enc_Encode(CBcj2Enc *p) +{ + PRF2("\n----") + if (p->tempPos != 0) + { + /* extra: number of bytes that were copied from (src) to (temp) buffer in this call */ + unsigned extra = 0; + /* We will touch only minimal required number of bytes in input (src) stream. + So we will add input bytes from (src) stream to temp[] with step of 1 byte. + We don't add new bytes to temp[] before Bcj2Enc_Encode_2() call + in first loop iteration because + - previous call of Bcj2Enc_Encode() could use another (finishMode), + - previous call could finish with (p->state != BCJ2_ENC_STATE_ORIG). + the case with full temp[] buffer (p->tempPos == 4) is possible here. + */ + for (;;) + { + // (0 < p->tempPos <= 5) // in non-marker version + /* p->src : the current src data position including extra bytes + that were copied to temp[] buffer in this call */ + const Byte *src = p->src; + const Byte *srcLim = p->srcLim; + const EBcj2Enc_FinishMode finishMode = p->finishMode; + if (src != srcLim) + { + /* if there are some src data after the data copied to temp[], + then we use MODE_CONTINUE for temp data */ + p->finishMode = BCJ2_ENC_FINISH_MODE_CONTINUE; + } + p->src = p->temp; + p->srcLim = p->temp + p->tempPos; + PRF2(" ") + Bcj2Enc_Encode_2(p); + { + const unsigned num = (unsigned)(p->src - p->temp); + const unsigned tempPos = p->tempPos - num; + unsigned i; + p->tempPos = tempPos; + for (i = 0; i < tempPos; i++) + p->temp[i] = p->temp[(SizeT)i + num]; + // tempPos : number of bytes in temp buffer + p->src = src; + p->srcLim = srcLim; + p->finishMode = finishMode; + if (p->state != BCJ2_ENC_STATE_ORIG) + { + // (p->tempPos <= 4) // in non-marker version + /* if (the reason of exit from Bcj2Enc_Encode_2() + is not BCJ2_ENC_STATE_ORIG), + then we exit from Bcj2Enc_Encode() with same reason */ + // optional code begin : we rollback (src) and tempPos, if it's possible: + if (extra >= tempPos) + extra = tempPos; + p->src = src - extra; + p->tempPos = tempPos - extra; + // optional code end : rollback of (src) and tempPos + return; + } + /* (p->tempPos <= 4) + (p->state == BCJ2_ENC_STATE_ORIG) + so encoder needs more data than in temp[] */ + if (src == srcLim) + return; // src buffer has no more input data. + /* (src != srcLim) + so we can provide more input data from src for Bcj2Enc_Encode_2() */ + if (extra >= tempPos) + { + /* (extra >= tempPos) means that temp buffer contains + only data from src buffer of this call. + So now we can encode without temp buffer */ + p->src = src - tempPos; // rollback (src) + p->tempPos = 0; + break; + } + // we append one additional extra byte from (src) to temp[] buffer: + p->temp[tempPos] = *src; + p->tempPos = tempPos + 1; + // (0 < p->tempPos <= 5) // in non-marker version + p->src = src + 1; + extra++; + } + } + } + + PRF2("++++") + // (p->tempPos == 0) + Bcj2Enc_Encode_2(p); + PRF2("====") + + if (p->state == BCJ2_ENC_STATE_ORIG) + { + const Byte *src = p->src; + const Byte *srcLim = p->srcLim; + const unsigned rem = (unsigned)(srcLim - src); + /* (rem <= 4) here. + if (p->src != p->srcLim), then + - we copy non-processed bytes from (p->src) to temp[] buffer, + - we set p->src equal to p->srcLim. + */ + if (rem) + { + unsigned i = 0; + p->src = srcLim; + p->tempPos = rem; + // (0 < p->tempPos <= 4) + do + p->temp[i] = src[i]; + while (++i != rem); + } + // (p->tempPos <= 4) + // (p->src == p->srcLim) + } +} + +#undef PRF2 +#undef CONV_FLAG +#undef MARKER_FLAG +#undef WRITE_CONTEXT +#undef WRITE_CONTEXT_AND_SRC +#undef ONE_ITER +#undef NUM_SHIFT_BITS +#undef kTopValue +#undef kNumBitModelTotalBits +#undef kBitModelTotal +#undef kNumMoveBits diff --git a/libraries/lzma/C/Bra.c b/libraries/lzma/C/Bra.c index 3b854d9ca..22e0e478e 100644 --- a/libraries/lzma/C/Bra.c +++ b/libraries/lzma/C/Bra.c @@ -1,230 +1,420 @@ -/* Bra.c -- Converters for RISC code -2021-02-09 : Igor Pavlov : Public domain */ +/* Bra.c -- Branch converters for RISC code +2023-04-02 : Igor Pavlov : Public domain */ #include "Precomp.h" -#include "CpuArch.h" #include "Bra.h" +#include "CpuArch.h" +#include "RotateDefs.h" -SizeT ARM_Convert(Byte *data, SizeT size, UInt32 ip, int encoding) +#if defined(MY_CPU_SIZEOF_POINTER) \ + && ( MY_CPU_SIZEOF_POINTER == 4 \ + || MY_CPU_SIZEOF_POINTER == 8) + #define BR_CONV_USE_OPT_PC_PTR +#endif + +#ifdef BR_CONV_USE_OPT_PC_PTR +#define BR_PC_INIT pc -= (UInt32)(SizeT)p; +#define BR_PC_GET (pc + (UInt32)(SizeT)p) +#else +#define BR_PC_INIT pc += (UInt32)size; +#define BR_PC_GET (pc - (UInt32)(SizeT)(lim - p)) +// #define BR_PC_INIT +// #define BR_PC_GET (pc + (UInt32)(SizeT)(p - data)) +#endif + +#define BR_CONVERT_VAL(v, c) if (encoding) v += c; else v -= c; +// #define BR_CONVERT_VAL(v, c) if (!encoding) c = (UInt32)0 - c; v += c; + +#define Z7_BRANCH_CONV(name) z7_BranchConv_ ## name + +#define Z7_BRANCH_FUNC_MAIN(name) \ +static \ +Z7_FORCE_INLINE \ +Z7_ATTRIB_NO_VECTOR \ +Byte *Z7_BRANCH_CONV(name)(Byte *p, SizeT size, UInt32 pc, int encoding) + +#define Z7_BRANCH_FUNC_IMP(name, m, encoding) \ +Z7_NO_INLINE \ +Z7_ATTRIB_NO_VECTOR \ +Byte *m(name)(Byte *data, SizeT size, UInt32 pc) \ + { return Z7_BRANCH_CONV(name)(data, size, pc, encoding); } \ + +#ifdef Z7_EXTRACT_ONLY +#define Z7_BRANCH_FUNCS_IMP(name) \ + Z7_BRANCH_FUNC_IMP(name, Z7_BRANCH_CONV_DEC, 0) +#else +#define Z7_BRANCH_FUNCS_IMP(name) \ + Z7_BRANCH_FUNC_IMP(name, Z7_BRANCH_CONV_DEC, 0) \ + Z7_BRANCH_FUNC_IMP(name, Z7_BRANCH_CONV_ENC, 1) +#endif + +#if defined(__clang__) +#define BR_EXTERNAL_FOR +#define BR_NEXT_ITERATION continue; +#else +#define BR_EXTERNAL_FOR for (;;) +#define BR_NEXT_ITERATION break; +#endif + +#if defined(__clang__) && (__clang_major__ >= 8) \ + || defined(__GNUC__) && (__GNUC__ >= 1000) \ + // GCC is not good for __builtin_expect() here + /* || defined(_MSC_VER) && (_MSC_VER >= 1920) */ + // #define Z7_unlikely [[unlikely]] + // #define Z7_LIKELY(x) (__builtin_expect((x), 1)) + #define Z7_UNLIKELY(x) (__builtin_expect((x), 0)) + // #define Z7_likely [[likely]] +#else + // #define Z7_LIKELY(x) (x) + #define Z7_UNLIKELY(x) (x) + // #define Z7_likely +#endif + + +Z7_BRANCH_FUNC_MAIN(ARM64) { - Byte *p; + // Byte *p = data; const Byte *lim; - size &= ~(size_t)3; - ip += 4; - p = data; - lim = data + size; - - if (encoding) - - for (;;) + const UInt32 flag = (UInt32)1 << (24 - 4); + const UInt32 mask = ((UInt32)1 << 24) - (flag << 1); + size &= ~(SizeT)3; + // if (size == 0) return p; + lim = p + size; + BR_PC_INIT + pc -= 4; // because (p) will point to next instruction + + BR_EXTERNAL_FOR { + // Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE for (;;) { - if (p >= lim) - return (SizeT)(p - data); + UInt32 v; + if Z7_UNLIKELY(p == lim) + return p; + v = GetUi32a(p); 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 (SizeT)(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); + if Z7_UNLIKELY(((v - 0x94000000) & 0xfc000000) == 0) + { + UInt32 c = BR_PC_GET >> 2; + BR_CONVERT_VAL(v, c) + v &= 0x03ffffff; + v |= 0x94000000; + SetUi32a(p - 4, v) + BR_NEXT_ITERATION + } + // v = rotlFixed(v, 8); v += (flag << 8) - 0x90; if Z7_UNLIKELY((v & ((mask << 8) + 0x9f)) == 0) + v -= 0x90000000; if Z7_UNLIKELY((v & 0x9f000000) == 0) + { + UInt32 z, c; + // v = rotrFixed(v, 8); + v += flag; if Z7_UNLIKELY(v & mask) continue; + z = (v & 0xffffffe0) | (v >> 26); + c = (BR_PC_GET >> (12 - 3)) & ~(UInt32)7; + BR_CONVERT_VAL(z, c) + v &= 0x1f; + v |= 0x90000000; + v |= z << 26; + v |= 0x00ffffe0 & ((z & (((flag << 1) - 1))) - flag); + SetUi32a(p - 4, v) + } } } } +Z7_BRANCH_FUNCS_IMP(ARM64) -SizeT ARMT_Convert(Byte *data, SizeT size, UInt32 ip, int encoding) +Z7_BRANCH_FUNC_MAIN(ARM) { - Byte *p; + // Byte *p = data; const Byte *lim; - size &= ~(size_t)1; - p = data; - lim = data + size - 4; - - if (encoding) + size &= ~(SizeT)3; + lim = p + size; + BR_PC_INIT + /* in ARM: branch offset is relative to the +2 instructions from current instruction. + (p) will point to next instruction */ + pc += 8 - 4; for (;;) { - UInt32 b1; for (;;) { - UInt32 b3; - if (p > lim) - return (SizeT)(p - data); - b1 = p[1]; - b3 = p[3]; - p += 2; - b1 ^= 8; - if ((b3 & b1) >= 0xF8) - break; + if Z7_UNLIKELY(p >= lim) { return p; } p += 4; if Z7_UNLIKELY(p[-1] == 0xeb) break; + if Z7_UNLIKELY(p >= lim) { return p; } p += 4; if Z7_UNLIKELY(p[-1] == 0xeb) 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 (SizeT)(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)); + UInt32 v = GetUi32a(p - 4); + UInt32 c = BR_PC_GET >> 2; + BR_CONVERT_VAL(v, c) + v &= 0x00ffffff; + v |= 0xeb000000; + SetUi32a(p - 4, v) } } } +Z7_BRANCH_FUNCS_IMP(ARM) -SizeT PPC_Convert(Byte *data, SizeT size, UInt32 ip, int encoding) +Z7_BRANCH_FUNC_MAIN(PPC) { - Byte *p; + // Byte *p = data; const Byte *lim; - size &= ~(size_t)3; - ip -= 4; - p = data; - lim = data + size; - + size &= ~(SizeT)3; + lim = p + size; + BR_PC_INIT + pc -= 4; // because (p) will point to next instruction + for (;;) { + UInt32 v; for (;;) { - if (p >= lim) - return (SizeT)(p - data); + if Z7_UNLIKELY(p == lim) + return p; + // v = GetBe32a(p); + v = *(UInt32 *)(void *)p; p += 4; - /* if ((v & 0xFC000003) == 0x48000001) */ - if ((p[-4] & 0xFC) == 0x48 && (p[-1] & 3) == 1) - break; + // if ((v & 0xfc000003) == 0x48000001) break; + // if ((p[-4] & 0xFC) == 0x48 && (p[-1] & 3) == 1) break; + if Z7_UNLIKELY( + ((v - Z7_CONV_BE_TO_NATIVE_CONST32(0x48000001)) + & Z7_CONV_BE_TO_NATIVE_CONST32(0xfc000003)) == 0) break; } { - UInt32 v = GetBe32(p - 4); - if (encoding) - v += ip + (UInt32)(p - data); - else - v -= ip + (UInt32)(p - data); - v &= 0x03FFFFFF; + v = Z7_CONV_NATIVE_TO_BE_32(v); + { + UInt32 c = BR_PC_GET; + BR_CONVERT_VAL(v, c) + } + v &= 0x03ffffff; v |= 0x48000000; - SetBe32(p - 4, v); + SetBe32a(p - 4, v) } } } +Z7_BRANCH_FUNCS_IMP(PPC) -SizeT SPARC_Convert(Byte *data, SizeT size, UInt32 ip, int encoding) +#ifdef Z7_CPU_FAST_ROTATE_SUPPORTED +#define BR_SPARC_USE_ROTATE +#endif + +Z7_BRANCH_FUNC_MAIN(SPARC) { - Byte *p; + // Byte *p = data; const Byte *lim; - size &= ~(size_t)3; - ip -= 4; - p = data; - lim = data + size; - + const UInt32 flag = (UInt32)1 << 22; + size &= ~(SizeT)3; + lim = p + size; + BR_PC_INIT + pc -= 4; // because (p) will point to next instruction for (;;) { + UInt32 v; for (;;) { - if (p >= lim) - return (SizeT)(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; + if Z7_UNLIKELY(p == lim) + return p; + /* // the code without GetBe32a(): + { const UInt32 v = GetUi16a(p) & 0xc0ff; p += 4; if (v == 0x40 || v == 0xc07f) break; } */ + v = GetBe32a(p); p += 4; - if ((p[-4] == 0x40 && (p[-3] & 0xC0) == 0) || - (p[-4] == 0x7F && (p[-3] >= 0xC0))) + #ifdef BR_SPARC_USE_ROTATE + v = rotlFixed(v, 2); + v += (flag << 2) - 1; + if Z7_UNLIKELY((v & (3 - (flag << 3))) == 0) + #else + v += (UInt32)5 << 29; + v ^= (UInt32)7 << 29; + v += flag; + if Z7_UNLIKELY((v & (0 - (flag << 1))) == 0) + #endif break; } { - UInt32 v = GetBe32(p - 4); + // UInt32 v = GetBe32a(p - 4); + #ifndef BR_SPARC_USE_ROTATE v <<= 2; - if (encoding) - v += ip + (UInt32)(p - data); - else - v -= ip + (UInt32)(p - data); - - v &= 0x01FFFFFF; - v -= (UInt32)1 << 24; - v ^= 0xFF000000; + #endif + { + UInt32 c = BR_PC_GET; + BR_CONVERT_VAL(v, c) + } + v &= (flag << 3) - 1; + #ifdef BR_SPARC_USE_ROTATE + v -= (flag << 2) - 1; + v = rotrFixed(v, 2); + #else + v -= (flag << 2); v >>= 2; - v |= 0x40000000; - SetBe32(p - 4, v); + v |= (UInt32)1 << 30; + #endif + SetBe32a(p - 4, v) } } } +Z7_BRANCH_FUNCS_IMP(SPARC) + + +Z7_BRANCH_FUNC_MAIN(ARMT) +{ + // Byte *p = data; + Byte *lim; + size &= ~(SizeT)1; + // if (size == 0) return p; + if (size <= 2) return p; + size -= 2; + lim = p + size; + BR_PC_INIT + /* in ARM: branch offset is relative to the +2 instructions from current instruction. + (p) will point to the +2 instructions from current instruction */ + // pc += 4 - 4; + // if (encoding) pc -= 0xf800 << 1; else pc += 0xf800 << 1; + // #define ARMT_TAIL_PROC { goto armt_tail; } + #define ARMT_TAIL_PROC { return p; } + + do + { + /* in MSVC 32-bit x86 compilers: + UInt32 version : it loads value from memory with movzx + Byte version : it loads value to 8-bit register (AL/CL) + movzx version is slightly faster in some cpus + */ + unsigned b1; + // Byte / unsigned + b1 = p[1]; + // optimized version to reduce one (p >= lim) check: + // unsigned a1 = p[1]; b1 = p[3]; p += 2; if Z7_LIKELY((b1 & (a1 ^ 8)) < 0xf8) + for (;;) + { + unsigned b3; // Byte / UInt32 + /* (Byte)(b3) normalization can use low byte computations in MSVC. + It gives smaller code, and no loss of speed in some compilers/cpus. + But new MSVC 32-bit x86 compilers use more slow load + from memory to low byte register in that case. + So we try to use full 32-bit computations for faster code. + */ + // if (p >= lim) { ARMT_TAIL_PROC } b3 = b1 + 8; b1 = p[3]; p += 2; if ((b3 & b1) >= 0xf8) break; + if Z7_UNLIKELY(p >= lim) { ARMT_TAIL_PROC } b3 = p[3]; p += 2; if Z7_UNLIKELY((b3 & (b1 ^ 8)) >= 0xf8) break; + if Z7_UNLIKELY(p >= lim) { ARMT_TAIL_PROC } b1 = p[3]; p += 2; if Z7_UNLIKELY((b1 & (b3 ^ 8)) >= 0xf8) break; + } + { + /* we can adjust pc for (0xf800) to rid of (& 0x7FF) operation. + But gcc/clang for arm64 can use bfi instruction for full code here */ + UInt32 v = + ((UInt32)GetUi16a(p - 2) << 11) | + ((UInt32)GetUi16a(p) & 0x7FF); + /* + UInt32 v = + ((UInt32)p[1 - 2] << 19) + + (((UInt32)p[1] & 0x7) << 8) + + (((UInt32)p[-2] << 11)) + + (p[0]); + */ + p += 2; + { + UInt32 c = BR_PC_GET >> 1; + BR_CONVERT_VAL(v, c) + } + SetUi16a(p - 4, (UInt16)(((v >> 11) & 0x7ff) | 0xf000)) + SetUi16a(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)); + */ + } + } + while (p < lim); + return p; + // armt_tail: + // if ((Byte)((lim[1] & 0xf8)) != 0xf0) { lim += 2; } return lim; + // return (Byte *)(lim + ((Byte)((lim[1] ^ 0xf0) & 0xf8) == 0 ? 0 : 2)); + // return (Byte *)(lim + (((lim[1] ^ ~0xfu) & ~7u) == 0 ? 0 : 2)); + // return (Byte *)(lim + 2 - (((((unsigned)lim[1] ^ 8) + 8) >> 7) & 2)); +} +Z7_BRANCH_FUNCS_IMP(ARMT) + + +// #define BR_IA64_NO_INLINE + +Z7_BRANCH_FUNC_MAIN(IA64) +{ + // Byte *p = data; + const Byte *lim; + size &= ~(SizeT)15; + lim = p + size; + pc -= 1 << 4; + pc >>= 4 - 1; + // pc -= 1 << 1; + + for (;;) + { + unsigned m; + for (;;) + { + if Z7_UNLIKELY(p == lim) + return p; + m = (unsigned)((UInt32)0x334b0000 >> (*p & 0x1e)); + p += 16; + pc += 1 << 1; + if (m &= 3) + break; + } + { + p += (ptrdiff_t)m * 5 - 20; // negative value is expected here. + do + { + const UInt32 t = + #if defined(MY_CPU_X86_OR_AMD64) + // we use 32-bit load here to reduce code size on x86: + GetUi32(p); + #else + GetUi16(p); + #endif + UInt32 z = GetUi32(p + 1) >> m; + p += 5; + if (((t >> m) & (0x70 << 1)) == 0 + && ((z - (0x5000000 << 1)) & (0xf000000 << 1)) == 0) + { + UInt32 v = (UInt32)((0x8fffff << 1) | 1) & z; + z ^= v; + #ifdef BR_IA64_NO_INLINE + v |= (v & ((UInt32)1 << (23 + 1))) >> 3; + { + UInt32 c = pc; + BR_CONVERT_VAL(v, c) + } + v &= (0x1fffff << 1) | 1; + #else + { + if (encoding) + { + // pc &= ~(0xc00000 << 1); // we just need to clear at least 2 bits + pc &= (0x1fffff << 1) | 1; + v += pc; + } + else + { + // pc |= 0xc00000 << 1; // we need to set at least 2 bits + pc |= ~(UInt32)((0x1fffff << 1) | 1); + v -= pc; + } + } + v &= ~(UInt32)(0x600000 << 1); + #endif + v += (0x700000 << 1); + v &= (0x8fffff << 1) | 1; + z |= v; + z <<= m; + SetUi32(p + 1 - 5, z) + } + m++; + } + while (m &= 3); // while (m < 4); + } + } +} +Z7_BRANCH_FUNCS_IMP(IA64) diff --git a/libraries/lzma/C/Bra.h b/libraries/lzma/C/Bra.h index 855e37a6b..a4ee568e2 100644 --- a/libraries/lzma/C/Bra.h +++ b/libraries/lzma/C/Bra.h @@ -1,64 +1,99 @@ /* Bra.h -- Branch converters for executables -2013-01-18 : Igor Pavlov : Public domain */ +2023-04-02 : Igor Pavlov : Public domain */ -#ifndef __BRA_H -#define __BRA_H +#ifndef ZIP7_INC_BRA_H +#define ZIP7_INC_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 +#define Z7_BRANCH_CONV_DEC(name) z7_BranchConv_ ## name ## _Dec +#define Z7_BRANCH_CONV_ENC(name) z7_BranchConv_ ## name ## _Enc +#define Z7_BRANCH_CONV_ST_DEC(name) z7_BranchConvSt_ ## name ## _Dec +#define Z7_BRANCH_CONV_ST_ENC(name) z7_BranchConvSt_ ## name ## _Enc - 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. +#define Z7_BRANCH_CONV_DECL(name) Byte * name(Byte *data, SizeT size, UInt32 pc) +#define Z7_BRANCH_CONV_ST_DECL(name) Byte * name(Byte *data, SizeT size, UInt32 pc, UInt32 *state) + +typedef Z7_BRANCH_CONV_DECL( (*z7_Func_BranchConv)); +typedef Z7_BRANCH_CONV_ST_DECL((*z7_Func_BranchConvSt)); + +#define Z7_BRANCH_CONV_ST_X86_STATE_INIT_VAL 0 +Z7_BRANCH_CONV_ST_DECL(Z7_BRANCH_CONV_ST_DEC(X86)); +Z7_BRANCH_CONV_ST_DECL(Z7_BRANCH_CONV_ST_ENC(X86)); + +#define Z7_BRANCH_FUNCS_DECL(name) \ +Z7_BRANCH_CONV_DECL(Z7_BRANCH_CONV_DEC(name)); \ +Z7_BRANCH_CONV_DECL(Z7_BRANCH_CONV_ENC(name)); + +Z7_BRANCH_FUNCS_DECL(ARM64) +Z7_BRANCH_FUNCS_DECL(ARM) +Z7_BRANCH_FUNCS_DECL(ARMT) +Z7_BRANCH_FUNCS_DECL(PPC) +Z7_BRANCH_FUNCS_DECL(SPARC) +Z7_BRANCH_FUNCS_DECL(IA64) + +/* +These functions convert data that contain CPU instructions. +Each such function converts relative addresses to absolute addresses in some +branch instructions: CALL (in all converters) and JUMP (X86 converter only). +Such conversion allows to increase compression ratio, if we compress that data. + +There are 2 types of converters: + Byte * Conv_RISC (Byte *data, SizeT size, UInt32 pc); + Byte * ConvSt_X86(Byte *data, SizeT size, UInt32 pc, UInt32 *state); +Each Converter supports 2 versions: one for encoding +and one for decoding (_Enc/_Dec postfixes in function name). + +In params: + data : data buffer + size : size of data + pc : current virtual Program Counter (Instruction Pinter) value +In/Out param: + state : pointer to state variable (for X86 converter only) + +Return: + The pointer to position in (data) buffer after last byte that was processed. + If the caller calls converter again, it must call it starting with that position. + But the caller is allowed to move data in buffer. so pointer to + current processed position also will be changed for next call. + Also the caller must increase internal (pc) value for next call. +Each converter has some characteristics: Endian, Alignment, LookAhead. Type Endian Alignment LookAhead - x86 little 1 4 + X86 little 1 4 ARMT little 2 2 ARM little 4 0 + ARM64 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. + (data) must be aligned for (Alignment). + processed size can be calculated as: + SizeT processed = Conv(data, size, pc) - data; + if (processed == 0) + it means that converter needs more data for processing. + If (size < Alignment + LookAhead) + then (processed == 0) is allowed. - 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; - } +Example code for conversion in loop: + UInt32 pc = 0; + size = 0; + for (;;) + { + size += Load_more_input_data(data + size); + SizeT processed = Conv(data, size, pc) - data; + if (processed == 0 && no_more_input_data_after_size) + break; // we stop convert loop + data += processed; + size -= processed; + pc += 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/libraries/lzma/C/Bra86.c b/libraries/lzma/C/Bra86.c index 10a0fbd16..d81f392ae 100644 --- a/libraries/lzma/C/Bra86.c +++ b/libraries/lzma/C/Bra86.c @@ -1,82 +1,187 @@ -/* Bra86.c -- Converter for x86 code (BCJ) -2021-02-09 : Igor Pavlov : Public domain */ +/* Bra86.c -- Branch converter for X86 code (BCJ) +2023-04-02 : Igor Pavlov : Public domain */ #include "Precomp.h" #include "Bra.h" +#include "CpuArch.h" -#define Test86MSByte(b) ((((b) + 1) & 0xFE) == 0) -SizeT x86_Convert(Byte *data, SizeT size, UInt32 ip, UInt32 *state, int encoding) +#if defined(MY_CPU_SIZEOF_POINTER) \ + && ( MY_CPU_SIZEOF_POINTER == 4 \ + || MY_CPU_SIZEOF_POINTER == 8) + #define BR_CONV_USE_OPT_PC_PTR +#endif + +#ifdef BR_CONV_USE_OPT_PC_PTR +#define BR_PC_INIT pc -= (UInt32)(SizeT)p; // (MY_uintptr_t) +#define BR_PC_GET (pc + (UInt32)(SizeT)p) +#else +#define BR_PC_INIT pc += (UInt32)size; +#define BR_PC_GET (pc - (UInt32)(SizeT)(lim - p)) +// #define BR_PC_INIT +// #define BR_PC_GET (pc + (UInt32)(SizeT)(p - data)) +#endif + +#define BR_CONVERT_VAL(v, c) if (encoding) v += c; else v -= c; +// #define BR_CONVERT_VAL(v, c) if (!encoding) c = (UInt32)0 - c; v += c; + +#define Z7_BRANCH_CONV_ST(name) z7_BranchConvSt_ ## name + +#define BR86_NEED_CONV_FOR_MS_BYTE(b) ((((b) + 1) & 0xfe) == 0) + +#ifdef MY_CPU_LE_UNALIGN + #define BR86_PREPARE_BCJ_SCAN const UInt32 v = GetUi32(p) ^ 0xe8e8e8e8; + #define BR86_IS_BCJ_BYTE(n) ((v & ((UInt32)0xfe << (n) * 8)) == 0) +#else + #define BR86_PREPARE_BCJ_SCAN + // bad for MSVC X86 (partial write to byte reg): + #define BR86_IS_BCJ_BYTE(n) ((p[n - 4] & 0xfe) == 0xe8) + // bad for old MSVC (partial write to byte reg): + // #define BR86_IS_BCJ_BYTE(n) (((*p ^ 0xe8) & 0xfe) == 0) +#endif + +static +Z7_FORCE_INLINE +Z7_ATTRIB_NO_VECTOR +Byte *Z7_BRANCH_CONV_ST(X86)(Byte *p, SizeT size, UInt32 pc, UInt32 *state, int encoding) { - SizeT pos = 0; - UInt32 mask = *state & 7; if (size < 5) - return 0; - size -= 4; - ip += 5; + return p; + { + // Byte *p = data; + const Byte *lim = p + size - 4; + unsigned mask = (unsigned)*state; // & 7; +#ifdef BR_CONV_USE_OPT_PC_PTR + /* if BR_CONV_USE_OPT_PC_PTR is defined: we need to adjust (pc) for (+4), + because call/jump offset is relative to the next instruction. + if BR_CONV_USE_OPT_PC_PTR is not defined : we don't need to adjust (pc) for (+4), + because BR_PC_GET uses (pc - (lim - p)), and lim was adjusted for (-4) before. + */ + pc += 4; +#endif + BR_PC_INIT + goto start; - for (;;) + for (;; mask |= 4) { - Byte *p = data + pos; - const Byte *limit = data + size; - for (; p < limit; p++) - if ((*p & 0xFE) == 0xE8) - break; - + // cont: mask |= 4; + start: + if (p >= lim) + goto fin; { - 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; - } - } + BR86_PREPARE_BCJ_SCAN + p += 4; + if (BR86_IS_BCJ_BYTE(0)) { goto m0; } mask >>= 1; + if (BR86_IS_BCJ_BYTE(1)) { goto m1; } mask >>= 1; + if (BR86_IS_BCJ_BYTE(2)) { goto m2; } mask = 0; + if (BR86_IS_BCJ_BYTE(3)) { goto a3; } } + goto main_loop; - if (Test86MSByte(p[4])) + m0: p--; + m1: p--; + m2: p--; + if (mask == 0) + goto a3; + if (p > lim) + goto fin_p; + + // if (((0x17u >> mask) & 1) == 0) + if (mask > 4 || mask == 3) { - 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) + mask >>= 1; + continue; // goto cont; + } + mask >>= 1; + if (BR86_NEED_CONV_FOR_MS_BYTE(p[mask])) + continue; // goto cont; + // if (!BR86_NEED_CONV_FOR_MS_BYTE(p[3])) continue; // goto cont; + { + UInt32 v = GetUi32(p); + UInt32 c; + v += (1 << 24); if (v & 0xfe000000) continue; // goto cont; + c = BR_PC_GET; + BR_CONVERT_VAL(v, c) { - unsigned sh = (mask & 6) << 2; - if (Test86MSByte((Byte)(v >> sh))) + mask <<= 3; + if (BR86_NEED_CONV_FOR_MS_BYTE(v >> mask)) { - v ^= (((UInt32)0x100 << sh) - 1); - if (encoding) - v += cur; - else - v -= cur; + v ^= (((UInt32)0x100 << mask) - 1); + #ifdef MY_CPU_X86 + // for X86 : we can recalculate (c) to reduce register pressure + c = BR_PC_GET; + #endif + BR_CONVERT_VAL(v, c) } mask = 0; } - p[1] = (Byte)v; - p[2] = (Byte)(v >> 8); - p[3] = (Byte)(v >> 16); - p[4] = (Byte)(0 - ((v >> 24) & 1)); + // v = (v & ((1 << 24) - 1)) - (v & (1 << 24)); + v &= (1 << 25) - 1; v -= (1 << 24); + SetUi32(p, v) + p += 4; + goto main_loop; } - else + + main_loop: + if (p >= lim) + goto fin; + for (;;) { - mask = (mask >> 1) | 4; - pos++; + BR86_PREPARE_BCJ_SCAN + p += 4; + if (BR86_IS_BCJ_BYTE(0)) { goto a0; } + if (BR86_IS_BCJ_BYTE(1)) { goto a1; } + if (BR86_IS_BCJ_BYTE(2)) { goto a2; } + if (BR86_IS_BCJ_BYTE(3)) { goto a3; } + if (p >= lim) + goto fin; + } + + a0: p--; + a1: p--; + a2: p--; + a3: + if (p > lim) + goto fin_p; + // if (!BR86_NEED_CONV_FOR_MS_BYTE(p[3])) continue; // goto cont; + { + UInt32 v = GetUi32(p); + UInt32 c; + v += (1 << 24); if (v & 0xfe000000) continue; // goto cont; + c = BR_PC_GET; + BR_CONVERT_VAL(v, c) + // v = (v & ((1 << 24) - 1)) - (v & (1 << 24)); + v &= (1 << 25) - 1; v -= (1 << 24); + SetUi32(p, v) + p += 4; + goto main_loop; } } + +fin_p: + p--; +fin: + // the following processing for tail is optional and can be commented + /* + lim += 4; + for (; p < lim; p++, mask >>= 1) + if ((*p & 0xfe) == 0xe8) + break; + */ + *state = (UInt32)mask; + return p; + } } + + +#define Z7_BRANCH_CONV_ST_FUNC_IMP(name, m, encoding) \ +Z7_NO_INLINE \ +Z7_ATTRIB_NO_VECTOR \ +Byte *m(name)(Byte *data, SizeT size, UInt32 pc, UInt32 *state) \ + { return Z7_BRANCH_CONV_ST(name)(data, size, pc, state, encoding); } + +Z7_BRANCH_CONV_ST_FUNC_IMP(X86, Z7_BRANCH_CONV_ST_DEC, 0) +#ifndef Z7_EXTRACT_ONLY +Z7_BRANCH_CONV_ST_FUNC_IMP(X86, Z7_BRANCH_CONV_ST_ENC, 1) +#endif diff --git a/libraries/lzma/C/BraIA64.c b/libraries/lzma/C/BraIA64.c index d1dbc62c5..9dfe3e289 100644 --- a/libraries/lzma/C/BraIA64.c +++ b/libraries/lzma/C/BraIA64.c @@ -1,53 +1,14 @@ /* BraIA64.c -- Converter for IA-64 code -2017-01-26 : Igor Pavlov : Public domain */ +2023-02-20 : Igor Pavlov : Public domain */ #include "Precomp.h" -#include "CpuArch.h" -#include "Bra.h" +// the code was moved to Bra.c -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; -} +#ifdef _MSC_VER +#pragma warning(disable : 4206) // nonstandard extension used : translation unit is empty +#endif + +#if defined(__clang__) +#pragma GCC diagnostic ignored "-Wempty-translation-unit" +#endif diff --git a/libraries/lzma/C/Compiler.h b/libraries/lzma/C/Compiler.h index a9816fa5a..185a52deb 100644 --- a/libraries/lzma/C/Compiler.h +++ b/libraries/lzma/C/Compiler.h @@ -1,12 +1,37 @@ -/* Compiler.h -2021-01-05 : Igor Pavlov : Public domain */ +/* Compiler.h : Compiler specific defines and pragmas +2023-04-02 : Igor Pavlov : Public domain */ -#ifndef __7Z_COMPILER_H -#define __7Z_COMPILER_H +#ifndef ZIP7_INC_COMPILER_H +#define ZIP7_INC_COMPILER_H + +#if defined(__clang__) +# define Z7_CLANG_VERSION (__clang_major__ * 10000 + __clang_minor__ * 100 + __clang_patchlevel__) +#endif +#if defined(__clang__) && defined(__apple_build_version__) +# define Z7_APPLE_CLANG_VERSION Z7_CLANG_VERSION +#elif defined(__clang__) +# define Z7_LLVM_CLANG_VERSION Z7_CLANG_VERSION +#elif defined(__GNUC__) +# define Z7_GCC_VERSION (__GNUC__ * 10000 + __GNUC_MINOR__ * 100 + __GNUC_PATCHLEVEL__) +#endif + +#ifdef _MSC_VER +#if !defined(__clang__) && !defined(__GNUC__) +#define Z7_MSC_VER_ORIGINAL _MSC_VER +#endif +#endif + +#if defined(__MINGW32__) || defined(__MINGW64__) +#define Z7_MINGW +#endif + +// #pragma GCC diagnostic ignored "-Wunknown-pragmas" + +#ifdef __clang__ +// padding size of '' with 4 bytes to alignment boundary +#pragma GCC diagnostic ignored "-Wpadded" +#endif - #ifdef __clang__ - #pragma clang diagnostic ignored "-Wunused-private-field" - #endif #ifdef _MSC_VER @@ -17,24 +42,115 @@ #pragma warning(disable : 4214) // nonstandard extension used : bit field types other than int #endif - #if _MSC_VER >= 1300 - #pragma warning(disable : 4996) // This function or variable may be unsafe - #else - #pragma warning(disable : 4511) // copy constructor could not be generated - #pragma warning(disable : 4512) // assignment operator could not be generated - #pragma warning(disable : 4514) // unreferenced inline function has been removed - #pragma warning(disable : 4702) // unreachable code - #pragma warning(disable : 4710) // not inlined - #pragma warning(disable : 4714) // function marked as __forceinline not inlined - #pragma warning(disable : 4786) // identifier was truncated to '255' characters in the debug information - #endif +#if defined(_MSC_VER) && _MSC_VER >= 1800 +#pragma warning(disable : 4464) // relative include path contains '..' +#endif - #ifdef __clang__ - #pragma clang diagnostic ignored "-Wdeprecated-declarations" - #pragma clang diagnostic ignored "-Wmicrosoft-exception-spec" - // #pragma clang diagnostic ignored "-Wreserved-id-macro" - #endif +// == 1200 : -O1 : for __forceinline +// >= 1900 : -O1 : for printf +#pragma warning(disable : 4710) // function not inlined +#if _MSC_VER < 1900 +// winnt.h: 'Int64ShllMod32' +#pragma warning(disable : 4514) // unreferenced inline function has been removed +#endif + +#if _MSC_VER < 1300 +// #pragma warning(disable : 4702) // unreachable code +// Bra.c : -O1: +#pragma warning(disable : 4714) // function marked as __forceinline not inlined +#endif + +/* +#if _MSC_VER > 1400 && _MSC_VER <= 1900 +// strcat: This function or variable may be unsafe +// sysinfoapi.h: kit10: GetVersion was declared deprecated +#pragma warning(disable : 4996) +#endif +*/ + +#if _MSC_VER > 1200 +// -Wall warnings + +#pragma warning(disable : 4711) // function selected for automatic inline expansion +#pragma warning(disable : 4820) // '2' bytes padding added after data member + +#if _MSC_VER >= 1400 && _MSC_VER < 1920 +// 1400: string.h: _DBG_MEMCPY_INLINE_ +// 1600 - 191x : smmintrin.h __cplusplus' +// is not defined as a preprocessor macro, replacing with '0' for '#if/#elif' +#pragma warning(disable : 4668) + +// 1400 - 1600 : WinDef.h : 'FARPROC' : +// 1900 - 191x : immintrin.h: _readfsbase_u32 +// no function prototype given : converting '()' to '(void)' +#pragma warning(disable : 4255) +#endif + +#if _MSC_VER >= 1914 +// Compiler will insert Spectre mitigation for memory load if /Qspectre switch specified +#pragma warning(disable : 5045) +#endif + +#endif // _MSC_VER > 1200 +#endif // _MSC_VER + + +#if defined(__clang__) && (__clang_major__ >= 4) + #define Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE \ + _Pragma("clang loop unroll(disable)") \ + _Pragma("clang loop vectorize(disable)") + #define Z7_ATTRIB_NO_VECTORIZE +#elif defined(__GNUC__) && (__GNUC__ >= 5) + #define Z7_ATTRIB_NO_VECTORIZE __attribute__((optimize("no-tree-vectorize"))) + // __attribute__((optimize("no-unroll-loops"))); + #define Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE +#elif defined(_MSC_VER) && (_MSC_VER >= 1920) + #define Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE \ + _Pragma("loop( no_vector )") + #define Z7_ATTRIB_NO_VECTORIZE +#else + #define Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE + #define Z7_ATTRIB_NO_VECTORIZE +#endif + +#if defined(MY_CPU_X86_OR_AMD64) && ( \ + defined(__clang__) && (__clang_major__ >= 4) \ + || defined(__GNUC__) && (__GNUC__ >= 5)) + #define Z7_ATTRIB_NO_SSE __attribute__((__target__("no-sse"))) +#else + #define Z7_ATTRIB_NO_SSE +#endif + +#define Z7_ATTRIB_NO_VECTOR \ + Z7_ATTRIB_NO_VECTORIZE \ + Z7_ATTRIB_NO_SSE + + +#if defined(__clang__) && (__clang_major__ >= 8) \ + || defined(__GNUC__) && (__GNUC__ >= 1000) \ + /* || defined(_MSC_VER) && (_MSC_VER >= 1920) */ + // GCC is not good for __builtin_expect() + #define Z7_LIKELY(x) (__builtin_expect((x), 1)) + #define Z7_UNLIKELY(x) (__builtin_expect((x), 0)) + // #define Z7_unlikely [[unlikely]] + // #define Z7_likely [[likely]] +#else + #define Z7_LIKELY(x) (x) + #define Z7_UNLIKELY(x) (x) + // #define Z7_likely +#endif + + +#if (defined(Z7_CLANG_VERSION) && (Z7_CLANG_VERSION >= 36000)) +#define Z7_DIAGNOSCTIC_IGNORE_BEGIN_RESERVED_MACRO_IDENTIFIER \ + _Pragma("GCC diagnostic push") \ + _Pragma("GCC diagnostic ignored \"-Wreserved-macro-identifier\"") +#define Z7_DIAGNOSCTIC_IGNORE_END_RESERVED_MACRO_IDENTIFIER \ + _Pragma("GCC diagnostic pop") +#else +#define Z7_DIAGNOSCTIC_IGNORE_BEGIN_RESERVED_MACRO_IDENTIFIER +#define Z7_DIAGNOSCTIC_IGNORE_END_RESERVED_MACRO_IDENTIFIER #endif #define UNUSED_VAR(x) (void)x; diff --git a/libraries/lzma/C/CpuArch.c b/libraries/lzma/C/CpuArch.c index fa9afe397..33f8a3ab4 100644 --- a/libraries/lzma/C/CpuArch.c +++ b/libraries/lzma/C/CpuArch.c @@ -1,187 +1,318 @@ /* CpuArch.c -- CPU specific code -2021-07-13 : Igor Pavlov : Public domain */ +2023-05-18 : Igor Pavlov : Public domain */ #include "Precomp.h" +// #include + #include "CpuArch.h" #ifdef MY_CPU_X86_OR_AMD64 -#if (defined(_MSC_VER) && !defined(MY_CPU_AMD64)) || defined(__GNUC__) -#define USE_ASM +#undef NEED_CHECK_FOR_CPUID +#if !defined(MY_CPU_AMD64) +#define NEED_CHECK_FOR_CPUID #endif -#if !defined(USE_ASM) && _MSC_VER >= 1500 -#include +/* + cpuid instruction supports (subFunction) parameter in ECX, + that is used only with some specific (function) parameter values. + But we always use only (subFunction==0). +*/ +/* + __cpuid(): MSVC and GCC/CLANG use same function/macro name + but parameters are different. + We use MSVC __cpuid() parameters style for our z7_x86_cpuid() function. +*/ + +#if defined(__GNUC__) /* && (__GNUC__ >= 10) */ \ + || defined(__clang__) /* && (__clang_major__ >= 10) */ + +/* there was some CLANG/GCC compilers that have issues with + rbx(ebx) handling in asm blocks in -fPIC mode (__PIC__ is defined). + compiler's contains the macro __cpuid() that is similar to our code. + The history of __cpuid() changes in CLANG/GCC: + GCC: + 2007: it preserved ebx for (__PIC__ && __i386__) + 2013: it preserved rbx and ebx for __PIC__ + 2014: it doesn't preserves rbx and ebx anymore + we suppose that (__GNUC__ >= 5) fixed that __PIC__ ebx/rbx problem. + CLANG: + 2014+: it preserves rbx, but only for 64-bit code. No __PIC__ check. + Why CLANG cares about 64-bit mode only, and doesn't care about ebx (in 32-bit)? + Do we need __PIC__ test for CLANG or we must care about rbx even if + __PIC__ is not defined? +*/ + +#define ASM_LN "\n" + +#if defined(MY_CPU_AMD64) && defined(__PIC__) \ + && ((defined (__GNUC__) && (__GNUC__ < 5)) || defined(__clang__)) + +#define x86_cpuid_MACRO(p, func) { \ + __asm__ __volatile__ ( \ + ASM_LN "mov %%rbx, %q1" \ + ASM_LN "cpuid" \ + ASM_LN "xchg %%rbx, %q1" \ + : "=a" ((p)[0]), "=&r" ((p)[1]), "=c" ((p)[2]), "=d" ((p)[3]) : "0" (func), "2"(0)); } + + /* "=&r" selects free register. It can select even rbx, if that register is free. + "=&D" for (RDI) also works, but the code can be larger with "=&D" + "2"(0) means (subFunction = 0), + 2 is (zero-based) index in the output constraint list "=c" (ECX). */ + +#elif defined(MY_CPU_X86) && defined(__PIC__) \ + && ((defined (__GNUC__) && (__GNUC__ < 5)) || defined(__clang__)) + +#define x86_cpuid_MACRO(p, func) { \ + __asm__ __volatile__ ( \ + ASM_LN "mov %%ebx, %k1" \ + ASM_LN "cpuid" \ + ASM_LN "xchg %%ebx, %k1" \ + : "=a" ((p)[0]), "=&r" ((p)[1]), "=c" ((p)[2]), "=d" ((p)[3]) : "0" (func), "2"(0)); } + +#else + +#define x86_cpuid_MACRO(p, func) { \ + __asm__ __volatile__ ( \ + ASM_LN "cpuid" \ + : "=a" ((p)[0]), "=b" ((p)[1]), "=c" ((p)[2]), "=d" ((p)[3]) : "0" (func), "2"(0)); } + #endif -#if defined(USE_ASM) && !defined(MY_CPU_AMD64) -static UInt32 CheckFlag(UInt32 flag) + +void Z7_FASTCALL z7_x86_cpuid(UInt32 p[4], UInt32 func) { - #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; + x86_cpuid_MACRO(p, func) } -#define CHECK_CPUID_IS_SUPPORTED if (CheckFlag(1 << 18) == 0 || CheckFlag(1 << 21) == 0) return False; + + +Z7_NO_INLINE +UInt32 Z7_FASTCALL z7_x86_cpuid_GetMaxFunc(void) +{ + #if defined(NEED_CHECK_FOR_CPUID) + #define EFALGS_CPUID_BIT 21 + UInt32 a; + __asm__ __volatile__ ( + ASM_LN "pushf" + ASM_LN "pushf" + ASM_LN "pop %0" + // ASM_LN "movl %0, %1" + // ASM_LN "xorl $0x200000, %0" + ASM_LN "btc %1, %0" + ASM_LN "push %0" + ASM_LN "popf" + ASM_LN "pushf" + ASM_LN "pop %0" + ASM_LN "xorl (%%esp), %0" + + ASM_LN "popf" + ASM_LN + : "=&r" (a) // "=a" + : "i" (EFALGS_CPUID_BIT) + ); + if ((a & (1 << EFALGS_CPUID_BIT)) == 0) + return 0; + #endif + { + UInt32 p[4]; + x86_cpuid_MACRO(p, 0) + return p[0]; + } +} + +#undef ASM_LN + +#elif !defined(_MSC_VER) + +/* +// for gcc/clang and other: we can try to use __cpuid macro: +#include +void Z7_FASTCALL z7_x86_cpuid(UInt32 p[4], UInt32 func) +{ + __cpuid(func, p[0], p[1], p[2], p[3]); +} +UInt32 Z7_FASTCALL z7_x86_cpuid_GetMaxFunc(void) +{ + return (UInt32)__get_cpuid_max(0, NULL); +} +*/ +// for unsupported cpuid: +void Z7_FASTCALL z7_x86_cpuid(UInt32 p[4], UInt32 func) +{ + UNUSED_VAR(func) + p[0] = p[1] = p[2] = p[3] = 0; +} +UInt32 Z7_FASTCALL z7_x86_cpuid_GetMaxFunc(void) +{ + return 0; +} + +#else // _MSC_VER + +#if !defined(MY_CPU_AMD64) + +UInt32 __declspec(naked) Z7_FASTCALL z7_x86_cpuid_GetMaxFunc(void) +{ + #if defined(NEED_CHECK_FOR_CPUID) + #define EFALGS_CPUID_BIT 21 + __asm pushfd + __asm pushfd + /* + __asm pop eax + // __asm mov edx, eax + __asm btc eax, EFALGS_CPUID_BIT + __asm push eax + */ + __asm btc dword ptr [esp], EFALGS_CPUID_BIT + __asm popfd + __asm pushfd + __asm pop eax + // __asm xor eax, edx + __asm xor eax, [esp] + // __asm push edx + __asm popfd + __asm and eax, (1 shl EFALGS_CPUID_BIT) + __asm jz end_func + #endif + __asm push ebx + __asm xor eax, eax // func + __asm xor ecx, ecx // subFunction (optional) for (func == 0) + __asm cpuid + __asm pop ebx + #if defined(NEED_CHECK_FOR_CPUID) + end_func: + #endif + __asm ret 0 +} + +void __declspec(naked) Z7_FASTCALL z7_x86_cpuid(UInt32 p[4], UInt32 func) +{ + UNUSED_VAR(p) + UNUSED_VAR(func) + __asm push ebx + __asm push edi + __asm mov edi, ecx // p + __asm mov eax, edx // func + __asm xor ecx, ecx // subfunction (optional) for (func == 0) + __asm cpuid + __asm mov [edi ], eax + __asm mov [edi + 4], ebx + __asm mov [edi + 8], ecx + __asm mov [edi + 12], edx + __asm pop edi + __asm pop ebx + __asm ret 0 +} + +#else // MY_CPU_AMD64 + + #if _MSC_VER >= 1600 + #include + #define MY_cpuidex __cpuidex + #else +/* + __cpuid (func == (0 or 7)) requires subfunction number in ECX. + MSDN: The __cpuid intrinsic clears the ECX register before calling the cpuid instruction. + __cpuid() in new MSVC clears ECX. + __cpuid() in old MSVC (14.00) x64 doesn't clear ECX + We still can use __cpuid for low (func) values that don't require ECX, + but __cpuid() in old MSVC will be incorrect for some func values: (func == 7). + So here we use the hack for old MSVC to send (subFunction) in ECX register to cpuid instruction, + where ECX value is first parameter for FASTCALL / NO_INLINE func, + So the caller of MY_cpuidex_HACK() sets ECX as subFunction, and + old MSVC for __cpuid() doesn't change ECX and cpuid instruction gets (subFunction) value. + +DON'T remove Z7_NO_INLINE and Z7_FASTCALL for MY_cpuidex_HACK(): !!! +*/ +static +Z7_NO_INLINE void Z7_FASTCALL MY_cpuidex_HACK(UInt32 subFunction, UInt32 func, int *CPUInfo) +{ + UNUSED_VAR(subFunction) + __cpuid(CPUInfo, func); +} + #define MY_cpuidex(info, func, func2) MY_cpuidex_HACK(func2, func, info) + #pragma message("======== MY_cpuidex_HACK WAS USED ========") + #endif // _MSC_VER >= 1600 + +#if !defined(MY_CPU_AMD64) +/* inlining for __cpuid() in MSVC x86 (32-bit) produces big ineffective code, + so we disable inlining here */ +Z7_NO_INLINE +#endif +void Z7_FASTCALL z7_x86_cpuid(UInt32 p[4], UInt32 func) +{ + MY_cpuidex((int *)p, (int)func, 0); +} + +Z7_NO_INLINE +UInt32 Z7_FASTCALL z7_x86_cpuid_GetMaxFunc(void) +{ + int a[4]; + MY_cpuidex(a, 0, 0); + return a[0]; +} + +#endif // MY_CPU_AMD64 +#endif // _MSC_VER + +#if defined(NEED_CHECK_FOR_CPUID) +#define CHECK_CPUID_IS_SUPPORTED { if (z7_x86_cpuid_GetMaxFunc() == 0) return 0; } #else #define CHECK_CPUID_IS_SUPPORTED #endif +#undef NEED_CHECK_FOR_CPUID -#ifndef USE_ASM - #ifdef _MSC_VER - #if _MSC_VER >= 1600 - #define MY__cpuidex __cpuidex - #else - -/* - __cpuid (function == 4) requires subfunction number in ECX. - MSDN: The __cpuid intrinsic clears the ECX register before calling the cpuid instruction. - __cpuid() in new MSVC clears ECX. - __cpuid() in old MSVC (14.00) doesn't clear ECX - We still can use __cpuid for low (function) values that don't require ECX, - but __cpuid() in old MSVC will be incorrect for some function values: (function == 4). - So here we use the hack for old MSVC to send (subFunction) in ECX register to cpuid instruction, - where ECX value is first parameter for FAST_CALL / NO_INLINE function, - So the caller of MY__cpuidex_HACK() sets ECX as subFunction, and - old MSVC for __cpuid() doesn't change ECX and cpuid instruction gets (subFunction) value. - - DON'T remove MY_NO_INLINE and MY_FAST_CALL for MY__cpuidex_HACK() !!! -*/ static -MY_NO_INLINE -void MY_FAST_CALL MY__cpuidex_HACK(UInt32 subFunction, int *CPUInfo, UInt32 function) -{ - UNUSED_VAR(subFunction); - __cpuid(CPUInfo, function); -} - - #define MY__cpuidex(info, func, func2) MY__cpuidex_HACK(func2, info, func) - #pragma message("======== MY__cpuidex_HACK WAS USED ========") - #endif - #else - #define MY__cpuidex(info, func, func2) __cpuid(info, func) - #pragma message("======== (INCORRECT ?) cpuid WAS USED ========") - #endif -#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), "c"(0) ) ; - - #endif - - #else - - int CPUInfo[4]; - - MY__cpuidex(CPUInfo, (int)function, 0); - - *a = (UInt32)CPUInfo[0]; - *b = (UInt32)CPUInfo[1]; - *c = (UInt32)CPUInfo[2]; - *d = (UInt32)CPUInfo[3]; - - #endif -} - -BoolInt x86cpuid_CheckAndRead(Cx86cpuid *p) +BoolInt x86cpuid_Func_1(UInt32 *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); + z7_x86_cpuid(p, 1); return True; } -static const UInt32 kVendors[][3] = +/* +static const UInt32 kVendors[][1] = { - { 0x756E6547, 0x49656E69, 0x6C65746E}, - { 0x68747541, 0x69746E65, 0x444D4163}, - { 0x746E6543, 0x48727561, 0x736C7561} + { 0x756E6547 }, // , 0x49656E69, 0x6C65746E }, + { 0x68747541 }, // , 0x69746E65, 0x444D4163 }, + { 0x746E6543 } // , 0x48727561, 0x736C7561 } }; +*/ + +/* +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 +}; +int x86cpuid_GetFirm(const Cx86cpuid *p); +#define x86cpuid_ver_GetFamily(ver) (((ver >> 16) & 0xff0) | ((ver >> 8) & 0xf)) +#define x86cpuid_ver_GetModel(ver) (((ver >> 12) & 0xf0) | ((ver >> 4) & 0xf)) +#define x86cpuid_ver_GetStepping(ver) (ver & 0xf) int x86cpuid_GetFirm(const Cx86cpuid *p) { unsigned i; - for (i = 0; i < sizeof(kVendors) / sizeof(kVendors[i]); i++) + for (i = 0; i < sizeof(kVendors) / sizeof(kVendors[0]); i++) { const UInt32 *v = kVendors[i]; - if (v[0] == p->vendor[0] && - v[1] == p->vendor[1] && - v[2] == p->vendor[2]) + if (v[0] == p->vendor[0] + // && v[1] == p->vendor[1] + // && v[2] == p->vendor[2] + ) return (int)i; } return -1; @@ -190,41 +321,55 @@ int x86cpuid_GetFirm(const Cx86cpuid *p) 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); + family = x86cpuid_ver_GetFamily(p.ver); + model = x86cpuid_ver_GetModel(p.ver); - switch (firm) + switch (x86cpuid_GetFirm(&p)) { 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 */ + // 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; + return False; // v23 : unknown processors are not In-Order } +*/ + +#ifdef _WIN32 +#include "7zWindows.h" +#endif #if !defined(MY_CPU_AMD64) && defined(_WIN32) -#include -static BoolInt CPU_Sys_Is_SSE_Supported() + +/* for legacy SSE ia32: there is no user-space cpu instruction to check + that OS supports SSE register storing/restoring on context switches. + So we need some OS-specific function to check that it's safe to use SSE registers. +*/ + +Z7_FORCE_INLINE +static BoolInt CPU_Sys_Is_SSE_Supported(void) { - OSVERSIONINFO vi; - vi.dwOSVersionInfoSize = sizeof(vi); - if (!GetVersionEx(&vi)) - return False; - return (vi.dwMajorVersion >= 5); +#ifdef _MSC_VER + #pragma warning(push) + #pragma warning(disable : 4996) // `GetVersion': was declared deprecated +#endif + /* low byte is major version of Windows + We suppose that any Windows version since + Windows2000 (major == 5) supports SSE registers */ + return (Byte)GetVersion() >= 5; +#if defined(_MSC_VER) + #pragma warning(pop) +#endif } #define CHECK_SYS_SSE_SUPPORT if (!CPU_Sys_Is_SSE_Supported()) return False; #else @@ -232,94 +377,300 @@ static BoolInt CPU_Sys_Is_SSE_Supported() #endif -static UInt32 X86_CPUID_ECX_Get_Flags() +#if !defined(MY_CPU_AMD64) + +BoolInt CPU_IsSupported_CMOV(void) { - Cx86cpuid p; - CHECK_SYS_SSE_SUPPORT - if (!x86cpuid_CheckAndRead(&p)) + UInt32 a[4]; + if (!x86cpuid_Func_1(&a[0])) return 0; - return p.c; + return (a[3] >> 15) & 1; } -BoolInt CPU_IsSupported_AES() +BoolInt CPU_IsSupported_SSE(void) { - return (X86_CPUID_ECX_Get_Flags() >> 25) & 1; -} - -BoolInt CPU_IsSupported_SSSE3() -{ - return (X86_CPUID_ECX_Get_Flags() >> 9) & 1; -} - -BoolInt CPU_IsSupported_SSE41() -{ - return (X86_CPUID_ECX_Get_Flags() >> 19) & 1; -} - -BoolInt CPU_IsSupported_SHA() -{ - Cx86cpuid p; + UInt32 a[4]; CHECK_SYS_SSE_SUPPORT - if (!x86cpuid_CheckAndRead(&p)) - return False; + if (!x86cpuid_Func_1(&a[0])) + return 0; + return (a[3] >> 25) & 1; +} - if (p.maxFunc < 7) +BoolInt CPU_IsSupported_SSE2(void) +{ + UInt32 a[4]; + CHECK_SYS_SSE_SUPPORT + if (!x86cpuid_Func_1(&a[0])) + return 0; + return (a[3] >> 26) & 1; +} + +#endif + + +static UInt32 x86cpuid_Func_1_ECX(void) +{ + UInt32 a[4]; + CHECK_SYS_SSE_SUPPORT + if (!x86cpuid_Func_1(&a[0])) + return 0; + return a[2]; +} + +BoolInt CPU_IsSupported_AES(void) +{ + return (x86cpuid_Func_1_ECX() >> 25) & 1; +} + +BoolInt CPU_IsSupported_SSSE3(void) +{ + return (x86cpuid_Func_1_ECX() >> 9) & 1; +} + +BoolInt CPU_IsSupported_SSE41(void) +{ + return (x86cpuid_Func_1_ECX() >> 19) & 1; +} + +BoolInt CPU_IsSupported_SHA(void) +{ + CHECK_SYS_SSE_SUPPORT + + if (z7_x86_cpuid_GetMaxFunc() < 7) return False; { - UInt32 d[4] = { 0 }; - MyCPUID(7, &d[0], &d[1], &d[2], &d[3]); + UInt32 d[4]; + z7_x86_cpuid(d, 7); return (d[1] >> 29) & 1; } } -// #include +/* +MSVC: _xgetbv() intrinsic is available since VS2010SP1. + MSVC also defines (_XCR_XFEATURE_ENABLED_MASK) macro in + that we can use or check. + For any 32-bit x86 we can use asm code in MSVC, + but MSVC asm code is huge after compilation. + So _xgetbv() is better -#ifdef _WIN32 -#include +ICC: _xgetbv() intrinsic is available (in what version of ICC?) + ICC defines (__GNUC___) and it supports gnu assembler + also ICC supports MASM style code with -use-msasm switch. + but ICC doesn't support __attribute__((__target__)) + +GCC/CLANG 9: + _xgetbv() is macro that works via __builtin_ia32_xgetbv() + and we need __attribute__((__target__("xsave")). + But with __target__("xsave") the function will be not + inlined to function that has no __target__("xsave") attribute. + If we want _xgetbv() call inlining, then we should use asm version + instead of calling _xgetbv(). + Note:intrinsic is broke before GCC 8.2: + https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85684 +*/ + +#if defined(__INTEL_COMPILER) && (__INTEL_COMPILER >= 1100) \ + || defined(_MSC_VER) && (_MSC_VER >= 1600) && (_MSC_FULL_VER >= 160040219) \ + || defined(__GNUC__) && (__GNUC__ >= 9) \ + || defined(__clang__) && (__clang_major__ >= 9) +// we define ATTRIB_XGETBV, if we want to use predefined _xgetbv() from compiler +#if defined(__INTEL_COMPILER) +#define ATTRIB_XGETBV +#elif defined(__GNUC__) || defined(__clang__) +// we don't define ATTRIB_XGETBV here, because asm version is better for inlining. +// #define ATTRIB_XGETBV __attribute__((__target__("xsave"))) +#else +#define ATTRIB_XGETBV +#endif #endif -BoolInt CPU_IsSupported_AVX2() -{ - Cx86cpuid p; - CHECK_SYS_SSE_SUPPORT +#if defined(ATTRIB_XGETBV) +#include +#endif + +// XFEATURE_ENABLED_MASK/XCR0 +#define MY_XCR_XFEATURE_ENABLED_MASK 0 + +#if defined(ATTRIB_XGETBV) +ATTRIB_XGETBV +#endif +static UInt64 x86_xgetbv_0(UInt32 num) +{ +#if defined(ATTRIB_XGETBV) + { + return + #if (defined(_MSC_VER)) + _xgetbv(num); + #else + __builtin_ia32_xgetbv( + #if !defined(__clang__) + (int) + #endif + num); + #endif + } + +#elif defined(__GNUC__) || defined(__clang__) || defined(__SUNPRO_CC) + + UInt32 a, d; + #if defined(__GNUC__) && (__GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 4)) + __asm__ + ( + "xgetbv" + : "=a"(a), "=d"(d) : "c"(num) : "cc" + ); + #else // is old gcc + __asm__ + ( + ".byte 0x0f, 0x01, 0xd0" "\n\t" + : "=a"(a), "=d"(d) : "c"(num) : "cc" + ); + #endif + return ((UInt64)d << 32) | a; + // return a; + +#elif defined(_MSC_VER) && !defined(MY_CPU_AMD64) + + UInt32 a, d; + __asm { + push eax + push edx + push ecx + mov ecx, num; + // xor ecx, ecx // = MY_XCR_XFEATURE_ENABLED_MASK + _emit 0x0f + _emit 0x01 + _emit 0xd0 + mov a, eax + mov d, edx + pop ecx + pop edx + pop eax + } + return ((UInt64)d << 32) | a; + // return a; + +#else // it's unknown compiler + // #error "Need xgetbv function" + UNUSED_VAR(num) + // for MSVC-X64 we could call external function from external file. + /* Actually we had checked OSXSAVE/AVX in cpuid before. + So it's expected that OS supports at least AVX and below. */ + // if (num != MY_XCR_XFEATURE_ENABLED_MASK) return 0; // if not XCR0 + return + // (1 << 0) | // x87 + (1 << 1) // SSE + | (1 << 2); // AVX + +#endif +} + +#ifdef _WIN32 +/* + Windows versions do not know about new ISA extensions that + can be introduced. But we still can use new extensions, + even if Windows doesn't report about supporting them, + But we can use new extensions, only if Windows knows about new ISA extension + that changes the number or size of registers: SSE, AVX/XSAVE, AVX512 + So it's enough to check + MY_PF_AVX_INSTRUCTIONS_AVAILABLE + instead of + MY_PF_AVX2_INSTRUCTIONS_AVAILABLE +*/ +#define MY_PF_XSAVE_ENABLED 17 +// #define MY_PF_SSSE3_INSTRUCTIONS_AVAILABLE 36 +// #define MY_PF_SSE4_1_INSTRUCTIONS_AVAILABLE 37 +// #define MY_PF_SSE4_2_INSTRUCTIONS_AVAILABLE 38 +// #define MY_PF_AVX_INSTRUCTIONS_AVAILABLE 39 +// #define MY_PF_AVX2_INSTRUCTIONS_AVAILABLE 40 +// #define MY_PF_AVX512F_INSTRUCTIONS_AVAILABLE 41 +#endif + +BoolInt CPU_IsSupported_AVX(void) +{ #ifdef _WIN32 - #define MY__PF_XSAVE_ENABLED 17 - if (!IsProcessorFeaturePresent(MY__PF_XSAVE_ENABLED)) + if (!IsProcessorFeaturePresent(MY_PF_XSAVE_ENABLED)) return False; + /* PF_AVX_INSTRUCTIONS_AVAILABLE probably is supported starting from + some latest Win10 revisions. But we need AVX in older Windows also. + So we don't use the following check: */ + /* + if (!IsProcessorFeaturePresent(MY_PF_AVX_INSTRUCTIONS_AVAILABLE)) + return False; + */ #endif - if (!x86cpuid_CheckAndRead(&p)) + /* + OS must use new special XSAVE/XRSTOR instructions to save + AVX registers when it required for context switching. + At OS statring: + OS sets CR4.OSXSAVE flag to signal the processor that OS supports the XSAVE extensions. + Also OS sets bitmask in XCR0 register that defines what + registers will be processed by XSAVE instruction: + XCR0.SSE[bit 0] - x87 registers and state + XCR0.SSE[bit 1] - SSE registers and state + XCR0.AVX[bit 2] - AVX registers and state + CR4.OSXSAVE is reflected to CPUID.1:ECX.OSXSAVE[bit 27]. + So we can read that bit in user-space. + XCR0 is available for reading in user-space by new XGETBV instruction. + */ + { + const UInt32 c = x86cpuid_Func_1_ECX(); + if (0 == (1 + & (c >> 28) // AVX instructions are supported by hardware + & (c >> 27))) // OSXSAVE bit: XSAVE and related instructions are enabled by OS. + return False; + } + + /* also we can check + CPUID.1:ECX.XSAVE [bit 26] : that shows that + XSAVE, XRESTOR, XSETBV, XGETBV instructions are supported by hardware. + But that check is redundant, because if OSXSAVE bit is set, then XSAVE is also set */ + + /* If OS have enabled XSAVE extension instructions (OSXSAVE == 1), + in most cases we expect that OS also will support storing/restoring + for AVX and SSE states at least. + But to be ensure for that we call user-space instruction + XGETBV(0) to get XCR0 value that contains bitmask that defines + what exact states(registers) OS have enabled for storing/restoring. + */ + + { + const UInt32 bm = (UInt32)x86_xgetbv_0(MY_XCR_XFEATURE_ENABLED_MASK); + // printf("\n=== XGetBV=%d\n", bm); + return 1 + & (bm >> 1) // SSE state is supported (set by OS) for storing/restoring + & (bm >> 2); // AVX state is supported (set by OS) for storing/restoring + } + // since Win7SP1: we can use GetEnabledXStateFeatures(); +} + + +BoolInt CPU_IsSupported_AVX2(void) +{ + if (!CPU_IsSupported_AVX()) return False; - if (p.maxFunc < 7) + if (z7_x86_cpuid_GetMaxFunc() < 7) return False; { - UInt32 d[4] = { 0 }; - MyCPUID(7, &d[0], &d[1], &d[2], &d[3]); + UInt32 d[4]; + z7_x86_cpuid(d, 7); // printf("\ncpuid(7): ebx=%8x ecx=%8x\n", d[1], d[2]); return 1 & (d[1] >> 5); // avx2 } } -BoolInt CPU_IsSupported_VAES_AVX2() +BoolInt CPU_IsSupported_VAES_AVX2(void) { - Cx86cpuid p; - CHECK_SYS_SSE_SUPPORT - - #ifdef _WIN32 - #define MY__PF_XSAVE_ENABLED 17 - if (!IsProcessorFeaturePresent(MY__PF_XSAVE_ENABLED)) + if (!CPU_IsSupported_AVX()) return False; - #endif - - if (!x86cpuid_CheckAndRead(&p)) - return False; - if (p.maxFunc < 7) + if (z7_x86_cpuid_GetMaxFunc() < 7) return False; { - UInt32 d[4] = { 0 }; - MyCPUID(7, &d[0], &d[1], &d[2], &d[3]); + UInt32 d[4]; + z7_x86_cpuid(d, 7); // printf("\ncpuid(7): ebx=%8x ecx=%8x\n", d[1], d[2]); return 1 & (d[1] >> 5) // avx2 @@ -328,20 +679,15 @@ BoolInt CPU_IsSupported_VAES_AVX2() } } -BoolInt CPU_IsSupported_PageGB() +BoolInt CPU_IsSupported_PageGB(void) { - Cx86cpuid cpuid; - if (!x86cpuid_CheckAndRead(&cpuid)) - return False; + CHECK_CPUID_IS_SUPPORTED { - UInt32 d[4] = { 0 }; - MyCPUID(0x80000000, &d[0], &d[1], &d[2], &d[3]); + UInt32 d[4]; + z7_x86_cpuid(d, 0x80000000); if (d[0] < 0x80000001) return False; - } - { - UInt32 d[4] = { 0 }; - MyCPUID(0x80000001, &d[0], &d[1], &d[2], &d[3]); + z7_x86_cpuid(d, 0x80000001); return (d[3] >> 26) & 1; } } @@ -351,11 +697,11 @@ BoolInt CPU_IsSupported_PageGB() #ifdef _WIN32 -#include +#include "7zWindows.h" -BoolInt CPU_IsSupported_CRC32() { return IsProcessorFeaturePresent(PF_ARM_V8_CRC32_INSTRUCTIONS_AVAILABLE) ? 1 : 0; } -BoolInt CPU_IsSupported_CRYPTO() { return IsProcessorFeaturePresent(PF_ARM_V8_CRYPTO_INSTRUCTIONS_AVAILABLE) ? 1 : 0; } -BoolInt CPU_IsSupported_NEON() { return IsProcessorFeaturePresent(PF_ARM_NEON_INSTRUCTIONS_AVAILABLE) ? 1 : 0; } +BoolInt CPU_IsSupported_CRC32(void) { return IsProcessorFeaturePresent(PF_ARM_V8_CRC32_INSTRUCTIONS_AVAILABLE) ? 1 : 0; } +BoolInt CPU_IsSupported_CRYPTO(void) { return IsProcessorFeaturePresent(PF_ARM_V8_CRYPTO_INSTRUCTIONS_AVAILABLE) ? 1 : 0; } +BoolInt CPU_IsSupported_NEON(void) { return IsProcessorFeaturePresent(PF_ARM_NEON_INSTRUCTIONS_AVAILABLE) ? 1 : 0; } #else @@ -378,28 +724,27 @@ static void Print_sysctlbyname(const char *name) } } */ +/* + Print_sysctlbyname("hw.pagesize"); + Print_sysctlbyname("machdep.cpu.brand_string"); +*/ -static BoolInt My_sysctlbyname_Get_BoolInt(const char *name) +static BoolInt z7_sysctlbyname_Get_BoolInt(const char *name) { UInt32 val = 0; - if (My_sysctlbyname_Get_UInt32(name, &val) == 0 && val == 1) + if (z7_sysctlbyname_Get_UInt32(name, &val) == 0 && val == 1) return 1; return 0; } - /* - Print_sysctlbyname("hw.pagesize"); - Print_sysctlbyname("machdep.cpu.brand_string"); - */ - BoolInt CPU_IsSupported_CRC32(void) { - return My_sysctlbyname_Get_BoolInt("hw.optional.armv8_crc32"); + return z7_sysctlbyname_Get_BoolInt("hw.optional.armv8_crc32"); } BoolInt CPU_IsSupported_NEON(void) { - return My_sysctlbyname_Get_BoolInt("hw.optional.neon"); + return z7_sysctlbyname_Get_BoolInt("hw.optional.neon"); } #ifdef MY_CPU_ARM64 @@ -461,15 +806,15 @@ MY_HWCAP_CHECK_FUNC (AES) #include -int My_sysctlbyname_Get(const char *name, void *buf, size_t *bufSize) +int z7_sysctlbyname_Get(const char *name, void *buf, size_t *bufSize) { return sysctlbyname(name, buf, bufSize, NULL, 0); } -int My_sysctlbyname_Get_UInt32(const char *name, UInt32 *val) +int z7_sysctlbyname_Get_UInt32(const char *name, UInt32 *val) { size_t bufSize = sizeof(*val); - int res = My_sysctlbyname_Get(name, val, &bufSize); + const int res = z7_sysctlbyname_Get(name, val, &bufSize); if (res == 0 && bufSize != sizeof(*val)) return EFAULT; return res; diff --git a/libraries/lzma/C/CpuArch.h b/libraries/lzma/C/CpuArch.h index 529d3a502..8e5d8a543 100644 --- a/libraries/lzma/C/CpuArch.h +++ b/libraries/lzma/C/CpuArch.h @@ -1,8 +1,8 @@ /* CpuArch.h -- CPU specific code -2021-07-13 : Igor Pavlov : Public domain */ +2023-04-02 : Igor Pavlov : Public domain */ -#ifndef __CPU_ARCH_H -#define __CPU_ARCH_H +#ifndef ZIP7_INC_CPU_ARCH_H +#define ZIP7_INC_CPU_ARCH_H #include "7zTypes.h" @@ -51,7 +51,13 @@ MY_CPU_64BIT means that processor can work with 64-bit registers. || defined(__AARCH64EB__) \ || defined(__aarch64__) #define MY_CPU_ARM64 - #define MY_CPU_NAME "arm64" + #ifdef __ILP32__ + #define MY_CPU_NAME "arm64-32" + #define MY_CPU_SIZEOF_POINTER 4 + #else + #define MY_CPU_NAME "arm64" + #define MY_CPU_SIZEOF_POINTER 8 + #endif #define MY_CPU_64BIT #endif @@ -68,8 +74,10 @@ MY_CPU_64BIT means that processor can work with 64-bit registers. #define MY_CPU_ARM #if defined(__thumb__) || defined(__THUMBEL__) || defined(_M_ARMT) + #define MY_CPU_ARMT #define MY_CPU_NAME "armt" #else + #define MY_CPU_ARM32 #define MY_CPU_NAME "arm" #endif /* #define MY_CPU_32BIT */ @@ -103,6 +111,8 @@ MY_CPU_64BIT means that processor can work with 64-bit registers. || defined(__PPC__) \ || defined(_POWER) +#define MY_CPU_PPC_OR_PPC64 + #if defined(__ppc64__) \ || defined(__powerpc64__) \ || defined(_LP64) \ @@ -123,12 +133,15 @@ MY_CPU_64BIT means that processor can work with 64-bit registers. #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 */ +#if defined(__riscv) \ + || defined(__riscv__) + #if __riscv_xlen == 32 + #define MY_CPU_NAME "riscv32" + #elif __riscv_xlen == 64 + #define MY_CPU_NAME "riscv64" + #else + #define MY_CPU_NAME "riscv" + #endif #endif @@ -194,6 +207,9 @@ MY_CPU_64BIT means that processor can work with 64-bit registers. #error Stop_Compiling_Bad_Endian #endif +#if !defined(MY_CPU_LE) && !defined(MY_CPU_BE) + #error Stop_Compiling_CPU_ENDIAN_must_be_detected_at_compile_time +#endif #if defined(MY_CPU_32BIT) && defined(MY_CPU_64BIT) #error Stop_Compiling_Bad_32_64_BIT @@ -250,6 +266,67 @@ MY_CPU_64BIT means that processor can work with 64-bit registers. +#ifdef __has_builtin + #define Z7_has_builtin(x) __has_builtin(x) +#else + #define Z7_has_builtin(x) 0 +#endif + + +#define Z7_BSWAP32_CONST(v) \ + ( (((UInt32)(v) << 24) ) \ + | (((UInt32)(v) << 8) & (UInt32)0xff0000) \ + | (((UInt32)(v) >> 8) & (UInt32)0xff00 ) \ + | (((UInt32)(v) >> 24) )) + + +#if defined(_MSC_VER) && (_MSC_VER >= 1300) + +#include + +/* Note: these macros will use bswap instruction (486), that is unsupported in 386 cpu */ + +#pragma intrinsic(_byteswap_ushort) +#pragma intrinsic(_byteswap_ulong) +#pragma intrinsic(_byteswap_uint64) + +#define Z7_BSWAP16(v) _byteswap_ushort(v) +#define Z7_BSWAP32(v) _byteswap_ulong (v) +#define Z7_BSWAP64(v) _byteswap_uint64(v) +#define Z7_CPU_FAST_BSWAP_SUPPORTED + +#elif (defined(__GNUC__) && (__GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 3))) \ + || (defined(__clang__) && Z7_has_builtin(__builtin_bswap16)) + +#define Z7_BSWAP16(v) __builtin_bswap16(v) +#define Z7_BSWAP32(v) __builtin_bswap32(v) +#define Z7_BSWAP64(v) __builtin_bswap64(v) +#define Z7_CPU_FAST_BSWAP_SUPPORTED + +#else + +#define Z7_BSWAP16(v) ((UInt16) \ + ( ((UInt32)(v) << 8) \ + | ((UInt32)(v) >> 8) \ + )) + +#define Z7_BSWAP32(v) Z7_BSWAP32_CONST(v) + +#define Z7_BSWAP64(v) \ + ( ( ( (UInt64)(v) ) << 8 * 7 ) \ + | ( ( (UInt64)(v) & ((UInt32)0xff << 8 * 1) ) << 8 * 5 ) \ + | ( ( (UInt64)(v) & ((UInt32)0xff << 8 * 2) ) << 8 * 3 ) \ + | ( ( (UInt64)(v) & ((UInt32)0xff << 8 * 3) ) << 8 * 1 ) \ + | ( ( (UInt64)(v) >> 8 * 1 ) & ((UInt32)0xff << 8 * 3) ) \ + | ( ( (UInt64)(v) >> 8 * 3 ) & ((UInt32)0xff << 8 * 2) ) \ + | ( ( (UInt64)(v) >> 8 * 5 ) & ((UInt32)0xff << 8 * 1) ) \ + | ( ( (UInt64)(v) >> 8 * 7 ) ) \ + ) + +#endif + + + #ifdef MY_CPU_LE #if defined(MY_CPU_X86_OR_AMD64) \ || defined(MY_CPU_ARM64) @@ -269,13 +346,11 @@ MY_CPU_64BIT means that processor can work with 64-bit registers. #define GetUi32(p) (*(const UInt32 *)(const void *)(p)) #ifdef MY_CPU_LE_UNALIGN_64 #define GetUi64(p) (*(const UInt64 *)(const void *)(p)) +#define SetUi64(p, v) { *(UInt64 *)(void *)(p) = (v); } #endif #define SetUi16(p, v) { *(UInt16 *)(void *)(p) = (v); } #define SetUi32(p, v) { *(UInt32 *)(void *)(p) = (v); } -#ifdef MY_CPU_LE_UNALIGN_64 -#define SetUi64(p, v) { *(UInt64 *)(void *)(p) = (v); } -#endif #else @@ -302,51 +377,26 @@ MY_CPU_64BIT means that processor can work with 64-bit registers. #endif -#ifndef MY_CPU_LE_UNALIGN_64 - +#ifndef GetUi64 #define GetUi64(p) (GetUi32(p) | ((UInt64)GetUi32(((const Byte *)(p)) + 4) << 32)) +#endif +#ifndef SetUi64 #define SetUi64(p, v) { Byte *_ppp2_ = (Byte *)(p); UInt64 _vvv2_ = (v); \ - SetUi32(_ppp2_ , (UInt32)_vvv2_); \ - SetUi32(_ppp2_ + 4, (UInt32)(_vvv2_ >> 32)); } - + SetUi32(_ppp2_ , (UInt32)_vvv2_) \ + SetUi32(_ppp2_ + 4, (UInt32)(_vvv2_ >> 32)) } #endif +#if defined(MY_CPU_LE_UNALIGN) && defined(Z7_CPU_FAST_BSWAP_SUPPORTED) +#define GetBe32(p) Z7_BSWAP32 (*(const UInt32 *)(const void *)(p)) +#define SetBe32(p, v) { (*(UInt32 *)(void *)(p)) = Z7_BSWAP32(v); } -#ifdef __has_builtin - #define MY__has_builtin(x) __has_builtin(x) -#else - #define MY__has_builtin(x) 0 +#if defined(MY_CPU_LE_UNALIGN_64) +#define GetBe64(p) Z7_BSWAP64 (*(const UInt64 *)(const void *)(p)) #endif -#if defined(MY_CPU_LE_UNALIGN) && /* defined(_WIN64) && */ defined(_MSC_VER) && (_MSC_VER >= 1300) - -/* Note: we use bswap instruction, that is unsupported in 386 cpu */ - -#include - -#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 void *)(p)) -#define GetBe64(p) _byteswap_uint64(*(const UInt64 *)(const void *)(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 void *)(p)) */ -#define GetBe32(p) __builtin_bswap32(*(const UInt32 *)(const void *)(p)) -#define GetBe64(p) __builtin_bswap64(*(const UInt64 *)(const void *)(p)) - -#define SetBe32(p, v) (*(UInt32 *)(void *)(p)) = __builtin_bswap32(v) - #else #define GetBe32(p) ( \ @@ -355,8 +405,6 @@ MY_CPU_64BIT means that processor can work with 64-bit registers. ((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); \ @@ -365,50 +413,83 @@ MY_CPU_64BIT means that processor can work with 64-bit registers. #endif +#ifndef GetBe64 +#define GetBe64(p) (((UInt64)GetBe32(p) << 32) | GetBe32(((const Byte *)(p)) + 4)) +#endif #ifndef GetBe16 - #define GetBe16(p) ( (UInt16) ( \ ((UInt16)((const Byte *)(p))[0] << 8) | \ ((const Byte *)(p))[1] )) - #endif +#if defined(MY_CPU_BE) +#define Z7_CONV_BE_TO_NATIVE_CONST32(v) (v) +#define Z7_CONV_LE_TO_NATIVE_CONST32(v) Z7_BSWAP32_CONST(v) +#define Z7_CONV_NATIVE_TO_BE_32(v) (v) +#elif defined(MY_CPU_LE) +#define Z7_CONV_BE_TO_NATIVE_CONST32(v) Z7_BSWAP32_CONST(v) +#define Z7_CONV_LE_TO_NATIVE_CONST32(v) (v) +#define Z7_CONV_NATIVE_TO_BE_32(v) Z7_BSWAP32(v) +#else +#error Stop_Compiling_Unknown_Endian_CONV +#endif + + +#if defined(MY_CPU_BE) + +#define GetBe32a(p) (*(const UInt32 *)(const void *)(p)) +#define GetBe16a(p) (*(const UInt16 *)(const void *)(p)) +#define SetBe32a(p, v) { *(UInt32 *)(void *)(p) = (v); } +#define SetBe16a(p, v) { *(UInt16 *)(void *)(p) = (v); } + +#define GetUi32a(p) GetUi32(p) +#define GetUi16a(p) GetUi16(p) +#define SetUi32a(p, v) SetUi32(p, v) +#define SetUi16a(p, v) SetUi16(p, v) + +#elif defined(MY_CPU_LE) + +#define GetUi32a(p) (*(const UInt32 *)(const void *)(p)) +#define GetUi16a(p) (*(const UInt16 *)(const void *)(p)) +#define SetUi32a(p, v) { *(UInt32 *)(void *)(p) = (v); } +#define SetUi16a(p, v) { *(UInt16 *)(void *)(p) = (v); } + +#define GetBe32a(p) GetBe32(p) +#define GetBe16a(p) GetBe16(p) +#define SetBe32a(p, v) SetBe32(p, v) +#define SetBe16a(p, v) SetBe16(p, v) + +#else +#error Stop_Compiling_Unknown_Endian_CPU_a +#endif + + +#if defined(MY_CPU_X86_OR_AMD64) \ + || defined(MY_CPU_ARM_OR_ARM64) \ + || defined(MY_CPU_PPC_OR_PPC64) + #define Z7_CPU_FAST_ROTATE_SUPPORTED +#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(void); +void Z7_FASTCALL z7_x86_cpuid(UInt32 a[4], UInt32 function); +UInt32 Z7_FASTCALL z7_x86_cpuid_GetMaxFunc(void); +#if defined(MY_CPU_AMD64) +#define Z7_IF_X86_CPUID_SUPPORTED +#else +#define Z7_IF_X86_CPUID_SUPPORTED if (z7_x86_cpuid_GetMaxFunc()) +#endif BoolInt CPU_IsSupported_AES(void); +BoolInt CPU_IsSupported_AVX(void); BoolInt CPU_IsSupported_AVX2(void); BoolInt CPU_IsSupported_VAES_AVX2(void); +BoolInt CPU_IsSupported_CMOV(void); +BoolInt CPU_IsSupported_SSE(void); +BoolInt CPU_IsSupported_SSE2(void); BoolInt CPU_IsSupported_SSSE3(void); BoolInt CPU_IsSupported_SSE41(void); BoolInt CPU_IsSupported_SHA(void); @@ -433,8 +514,8 @@ BoolInt CPU_IsSupported_AES(void); #endif #if defined(__APPLE__) -int My_sysctlbyname_Get(const char *name, void *buf, size_t *bufSize); -int My_sysctlbyname_Get_UInt32(const char *name, UInt32 *val); +int z7_sysctlbyname_Get(const char *name, void *buf, size_t *bufSize); +int z7_sysctlbyname_Get_UInt32(const char *name, UInt32 *val); #endif EXTERN_C_END diff --git a/libraries/lzma/C/Delta.h b/libraries/lzma/C/Delta.h index 2fa54ad67..706095417 100644 --- a/libraries/lzma/C/Delta.h +++ b/libraries/lzma/C/Delta.h @@ -1,8 +1,8 @@ /* Delta.h -- Delta converter -2013-01-18 : Igor Pavlov : Public domain */ +2023-03-03 : Igor Pavlov : Public domain */ -#ifndef __DELTA_H -#define __DELTA_H +#ifndef ZIP7_INC_DELTA_H +#define ZIP7_INC_DELTA_H #include "7zTypes.h" diff --git a/libraries/lzma/C/DllSecur.c b/libraries/lzma/C/DllSecur.c new file mode 100644 index 000000000..02a0f977e --- /dev/null +++ b/libraries/lzma/C/DllSecur.c @@ -0,0 +1,111 @@ +/* DllSecur.c -- DLL loading security +2023-04-02 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#ifdef _WIN32 + +#include "7zWindows.h" + +#include "DllSecur.h" + +#ifndef UNDER_CE + +#if (defined(__GNUC__) && (__GNUC__ >= 8)) || defined(__clang__) + // #pragma GCC diagnostic ignored "-Wcast-function-type" +#endif + +#if defined(__clang__) || defined(__GNUC__) +typedef void (*Z7_voidFunction)(void); +#define MY_CAST_FUNC (Z7_voidFunction) +#elif defined(_MSC_VER) && _MSC_VER > 1920 +#define MY_CAST_FUNC (void *) +// #pragma warning(disable : 4191) // 'type cast': unsafe conversion from 'FARPROC' to 'void (__cdecl *)()' +#else +#define MY_CAST_FUNC +#endif + +typedef BOOL (WINAPI *Func_SetDefaultDllDirectories)(DWORD DirectoryFlags); + +#define MY_LOAD_LIBRARY_SEARCH_USER_DIRS 0x400 +#define MY_LOAD_LIBRARY_SEARCH_SYSTEM32 0x800 + +#define DELIM "\0" + +static const char * const g_Dlls = + "userenv" + DELIM "setupapi" + DELIM "apphelp" + DELIM "propsys" + DELIM "dwmapi" + DELIM "cryptbase" + DELIM "oleacc" + DELIM "clbcatq" + DELIM "version" + #ifndef _CONSOLE + DELIM "uxtheme" + #endif + DELIM; + +#endif + +#ifdef __clang__ + #pragma GCC diagnostic ignored "-Wdeprecated-declarations" +#endif +#if defined (_MSC_VER) && _MSC_VER >= 1900 +// sysinfoapi.h: kit10: GetVersion was declared deprecated +#pragma warning(disable : 4996) +#endif + +#define IF_NON_VISTA_SET_DLL_DIRS_AND_RETURN \ + if ((UInt16)GetVersion() != 6) { \ + const \ + Func_SetDefaultDllDirectories setDllDirs = \ + (Func_SetDefaultDllDirectories) MY_CAST_FUNC GetProcAddress(GetModuleHandle(TEXT("kernel32.dll")), \ + "SetDefaultDllDirectories"); \ + if (setDllDirs) if (setDllDirs(MY_LOAD_LIBRARY_SEARCH_SYSTEM32 | MY_LOAD_LIBRARY_SEARCH_USER_DIRS)) return; } + +void My_SetDefaultDllDirectories(void) +{ + #ifndef UNDER_CE + IF_NON_VISTA_SET_DLL_DIRS_AND_RETURN + #endif +} + + +void LoadSecurityDlls(void) +{ + #ifndef UNDER_CE + // at Vista (ver 6.0) : CoCreateInstance(CLSID_ShellLink, ...) doesn't work after SetDefaultDllDirectories() : Check it ??? + IF_NON_VISTA_SET_DLL_DIRS_AND_RETURN + { + wchar_t buf[MAX_PATH + 100]; + const char *dll; + unsigned pos = GetSystemDirectoryW(buf, MAX_PATH + 2); + if (pos == 0 || pos > MAX_PATH) + return; + if (buf[pos - 1] != '\\') + buf[pos++] = '\\'; + for (dll = g_Dlls; *dll != 0;) + { + wchar_t *dest = &buf[pos]; + for (;;) + { + const char c = *dll++; + if (c == 0) + break; + *dest++ = (Byte)c; + } + dest[0] = '.'; + dest[1] = 'd'; + dest[2] = 'l'; + dest[3] = 'l'; + dest[4] = 0; + // lstrcatW(buf, L".dll"); + LoadLibraryExW(buf, NULL, LOAD_WITH_ALTERED_SEARCH_PATH); + } + } + #endif +} + +#endif // _WIN32 diff --git a/libraries/lzma/C/DllSecur.h b/libraries/lzma/C/DllSecur.h new file mode 100644 index 000000000..9fa415382 --- /dev/null +++ b/libraries/lzma/C/DllSecur.h @@ -0,0 +1,20 @@ +/* DllSecur.h -- DLL loading for security +2023-03-03 : Igor Pavlov : Public domain */ + +#ifndef ZIP7_INC_DLL_SECUR_H +#define ZIP7_INC_DLL_SECUR_H + +#include "7zTypes.h" + +EXTERN_C_BEGIN + +#ifdef _WIN32 + +void My_SetDefaultDllDirectories(void); +void LoadSecurityDlls(void); + +#endif + +EXTERN_C_END + +#endif diff --git a/libraries/lzma/C/LzFind.c b/libraries/lzma/C/LzFind.c index 1b73c2848..0fbd5aae5 100644 --- a/libraries/lzma/C/LzFind.c +++ b/libraries/lzma/C/LzFind.c @@ -1,5 +1,5 @@ /* LzFind.c -- Match finder for LZ algorithms -2021-11-29 : Igor Pavlov : Public domain */ +2023-03-14 : Igor Pavlov : Public domain */ #include "Precomp.h" @@ -17,7 +17,7 @@ #define kEmptyHashValue 0 #define kMaxValForNormalize ((UInt32)0) -// #define kMaxValForNormalize ((UInt32)(1 << 20) + 0xFFF) // for debug +// #define kMaxValForNormalize ((UInt32)(1 << 20) + 0xfff) // for debug // #define kNormalizeAlign (1 << 7) // alignment for speculated accesses @@ -67,10 +67,10 @@ static void LzInWindow_Free(CMatchFinder *p, ISzAllocPtr alloc) { - if (!p->directInput) + // if (!p->directInput) { - ISzAlloc_Free(alloc, p->bufferBase); - p->bufferBase = NULL; + ISzAlloc_Free(alloc, p->bufBase); + p->bufBase = NULL; } } @@ -79,7 +79,7 @@ static int LzInWindow_Create2(CMatchFinder *p, UInt32 blockSize, ISzAllocPtr all { if (blockSize == 0) return 0; - if (!p->bufferBase || p->blockSize != blockSize) + if (!p->bufBase || p->blockSize != blockSize) { // size_t blockSizeT; LzInWindow_Free(p, alloc); @@ -101,11 +101,11 @@ static int LzInWindow_Create2(CMatchFinder *p, UInt32 blockSize, ISzAllocPtr all #endif */ - p->bufferBase = (Byte *)ISzAlloc_Alloc(alloc, blockSize); - // printf("\nbufferBase = %p\n", p->bufferBase); + p->bufBase = (Byte *)ISzAlloc_Alloc(alloc, blockSize); + // printf("\nbufferBase = %p\n", p->bufBase); // return 0; // for debug } - return (p->bufferBase != NULL); + return (p->bufBase != NULL); } static const Byte *MatchFinder_GetPointerToCurrentPos(CMatchFinder *p) { return p->buffer; } @@ -113,7 +113,7 @@ static const Byte *MatchFinder_GetPointerToCurrentPos(CMatchFinder *p) { return static UInt32 MatchFinder_GetNumAvailableBytes(CMatchFinder *p) { return GET_AVAIL_BYTES(p); } -MY_NO_INLINE +Z7_NO_INLINE static void MatchFinder_ReadBlock(CMatchFinder *p) { if (p->streamEndWasReached || p->result != SZ_OK) @@ -127,8 +127,8 @@ static void MatchFinder_ReadBlock(CMatchFinder *p) UInt32 curSize = 0xFFFFFFFF - GET_AVAIL_BYTES(p); if (curSize > p->directInputRem) curSize = (UInt32)p->directInputRem; - p->directInputRem -= curSize; p->streamPos += curSize; + p->directInputRem -= curSize; if (p->directInputRem == 0) p->streamEndWasReached = 1; return; @@ -136,8 +136,8 @@ static void MatchFinder_ReadBlock(CMatchFinder *p) for (;;) { - Byte *dest = p->buffer + GET_AVAIL_BYTES(p); - size_t size = (size_t)(p->bufferBase + p->blockSize - dest); + const Byte *dest = p->buffer + GET_AVAIL_BYTES(p); + size_t size = (size_t)(p->bufBase + p->blockSize - dest); if (size == 0) { /* we call ReadBlock() after NeedMove() and MoveBlock(). @@ -153,7 +153,14 @@ static void MatchFinder_ReadBlock(CMatchFinder *p) // #define kRead 3 // if (size > kRead) size = kRead; // for debug - p->result = ISeqInStream_Read(p->stream, dest, &size); + /* + // we need cast (Byte *)dest. + #ifdef __clang__ + #pragma GCC diagnostic ignored "-Wcast-qual" + #endif + */ + p->result = ISeqInStream_Read(p->stream, + p->bufBase + (dest - p->bufBase), &size); if (p->result != SZ_OK) return; if (size == 0) @@ -173,14 +180,14 @@ static void MatchFinder_ReadBlock(CMatchFinder *p) -MY_NO_INLINE +Z7_NO_INLINE void MatchFinder_MoveBlock(CMatchFinder *p) { - const size_t offset = (size_t)(p->buffer - p->bufferBase) - p->keepSizeBefore; + const size_t offset = (size_t)(p->buffer - p->bufBase) - p->keepSizeBefore; const size_t keepBefore = (offset & (kBlockMoveAlign - 1)) + p->keepSizeBefore; - p->buffer = p->bufferBase + keepBefore; - memmove(p->bufferBase, - p->bufferBase + (offset & ~((size_t)kBlockMoveAlign - 1)), + p->buffer = p->bufBase + keepBefore; + memmove(p->bufBase, + p->bufBase + (offset & ~((size_t)kBlockMoveAlign - 1)), keepBefore + (size_t)GET_AVAIL_BYTES(p)); } @@ -198,7 +205,7 @@ int MatchFinder_NeedMove(CMatchFinder *p) return 0; if (p->streamEndWasReached || p->result != SZ_OK) return 0; - return ((size_t)(p->bufferBase + p->blockSize - p->buffer) <= p->keepSizeAfter); + return ((size_t)(p->bufBase + p->blockSize - p->buffer) <= p->keepSizeAfter); } void MatchFinder_ReadIfRequired(CMatchFinder *p) @@ -214,6 +221,8 @@ static void MatchFinder_SetDefaultSettings(CMatchFinder *p) p->cutValue = 32; p->btMode = 1; p->numHashBytes = 4; + p->numHashBytes_Min = 2; + p->numHashOutBits = 0; p->bigHash = 0; } @@ -222,8 +231,10 @@ static void MatchFinder_SetDefaultSettings(CMatchFinder *p) void MatchFinder_Construct(CMatchFinder *p) { unsigned i; - p->bufferBase = NULL; + p->buffer = NULL; + p->bufBase = NULL; p->directInput = 0; + p->stream = NULL; p->hash = NULL; p->expectedDataSize = (UInt64)(Int64)-1; MatchFinder_SetDefaultSettings(p); @@ -238,6 +249,8 @@ void MatchFinder_Construct(CMatchFinder *p) } } +#undef kCrcPoly + static void MatchFinder_FreeThisClassMemory(CMatchFinder *p, ISzAllocPtr alloc) { ISzAlloc_Free(alloc, p->hash); @@ -252,7 +265,7 @@ void MatchFinder_Free(CMatchFinder *p, ISzAllocPtr alloc) static CLzRef* AllocRefs(size_t num, ISzAllocPtr alloc) { - size_t sizeInBytes = (size_t)num * sizeof(CLzRef); + const size_t sizeInBytes = (size_t)num * sizeof(CLzRef); if (sizeInBytes / sizeof(CLzRef) != num) return NULL; return (CLzRef *)ISzAlloc_Alloc(alloc, sizeInBytes); @@ -298,6 +311,62 @@ static UInt32 GetBlockSize(CMatchFinder *p, UInt32 historySize) } +// input is historySize +static UInt32 MatchFinder_GetHashMask2(CMatchFinder *p, UInt32 hs) +{ + if (p->numHashBytes == 2) + return (1 << 16) - 1; + if (hs != 0) + hs--; + hs |= (hs >> 1); + hs |= (hs >> 2); + hs |= (hs >> 4); + hs |= (hs >> 8); + // we propagated 16 bits in (hs). Low 16 bits must be set later + if (hs >= (1 << 24)) + { + if (p->numHashBytes == 3) + hs = (1 << 24) - 1; + /* if (bigHash) mode, GetHeads4b() in LzFindMt.c needs (hs >= ((1 << 24) - 1))) */ + } + // (hash_size >= (1 << 16)) : Required for (numHashBytes > 2) + hs |= (1 << 16) - 1; /* don't change it! */ + // bt5: we adjust the size with recommended minimum size + if (p->numHashBytes >= 5) + hs |= (256 << kLzHash_CrcShift_2) - 1; + return hs; +} + +// input is historySize +static UInt32 MatchFinder_GetHashMask(CMatchFinder *p, UInt32 hs) +{ + if (p->numHashBytes == 2) + return (1 << 16) - 1; + if (hs != 0) + hs--; + hs |= (hs >> 1); + hs |= (hs >> 2); + hs |= (hs >> 4); + hs |= (hs >> 8); + // we propagated 16 bits in (hs). Low 16 bits must be set later + hs >>= 1; + if (hs >= (1 << 24)) + { + if (p->numHashBytes == 3) + hs = (1 << 24) - 1; + else + hs >>= 1; + /* if (bigHash) mode, GetHeads4b() in LzFindMt.c needs (hs >= ((1 << 24) - 1))) */ + } + // (hash_size >= (1 << 16)) : Required for (numHashBytes > 2) + hs |= (1 << 16) - 1; /* don't change it! */ + // bt5: we adjust the size with recommended minimum size + if (p->numHashBytes >= 5) + hs |= (256 << kLzHash_CrcShift_2) - 1; + return hs; +} + + int MatchFinder_Create(CMatchFinder *p, UInt32 historySize, UInt32 keepAddBufferBefore, UInt32 matchMaxLen, UInt32 keepAddBufferAfter, ISzAllocPtr alloc) @@ -318,78 +387,91 @@ int MatchFinder_Create(CMatchFinder *p, UInt32 historySize, p->blockSize = 0; if (p->directInput || LzInWindow_Create2(p, GetBlockSize(p, historySize), alloc)) { - const UInt32 newCyclicBufferSize = historySize + 1; // do not change it - UInt32 hs; - p->matchMaxLen = matchMaxLen; + size_t hashSizeSum; { - // UInt32 hs4; - p->fixedHashSize = 0; - hs = (1 << 16) - 1; - if (p->numHashBytes != 2) + UInt32 hs; + UInt32 hsCur; + + if (p->numHashOutBits != 0) { - hs = historySize; - if (hs > p->expectedDataSize) - hs = (UInt32)p->expectedDataSize; - if (hs != 0) - hs--; - hs |= (hs >> 1); - hs |= (hs >> 2); - hs |= (hs >> 4); - hs |= (hs >> 8); - // we propagated 16 bits in (hs). Low 16 bits must be set later - hs >>= 1; - if (hs >= (1 << 24)) - { - if (p->numHashBytes == 3) - hs = (1 << 24) - 1; - else - hs >>= 1; - /* if (bigHash) mode, GetHeads4b() in LzFindMt.c needs (hs >= ((1 << 24) - 1))) */ - } - - // hs = ((UInt32)1 << 25) - 1; // for test - + unsigned numBits = p->numHashOutBits; + const unsigned nbMax = + (p->numHashBytes == 2 ? 16 : + (p->numHashBytes == 3 ? 24 : 32)); + if (numBits > nbMax) + numBits = nbMax; + if (numBits >= 32) + hs = (UInt32)0 - 1; + else + hs = ((UInt32)1 << numBits) - 1; // (hash_size >= (1 << 16)) : Required for (numHashBytes > 2) hs |= (1 << 16) - 1; /* don't change it! */ - - // bt5: we adjust the size with recommended minimum size if (p->numHashBytes >= 5) hs |= (256 << kLzHash_CrcShift_2) - 1; + { + const UInt32 hs2 = MatchFinder_GetHashMask2(p, historySize); + if (hs > hs2) + hs = hs2; + } + hsCur = hs; + if (p->expectedDataSize < historySize) + { + const UInt32 hs2 = MatchFinder_GetHashMask2(p, (UInt32)p->expectedDataSize); + if (hsCur > hs2) + hsCur = hs2; + } + } + else + { + hs = MatchFinder_GetHashMask(p, historySize); + hsCur = hs; + if (p->expectedDataSize < historySize) + { + hsCur = MatchFinder_GetHashMask(p, (UInt32)p->expectedDataSize); + if (hsCur > hs) // is it possible? + hsCur = hs; + } } - p->hashMask = hs; - hs++; - /* - hs4 = (1 << 20); - if (hs4 > hs) - hs4 = hs; - // hs4 = (1 << 16); // for test - p->hash4Mask = hs4 - 1; - */ + p->hashMask = hsCur; - if (p->numHashBytes > 2) p->fixedHashSize += kHash2Size; - if (p->numHashBytes > 3) p->fixedHashSize += kHash3Size; - // if (p->numHashBytes > 4) p->fixedHashSize += hs4; // kHash4Size; - hs += p->fixedHashSize; + hashSizeSum = hs; + hashSizeSum++; + if (hashSizeSum < hs) + return 0; + { + UInt32 fixedHashSize = 0; + if (p->numHashBytes > 2 && p->numHashBytes_Min <= 2) fixedHashSize += kHash2Size; + if (p->numHashBytes > 3 && p->numHashBytes_Min <= 3) fixedHashSize += kHash3Size; + // if (p->numHashBytes > 4) p->fixedHashSize += hs4; // kHash4Size; + hashSizeSum += fixedHashSize; + p->fixedHashSize = fixedHashSize; + } } + p->matchMaxLen = matchMaxLen; + { size_t newSize; size_t numSons; + const UInt32 newCyclicBufferSize = historySize + 1; // do not change it p->historySize = historySize; - p->hashSizeSum = hs; p->cyclicBufferSize = newCyclicBufferSize; // it must be = (historySize + 1) numSons = newCyclicBufferSize; if (p->btMode) numSons <<= 1; - newSize = hs + numSons; + newSize = hashSizeSum + numSons; + + if (numSons < newCyclicBufferSize || newSize < numSons) + return 0; // aligned size is not required here, but it can be better for some loops #define NUM_REFS_ALIGN_MASK 0xF newSize = (newSize + NUM_REFS_ALIGN_MASK) & ~(size_t)NUM_REFS_ALIGN_MASK; - if (p->hash && p->numRefs == newSize) + // 22.02: we don't reallocate buffer, if old size is enough + if (p->hash && p->numRefs >= newSize) return 1; MatchFinder_FreeThisClassMemory(p, alloc); @@ -398,7 +480,7 @@ int MatchFinder_Create(CMatchFinder *p, UInt32 historySize, if (p->hash) { - p->son = p->hash + p->hashSizeSum; + p->son = p->hash + hashSizeSum; return 1; } } @@ -470,7 +552,8 @@ void MatchFinder_Init_HighHash(CMatchFinder *p) void MatchFinder_Init_4(CMatchFinder *p) { - p->buffer = p->bufferBase; + if (!p->directInput) + p->buffer = p->bufBase; { /* kEmptyHashValue = 0 (Zero) is used in hash tables as NO-VALUE marker. the code in CMatchFinderMt expects (pos = 1) */ @@ -507,20 +590,20 @@ void MatchFinder_Init(CMatchFinder *p) #ifdef MY_CPU_X86_OR_AMD64 - #if defined(__clang__) && (__clang_major__ >= 8) \ - || defined(__GNUC__) && (__GNUC__ >= 8) \ - || defined(__INTEL_COMPILER) && (__INTEL_COMPILER >= 1900) - #define USE_SATUR_SUB_128 - #define USE_AVX2 - #define ATTRIB_SSE41 __attribute__((__target__("sse4.1"))) - #define ATTRIB_AVX2 __attribute__((__target__("avx2"))) + #if defined(__clang__) && (__clang_major__ >= 4) \ + || defined(Z7_GCC_VERSION) && (Z7_GCC_VERSION >= 40701) + // || defined(__INTEL_COMPILER) && (__INTEL_COMPILER >= 1900) + + #define USE_LZFIND_SATUR_SUB_128 + #define USE_LZFIND_SATUR_SUB_256 + #define LZFIND_ATTRIB_SSE41 __attribute__((__target__("sse4.1"))) + #define LZFIND_ATTRIB_AVX2 __attribute__((__target__("avx2"))) #elif defined(_MSC_VER) #if (_MSC_VER >= 1600) - #define USE_SATUR_SUB_128 - #if (_MSC_VER >= 1900) - #define USE_AVX2 - #include // avx - #endif + #define USE_LZFIND_SATUR_SUB_128 + #endif + #if (_MSC_VER >= 1900) + #define USE_LZFIND_SATUR_SUB_256 #endif #endif @@ -529,16 +612,16 @@ void MatchFinder_Init(CMatchFinder *p) #if defined(__clang__) && (__clang_major__ >= 8) \ || defined(__GNUC__) && (__GNUC__ >= 8) - #define USE_SATUR_SUB_128 + #define USE_LZFIND_SATUR_SUB_128 #ifdef MY_CPU_ARM64 - // #define ATTRIB_SSE41 __attribute__((__target__(""))) + // #define LZFIND_ATTRIB_SSE41 __attribute__((__target__(""))) #else - // #define ATTRIB_SSE41 __attribute__((__target__("fpu=crypto-neon-fp-armv8"))) + // #define LZFIND_ATTRIB_SSE41 __attribute__((__target__("fpu=crypto-neon-fp-armv8"))) #endif #elif defined(_MSC_VER) #if (_MSC_VER >= 1910) - #define USE_SATUR_SUB_128 + #define USE_LZFIND_SATUR_SUB_128 #endif #endif @@ -550,121 +633,130 @@ void MatchFinder_Init(CMatchFinder *p) #endif -/* -#ifndef ATTRIB_SSE41 - #define ATTRIB_SSE41 -#endif -#ifndef ATTRIB_AVX2 - #define ATTRIB_AVX2 -#endif -*/ -#ifdef USE_SATUR_SUB_128 +#ifdef USE_LZFIND_SATUR_SUB_128 -// #define _SHOW_HW_STATUS +// #define Z7_SHOW_HW_STATUS -#ifdef _SHOW_HW_STATUS +#ifdef Z7_SHOW_HW_STATUS #include -#define _PRF(x) x -_PRF(;) +#define PRF(x) x +PRF(;) #else -#define _PRF(x) +#define PRF(x) #endif + #ifdef MY_CPU_ARM_OR_ARM64 #ifdef MY_CPU_ARM64 -// #define FORCE_SATUR_SUB_128 +// #define FORCE_LZFIND_SATUR_SUB_128 #endif +typedef uint32x4_t LzFind_v128; +#define SASUB_128_V(v, s) \ + vsubq_u32(vmaxq_u32(v, s), s) -typedef uint32x4_t v128; -#define SASUB_128(i) \ - *(v128 *)(void *)(items + (i) * 4) = \ - vsubq_u32(vmaxq_u32(*(const v128 *)(const void *)(items + (i) * 4), sub2), sub2); - -#else +#else // MY_CPU_ARM_OR_ARM64 #include // sse4.1 -typedef __m128i v128; +typedef __m128i LzFind_v128; +// SSE 4.1 +#define SASUB_128_V(v, s) \ + _mm_sub_epi32(_mm_max_epu32(v, s), s) + +#endif // MY_CPU_ARM_OR_ARM64 + + #define SASUB_128(i) \ - *(v128 *)(void *)(items + (i) * 4) = \ - _mm_sub_epi32(_mm_max_epu32(*(const v128 *)(const void *)(items + (i) * 4), sub2), sub2); // SSE 4.1 - -#endif + *( LzFind_v128 *)( void *)(items + (i) * 4) = SASUB_128_V( \ + *(const LzFind_v128 *)(const void *)(items + (i) * 4), sub2); - -MY_NO_INLINE +Z7_NO_INLINE static -#ifdef ATTRIB_SSE41 -ATTRIB_SSE41 +#ifdef LZFIND_ATTRIB_SSE41 +LZFIND_ATTRIB_SSE41 #endif void -MY_FAST_CALL +Z7_FASTCALL LzFind_SaturSub_128(UInt32 subValue, CLzRef *items, const CLzRef *lim) { - v128 sub2 = + const LzFind_v128 sub2 = #ifdef MY_CPU_ARM_OR_ARM64 vdupq_n_u32(subValue); #else _mm_set_epi32((Int32)subValue, (Int32)subValue, (Int32)subValue, (Int32)subValue); #endif + Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE do { - SASUB_128(0) - SASUB_128(1) - SASUB_128(2) - SASUB_128(3) - items += 4 * 4; + SASUB_128(0) SASUB_128(1) items += 2 * 4; + SASUB_128(0) SASUB_128(1) items += 2 * 4; } while (items != lim); } -#ifdef USE_AVX2 +#ifdef USE_LZFIND_SATUR_SUB_256 #include // avx +/* +clang :immintrin.h uses +#if !(defined(_MSC_VER) || defined(__SCE__)) || __has_feature(modules) || \ + defined(__AVX2__) +#include +#endif +so we need for clang-cl */ -#define SASUB_256(i) *(__m256i *)(void *)(items + (i) * 8) = _mm256_sub_epi32(_mm256_max_epu32(*(const __m256i *)(const void *)(items + (i) * 8), sub2), sub2); // AVX2 +#if defined(__clang__) +#include +#include +#endif -MY_NO_INLINE +// AVX2: +#define SASUB_256(i) \ + *( __m256i *)( void *)(items + (i) * 8) = \ + _mm256_sub_epi32(_mm256_max_epu32( \ + *(const __m256i *)(const void *)(items + (i) * 8), sub2), sub2); + +Z7_NO_INLINE static -#ifdef ATTRIB_AVX2 -ATTRIB_AVX2 +#ifdef LZFIND_ATTRIB_AVX2 +LZFIND_ATTRIB_AVX2 #endif void -MY_FAST_CALL +Z7_FASTCALL LzFind_SaturSub_256(UInt32 subValue, CLzRef *items, const CLzRef *lim) { - __m256i sub2 = _mm256_set_epi32( + const __m256i sub2 = _mm256_set_epi32( (Int32)subValue, (Int32)subValue, (Int32)subValue, (Int32)subValue, (Int32)subValue, (Int32)subValue, (Int32)subValue, (Int32)subValue); + Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE do { - SASUB_256(0) - SASUB_256(1) - items += 2 * 8; + SASUB_256(0) SASUB_256(1) items += 2 * 8; + SASUB_256(0) SASUB_256(1) items += 2 * 8; } while (items != lim); } -#endif // USE_AVX2 +#endif // USE_LZFIND_SATUR_SUB_256 -#ifndef FORCE_SATUR_SUB_128 -typedef void (MY_FAST_CALL *LZFIND_SATUR_SUB_CODE_FUNC)( +#ifndef FORCE_LZFIND_SATUR_SUB_128 +typedef void (Z7_FASTCALL *LZFIND_SATUR_SUB_CODE_FUNC)( UInt32 subValue, CLzRef *items, const CLzRef *lim); static LZFIND_SATUR_SUB_CODE_FUNC g_LzFind_SaturSub; -#endif // FORCE_SATUR_SUB_128 +#endif // FORCE_LZFIND_SATUR_SUB_128 -#endif // USE_SATUR_SUB_128 +#endif // USE_LZFIND_SATUR_SUB_128 // kEmptyHashValue must be zero -// #define SASUB_32(i) v = items[i]; m = v - subValue; if (v < subValue) m = kEmptyHashValue; items[i] = m; -#define SASUB_32(i) v = items[i]; if (v < subValue) v = subValue; items[i] = v - subValue; +// #define SASUB_32(i) { UInt32 v = items[i]; UInt32 m = v - subValue; if (v < subValue) m = kEmptyHashValue; items[i] = m; } +#define SASUB_32(i) { UInt32 v = items[i]; if (v < subValue) v = subValue; items[i] = v - subValue; } -#ifdef FORCE_SATUR_SUB_128 +#ifdef FORCE_LZFIND_SATUR_SUB_128 #define DEFAULT_SaturSub LzFind_SaturSub_128 @@ -672,24 +764,19 @@ static LZFIND_SATUR_SUB_CODE_FUNC g_LzFind_SaturSub; #define DEFAULT_SaturSub LzFind_SaturSub_32 -MY_NO_INLINE +Z7_NO_INLINE static void -MY_FAST_CALL +Z7_FASTCALL LzFind_SaturSub_32(UInt32 subValue, CLzRef *items, const CLzRef *lim) { + Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE do { - UInt32 v; - SASUB_32(0) - SASUB_32(1) - SASUB_32(2) - SASUB_32(3) - SASUB_32(4) - SASUB_32(5) - SASUB_32(6) - SASUB_32(7) - items += 8; + SASUB_32(0) SASUB_32(1) items += 2; + SASUB_32(0) SASUB_32(1) items += 2; + SASUB_32(0) SASUB_32(1) items += 2; + SASUB_32(0) SASUB_32(1) items += 2; } while (items != lim); } @@ -697,27 +784,23 @@ LzFind_SaturSub_32(UInt32 subValue, CLzRef *items, const CLzRef *lim) #endif -MY_NO_INLINE +Z7_NO_INLINE void MatchFinder_Normalize3(UInt32 subValue, CLzRef *items, size_t numItems) { - #define K_NORM_ALIGN_BLOCK_SIZE (1 << 6) - - CLzRef *lim; - - for (; numItems != 0 && ((unsigned)(ptrdiff_t)items & (K_NORM_ALIGN_BLOCK_SIZE - 1)) != 0; numItems--) + #define LZFIND_NORM_ALIGN_BLOCK_SIZE (1 << 7) + Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE + for (; numItems != 0 && ((unsigned)(ptrdiff_t)items & (LZFIND_NORM_ALIGN_BLOCK_SIZE - 1)) != 0; numItems--) { - UInt32 v; - SASUB_32(0); + SASUB_32(0) items++; } - { - #define K_NORM_ALIGN_MASK (K_NORM_ALIGN_BLOCK_SIZE / 4 - 1) - lim = items + (numItems & ~(size_t)K_NORM_ALIGN_MASK); - numItems &= K_NORM_ALIGN_MASK; + const size_t k_Align_Mask = (LZFIND_NORM_ALIGN_BLOCK_SIZE / 4 - 1); + CLzRef *lim = items + (numItems & ~(size_t)k_Align_Mask); + numItems &= k_Align_Mask; if (items != lim) { - #if defined(USE_SATUR_SUB_128) && !defined(FORCE_SATUR_SUB_128) + #if defined(USE_LZFIND_SATUR_SUB_128) && !defined(FORCE_LZFIND_SATUR_SUB_128) if (g_LzFind_SaturSub) g_LzFind_SaturSub(subValue, items, lim); else @@ -726,12 +809,10 @@ void MatchFinder_Normalize3(UInt32 subValue, CLzRef *items, size_t numItems) } items = lim; } - - + Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE for (; numItems != 0; numItems--) { - UInt32 v; - SASUB_32(0); + SASUB_32(0) items++; } } @@ -740,7 +821,7 @@ void MatchFinder_Normalize3(UInt32 subValue, CLzRef *items, size_t numItems) // call MatchFinder_CheckLimits() only after (p->pos++) update -MY_NO_INLINE +Z7_NO_INLINE static void MatchFinder_CheckLimits(CMatchFinder *p) { if (// !p->streamEndWasReached && p->result == SZ_OK && @@ -768,11 +849,14 @@ static void MatchFinder_CheckLimits(CMatchFinder *p) const UInt32 subValue = (p->pos - p->historySize - 1) /* & ~(UInt32)(kNormalizeAlign - 1) */; // const UInt32 subValue = (1 << 15); // for debug // printf("\nMatchFinder_Normalize() subValue == 0x%x\n", subValue); - size_t numSonRefs = p->cyclicBufferSize; - if (p->btMode) - numSonRefs <<= 1; - Inline_MatchFinder_ReduceOffsets(p, subValue); - MatchFinder_Normalize3(subValue, p->hash, (size_t)p->hashSizeSum + numSonRefs); + MatchFinder_REDUCE_OFFSETS(p, subValue) + MatchFinder_Normalize3(subValue, p->hash, (size_t)p->hashMask + 1 + p->fixedHashSize); + { + size_t numSonRefs = p->cyclicBufferSize; + if (p->btMode) + numSonRefs <<= 1; + MatchFinder_Normalize3(subValue, p->son, numSonRefs); + } } if (p->cyclicBufferPos == p->cyclicBufferSize) @@ -785,7 +869,7 @@ static void MatchFinder_CheckLimits(CMatchFinder *p) /* (lenLimit > maxLen) */ -MY_FORCE_INLINE +Z7_FORCE_INLINE static UInt32 * Hc_GetMatchesSpec(size_t lenLimit, UInt32 curMatch, UInt32 pos, const Byte *cur, CLzRef *son, size_t _cyclicBufferPos, UInt32 _cyclicBufferSize, UInt32 cutValue, UInt32 *d, unsigned maxLen) @@ -867,7 +951,7 @@ static UInt32 * Hc_GetMatchesSpec(size_t lenLimit, UInt32 curMatch, UInt32 pos, } -MY_FORCE_INLINE +Z7_FORCE_INLINE UInt32 * GetMatchesSpec1(UInt32 lenLimit, UInt32 curMatch, UInt32 pos, const Byte *cur, CLzRef *son, size_t _cyclicBufferPos, UInt32 _cyclicBufferSize, UInt32 cutValue, UInt32 *d, UInt32 maxLen) @@ -1004,7 +1088,7 @@ static void SkipMatchesSpec(UInt32 lenLimit, UInt32 curMatch, UInt32 pos, const #define MOVE_POS_RET MOVE_POS return distances; -MY_NO_INLINE +Z7_NO_INLINE static void MatchFinder_MovePos(CMatchFinder *p) { /* we go here at the end of stream data, when (avail < num_hash_bytes) @@ -1015,11 +1099,11 @@ static void MatchFinder_MovePos(CMatchFinder *p) if (p->btMode) p->sons[(p->cyclicBufferPos << p->btMode) + 1] = 0; // kEmptyHashValue */ - MOVE_POS; + MOVE_POS } #define GET_MATCHES_HEADER2(minLen, ret_op) \ - unsigned lenLimit; UInt32 hv; Byte *cur; UInt32 curMatch; \ + unsigned lenLimit; UInt32 hv; const Byte *cur; UInt32 curMatch; \ lenLimit = (unsigned)p->lenLimit; { if (lenLimit < minLen) { MatchFinder_MovePos(p); ret_op; }} \ cur = p->buffer; @@ -1028,11 +1112,11 @@ static void MatchFinder_MovePos(CMatchFinder *p) #define MF_PARAMS(p) lenLimit, curMatch, p->pos, p->buffer, p->son, p->cyclicBufferPos, p->cyclicBufferSize, p->cutValue -#define SKIP_FOOTER SkipMatchesSpec(MF_PARAMS(p)); MOVE_POS; } while (--num); +#define SKIP_FOOTER SkipMatchesSpec(MF_PARAMS(p)); MOVE_POS } while (--num); #define GET_MATCHES_FOOTER_BASE(_maxLen_, func) \ distances = func(MF_PARAMS(p), \ - distances, (UInt32)_maxLen_); MOVE_POS_RET; + distances, (UInt32)_maxLen_); MOVE_POS_RET #define GET_MATCHES_FOOTER_BT(_maxLen_) \ GET_MATCHES_FOOTER_BASE(_maxLen_, GetMatchesSpec1) @@ -1052,7 +1136,7 @@ static void MatchFinder_MovePos(CMatchFinder *p) static UInt32* Bt2_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances) { GET_MATCHES_HEADER(2) - HASH2_CALC; + HASH2_CALC curMatch = p->hash[hv]; p->hash[hv] = p->pos; GET_MATCHES_FOOTER_BT(1) @@ -1061,7 +1145,7 @@ static UInt32* Bt2_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances) UInt32* Bt3Zip_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances) { GET_MATCHES_HEADER(3) - HASH_ZIP_CALC; + HASH_ZIP_CALC curMatch = p->hash[hv]; p->hash[hv] = p->pos; GET_MATCHES_FOOTER_BT(2) @@ -1082,7 +1166,7 @@ static UInt32* Bt3_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances) UInt32 *hash; GET_MATCHES_HEADER(3) - HASH3_CALC; + HASH3_CALC hash = p->hash; pos = p->pos; @@ -1107,7 +1191,7 @@ static UInt32* Bt3_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances) if (maxLen == lenLimit) { SkipMatchesSpec(MF_PARAMS(p)); - MOVE_POS_RET; + MOVE_POS_RET } } @@ -1123,7 +1207,7 @@ static UInt32* Bt4_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances) UInt32 *hash; GET_MATCHES_HEADER(4) - HASH4_CALC; + HASH4_CALC hash = p->hash; pos = p->pos; @@ -1190,7 +1274,7 @@ static UInt32* Bt5_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances) UInt32 *hash; GET_MATCHES_HEADER(5) - HASH5_CALC; + HASH5_CALC hash = p->hash; pos = p->pos; @@ -1246,7 +1330,7 @@ static UInt32* Bt5_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances) if (maxLen == lenLimit) { SkipMatchesSpec(MF_PARAMS(p)); - MOVE_POS_RET; + MOVE_POS_RET } break; } @@ -1263,7 +1347,7 @@ static UInt32* Hc4_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances) UInt32 *hash; GET_MATCHES_HEADER(4) - HASH4_CALC; + HASH4_CALC hash = p->hash; pos = p->pos; @@ -1314,12 +1398,12 @@ static UInt32* Hc4_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances) if (maxLen == lenLimit) { p->son[p->cyclicBufferPos] = curMatch; - MOVE_POS_RET; + MOVE_POS_RET } break; } - GET_MATCHES_FOOTER_HC(maxLen); + GET_MATCHES_FOOTER_HC(maxLen) } @@ -1330,7 +1414,7 @@ static UInt32 * Hc5_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances) UInt32 *hash; GET_MATCHES_HEADER(5) - HASH5_CALC; + HASH5_CALC hash = p->hash; pos = p->pos; @@ -1386,19 +1470,19 @@ static UInt32 * Hc5_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances) if (maxLen == lenLimit) { p->son[p->cyclicBufferPos] = curMatch; - MOVE_POS_RET; + MOVE_POS_RET } break; } - GET_MATCHES_FOOTER_HC(maxLen); + GET_MATCHES_FOOTER_HC(maxLen) } UInt32* Hc3Zip_MatchFinder_GetMatches(CMatchFinder *p, UInt32 *distances) { GET_MATCHES_HEADER(3) - HASH_ZIP_CALC; + HASH_ZIP_CALC curMatch = p->hash[hv]; p->hash[hv] = p->pos; GET_MATCHES_FOOTER_HC(2) @@ -1409,7 +1493,7 @@ static void Bt2_MatchFinder_Skip(CMatchFinder *p, UInt32 num) { SKIP_HEADER(2) { - HASH2_CALC; + HASH2_CALC curMatch = p->hash[hv]; p->hash[hv] = p->pos; } @@ -1420,7 +1504,7 @@ void Bt3Zip_MatchFinder_Skip(CMatchFinder *p, UInt32 num) { SKIP_HEADER(3) { - HASH_ZIP_CALC; + HASH_ZIP_CALC curMatch = p->hash[hv]; p->hash[hv] = p->pos; } @@ -1433,7 +1517,7 @@ static void Bt3_MatchFinder_Skip(CMatchFinder *p, UInt32 num) { UInt32 h2; UInt32 *hash; - HASH3_CALC; + HASH3_CALC hash = p->hash; curMatch = (hash + kFix3HashSize)[hv]; hash[h2] = @@ -1448,7 +1532,7 @@ static void Bt4_MatchFinder_Skip(CMatchFinder *p, UInt32 num) { UInt32 h2, h3; UInt32 *hash; - HASH4_CALC; + HASH4_CALC hash = p->hash; curMatch = (hash + kFix4HashSize)[hv]; hash [h2] = @@ -1464,7 +1548,7 @@ static void Bt5_MatchFinder_Skip(CMatchFinder *p, UInt32 num) { UInt32 h2, h3; UInt32 *hash; - HASH5_CALC; + HASH5_CALC hash = p->hash; curMatch = (hash + kFix5HashSize)[hv]; hash [h2] = @@ -1478,7 +1562,7 @@ static void Bt5_MatchFinder_Skip(CMatchFinder *p, UInt32 num) #define HC_SKIP_HEADER(minLen) \ do { if (p->lenLimit < minLen) { MatchFinder_MovePos(p); num--; continue; } { \ - Byte *cur; \ + const Byte *cur; \ UInt32 *hash; \ UInt32 *son; \ UInt32 pos = p->pos; \ @@ -1510,7 +1594,7 @@ static void Hc4_MatchFinder_Skip(CMatchFinder *p, UInt32 num) HC_SKIP_HEADER(4) UInt32 h2, h3; - HASH4_CALC; + HASH4_CALC curMatch = (hash + kFix4HashSize)[hv]; hash [h2] = (hash + kFix3HashSize)[h3] = @@ -1540,7 +1624,7 @@ void Hc3Zip_MatchFinder_Skip(CMatchFinder *p, UInt32 num) { HC_SKIP_HEADER(3) - HASH_ZIP_CALC; + HASH_ZIP_CALC curMatch = hash[hv]; hash[hv] = pos; @@ -1590,17 +1674,17 @@ void MatchFinder_CreateVTable(CMatchFinder *p, IMatchFinder2 *vTable) -void LzFindPrepare() +void LzFindPrepare(void) { - #ifndef FORCE_SATUR_SUB_128 - #ifdef USE_SATUR_SUB_128 + #ifndef FORCE_LZFIND_SATUR_SUB_128 + #ifdef USE_LZFIND_SATUR_SUB_128 LZFIND_SATUR_SUB_CODE_FUNC f = NULL; #ifdef MY_CPU_ARM_OR_ARM64 { if (CPU_IsSupported_NEON()) { // #pragma message ("=== LzFind NEON") - _PRF(printf("\n=== LzFind NEON\n")); + PRF(printf("\n=== LzFind NEON\n")); f = LzFind_SaturSub_128; } // f = 0; // for debug @@ -1609,20 +1693,25 @@ void LzFindPrepare() if (CPU_IsSupported_SSE41()) { // #pragma message ("=== LzFind SSE41") - _PRF(printf("\n=== LzFind SSE41\n")); + PRF(printf("\n=== LzFind SSE41\n")); f = LzFind_SaturSub_128; - #ifdef USE_AVX2 + #ifdef USE_LZFIND_SATUR_SUB_256 if (CPU_IsSupported_AVX2()) { // #pragma message ("=== LzFind AVX2") - _PRF(printf("\n=== LzFind AVX2\n")); + PRF(printf("\n=== LzFind AVX2\n")); f = LzFind_SaturSub_256; } #endif } #endif // MY_CPU_ARM_OR_ARM64 g_LzFind_SaturSub = f; - #endif // USE_SATUR_SUB_128 - #endif // FORCE_SATUR_SUB_128 + #endif // USE_LZFIND_SATUR_SUB_128 + #endif // FORCE_LZFIND_SATUR_SUB_128 } + + +#undef MOVE_POS +#undef MOVE_POS_RET +#undef PRF diff --git a/libraries/lzma/C/LzFind.h b/libraries/lzma/C/LzFind.h index eea873ff6..a3f72c987 100644 --- a/libraries/lzma/C/LzFind.h +++ b/libraries/lzma/C/LzFind.h @@ -1,8 +1,8 @@ /* LzFind.h -- Match finder for LZ algorithms -2021-07-13 : Igor Pavlov : Public domain */ +2023-03-04 : Igor Pavlov : Public domain */ -#ifndef __LZ_FIND_H -#define __LZ_FIND_H +#ifndef ZIP7_INC_LZ_FIND_H +#define ZIP7_INC_LZ_FIND_H #include "7zTypes.h" @@ -10,9 +10,9 @@ EXTERN_C_BEGIN typedef UInt32 CLzRef; -typedef struct _CMatchFinder +typedef struct { - Byte *buffer; + const Byte *buffer; UInt32 pos; UInt32 posLimit; UInt32 streamPos; /* wrap over Zero is allowed (streamPos < pos). Use (UInt32)(streamPos - pos) */ @@ -32,8 +32,8 @@ typedef struct _CMatchFinder UInt32 hashMask; UInt32 cutValue; - Byte *bufferBase; - ISeqInStream *stream; + Byte *bufBase; + ISeqInStreamPtr stream; UInt32 blockSize; UInt32 keepSizeBefore; @@ -43,7 +43,9 @@ typedef struct _CMatchFinder size_t directInputRem; UInt32 historySize; UInt32 fixedHashSize; - UInt32 hashSizeSum; + Byte numHashBytes_Min; + Byte numHashOutBits; + Byte _pad2_[2]; SRes result; UInt32 crc[256]; size_t numRefs; @@ -69,24 +71,45 @@ void MatchFinder_ReadIfRequired(CMatchFinder *p); void MatchFinder_Construct(CMatchFinder *p); -/* Conditions: - historySize <= 3 GB - keepAddBufferBefore + matchMaxLen + keepAddBufferAfter < 511MB +/* (directInput = 0) is default value. + It's required to provide correct (directInput) value + before calling MatchFinder_Create(). + You can set (directInput) by any of the following calls: + - MatchFinder_SET_DIRECT_INPUT_BUF() + - MatchFinder_SET_STREAM() + - MatchFinder_SET_STREAM_MODE() */ + +#define MatchFinder_SET_DIRECT_INPUT_BUF(p, _src_, _srcLen_) { \ + (p)->stream = NULL; \ + (p)->directInput = 1; \ + (p)->buffer = (_src_); \ + (p)->directInputRem = (_srcLen_); } + +/* +#define MatchFinder_SET_STREAM_MODE(p) { \ + (p)->directInput = 0; } +*/ + +#define MatchFinder_SET_STREAM(p, _stream_) { \ + (p)->stream = _stream_; \ + (p)->directInput = 0; } + + int MatchFinder_Create(CMatchFinder *p, UInt32 historySize, UInt32 keepAddBufferBefore, UInt32 matchMaxLen, UInt32 keepAddBufferAfter, ISzAllocPtr alloc); void MatchFinder_Free(CMatchFinder *p, ISzAllocPtr alloc); void MatchFinder_Normalize3(UInt32 subValue, CLzRef *items, size_t numItems); -// void MatchFinder_ReduceOffsets(CMatchFinder *p, UInt32 subValue); /* -#define Inline_MatchFinder_InitPos(p, val) \ +#define MatchFinder_INIT_POS(p, val) \ (p)->pos = (val); \ (p)->streamPos = (val); */ -#define Inline_MatchFinder_ReduceOffsets(p, subValue) \ +// void MatchFinder_ReduceOffsets(CMatchFinder *p, UInt32 subValue); +#define MatchFinder_REDUCE_OFFSETS(p, subValue) \ (p)->pos -= (subValue); \ (p)->streamPos -= (subValue); @@ -107,7 +130,7 @@ typedef const Byte * (*Mf_GetPointerToCurrentPos_Func)(void *object); typedef UInt32 * (*Mf_GetMatches_Func)(void *object, UInt32 *distances); typedef void (*Mf_Skip_Func)(void *object, UInt32); -typedef struct _IMatchFinder +typedef struct { Mf_Init_Func Init; Mf_GetNumAvailableBytes_Func GetNumAvailableBytes; diff --git a/libraries/lzma/C/LzFindMt.c b/libraries/lzma/C/LzFindMt.c index 4e67fc3f2..5253e6ebb 100644 --- a/libraries/lzma/C/LzFindMt.c +++ b/libraries/lzma/C/LzFindMt.c @@ -1,5 +1,5 @@ /* LzFindMt.c -- multithreaded Match finder for LZ algorithms -2021-12-21 : Igor Pavlov : Public domain */ +2023-04-02 : Igor Pavlov : Public domain */ #include "Precomp.h" @@ -69,7 +69,7 @@ extern UInt64 g_NumIters_Bytes; UInt32 temp = p->crc[cur[0]] ^ cur[1]; \ h3 = (temp ^ ((UInt32)cur[2] << 8)) & (kHash3Size - 1); } -#define __MT_HASH4_CALC { \ +#define MT_HASH4_CALC { \ UInt32 temp = p->crc[cur[0]] ^ cur[1]; \ h2 = temp & (kHash2Size - 1); \ temp ^= ((UInt32)cur[2] << 8); \ @@ -79,14 +79,14 @@ extern UInt64 g_NumIters_Bytes; */ -MY_NO_INLINE +Z7_NO_INLINE static void MtSync_Construct(CMtSync *p) { p->affinity = 0; p->wasCreated = False; p->csWasInitialized = False; p->csWasEntered = False; - Thread_Construct(&p->thread); + Thread_CONSTRUCT(&p->thread) Event_Construct(&p->canStart); Event_Construct(&p->wasStopped); Semaphore_Construct(&p->freeSemaphore); @@ -116,7 +116,7 @@ static void MtSync_Construct(CMtSync *p) (p)->csWasEntered = False; } -MY_NO_INLINE +Z7_NO_INLINE static UInt32 MtSync_GetNextBlock(CMtSync *p) { UInt32 numBlocks = 0; @@ -140,14 +140,14 @@ static UInt32 MtSync_GetNextBlock(CMtSync *p) // buffer is UNLOCKED here Semaphore_Wait(&p->filledSemaphore); - LOCK_BUFFER(p); + LOCK_BUFFER(p) return numBlocks; } /* if Writing (Processing) thread was started, we must call MtSync_StopWriting() */ -MY_NO_INLINE +Z7_NO_INLINE static void MtSync_StopWriting(CMtSync *p) { if (!Thread_WasCreated(&p->thread) || p->needStart) @@ -185,7 +185,7 @@ static void MtSync_StopWriting(CMtSync *p) } -MY_NO_INLINE +Z7_NO_INLINE static void MtSync_Destruct(CMtSync *p) { PRF(printf("\nMtSync_Destruct %p\n", p)); @@ -220,11 +220,11 @@ static void MtSync_Destruct(CMtSync *p) // #define RINOK_THREAD(x) { if ((x) != 0) return SZ_ERROR_THREAD; } // we want to get real system error codes here instead of SZ_ERROR_THREAD -#define RINOK_THREAD(x) RINOK(x) +#define RINOK_THREAD(x) RINOK_WRes(x) // call it before each new file (when new starting is required): -MY_NO_INLINE +Z7_NO_INLINE static SRes MtSync_Init(CMtSync *p, UInt32 numBlocks) { WRes wres; @@ -245,12 +245,12 @@ static WRes MtSync_Create_WRes(CMtSync *p, THREAD_FUNC_TYPE startAddress, void * if (p->wasCreated) return SZ_OK; - RINOK_THREAD(CriticalSection_Init(&p->cs)); + RINOK_THREAD(CriticalSection_Init(&p->cs)) p->csWasInitialized = True; p->csWasEntered = False; - RINOK_THREAD(AutoResetEvent_CreateNotSignaled(&p->canStart)); - RINOK_THREAD(AutoResetEvent_CreateNotSignaled(&p->wasStopped)); + RINOK_THREAD(AutoResetEvent_CreateNotSignaled(&p->canStart)) + RINOK_THREAD(AutoResetEvent_CreateNotSignaled(&p->wasStopped)) p->needStart = True; p->exit = True; /* p->exit is unused before (canStart) Event. @@ -264,13 +264,13 @@ static WRes MtSync_Create_WRes(CMtSync *p, THREAD_FUNC_TYPE startAddress, void * else wres = Thread_Create(&p->thread, startAddress, obj); - RINOK_THREAD(wres); + RINOK_THREAD(wres) p->wasCreated = True; return SZ_OK; } -MY_NO_INLINE +Z7_NO_INLINE static SRes MtSync_Create(CMtSync *p, THREAD_FUNC_TYPE startAddress, void *obj) { const WRes wres = MtSync_Create_WRes(p, startAddress, obj); @@ -519,7 +519,7 @@ static void HashThreadFunc(CMatchFinderMt *mt) if (mf->pos > (UInt32)kMtMaxValForNormalize - num) { const UInt32 subValue = (mf->pos - mf->historySize - 1); // & ~(UInt32)(kNormalizeAlign - 1); - Inline_MatchFinder_ReduceOffsets(mf, subValue); + MatchFinder_REDUCE_OFFSETS(mf, subValue) MatchFinder_Normalize3(subValue, mf->hash + mf->fixedHashSize, (size_t)mf->hashMask + 1); } @@ -560,7 +560,7 @@ static void HashThreadFunc(CMatchFinderMt *mt) */ -UInt32 * MY_FAST_CALL GetMatchesSpecN_2(const Byte *lenLimit, size_t pos, const Byte *cur, CLzRef *son, +UInt32 * Z7_FASTCALL GetMatchesSpecN_2(const Byte *lenLimit, size_t pos, const Byte *cur, CLzRef *son, UInt32 _cutValue, UInt32 *d, size_t _maxLen, const UInt32 *hash, const UInt32 *limit, const UInt32 *size, size_t _cyclicBufferPos, UInt32 _cyclicBufferSize, UInt32 *posRes); @@ -749,7 +749,7 @@ static void BtFillBlock(CMatchFinderMt *p, UInt32 globalBlockIndex) } -MY_NO_INLINE +Z7_NO_INLINE static void BtThreadFunc(CMatchFinderMt *mt) { CMtSync *p = &mt->btSync; @@ -864,15 +864,15 @@ SRes MatchFinderMt_Create(CMatchFinderMt *p, UInt32 historySize, UInt32 keepAddB if (!MatchFinder_Create(mf, historySize, keepAddBufferBefore, matchMaxLen, keepAddBufferAfter, alloc)) return SZ_ERROR_MEM; - RINOK(MtSync_Create(&p->hashSync, HashThreadFunc2, p)); - RINOK(MtSync_Create(&p->btSync, BtThreadFunc2, p)); + RINOK(MtSync_Create(&p->hashSync, HashThreadFunc2, p)) + RINOK(MtSync_Create(&p->btSync, BtThreadFunc2, p)) return SZ_OK; } SRes MatchFinderMt_InitMt(CMatchFinderMt *p) { - RINOK(MtSync_Init(&p->hashSync, kMtHashNumBlocks)); + RINOK(MtSync_Init(&p->hashSync, kMtHashNumBlocks)) return MtSync_Init(&p->btSync, kMtBtNumBlocks); } @@ -941,7 +941,7 @@ void MatchFinderMt_ReleaseStream(CMatchFinderMt *p) } -MY_NO_INLINE +Z7_NO_INLINE static UInt32 MatchFinderMt_GetNextBlock_Bt(CMatchFinderMt *p) { if (p->failure_LZ_BT) @@ -1163,7 +1163,7 @@ UInt32* MatchFinderMt_GetMatches_Bt4(CMatchFinderMt *p, UInt32 *d) */ -static UInt32 *MixMatches4(CMatchFinderMt *p, UInt32 matchMinPos, UInt32 *d) +static UInt32 * MixMatches4(CMatchFinderMt *p, UInt32 matchMinPos, UInt32 *d) { UInt32 h2, h3, /* h4, */ c2, c3 /* , c4 */; UInt32 *hash = p->hash; @@ -1179,9 +1179,8 @@ static UInt32 *MixMatches4(CMatchFinderMt *p, UInt32 matchMinPos, UInt32 *d) (hash + kFix3HashSize)[h3] = m; // (hash + kFix4HashSize)[h4] = m; - #define _USE_H2 - - #ifdef _USE_H2 + // #define BT5_USE_H2 + // #ifdef BT5_USE_H2 if (c2 >= matchMinPos && cur[(ptrdiff_t)c2 - (ptrdiff_t)m] == cur[0]) { d[1] = m - c2 - 1; @@ -1197,8 +1196,8 @@ static UInt32 *MixMatches4(CMatchFinderMt *p, UInt32 matchMinPos, UInt32 *d) } d[0] = 3; d += 2; - - #ifdef _USE_H4 + + #ifdef BT5_USE_H4 if (c4 >= matchMinPos) if ( cur[(ptrdiff_t)c4 - (ptrdiff_t)m] == cur[0] && @@ -1214,7 +1213,7 @@ static UInt32 *MixMatches4(CMatchFinderMt *p, UInt32 matchMinPos, UInt32 *d) d[0] = 2; d += 2; } - #endif + // #endif if (c3 >= matchMinPos && cur[(ptrdiff_t)c3 - (ptrdiff_t)m] == cur[0]) { @@ -1228,7 +1227,7 @@ static UInt32 *MixMatches4(CMatchFinderMt *p, UInt32 matchMinPos, UInt32 *d) d += 2; } - #ifdef _USE_H4 + #ifdef BT5_USE_H4 if (c4 >= matchMinPos) if ( cur[(ptrdiff_t)c4 - (ptrdiff_t)m] == cur[0] && @@ -1244,7 +1243,7 @@ static UInt32 *MixMatches4(CMatchFinderMt *p, UInt32 matchMinPos, UInt32 *d) } -static UInt32* MatchFinderMt2_GetMatches(CMatchFinderMt *p, UInt32 *d) +static UInt32 * MatchFinderMt2_GetMatches(CMatchFinderMt *p, UInt32 *d) { const UInt32 *bt = p->btBufPos; const UInt32 len = *bt++; @@ -1268,7 +1267,7 @@ static UInt32* MatchFinderMt2_GetMatches(CMatchFinderMt *p, UInt32 *d) -static UInt32* MatchFinderMt_GetMatches(CMatchFinderMt *p, UInt32 *d) +static UInt32 * MatchFinderMt_GetMatches(CMatchFinderMt *p, UInt32 *d) { const UInt32 *bt = p->btBufPos; UInt32 len = *bt++; @@ -1398,3 +1397,10 @@ void MatchFinderMt_CreateVTable(CMatchFinderMt *p, IMatchFinder2 *vTable) break; } } + +#undef RINOK_THREAD +#undef PRF +#undef MF +#undef GetUi24hi_from32 +#undef LOCK_BUFFER +#undef UNLOCK_BUFFER diff --git a/libraries/lzma/C/LzFindMt.h b/libraries/lzma/C/LzFindMt.h index 660b7244d..db5923ea0 100644 --- a/libraries/lzma/C/LzFindMt.h +++ b/libraries/lzma/C/LzFindMt.h @@ -1,15 +1,15 @@ /* LzFindMt.h -- multithreaded Match finder for LZ algorithms -2021-07-12 : Igor Pavlov : Public domain */ +2023-03-05 : Igor Pavlov : Public domain */ -#ifndef __LZ_FIND_MT_H -#define __LZ_FIND_MT_H +#ifndef ZIP7_INC_LZ_FIND_MT_H +#define ZIP7_INC_LZ_FIND_MT_H #include "LzFind.h" #include "Threads.h" EXTERN_C_BEGIN -typedef struct _CMtSync +typedef struct { UInt32 numProcessedBlocks; CThread thread; @@ -39,7 +39,7 @@ typedef UInt32 * (*Mf_Mix_Matches)(void *p, UInt32 matchMinPos, UInt32 *distance typedef void (*Mf_GetHeads)(const Byte *buffer, UInt32 pos, UInt32 *hash, UInt32 hashMask, UInt32 *heads, UInt32 numHeads, const UInt32 *crc); -typedef struct _CMatchFinderMt +typedef struct { /* LZ */ const Byte *pointerToCurPos; diff --git a/libraries/lzma/C/LzFindOpt.c b/libraries/lzma/C/LzFindOpt.c index 8ff006e07..85bdc136d 100644 --- a/libraries/lzma/C/LzFindOpt.c +++ b/libraries/lzma/C/LzFindOpt.c @@ -1,5 +1,5 @@ /* LzFindOpt.c -- multithreaded Match finder for LZ algorithms -2021-07-13 : Igor Pavlov : Public domain */ +2023-04-02 : Igor Pavlov : Public domain */ #include "Precomp.h" @@ -41,8 +41,8 @@ UInt64 g_NumIters_Bytes; // #define CYC_TO_POS_OFFSET 1 // for debug /* -MY_NO_INLINE -UInt32 * MY_FAST_CALL GetMatchesSpecN_1(const Byte *lenLimit, size_t pos, const Byte *cur, CLzRef *son, +Z7_NO_INLINE +UInt32 * Z7_FASTCALL GetMatchesSpecN_1(const Byte *lenLimit, size_t pos, const Byte *cur, CLzRef *son, UInt32 _cutValue, UInt32 *d, size_t _maxLen, const UInt32 *hash, const UInt32 *limit, const UInt32 *size, UInt32 *posRes) { do @@ -214,13 +214,13 @@ else to eliminate "movsx" BUG in old MSVC x64 compiler. */ -UInt32 * MY_FAST_CALL GetMatchesSpecN_2(const Byte *lenLimit, size_t pos, const Byte *cur, CLzRef *son, +UInt32 * Z7_FASTCALL GetMatchesSpecN_2(const Byte *lenLimit, size_t pos, const Byte *cur, CLzRef *son, UInt32 _cutValue, UInt32 *d, size_t _maxLen, const UInt32 *hash, const UInt32 *limit, const UInt32 *size, size_t _cyclicBufferPos, UInt32 _cyclicBufferSize, UInt32 *posRes); -MY_NO_INLINE -UInt32 * MY_FAST_CALL GetMatchesSpecN_2(const Byte *lenLimit, size_t pos, const Byte *cur, CLzRef *son, +Z7_NO_INLINE +UInt32 * Z7_FASTCALL GetMatchesSpecN_2(const Byte *lenLimit, size_t pos, const Byte *cur, CLzRef *son, UInt32 _cutValue, UInt32 *d, size_t _maxLen, const UInt32 *hash, const UInt32 *limit, const UInt32 *size, size_t _cyclicBufferPos, UInt32 _cyclicBufferSize, UInt32 *posRes) @@ -404,7 +404,7 @@ else /* typedef UInt32 uint32plus; // size_t -UInt32 * MY_FAST_CALL GetMatchesSpecN_3(uint32plus lenLimit, size_t pos, const Byte *cur, CLzRef *son, +UInt32 * Z7_FASTCALL GetMatchesSpecN_3(uint32plus lenLimit, size_t pos, const Byte *cur, CLzRef *son, UInt32 _cutValue, UInt32 *d, uint32plus _maxLen, const UInt32 *hash, const UInt32 *limit, const UInt32 *size, size_t _cyclicBufferPos, UInt32 _cyclicBufferSize, UInt32 *posRes) diff --git a/libraries/lzma/C/LzHash.h b/libraries/lzma/C/LzHash.h index 77b898cfa..2b6290b64 100644 --- a/libraries/lzma/C/LzHash.h +++ b/libraries/lzma/C/LzHash.h @@ -1,8 +1,8 @@ -/* LzHash.h -- HASH functions for LZ algorithms -2019-10-30 : Igor Pavlov : Public domain */ +/* LzHash.h -- HASH constants for LZ algorithms +2023-03-05 : Igor Pavlov : Public domain */ -#ifndef __LZ_HASH_H -#define __LZ_HASH_H +#ifndef ZIP7_INC_LZ_HASH_H +#define ZIP7_INC_LZ_HASH_H /* (kHash2Size >= (1 << 8)) : Required diff --git a/libraries/lzma/C/Lzma2Dec.c b/libraries/lzma/C/Lzma2Dec.c index ac970a843..388cbc717 100644 --- a/libraries/lzma/C/Lzma2Dec.c +++ b/libraries/lzma/C/Lzma2Dec.c @@ -1,5 +1,5 @@ /* Lzma2Dec.c -- LZMA2 Decoder -2021-02-09 : Igor Pavlov : Public domain */ +2023-03-03 : Igor Pavlov : Public domain */ /* #define SHOW_DEBUG_INFO */ @@ -71,14 +71,14 @@ static SRes Lzma2Dec_GetOldProps(Byte prop, Byte *props) SRes Lzma2Dec_AllocateProbs(CLzma2Dec *p, Byte prop, ISzAllocPtr alloc) { Byte props[LZMA_PROPS_SIZE]; - RINOK(Lzma2Dec_GetOldProps(prop, props)); + 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)); + RINOK(Lzma2Dec_GetOldProps(prop, props)) return LzmaDec_Allocate(&p->decoder, props, LZMA_PROPS_SIZE, alloc); } @@ -474,8 +474,8 @@ SRes Lzma2Decode(Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen, SizeT outSize = *destLen, inSize = *srcLen; *destLen = *srcLen = 0; *status = LZMA_STATUS_NOT_SPECIFIED; - Lzma2Dec_Construct(&p); - RINOK(Lzma2Dec_AllocateProbs(&p, prop, alloc)); + Lzma2Dec_CONSTRUCT(&p) + RINOK(Lzma2Dec_AllocateProbs(&p, prop, alloc)) p.decoder.dic = dest; p.decoder.dicBufSize = outSize; Lzma2Dec_Init(&p); @@ -487,3 +487,5 @@ SRes Lzma2Decode(Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen, Lzma2Dec_FreeProbs(&p, alloc); return res; } + +#undef PRF diff --git a/libraries/lzma/C/Lzma2Dec.h b/libraries/lzma/C/Lzma2Dec.h index b8ddeac89..1f5233a72 100644 --- a/libraries/lzma/C/Lzma2Dec.h +++ b/libraries/lzma/C/Lzma2Dec.h @@ -1,8 +1,8 @@ /* Lzma2Dec.h -- LZMA2 Decoder -2018-02-19 : Igor Pavlov : Public domain */ +2023-03-03 : Igor Pavlov : Public domain */ -#ifndef __LZMA2_DEC_H -#define __LZMA2_DEC_H +#ifndef ZIP7_INC_LZMA2_DEC_H +#define ZIP7_INC_LZMA2_DEC_H #include "LzmaDec.h" @@ -22,9 +22,10 @@ typedef struct 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) +#define Lzma2Dec_CONSTRUCT(p) LzmaDec_CONSTRUCT(&(p)->decoder) +#define Lzma2Dec_Construct(p) Lzma2Dec_CONSTRUCT(p) +#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); @@ -90,7 +91,7 @@ Lzma2Dec_GetUnpackExtra() returns the value that shows at current input positon. */ -#define Lzma2Dec_GetUnpackExtra(p) ((p)->isExtraMode ? (p)->unpackSize : 0); +#define Lzma2Dec_GetUnpackExtra(p) ((p)->isExtraMode ? (p)->unpackSize : 0) /* ---------- One Call Interface ---------- */ diff --git a/libraries/lzma/C/Lzma2DecMt.c b/libraries/lzma/C/Lzma2DecMt.c new file mode 100644 index 000000000..4bc4ddeb5 --- /dev/null +++ b/libraries/lzma/C/Lzma2DecMt.c @@ -0,0 +1,1095 @@ +/* Lzma2DecMt.c -- LZMA2 Decoder Multi-thread +2023-04-13 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +// #define SHOW_DEBUG_INFO +// #define Z7_ST + +#ifdef SHOW_DEBUG_INFO +#include +#endif + +#include "Alloc.h" + +#include "Lzma2Dec.h" +#include "Lzma2DecMt.h" + +#ifndef Z7_ST +#include "MtDec.h" + +#define LZMA2DECMT_OUT_BLOCK_MAX_DEFAULT (1 << 28) +#endif + + +#ifndef Z7_ST +#ifdef SHOW_DEBUG_INFO +#define PRF(x) x +#else +#define PRF(x) +#endif +#define PRF_STR(s) PRF(printf("\n" s "\n");) +#define PRF_STR_INT_2(s, d1, d2) PRF(printf("\n" s " %d %d\n", (unsigned)d1, (unsigned)d2);) +#endif + + +void Lzma2DecMtProps_Init(CLzma2DecMtProps *p) +{ + p->inBufSize_ST = 1 << 20; + p->outStep_ST = 1 << 20; + + #ifndef Z7_ST + p->numThreads = 1; + p->inBufSize_MT = 1 << 18; + p->outBlockMax = LZMA2DECMT_OUT_BLOCK_MAX_DEFAULT; + p->inBlockMax = p->outBlockMax + p->outBlockMax / 16; + #endif +} + + + +#ifndef Z7_ST + +/* ---------- CLzma2DecMtThread ---------- */ + +typedef struct +{ + CLzma2Dec dec; + Byte dec_created; + Byte needInit; + + Byte *outBuf; + size_t outBufSize; + + EMtDecParseState state; + ELzma2ParseStatus parseStatus; + + size_t inPreSize; + size_t outPreSize; + + size_t inCodeSize; + size_t outCodeSize; + SRes codeRes; + + CAlignOffsetAlloc alloc; + + Byte mtPad[1 << 7]; +} CLzma2DecMtThread; + +#endif + + +/* ---------- CLzma2DecMt ---------- */ + +struct CLzma2DecMt +{ + // ISzAllocPtr alloc; + ISzAllocPtr allocMid; + + CAlignOffsetAlloc alignOffsetAlloc; + CLzma2DecMtProps props; + Byte prop; + + ISeqInStreamPtr inStream; + ISeqOutStreamPtr outStream; + ICompressProgressPtr progress; + + BoolInt finishMode; + BoolInt outSize_Defined; + UInt64 outSize; + + UInt64 outProcessed; + UInt64 inProcessed; + BoolInt readWasFinished; + SRes readRes; + + Byte *inBuf; + size_t inBufSize; + Byte dec_created; + CLzma2Dec dec; + + size_t inPos; + size_t inLim; + + #ifndef Z7_ST + UInt64 outProcessed_Parse; + BoolInt mtc_WasConstructed; + CMtDec mtc; + CLzma2DecMtThread coders[MTDEC_THREADS_MAX]; + #endif +}; + + + +CLzma2DecMtHandle Lzma2DecMt_Create(ISzAllocPtr alloc, ISzAllocPtr allocMid) +{ + CLzma2DecMt *p = (CLzma2DecMt *)ISzAlloc_Alloc(alloc, sizeof(CLzma2DecMt)); + if (!p) + return NULL; + + // p->alloc = alloc; + p->allocMid = allocMid; + + AlignOffsetAlloc_CreateVTable(&p->alignOffsetAlloc); + p->alignOffsetAlloc.numAlignBits = 7; + p->alignOffsetAlloc.offset = 0; + p->alignOffsetAlloc.baseAlloc = alloc; + + p->inBuf = NULL; + p->inBufSize = 0; + p->dec_created = False; + + // Lzma2DecMtProps_Init(&p->props); + + #ifndef Z7_ST + p->mtc_WasConstructed = False; + { + unsigned i; + for (i = 0; i < MTDEC_THREADS_MAX; i++) + { + CLzma2DecMtThread *t = &p->coders[i]; + t->dec_created = False; + t->outBuf = NULL; + t->outBufSize = 0; + } + } + #endif + + return (CLzma2DecMtHandle)(void *)p; +} + + +#ifndef Z7_ST + +static void Lzma2DecMt_FreeOutBufs(CLzma2DecMt *p) +{ + unsigned i; + for (i = 0; i < MTDEC_THREADS_MAX; i++) + { + CLzma2DecMtThread *t = &p->coders[i]; + if (t->outBuf) + { + ISzAlloc_Free(p->allocMid, t->outBuf); + t->outBuf = NULL; + t->outBufSize = 0; + } + } +} + +#endif + + +static void Lzma2DecMt_FreeSt(CLzma2DecMt *p) +{ + if (p->dec_created) + { + Lzma2Dec_Free(&p->dec, &p->alignOffsetAlloc.vt); + p->dec_created = False; + } + if (p->inBuf) + { + ISzAlloc_Free(p->allocMid, p->inBuf); + p->inBuf = NULL; + } + p->inBufSize = 0; +} + + +// #define GET_CLzma2DecMt_p CLzma2DecMt *p = (CLzma2DecMt *)(void *)pp; + +void Lzma2DecMt_Destroy(CLzma2DecMtHandle p) +{ + // GET_CLzma2DecMt_p + + Lzma2DecMt_FreeSt(p); + + #ifndef Z7_ST + + if (p->mtc_WasConstructed) + { + MtDec_Destruct(&p->mtc); + p->mtc_WasConstructed = False; + } + { + unsigned i; + for (i = 0; i < MTDEC_THREADS_MAX; i++) + { + CLzma2DecMtThread *t = &p->coders[i]; + if (t->dec_created) + { + // we don't need to free dict here + Lzma2Dec_FreeProbs(&t->dec, &t->alloc.vt); // p->alloc !!! + t->dec_created = False; + } + } + } + Lzma2DecMt_FreeOutBufs(p); + + #endif + + ISzAlloc_Free(p->alignOffsetAlloc.baseAlloc, p); +} + + + +#ifndef Z7_ST + +static void Lzma2DecMt_MtCallback_Parse(void *obj, unsigned coderIndex, CMtDecCallbackInfo *cc) +{ + CLzma2DecMt *me = (CLzma2DecMt *)obj; + CLzma2DecMtThread *t = &me->coders[coderIndex]; + + PRF_STR_INT_2("Parse", coderIndex, cc->srcSize) + + cc->state = MTDEC_PARSE_CONTINUE; + + if (cc->startCall) + { + if (!t->dec_created) + { + Lzma2Dec_CONSTRUCT(&t->dec) + t->dec_created = True; + AlignOffsetAlloc_CreateVTable(&t->alloc); + { + /* (1 << 12) is expected size of one way in data cache. + We optimize alignment for cache line size of 128 bytes and smaller */ + const unsigned kNumAlignBits = 12; + const unsigned kNumCacheLineBits = 7; /* <= kNumAlignBits */ + t->alloc.numAlignBits = kNumAlignBits; + t->alloc.offset = ((UInt32)coderIndex * (((unsigned)1 << 11) + (1 << 8) + (1 << 6))) & (((unsigned)1 << kNumAlignBits) - ((unsigned)1 << kNumCacheLineBits)); + t->alloc.baseAlloc = me->alignOffsetAlloc.baseAlloc; + } + } + Lzma2Dec_Init(&t->dec); + + t->inPreSize = 0; + t->outPreSize = 0; + // t->blockWasFinished = False; + // t->finishedWithMark = False; + t->parseStatus = (ELzma2ParseStatus)LZMA_STATUS_NOT_SPECIFIED; + t->state = MTDEC_PARSE_CONTINUE; + + t->inCodeSize = 0; + t->outCodeSize = 0; + t->codeRes = SZ_OK; + + // (cc->srcSize == 0) is allowed + } + + { + ELzma2ParseStatus status; + BoolInt overflow; + UInt32 unpackRem = 0; + + int checkFinishBlock = True; + size_t limit = me->props.outBlockMax; + if (me->outSize_Defined) + { + UInt64 rem = me->outSize - me->outProcessed_Parse; + if (limit >= rem) + { + limit = (size_t)rem; + if (!me->finishMode) + checkFinishBlock = False; + } + } + + // checkFinishBlock = False, if we want to decode partial data + // that must be finished at position <= outBlockMax. + + { + const size_t srcOrig = cc->srcSize; + SizeT srcSize_Point = 0; + SizeT dicPos_Point = 0; + + cc->srcSize = 0; + overflow = False; + + for (;;) + { + SizeT srcCur = (SizeT)(srcOrig - cc->srcSize); + + status = Lzma2Dec_Parse(&t->dec, + (SizeT)limit - t->dec.decoder.dicPos, + cc->src + cc->srcSize, &srcCur, + checkFinishBlock); + + cc->srcSize += srcCur; + + if (status == LZMA2_PARSE_STATUS_NEW_CHUNK) + { + if (t->dec.unpackSize > me->props.outBlockMax - t->dec.decoder.dicPos) + { + overflow = True; + break; + } + continue; + } + + if (status == LZMA2_PARSE_STATUS_NEW_BLOCK) + { + if (t->dec.decoder.dicPos == 0) + continue; + // we decode small blocks in one thread + if (t->dec.decoder.dicPos >= (1 << 14)) + break; + dicPos_Point = t->dec.decoder.dicPos; + srcSize_Point = (SizeT)cc->srcSize; + continue; + } + + if ((int)status == LZMA_STATUS_NOT_FINISHED && checkFinishBlock + // && limit == t->dec.decoder.dicPos + // && limit == me->props.outBlockMax + ) + { + overflow = True; + break; + } + + unpackRem = Lzma2Dec_GetUnpackExtra(&t->dec); + break; + } + + if (dicPos_Point != 0 + && (int)status != LZMA2_PARSE_STATUS_NEW_BLOCK + && (int)status != LZMA_STATUS_FINISHED_WITH_MARK + && (int)status != LZMA_STATUS_NOT_SPECIFIED) + { + // we revert to latest newBlock state + status = LZMA2_PARSE_STATUS_NEW_BLOCK; + unpackRem = 0; + t->dec.decoder.dicPos = dicPos_Point; + cc->srcSize = srcSize_Point; + overflow = False; + } + } + + t->inPreSize += cc->srcSize; + t->parseStatus = status; + + if (overflow) + cc->state = MTDEC_PARSE_OVERFLOW; + else + { + size_t dicPos = t->dec.decoder.dicPos; + + if ((int)status != LZMA_STATUS_NEEDS_MORE_INPUT) + { + if (status == LZMA2_PARSE_STATUS_NEW_BLOCK) + { + cc->state = MTDEC_PARSE_NEW; + cc->srcSize--; // we don't need control byte of next block + t->inPreSize--; + } + else + { + cc->state = MTDEC_PARSE_END; + if ((int)status != LZMA_STATUS_FINISHED_WITH_MARK) + { + // (status == LZMA_STATUS_NOT_SPECIFIED) + // (status == LZMA_STATUS_NOT_FINISHED) + if (unpackRem != 0) + { + /* we also reserve space for max possible number of output bytes of current LZMA chunk */ + size_t rem = limit - dicPos; + if (rem > unpackRem) + rem = unpackRem; + dicPos += rem; + } + } + } + + me->outProcessed_Parse += dicPos; + } + + cc->outPos = dicPos; + t->outPreSize = (size_t)dicPos; + } + + t->state = cc->state; + return; + } +} + + +static SRes Lzma2DecMt_MtCallback_PreCode(void *pp, unsigned coderIndex) +{ + CLzma2DecMt *me = (CLzma2DecMt *)pp; + CLzma2DecMtThread *t = &me->coders[coderIndex]; + Byte *dest = t->outBuf; + + if (t->inPreSize == 0) + { + t->codeRes = SZ_ERROR_DATA; + return t->codeRes; + } + + if (!dest || t->outBufSize < t->outPreSize) + { + if (dest) + { + ISzAlloc_Free(me->allocMid, dest); + t->outBuf = NULL; + t->outBufSize = 0; + } + + dest = (Byte *)ISzAlloc_Alloc(me->allocMid, t->outPreSize + // + (1 << 28) + ); + // Sleep(200); + if (!dest) + return SZ_ERROR_MEM; + t->outBuf = dest; + t->outBufSize = t->outPreSize; + } + + t->dec.decoder.dic = dest; + t->dec.decoder.dicBufSize = (SizeT)t->outPreSize; + + t->needInit = True; + + return Lzma2Dec_AllocateProbs(&t->dec, me->prop, &t->alloc.vt); // alloc.vt +} + + +static SRes Lzma2DecMt_MtCallback_Code(void *pp, unsigned coderIndex, + const Byte *src, size_t srcSize, int srcFinished, + // int finished, int blockFinished, + UInt64 *inCodePos, UInt64 *outCodePos, int *stop) +{ + CLzma2DecMt *me = (CLzma2DecMt *)pp; + CLzma2DecMtThread *t = &me->coders[coderIndex]; + + UNUSED_VAR(srcFinished) + + PRF_STR_INT_2("Code", coderIndex, srcSize) + + *inCodePos = t->inCodeSize; + *outCodePos = 0; + *stop = True; + + if (t->needInit) + { + Lzma2Dec_Init(&t->dec); + t->needInit = False; + } + + { + ELzmaStatus status; + SizeT srcProcessed = (SizeT)srcSize; + BoolInt blockWasFinished = + ((int)t->parseStatus == LZMA_STATUS_FINISHED_WITH_MARK + || t->parseStatus == LZMA2_PARSE_STATUS_NEW_BLOCK); + + SRes res = Lzma2Dec_DecodeToDic(&t->dec, + (SizeT)t->outPreSize, + src, &srcProcessed, + blockWasFinished ? LZMA_FINISH_END : LZMA_FINISH_ANY, + &status); + + t->codeRes = res; + + t->inCodeSize += srcProcessed; + *inCodePos = t->inCodeSize; + t->outCodeSize = t->dec.decoder.dicPos; + *outCodePos = t->dec.decoder.dicPos; + + if (res != SZ_OK) + return res; + + if (srcProcessed == srcSize) + *stop = False; + + if (blockWasFinished) + { + if (srcSize != srcProcessed) + return SZ_ERROR_FAIL; + + if (t->inPreSize == t->inCodeSize) + { + if (t->outPreSize != t->outCodeSize) + return SZ_ERROR_FAIL; + *stop = True; + } + } + else + { + if (t->outPreSize == t->outCodeSize) + *stop = True; + } + + return SZ_OK; + } +} + + +#define LZMA2DECMT_STREAM_WRITE_STEP (1 << 24) + +static SRes Lzma2DecMt_MtCallback_Write(void *pp, unsigned coderIndex, + BoolInt needWriteToStream, + const Byte *src, size_t srcSize, BoolInt isCross, + BoolInt *needContinue, BoolInt *canRecode) +{ + CLzma2DecMt *me = (CLzma2DecMt *)pp; + const CLzma2DecMtThread *t = &me->coders[coderIndex]; + size_t size = t->outCodeSize; + const Byte *data = t->outBuf; + BoolInt needContinue2 = True; + + UNUSED_VAR(src) + UNUSED_VAR(srcSize) + UNUSED_VAR(isCross) + + PRF_STR_INT_2("Write", coderIndex, srcSize) + + *needContinue = False; + *canRecode = True; + + if ( + // t->parseStatus == LZMA_STATUS_FINISHED_WITH_MARK + t->state == MTDEC_PARSE_OVERFLOW + || t->state == MTDEC_PARSE_END) + needContinue2 = False; + + + if (!needWriteToStream) + return SZ_OK; + + me->mtc.inProcessed += t->inCodeSize; + + if (t->codeRes == SZ_OK) + if ((int)t->parseStatus == LZMA_STATUS_FINISHED_WITH_MARK + || t->parseStatus == LZMA2_PARSE_STATUS_NEW_BLOCK) + if (t->outPreSize != t->outCodeSize + || t->inPreSize != t->inCodeSize) + return SZ_ERROR_FAIL; + + *canRecode = False; + + if (me->outStream) + { + for (;;) + { + size_t cur = size; + size_t written; + if (cur > LZMA2DECMT_STREAM_WRITE_STEP) + cur = LZMA2DECMT_STREAM_WRITE_STEP; + + written = ISeqOutStream_Write(me->outStream, data, cur); + + me->outProcessed += written; + // me->mtc.writtenTotal += written; + if (written != cur) + return SZ_ERROR_WRITE; + data += cur; + size -= cur; + if (size == 0) + { + *needContinue = needContinue2; + return SZ_OK; + } + RINOK(MtProgress_ProgressAdd(&me->mtc.mtProgress, 0, 0)) + } + } + + return SZ_ERROR_FAIL; + /* + if (size > me->outBufSize) + return SZ_ERROR_OUTPUT_EOF; + memcpy(me->outBuf, data, size); + me->outBufSize -= size; + me->outBuf += size; + *needContinue = needContinue2; + return SZ_OK; + */ +} + +#endif + + +static SRes Lzma2Dec_Prepare_ST(CLzma2DecMt *p) +{ + if (!p->dec_created) + { + Lzma2Dec_CONSTRUCT(&p->dec) + p->dec_created = True; + } + + RINOK(Lzma2Dec_Allocate(&p->dec, p->prop, &p->alignOffsetAlloc.vt)) + + if (!p->inBuf || p->inBufSize != p->props.inBufSize_ST) + { + ISzAlloc_Free(p->allocMid, p->inBuf); + p->inBufSize = 0; + p->inBuf = (Byte *)ISzAlloc_Alloc(p->allocMid, p->props.inBufSize_ST); + if (!p->inBuf) + return SZ_ERROR_MEM; + p->inBufSize = p->props.inBufSize_ST; + } + + Lzma2Dec_Init(&p->dec); + + return SZ_OK; +} + + +static SRes Lzma2Dec_Decode_ST(CLzma2DecMt *p + #ifndef Z7_ST + , BoolInt tMode + #endif + ) +{ + SizeT wrPos; + size_t inPos, inLim; + const Byte *inData; + UInt64 inPrev, outPrev; + + CLzma2Dec *dec; + + #ifndef Z7_ST + if (tMode) + { + Lzma2DecMt_FreeOutBufs(p); + tMode = MtDec_PrepareRead(&p->mtc); + } + #endif + + RINOK(Lzma2Dec_Prepare_ST(p)) + + dec = &p->dec; + + inPrev = p->inProcessed; + outPrev = p->outProcessed; + + inPos = 0; + inLim = 0; + inData = NULL; + wrPos = dec->decoder.dicPos; + + for (;;) + { + SizeT dicPos; + SizeT size; + ELzmaFinishMode finishMode; + SizeT inProcessed; + ELzmaStatus status; + SRes res; + + SizeT outProcessed; + BoolInt outFinished; + BoolInt needStop; + + if (inPos == inLim) + { + #ifndef Z7_ST + if (tMode) + { + inData = MtDec_Read(&p->mtc, &inLim); + inPos = 0; + if (inData) + continue; + tMode = False; + inLim = 0; + } + #endif + + if (!p->readWasFinished) + { + inPos = 0; + inLim = p->inBufSize; + inData = p->inBuf; + p->readRes = ISeqInStream_Read(p->inStream, (void *)(p->inBuf), &inLim); + // p->readProcessed += inLim; + // inLim -= 5; p->readWasFinished = True; // for test + if (inLim == 0 || p->readRes != SZ_OK) + p->readWasFinished = True; + } + } + + dicPos = dec->decoder.dicPos; + { + SizeT next = dec->decoder.dicBufSize; + if (next - wrPos > p->props.outStep_ST) + next = wrPos + (SizeT)p->props.outStep_ST; + size = next - dicPos; + } + + finishMode = LZMA_FINISH_ANY; + if (p->outSize_Defined) + { + const UInt64 rem = p->outSize - p->outProcessed; + if (size >= rem) + { + size = (SizeT)rem; + if (p->finishMode) + finishMode = LZMA_FINISH_END; + } + } + + inProcessed = (SizeT)(inLim - inPos); + + res = Lzma2Dec_DecodeToDic(dec, dicPos + size, inData + inPos, &inProcessed, finishMode, &status); + + inPos += inProcessed; + p->inProcessed += inProcessed; + outProcessed = dec->decoder.dicPos - dicPos; + p->outProcessed += outProcessed; + + outFinished = (p->outSize_Defined && p->outSize <= p->outProcessed); + + needStop = (res != SZ_OK + || (inProcessed == 0 && outProcessed == 0) + || status == LZMA_STATUS_FINISHED_WITH_MARK + || (!p->finishMode && outFinished)); + + if (needStop || outProcessed >= size) + { + SRes res2; + { + size_t writeSize = dec->decoder.dicPos - wrPos; + size_t written = ISeqOutStream_Write(p->outStream, dec->decoder.dic + wrPos, writeSize); + res2 = (written == writeSize) ? SZ_OK : SZ_ERROR_WRITE; + } + + if (dec->decoder.dicPos == dec->decoder.dicBufSize) + dec->decoder.dicPos = 0; + wrPos = dec->decoder.dicPos; + + RINOK(res2) + + if (needStop) + { + if (res != SZ_OK) + return res; + + if (status == LZMA_STATUS_FINISHED_WITH_MARK) + { + if (p->finishMode) + { + if (p->outSize_Defined && p->outSize != p->outProcessed) + return SZ_ERROR_DATA; + } + return SZ_OK; + } + + if (!p->finishMode && outFinished) + return SZ_OK; + + if (status == LZMA_STATUS_NEEDS_MORE_INPUT) + return SZ_ERROR_INPUT_EOF; + + return SZ_ERROR_DATA; + } + } + + if (p->progress) + { + UInt64 inDelta = p->inProcessed - inPrev; + UInt64 outDelta = p->outProcessed - outPrev; + if (inDelta >= (1 << 22) || outDelta >= (1 << 22)) + { + RINOK(ICompressProgress_Progress(p->progress, p->inProcessed, p->outProcessed)) + inPrev = p->inProcessed; + outPrev = p->outProcessed; + } + } + } +} + + + +SRes Lzma2DecMt_Decode(CLzma2DecMtHandle p, + Byte prop, + const CLzma2DecMtProps *props, + ISeqOutStreamPtr outStream, const UInt64 *outDataSize, int finishMode, + // Byte *outBuf, size_t *outBufSize, + ISeqInStreamPtr inStream, + // const Byte *inData, size_t inDataSize, + UInt64 *inProcessed, + // UInt64 *outProcessed, + int *isMT, + ICompressProgressPtr progress) +{ + // GET_CLzma2DecMt_p + #ifndef Z7_ST + BoolInt tMode; + #endif + + *inProcessed = 0; + + if (prop > 40) + return SZ_ERROR_UNSUPPORTED; + + p->prop = prop; + p->props = *props; + + p->inStream = inStream; + p->outStream = outStream; + p->progress = progress; + + p->outSize = 0; + p->outSize_Defined = False; + if (outDataSize) + { + p->outSize_Defined = True; + p->outSize = *outDataSize; + } + p->finishMode = finishMode; + + p->outProcessed = 0; + p->inProcessed = 0; + + p->readWasFinished = False; + p->readRes = SZ_OK; + + *isMT = False; + + + #ifndef Z7_ST + + tMode = False; + + // p->mtc.parseRes = SZ_OK; + + // p->mtc.numFilledThreads = 0; + // p->mtc.crossStart = 0; + // p->mtc.crossEnd = 0; + // p->mtc.allocError_for_Read_BlockIndex = 0; + // p->mtc.isAllocError = False; + + if (p->props.numThreads > 1) + { + IMtDecCallback2 vt; + + Lzma2DecMt_FreeSt(p); + + p->outProcessed_Parse = 0; + + if (!p->mtc_WasConstructed) + { + p->mtc_WasConstructed = True; + MtDec_Construct(&p->mtc); + } + + p->mtc.progress = progress; + p->mtc.inStream = inStream; + + // p->outBuf = NULL; + // p->outBufSize = 0; + /* + if (!outStream) + { + // p->outBuf = outBuf; + // p->outBufSize = *outBufSize; + // *outBufSize = 0; + return SZ_ERROR_PARAM; + } + */ + + // p->mtc.inBlockMax = p->props.inBlockMax; + p->mtc.alloc = &p->alignOffsetAlloc.vt; + // p->alignOffsetAlloc.baseAlloc; + // p->mtc.inData = inData; + // p->mtc.inDataSize = inDataSize; + p->mtc.mtCallback = &vt; + p->mtc.mtCallbackObject = p; + + p->mtc.inBufSize = p->props.inBufSize_MT; + + p->mtc.numThreadsMax = p->props.numThreads; + + *isMT = True; + + vt.Parse = Lzma2DecMt_MtCallback_Parse; + vt.PreCode = Lzma2DecMt_MtCallback_PreCode; + vt.Code = Lzma2DecMt_MtCallback_Code; + vt.Write = Lzma2DecMt_MtCallback_Write; + + { + BoolInt needContinue = False; + + SRes res = MtDec_Code(&p->mtc); + + /* + if (!outStream) + *outBufSize = p->outBuf - outBuf; + */ + + *inProcessed = p->mtc.inProcessed; + + needContinue = False; + + if (res == SZ_OK) + { + if (p->mtc.mtProgress.res != SZ_OK) + res = p->mtc.mtProgress.res; + else + needContinue = p->mtc.needContinue; + } + + if (!needContinue) + { + if (res == SZ_OK) + return p->mtc.readRes; + return res; + } + + tMode = True; + p->readRes = p->mtc.readRes; + p->readWasFinished = p->mtc.readWasFinished; + p->inProcessed = p->mtc.inProcessed; + + PRF_STR("----- decoding ST -----") + } + } + + #endif + + + *isMT = False; + + { + SRes res = Lzma2Dec_Decode_ST(p + #ifndef Z7_ST + , tMode + #endif + ); + + *inProcessed = p->inProcessed; + + // res = SZ_OK; // for test + if (res == SZ_ERROR_INPUT_EOF) + { + if (p->readRes != SZ_OK) + res = p->readRes; + } + else if (res == SZ_OK && p->readRes != SZ_OK) + res = p->readRes; + + /* + #ifndef Z7_ST + if (res == SZ_OK && tMode && p->mtc.parseRes != SZ_OK) + res = p->mtc.parseRes; + #endif + */ + + return res; + } +} + + +/* ---------- Read from CLzma2DecMtHandle Interface ---------- */ + +SRes Lzma2DecMt_Init(CLzma2DecMtHandle p, + Byte prop, + const CLzma2DecMtProps *props, + const UInt64 *outDataSize, int finishMode, + ISeqInStreamPtr inStream) +{ + // GET_CLzma2DecMt_p + + if (prop > 40) + return SZ_ERROR_UNSUPPORTED; + + p->prop = prop; + p->props = *props; + + p->inStream = inStream; + + p->outSize = 0; + p->outSize_Defined = False; + if (outDataSize) + { + p->outSize_Defined = True; + p->outSize = *outDataSize; + } + p->finishMode = finishMode; + + p->outProcessed = 0; + p->inProcessed = 0; + + p->inPos = 0; + p->inLim = 0; + + return Lzma2Dec_Prepare_ST(p); +} + + +SRes Lzma2DecMt_Read(CLzma2DecMtHandle p, + Byte *data, size_t *outSize, + UInt64 *inStreamProcessed) +{ + // GET_CLzma2DecMt_p + ELzmaFinishMode finishMode; + SRes readRes; + size_t size = *outSize; + + *outSize = 0; + *inStreamProcessed = 0; + + finishMode = LZMA_FINISH_ANY; + if (p->outSize_Defined) + { + const UInt64 rem = p->outSize - p->outProcessed; + if (size >= rem) + { + size = (size_t)rem; + if (p->finishMode) + finishMode = LZMA_FINISH_END; + } + } + + readRes = SZ_OK; + + for (;;) + { + SizeT inCur; + SizeT outCur; + ELzmaStatus status; + SRes res; + + if (p->inPos == p->inLim && readRes == SZ_OK) + { + p->inPos = 0; + p->inLim = p->props.inBufSize_ST; + readRes = ISeqInStream_Read(p->inStream, p->inBuf, &p->inLim); + } + + inCur = (SizeT)(p->inLim - p->inPos); + outCur = (SizeT)size; + + res = Lzma2Dec_DecodeToBuf(&p->dec, data, &outCur, + p->inBuf + p->inPos, &inCur, finishMode, &status); + + p->inPos += inCur; + p->inProcessed += inCur; + *inStreamProcessed += inCur; + p->outProcessed += outCur; + *outSize += outCur; + size -= outCur; + data += outCur; + + if (res != 0) + return res; + + /* + if (status == LZMA_STATUS_FINISHED_WITH_MARK) + return readRes; + + if (size == 0 && status != LZMA_STATUS_NEEDS_MORE_INPUT) + { + if (p->finishMode && p->outSize_Defined && p->outProcessed >= p->outSize) + return SZ_ERROR_DATA; + return readRes; + } + */ + + if (inCur == 0 && outCur == 0) + return readRes; + } +} + +#undef PRF +#undef PRF_STR +#undef PRF_STR_INT_2 diff --git a/libraries/lzma/C/Lzma2DecMt.h b/libraries/lzma/C/Lzma2DecMt.h new file mode 100644 index 000000000..93a5cd59e --- /dev/null +++ b/libraries/lzma/C/Lzma2DecMt.h @@ -0,0 +1,81 @@ +/* Lzma2DecMt.h -- LZMA2 Decoder Multi-thread +2023-04-13 : Igor Pavlov : Public domain */ + +#ifndef ZIP7_INC_LZMA2_DEC_MT_H +#define ZIP7_INC_LZMA2_DEC_MT_H + +#include "7zTypes.h" + +EXTERN_C_BEGIN + +typedef struct +{ + size_t inBufSize_ST; + size_t outStep_ST; + + #ifndef Z7_ST + unsigned numThreads; + size_t inBufSize_MT; + size_t outBlockMax; + size_t inBlockMax; + #endif +} CLzma2DecMtProps; + +/* init to single-thread mode */ +void Lzma2DecMtProps_Init(CLzma2DecMtProps *p); + + +/* ---------- CLzma2DecMtHandle Interface ---------- */ + +/* Lzma2DecMt_ * 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 struct CLzma2DecMt CLzma2DecMt; +typedef CLzma2DecMt * CLzma2DecMtHandle; +// Z7_DECLARE_HANDLE(CLzma2DecMtHandle) + +CLzma2DecMtHandle Lzma2DecMt_Create(ISzAllocPtr alloc, ISzAllocPtr allocMid); +void Lzma2DecMt_Destroy(CLzma2DecMtHandle p); + +SRes Lzma2DecMt_Decode(CLzma2DecMtHandle p, + Byte prop, + const CLzma2DecMtProps *props, + ISeqOutStreamPtr outStream, + const UInt64 *outDataSize, // NULL means undefined + int finishMode, // 0 - partial unpacking is allowed, 1 - if lzma2 stream must be finished + // Byte *outBuf, size_t *outBufSize, + ISeqInStreamPtr inStream, + // const Byte *inData, size_t inDataSize, + + // out variables: + UInt64 *inProcessed, + int *isMT, /* out: (*isMT == 0), if single thread decoding was used */ + + // UInt64 *outProcessed, + ICompressProgressPtr progress); + + +/* ---------- Read from CLzma2DecMtHandle Interface ---------- */ + +SRes Lzma2DecMt_Init(CLzma2DecMtHandle pp, + Byte prop, + const CLzma2DecMtProps *props, + const UInt64 *outDataSize, int finishMode, + ISeqInStreamPtr inStream); + +SRes Lzma2DecMt_Read(CLzma2DecMtHandle pp, + Byte *data, size_t *outSize, + UInt64 *inStreamProcessed); + + +EXTERN_C_END + +#endif diff --git a/libraries/lzma/C/Lzma2Enc.c b/libraries/lzma/C/Lzma2Enc.c new file mode 100644 index 000000000..703e146b5 --- /dev/null +++ b/libraries/lzma/C/Lzma2Enc.c @@ -0,0 +1,805 @@ +/* Lzma2Enc.c -- LZMA2 Encoder +2023-04-13 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include + +/* #define Z7_ST */ + +#include "Lzma2Enc.h" + +#ifndef Z7_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; + ISeqInStreamPtr 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(ISeqInStreamPtr pp, void *data, size_t *size) +{ + Z7_CONTAINER_FROM_VTBL_TO_DECL_VAR_pp_vt_p(CLimitedSeqInStream) + size_t size2 = *size; + SRes res = SZ_OK; + + if (p->limit != (UInt64)(Int64)-1) + { + const 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 p, ISeqInStreamPtr inStream, UInt32 keepWindowSize, + ISzAllocPtr alloc, ISzAllocPtr allocBig); +SRes LzmaEnc_MemPrepare(CLzmaEncHandle p, const Byte *src, SizeT srcLen, + UInt32 keepWindowSize, ISzAllocPtr alloc, ISzAllocPtr allocBig); +SRes LzmaEnc_CodeOneMemBlock(CLzmaEncHandle p, BoolInt reInit, + Byte *dest, size_t *destLen, UInt32 desiredPackSize, UInt32 *unpackSize); +const Byte *LzmaEnc_GetCurBuf(CLzmaEncHandle p); +void LzmaEnc_Finish(CLzmaEncHandle p); +void LzmaEnc_SaveState(CLzmaEncHandle p); +void LzmaEnc_RestoreState(CLzmaEncHandle p); + +/* +UInt32 LzmaEnc_GetNumAvailableBytes(CLzmaEncHandle p); +*/ + +static SRes Lzma2EncInt_EncodeSubblock(CLzma2EncInt *p, Byte *outBuf, + size_t *packSizeRes, ISeqOutStreamPtr 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) + { + const 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; + const UInt32 u = unpackSize - 1; + const UInt32 pm = (UInt32)(packSize - 1); + const 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 = (int)numBlocks; + if (t2r == 0) + t2r = 1; + t3 = t1 * t2r; + } + } + } + + p->numBlockThreads_Max = t2; + p->numBlockThreads_Reduced = t2r; + p->numTotalThreads = t3; +} + + +static SRes Progress(ICompressProgressPtr p, UInt64 inSize, UInt64 outSize) +{ + return (p && ICompressProgress_Progress(p, inSize, outSize) != SZ_OK) ? SZ_ERROR_PROGRESS : SZ_OK; +} + + +/* ---------- Lzma2 ---------- */ + +struct CLzma2Enc +{ + Byte propEncoded; + CLzma2EncProps props; + UInt64 expectedDataSize; + + Byte *tempBufLzma; + + ISzAllocPtr alloc; + ISzAllocPtr allocBig; + + CLzma2EncInt coders[MTCODER_THREADS_MAX]; + + #ifndef Z7_ST + + ISeqOutStreamPtr 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 +}; + + + +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 Z7_ST + p->mtCoder_WasConstructed = False; + { + unsigned i; + for (i = 0; i < MTCODER_BLOCKS_MAX; i++) + p->outBufs[i] = NULL; + p->outBufSize = 0; + } + #endif + + return (CLzma2EncHandle)p; +} + + +#ifndef Z7_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 + +// #define GET_CLzma2Enc_p CLzma2Enc *p = (CLzma2Enc *)(void *)p; + +void Lzma2Enc_Destroy(CLzma2EncHandle p) +{ + // GET_CLzma2Enc_p + 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 Z7_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, p); +} + + +SRes Lzma2Enc_SetProps(CLzma2EncHandle p, const CLzma2EncProps *props) +{ + // GET_CLzma2Enc_p + 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(CLzma2EncHandle p, UInt64 expectedDataSiize) +{ + // GET_CLzma2Enc_p + p->expectedDataSize = expectedDataSiize; +} + + +Byte Lzma2Enc_WriteProperties(CLzma2EncHandle p) +{ + // GET_CLzma2Enc_p + 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, + ISeqOutStreamPtr outStream, + Byte *outBuf, size_t *outBufSize, + ISeqInStreamPtr inStream, + const Byte *inData, size_t inDataSize, + int finished, + ICompressProgressPtr 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; + SizeT 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 = (SizeT)(inDataSize - (size_t)unpackTotal); + if (me->props.blockSize != LZMA2_ENC_PROPS_BLOCK_SIZE_SOLID + && inSizeCur > me->props.blockSize) + inSizeCur = (SizeT)(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) + { + const size_t destPos = *outBufSize; + if (destPos >= outLim) + return SZ_ERROR_OUTPUT_EOF; + outBuf[destPos] = LZMA2_CONTROL_EOF; // 0 + *outBufSize = destPos + 1; + } + else + { + const Byte b = LZMA2_CONTROL_EOF; // 0; + if (ISeqOutStream_Write(outStream, &b, 1) != 1) + return SZ_ERROR_WRITE; + } + } + return SZ_OK; + } + } +} + + + +#ifndef Z7_ST + +static SRes Lzma2Enc_MtCallback_Code(void *p, unsigned coderIndex, unsigned outBufIndex, + const Byte *src, size_t srcSize, int finished) +{ + CLzma2Enc *me = (CLzma2Enc *)p; + 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 *p, unsigned outBufIndex) +{ + CLzma2Enc *me = (CLzma2Enc *)p; + 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 p, + ISeqOutStreamPtr outStream, + Byte *outBuf, size_t *outBufSize, + ISeqInStreamPtr inStream, + const Byte *inData, size_t inDataSize, + ICompressProgressPtr progress) +{ + // GET_CLzma2Enc_p + + 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 Z7_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 */ + + { + const 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 = (unsigned)p->props.numBlockThreads_Max; + p->mtCoder.expectedDataSize = p->expectedDataSize; + + { + const SRes res = MtCoder_Code(&p->mtCoder); + if (!outStream) + *outBufSize = (size_t)(p->outBuf - outBuf); + return res; + } + } + + #endif + + + return Lzma2Enc_EncodeMt1(p, + &p->coders[0], + outStream, outBuf, outBufSize, + inStream, inData, inDataSize, + True, /* finished */ + progress); +} + +#undef PRF diff --git a/libraries/lzma/C/Lzma2Enc.h b/libraries/lzma/C/Lzma2Enc.h new file mode 100644 index 000000000..cb25275c6 --- /dev/null +++ b/libraries/lzma/C/Lzma2Enc.h @@ -0,0 +1,57 @@ +/* Lzma2Enc.h -- LZMA2 Encoder +2023-04-13 : Igor Pavlov : Public domain */ + +#ifndef ZIP7_INC_LZMA2_ENC_H +#define ZIP7_INC_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 struct CLzma2Enc CLzma2Enc; +typedef CLzma2Enc * CLzma2EncHandle; +// Z7_DECLARE_HANDLE(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, + ISeqOutStreamPtr outStream, + Byte *outBuf, size_t *outBufSize, + ISeqInStreamPtr inStream, + const Byte *inData, size_t inDataSize, + ICompressProgressPtr progress); + +EXTERN_C_END + +#endif diff --git a/libraries/lzma/C/Lzma86.h b/libraries/lzma/C/Lzma86.h new file mode 100644 index 000000000..e7707e2c4 --- /dev/null +++ b/libraries/lzma/C/Lzma86.h @@ -0,0 +1,111 @@ +/* Lzma86.h -- LZMA + x86 (BCJ) Filter +2023-03-03 : Igor Pavlov : Public domain */ + +#ifndef ZIP7_INC_LZMA86_H +#define ZIP7_INC_LZMA86_H + +#include "7zTypes.h" + +EXTERN_C_BEGIN + +#define LZMA86_SIZE_OFFSET (1 + 5) +#define LZMA86_HEADER_SIZE (LZMA86_SIZE_OFFSET + 8) + +/* +It's an example for LZMA + x86 Filter use. +You can use .lzma86 extension, if you write that stream to file. +.lzma86 header adds one additional byte to standard .lzma header. +.lzma86 header (14 bytes): + Offset Size Description + 0 1 = 0 - no filter, pure LZMA + = 1 - x86 filter + LZMA + 1 1 lc, lp and pb in encoded form + 2 4 dictSize (little endian) + 6 8 uncompressed size (little endian) + + +Lzma86_Encode +------------- +level - compression level: 0 <= level <= 9, the default value for "level" is 5. + +dictSize - The dictionary size in bytes. The maximum value is + 128 MB = (1 << 27) bytes for 32-bit version + 1 GB = (1 << 30) bytes for 64-bit version + The default value is 16 MB = (1 << 24) bytes, for level = 5. + It's recommended to use the dictionary that is larger than 4 KB and + that can be calculated as (1 << N) or (3 << N) sizes. + For better compression ratio dictSize must be >= inSize. + +filterMode: + SZ_FILTER_NO - no Filter + SZ_FILTER_YES - x86 Filter + SZ_FILTER_AUTO - it tries both alternatives to select best. + Encoder will use 2 or 3 passes: + 2 passes when FILTER_NO provides better compression. + 3 passes when FILTER_YES provides better compression. + +Lzma86Encode allocates Data with MyAlloc functions. +RAM Requirements for compressing: + RamSize = dictionarySize * 11.5 + 6MB + FilterBlockSize + filterMode FilterBlockSize + SZ_FILTER_NO 0 + SZ_FILTER_YES inSize + SZ_FILTER_AUTO inSize + + +Return code: + SZ_OK - OK + SZ_ERROR_MEM - Memory allocation error + SZ_ERROR_PARAM - Incorrect paramater + SZ_ERROR_OUTPUT_EOF - output buffer overflow + SZ_ERROR_THREAD - errors in multithreading functions (only for Mt version) +*/ + +enum ESzFilterMode +{ + SZ_FILTER_NO, + SZ_FILTER_YES, + SZ_FILTER_AUTO +}; + +SRes Lzma86_Encode(Byte *dest, size_t *destLen, const Byte *src, size_t srcLen, + int level, UInt32 dictSize, int filterMode); + + +/* +Lzma86_GetUnpackSize: + In: + src - input data + srcLen - input data size + Out: + unpackSize - size of uncompressed stream + Return code: + SZ_OK - OK + SZ_ERROR_INPUT_EOF - Error in headers +*/ + +SRes Lzma86_GetUnpackSize(const Byte *src, SizeT srcLen, UInt64 *unpackSize); + +/* +Lzma86_Decode: + In: + dest - output data + destLen - output data size + src - input data + srcLen - input data size + Out: + destLen - processed output size + srcLen - processed input size + Return code: + SZ_OK - OK + SZ_ERROR_DATA - Data error + SZ_ERROR_MEM - Memory allocation error + SZ_ERROR_UNSUPPORTED - unsupported file + SZ_ERROR_INPUT_EOF - it needs more bytes in input buffer +*/ + +SRes Lzma86_Decode(Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen); + +EXTERN_C_END + +#endif diff --git a/libraries/lzma/C/Lzma86Dec.c b/libraries/lzma/C/Lzma86Dec.c new file mode 100644 index 000000000..f094d4c35 --- /dev/null +++ b/libraries/lzma/C/Lzma86Dec.c @@ -0,0 +1,53 @@ +/* Lzma86Dec.c -- LZMA + x86 (BCJ) Filter Decoder +2023-03-03 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include "Lzma86.h" + +#include "Alloc.h" +#include "Bra.h" +#include "LzmaDec.h" + +SRes Lzma86_GetUnpackSize(const Byte *src, SizeT srcLen, UInt64 *unpackSize) +{ + unsigned i; + if (srcLen < LZMA86_HEADER_SIZE) + return SZ_ERROR_INPUT_EOF; + *unpackSize = 0; + for (i = 0; i < sizeof(UInt64); i++) + *unpackSize += ((UInt64)src[LZMA86_SIZE_OFFSET + i]) << (8 * i); + return SZ_OK; +} + +SRes Lzma86_Decode(Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen) +{ + SRes res; + int useFilter; + SizeT inSizePure; + ELzmaStatus status; + + if (*srcLen < LZMA86_HEADER_SIZE) + return SZ_ERROR_INPUT_EOF; + + useFilter = src[0]; + + if (useFilter > 1) + { + *destLen = 0; + return SZ_ERROR_UNSUPPORTED; + } + + inSizePure = *srcLen - LZMA86_HEADER_SIZE; + res = LzmaDecode(dest, destLen, src + LZMA86_HEADER_SIZE, &inSizePure, + src + 1, LZMA_PROPS_SIZE, LZMA_FINISH_ANY, &status, &g_Alloc); + *srcLen = inSizePure + LZMA86_HEADER_SIZE; + if (res != SZ_OK) + return res; + if (useFilter == 1) + { + UInt32 x86State = Z7_BRANCH_CONV_ST_X86_STATE_INIT_VAL; + z7_BranchConvSt_X86_Dec(dest, *destLen, 0, &x86State); + } + return SZ_OK; +} diff --git a/libraries/lzma/C/Lzma86Enc.c b/libraries/lzma/C/Lzma86Enc.c new file mode 100644 index 000000000..0cdde1c9a --- /dev/null +++ b/libraries/lzma/C/Lzma86Enc.c @@ -0,0 +1,103 @@ +/* Lzma86Enc.c -- LZMA + x86 (BCJ) Filter Encoder +2023-03-03 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include + +#include "Lzma86.h" + +#include "Alloc.h" +#include "Bra.h" +#include "LzmaEnc.h" + +int Lzma86_Encode(Byte *dest, size_t *destLen, const Byte *src, size_t srcLen, + int level, UInt32 dictSize, int filterMode) +{ + size_t outSize2 = *destLen; + Byte *filteredStream; + BoolInt useFilter; + int mainResult = SZ_ERROR_OUTPUT_EOF; + CLzmaEncProps props; + LzmaEncProps_Init(&props); + props.level = level; + props.dictSize = dictSize; + + *destLen = 0; + if (outSize2 < LZMA86_HEADER_SIZE) + return SZ_ERROR_OUTPUT_EOF; + + { + int i; + UInt64 t = srcLen; + for (i = 0; i < 8; i++, t >>= 8) + dest[LZMA86_SIZE_OFFSET + i] = (Byte)t; + } + + filteredStream = 0; + useFilter = (filterMode != SZ_FILTER_NO); + if (useFilter) + { + if (srcLen != 0) + { + filteredStream = (Byte *)MyAlloc(srcLen); + if (filteredStream == 0) + return SZ_ERROR_MEM; + memcpy(filteredStream, src, srcLen); + } + { + UInt32 x86State = Z7_BRANCH_CONV_ST_X86_STATE_INIT_VAL; + z7_BranchConvSt_X86_Enc(filteredStream, srcLen, 0, &x86State); + } + } + + { + size_t minSize = 0; + BoolInt bestIsFiltered = False; + + /* passes for SZ_FILTER_AUTO: + 0 - BCJ + LZMA + 1 - LZMA + 2 - BCJ + LZMA agaian, if pass 0 (BCJ + LZMA) is better. + */ + int numPasses = (filterMode == SZ_FILTER_AUTO) ? 3 : 1; + + int i; + for (i = 0; i < numPasses; i++) + { + size_t outSizeProcessed = outSize2 - LZMA86_HEADER_SIZE; + size_t outPropsSize = 5; + SRes curRes; + BoolInt curModeIsFiltered = (numPasses > 1 && i == numPasses - 1); + if (curModeIsFiltered && !bestIsFiltered) + break; + if (useFilter && i == 0) + curModeIsFiltered = True; + + curRes = LzmaEncode(dest + LZMA86_HEADER_SIZE, &outSizeProcessed, + curModeIsFiltered ? filteredStream : src, srcLen, + &props, dest + 1, &outPropsSize, 0, + NULL, &g_Alloc, &g_Alloc); + + if (curRes != SZ_ERROR_OUTPUT_EOF) + { + if (curRes != SZ_OK) + { + mainResult = curRes; + break; + } + if (outSizeProcessed <= minSize || mainResult != SZ_OK) + { + minSize = outSizeProcessed; + bestIsFiltered = curModeIsFiltered; + mainResult = SZ_OK; + } + } + } + dest[0] = (Byte)(bestIsFiltered ? 1 : 0); + *destLen = LZMA86_HEADER_SIZE + minSize; + } + if (useFilter) + MyFree(filteredStream); + return mainResult; +} diff --git a/libraries/lzma/C/LzmaDec.c b/libraries/lzma/C/LzmaDec.c index d6742e5af..69bb8bba9 100644 --- a/libraries/lzma/C/LzmaDec.c +++ b/libraries/lzma/C/LzmaDec.c @@ -1,5 +1,5 @@ /* LzmaDec.c -- LZMA Decoder -2021-04-01 : Igor Pavlov : Public domain */ +2023-04-07 : Igor Pavlov : Public domain */ #include "Precomp.h" @@ -8,15 +8,15 @@ /* #include "CpuArch.h" */ #include "LzmaDec.h" -#define kNumTopBits 24 -#define kTopValue ((UInt32)1 << kNumTopBits) +// #define kNumTopBits 24 +#define kTopValue ((UInt32)1 << 24) #define kNumBitModelTotalBits 11 #define kBitModelTotal (1 << kNumBitModelTotalBits) #define RC_INIT_SIZE 5 -#ifndef _LZMA_DEC_OPT +#ifndef Z7_LZMA_DEC_OPT #define kNumMoveBits 5 #define NORMALIZE if (range < kTopValue) { range <<= 8; code = (code << 8) | (*buf++); } @@ -25,14 +25,14 @@ #define UPDATE_0(p) range = bound; *(p) = (CLzmaProb)(ttt + ((kBitModelTotal - ttt) >> kNumMoveBits)); #define UPDATE_1(p) range -= bound; code -= bound; *(p) = (CLzmaProb)(ttt - (ttt >> kNumMoveBits)); #define GET_BIT2(p, i, A0, A1) IF_BIT_0(p) \ - { UPDATE_0(p); i = (i + i); A0; } else \ - { UPDATE_1(p); i = (i + i) + 1; A1; } + { UPDATE_0(p) i = (i + i); A0; } else \ + { UPDATE_1(p) i = (i + i) + 1; A1; } #define TREE_GET_BIT(probs, i) { GET_BIT2(probs + i, i, ;, ;); } #define REV_BIT(p, i, A0, A1) IF_BIT_0(p + i) \ - { UPDATE_0(p + i); A0; } else \ - { UPDATE_1(p + i); A1; } + { UPDATE_0(p + i) A0; } else \ + { UPDATE_1(p + i) A1; } #define REV_BIT_VAR( p, i, m) REV_BIT(p, i, i += m; m += m, m += m; i += m; ) #define REV_BIT_CONST(p, i, m) REV_BIT(p, i, i += m; , i += m * 2; ) #define REV_BIT_LAST( p, i, m) REV_BIT(p, i, i -= m , ; ) @@ -40,19 +40,19 @@ #define TREE_DECODE(probs, limit, i) \ { i = 1; do { TREE_GET_BIT(probs, i); } while (i < limit); i -= limit; } -/* #define _LZMA_SIZE_OPT */ +/* #define Z7_LZMA_SIZE_OPT */ -#ifdef _LZMA_SIZE_OPT +#ifdef Z7_LZMA_SIZE_OPT #define TREE_6_DECODE(probs, i) TREE_DECODE(probs, (1 << 6), i) #else #define TREE_6_DECODE(probs, i) \ { i = 1; \ - TREE_GET_BIT(probs, i); \ - TREE_GET_BIT(probs, i); \ - TREE_GET_BIT(probs, i); \ - TREE_GET_BIT(probs, i); \ - TREE_GET_BIT(probs, i); \ - TREE_GET_BIT(probs, i); \ + TREE_GET_BIT(probs, i) \ + TREE_GET_BIT(probs, i) \ + TREE_GET_BIT(probs, i) \ + TREE_GET_BIT(probs, i) \ + TREE_GET_BIT(probs, i) \ + TREE_GET_BIT(probs, i) \ i -= 0x40; } #endif @@ -64,25 +64,25 @@ probLit = prob + (offs + bit + symbol); \ GET_BIT2(probLit, symbol, offs ^= bit; , ;) -#endif // _LZMA_DEC_OPT +#endif // Z7_LZMA_DEC_OPT #define NORMALIZE_CHECK if (range < kTopValue) { if (buf >= bufLimit) return DUMMY_INPUT_EOF; range <<= 8; code = (code << 8) | (*buf++); } -#define IF_BIT_0_CHECK(p) ttt = *(p); NORMALIZE_CHECK; bound = (range >> kNumBitModelTotalBits) * (UInt32)ttt; if (code < bound) +#define IF_BIT_0_CHECK(p) ttt = *(p); NORMALIZE_CHECK bound = (range >> kNumBitModelTotalBits) * (UInt32)ttt; if (code < bound) #define UPDATE_0_CHECK range = bound; #define UPDATE_1_CHECK range -= bound; code -= bound; #define GET_BIT2_CHECK(p, i, A0, A1) IF_BIT_0_CHECK(p) \ - { UPDATE_0_CHECK; i = (i + i); A0; } else \ - { UPDATE_1_CHECK; i = (i + i) + 1; A1; } + { UPDATE_0_CHECK i = (i + i); A0; } else \ + { UPDATE_1_CHECK i = (i + i) + 1; A1; } #define GET_BIT_CHECK(p, i) GET_BIT2_CHECK(p, i, ; , ;) #define TREE_DECODE_CHECK(probs, limit, i) \ { i = 1; do { GET_BIT_CHECK(probs + i, i) } while (i < limit); i -= limit; } #define REV_BIT_CHECK(p, i, m) IF_BIT_0_CHECK(p + i) \ - { UPDATE_0_CHECK; i += m; m += m; } else \ - { UPDATE_1_CHECK; m += m; i += m; } + { UPDATE_0_CHECK i += m; m += m; } else \ + { UPDATE_1_CHECK m += m; i += m; } #define kNumPosBitsMax 4 @@ -224,14 +224,14 @@ Out: */ -#ifdef _LZMA_DEC_OPT +#ifdef Z7_LZMA_DEC_OPT -int MY_FAST_CALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit); +int Z7_FASTCALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit); #else static -int MY_FAST_CALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit) +int Z7_FASTCALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit) { CLzmaProb *probs = GET_PROBS; unsigned state = (unsigned)p->state; @@ -263,7 +263,7 @@ int MY_FAST_CALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit IF_BIT_0(prob) { unsigned symbol; - UPDATE_0(prob); + UPDATE_0(prob) prob = probs + Literal; if (processedPos != 0 || checkDicSize != 0) prob += (UInt32)3 * ((((processedPos << 8) + dic[(dicPos == 0 ? dicBufSize : dicPos) - 1]) & lpMask) << lc); @@ -273,7 +273,7 @@ int MY_FAST_CALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit { state -= (state < 4) ? state : 3; symbol = 1; - #ifdef _LZMA_SIZE_OPT + #ifdef Z7_LZMA_SIZE_OPT do { NORMAL_LITER_DEC } while (symbol < 0x100); #else NORMAL_LITER_DEC @@ -292,7 +292,7 @@ int MY_FAST_CALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit unsigned offs = 0x100; state -= (state < 10) ? 3 : 6; symbol = 1; - #ifdef _LZMA_SIZE_OPT + #ifdef Z7_LZMA_SIZE_OPT do { unsigned bit; @@ -321,25 +321,25 @@ int MY_FAST_CALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit } { - UPDATE_1(prob); + UPDATE_1(prob) prob = probs + IsRep + state; IF_BIT_0(prob) { - UPDATE_0(prob); + UPDATE_0(prob) state += kNumStates; prob = probs + LenCoder; } else { - UPDATE_1(prob); + UPDATE_1(prob) prob = probs + IsRepG0 + state; IF_BIT_0(prob) { - UPDATE_0(prob); + UPDATE_0(prob) prob = probs + IsRep0Long + COMBINED_PS_STATE; IF_BIT_0(prob) { - UPDATE_0(prob); + UPDATE_0(prob) // that case was checked before with kBadRepCode // if (checkDicSize == 0 && processedPos == 0) { len = kMatchSpecLen_Error_Data + 1; break; } @@ -353,30 +353,30 @@ int MY_FAST_CALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit state = state < kNumLitStates ? 9 : 11; continue; } - UPDATE_1(prob); + UPDATE_1(prob) } else { UInt32 distance; - UPDATE_1(prob); + UPDATE_1(prob) prob = probs + IsRepG1 + state; IF_BIT_0(prob) { - UPDATE_0(prob); + UPDATE_0(prob) distance = rep1; } else { - UPDATE_1(prob); + UPDATE_1(prob) prob = probs + IsRepG2 + state; IF_BIT_0(prob) { - UPDATE_0(prob); + UPDATE_0(prob) distance = rep2; } else { - UPDATE_1(prob); + UPDATE_1(prob) distance = rep3; rep3 = rep2; } @@ -389,37 +389,37 @@ int MY_FAST_CALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit prob = probs + RepLenCoder; } - #ifdef _LZMA_SIZE_OPT + #ifdef Z7_LZMA_SIZE_OPT { unsigned lim, offset; CLzmaProb *probLen = prob + LenChoice; IF_BIT_0(probLen) { - UPDATE_0(probLen); + UPDATE_0(probLen) probLen = prob + LenLow + GET_LEN_STATE; offset = 0; lim = (1 << kLenNumLowBits); } else { - UPDATE_1(probLen); + UPDATE_1(probLen) probLen = prob + LenChoice2; IF_BIT_0(probLen) { - UPDATE_0(probLen); + UPDATE_0(probLen) probLen = prob + LenLow + GET_LEN_STATE + (1 << kLenNumLowBits); offset = kLenNumLowSymbols; lim = (1 << kLenNumLowBits); } else { - UPDATE_1(probLen); + UPDATE_1(probLen) probLen = prob + LenHigh; offset = kLenNumLowSymbols * 2; lim = (1 << kLenNumHighBits); } } - TREE_DECODE(probLen, lim, len); + TREE_DECODE(probLen, lim, len) len += offset; } #else @@ -427,32 +427,32 @@ int MY_FAST_CALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit CLzmaProb *probLen = prob + LenChoice; IF_BIT_0(probLen) { - UPDATE_0(probLen); + UPDATE_0(probLen) probLen = prob + LenLow + GET_LEN_STATE; len = 1; - TREE_GET_BIT(probLen, len); - TREE_GET_BIT(probLen, len); - TREE_GET_BIT(probLen, len); + TREE_GET_BIT(probLen, len) + TREE_GET_BIT(probLen, len) + TREE_GET_BIT(probLen, len) len -= 8; } else { - UPDATE_1(probLen); + UPDATE_1(probLen) probLen = prob + LenChoice2; IF_BIT_0(probLen) { - UPDATE_0(probLen); + UPDATE_0(probLen) probLen = prob + LenLow + GET_LEN_STATE + (1 << kLenNumLowBits); len = 1; - TREE_GET_BIT(probLen, len); - TREE_GET_BIT(probLen, len); - TREE_GET_BIT(probLen, len); + TREE_GET_BIT(probLen, len) + TREE_GET_BIT(probLen, len) + TREE_GET_BIT(probLen, len) } else { - UPDATE_1(probLen); + UPDATE_1(probLen) probLen = prob + LenHigh; - TREE_DECODE(probLen, (1 << kLenNumHighBits), len); + TREE_DECODE(probLen, (1 << kLenNumHighBits), len) len += kLenNumLowSymbols * 2; } } @@ -464,7 +464,7 @@ int MY_FAST_CALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit UInt32 distance; prob = probs + PosSlot + ((len < kNumLenToPosStates ? len : kNumLenToPosStates - 1) << kNumPosSlotBits); - TREE_6_DECODE(prob, distance); + TREE_6_DECODE(prob, distance) if (distance >= kStartPosModelIndex) { unsigned posSlot = (unsigned)distance; @@ -479,7 +479,7 @@ int MY_FAST_CALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit distance++; do { - REV_BIT_VAR(prob, distance, m); + REV_BIT_VAR(prob, distance, m) } while (--numDirectBits); distance -= m; @@ -514,10 +514,10 @@ int MY_FAST_CALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit distance <<= kNumAlignBits; { unsigned i = 1; - REV_BIT_CONST(prob, i, 1); - REV_BIT_CONST(prob, i, 2); - REV_BIT_CONST(prob, i, 4); - REV_BIT_LAST (prob, i, 8); + REV_BIT_CONST(prob, i, 1) + REV_BIT_CONST(prob, i, 2) + REV_BIT_CONST(prob, i, 4) + REV_BIT_LAST (prob, i, 8) distance |= i; } if (distance == (UInt32)0xFFFFFFFF) @@ -592,7 +592,7 @@ int MY_FAST_CALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit } while (dicPos < limit && buf < bufLimit); - NORMALIZE; + NORMALIZE p->buf = buf; p->range = range; @@ -613,7 +613,7 @@ int MY_FAST_CALL LZMA_DECODE_REAL(CLzmaDec *p, SizeT limit, const Byte *bufLimit -static void MY_FAST_CALL LzmaDec_WriteRem(CLzmaDec *p, SizeT limit) +static void Z7_FASTCALL LzmaDec_WriteRem(CLzmaDec *p, SizeT limit) { unsigned len = (unsigned)p->remainLen; if (len == 0 /* || len >= kMatchSpecLenStart */) @@ -683,7 +683,7 @@ and we support the following state of (p->checkDicSize): (p->checkDicSize == p->prop.dicSize) */ -static int MY_FAST_CALL LzmaDec_DecodeReal2(CLzmaDec *p, SizeT limit, const Byte *bufLimit) +static int Z7_FASTCALL LzmaDec_DecodeReal2(CLzmaDec *p, SizeT limit, const Byte *bufLimit) { if (p->checkDicSize == 0) { @@ -767,54 +767,54 @@ static ELzmaDummy LzmaDec_TryDummy(const CLzmaDec *p, const Byte *buf, const Byt else { unsigned len; - UPDATE_1_CHECK; + UPDATE_1_CHECK prob = probs + IsRep + state; IF_BIT_0_CHECK(prob) { - UPDATE_0_CHECK; + UPDATE_0_CHECK state = 0; prob = probs + LenCoder; res = DUMMY_MATCH; } else { - UPDATE_1_CHECK; + UPDATE_1_CHECK res = DUMMY_REP; prob = probs + IsRepG0 + state; IF_BIT_0_CHECK(prob) { - UPDATE_0_CHECK; + UPDATE_0_CHECK prob = probs + IsRep0Long + COMBINED_PS_STATE; IF_BIT_0_CHECK(prob) { - UPDATE_0_CHECK; + UPDATE_0_CHECK break; } else { - UPDATE_1_CHECK; + UPDATE_1_CHECK } } else { - UPDATE_1_CHECK; + UPDATE_1_CHECK prob = probs + IsRepG1 + state; IF_BIT_0_CHECK(prob) { - UPDATE_0_CHECK; + UPDATE_0_CHECK } else { - UPDATE_1_CHECK; + UPDATE_1_CHECK prob = probs + IsRepG2 + state; IF_BIT_0_CHECK(prob) { - UPDATE_0_CHECK; + UPDATE_0_CHECK } else { - UPDATE_1_CHECK; + UPDATE_1_CHECK } } } @@ -826,31 +826,31 @@ static ELzmaDummy LzmaDec_TryDummy(const CLzmaDec *p, const Byte *buf, const Byt const CLzmaProb *probLen = prob + LenChoice; IF_BIT_0_CHECK(probLen) { - UPDATE_0_CHECK; + UPDATE_0_CHECK probLen = prob + LenLow + GET_LEN_STATE; offset = 0; limit = 1 << kLenNumLowBits; } else { - UPDATE_1_CHECK; + UPDATE_1_CHECK probLen = prob + LenChoice2; IF_BIT_0_CHECK(probLen) { - UPDATE_0_CHECK; + UPDATE_0_CHECK probLen = prob + LenLow + GET_LEN_STATE + (1 << kLenNumLowBits); offset = kLenNumLowSymbols; limit = 1 << kLenNumLowBits; } else { - UPDATE_1_CHECK; + UPDATE_1_CHECK probLen = prob + LenHigh; offset = kLenNumLowSymbols * 2; limit = 1 << kLenNumHighBits; } } - TREE_DECODE_CHECK(probLen, limit, len); + TREE_DECODE_CHECK(probLen, limit, len) len += offset; } @@ -860,7 +860,7 @@ static ELzmaDummy LzmaDec_TryDummy(const CLzmaDec *p, const Byte *buf, const Byt prob = probs + PosSlot + ((len < kNumLenToPosStates - 1 ? len : kNumLenToPosStates - 1) << kNumPosSlotBits); - TREE_DECODE_CHECK(prob, 1 << kNumPosSlotBits, posSlot); + TREE_DECODE_CHECK(prob, 1 << kNumPosSlotBits, posSlot) if (posSlot >= kStartPosModelIndex) { unsigned numDirectBits = ((posSlot >> 1) - 1); @@ -888,7 +888,7 @@ static ELzmaDummy LzmaDec_TryDummy(const CLzmaDec *p, const Byte *buf, const Byt unsigned m = 1; do { - REV_BIT_CHECK(prob, i, m); + REV_BIT_CHECK(prob, i, m) } while (--numDirectBits); } @@ -897,7 +897,7 @@ static ELzmaDummy LzmaDec_TryDummy(const CLzmaDec *p, const Byte *buf, const Byt } break; } - NORMALIZE_CHECK; + NORMALIZE_CHECK *bufOut = buf; return res; @@ -943,7 +943,7 @@ When the decoder lookahead, and the lookahead symbol is not end_marker, we have */ -#define RETURN__NOT_FINISHED__FOR_FINISH \ +#define RETURN_NOT_FINISHED_FOR_FINISH \ *status = LZMA_STATUS_NOT_FINISHED; \ return SZ_ERROR_DATA; // for strict mode // return SZ_OK; // for relaxed mode @@ -1029,7 +1029,7 @@ SRes LzmaDec_DecodeToDic(CLzmaDec *p, SizeT dicLimit, const Byte *src, SizeT *sr } if (p->remainLen != 0) { - RETURN__NOT_FINISHED__FOR_FINISH; + RETURN_NOT_FINISHED_FOR_FINISH } checkEndMarkNow = 1; } @@ -1072,7 +1072,7 @@ SRes LzmaDec_DecodeToDic(CLzmaDec *p, SizeT dicLimit, const Byte *src, SizeT *sr for (i = 0; i < (unsigned)dummyProcessed; i++) p->tempBuf[i] = src[i]; // p->remainLen = kMatchSpecLen_Error_Data; - RETURN__NOT_FINISHED__FOR_FINISH; + RETURN_NOT_FINISHED_FOR_FINISH } bufLimit = src; @@ -1150,7 +1150,7 @@ SRes LzmaDec_DecodeToDic(CLzmaDec *p, SizeT dicLimit, const Byte *src, SizeT *sr (*srcLen) += (unsigned)dummyProcessed - p->tempBufSize; p->tempBufSize = (unsigned)dummyProcessed; // p->remainLen = kMatchSpecLen_Error_Data; - RETURN__NOT_FINISHED__FOR_FINISH; + RETURN_NOT_FINISHED_FOR_FINISH } } @@ -1299,8 +1299,8 @@ static SRes LzmaDec_AllocateProbs2(CLzmaDec *p, const CLzmaProps *propNew, ISzAl SRes LzmaDec_AllocateProbs(CLzmaDec *p, const Byte *props, unsigned propsSize, ISzAllocPtr alloc) { CLzmaProps propNew; - RINOK(LzmaProps_Decode(&propNew, props, propsSize)); - RINOK(LzmaDec_AllocateProbs2(p, &propNew, alloc)); + RINOK(LzmaProps_Decode(&propNew, props, propsSize)) + RINOK(LzmaDec_AllocateProbs2(p, &propNew, alloc)) p->prop = propNew; return SZ_OK; } @@ -1309,14 +1309,14 @@ SRes LzmaDec_Allocate(CLzmaDec *p, const Byte *props, unsigned propsSize, ISzAll { CLzmaProps propNew; SizeT dicBufSize; - RINOK(LzmaProps_Decode(&propNew, props, propsSize)); - RINOK(LzmaDec_AllocateProbs2(p, &propNew, alloc)); + RINOK(LzmaProps_Decode(&propNew, props, propsSize)) + RINOK(LzmaDec_AllocateProbs2(p, &propNew, alloc)) { UInt32 dictSize = propNew.dicSize; SizeT mask = ((UInt32)1 << 12) - 1; if (dictSize >= ((UInt32)1 << 30)) mask = ((UInt32)1 << 22) - 1; - else if (dictSize >= ((UInt32)1 << 22)) mask = ((UInt32)1 << 20) - 1;; + else if (dictSize >= ((UInt32)1 << 22)) mask = ((UInt32)1 << 20) - 1; dicBufSize = ((SizeT)dictSize + mask) & ~mask; if (dicBufSize < dictSize) dicBufSize = dictSize; @@ -1348,8 +1348,8 @@ SRes LzmaDecode(Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen, *status = LZMA_STATUS_NOT_SPECIFIED; if (inSize < RC_INIT_SIZE) return SZ_ERROR_INPUT_EOF; - LzmaDec_Construct(&p); - RINOK(LzmaDec_AllocateProbs(&p, propData, propSize, alloc)); + LzmaDec_CONSTRUCT(&p) + RINOK(LzmaDec_AllocateProbs(&p, propData, propSize, alloc)) p.dic = dest; p.dicBufSize = outSize; LzmaDec_Init(&p); diff --git a/libraries/lzma/C/LzmaDec.h b/libraries/lzma/C/LzmaDec.h index 6f1296250..b0ce28fa0 100644 --- a/libraries/lzma/C/LzmaDec.h +++ b/libraries/lzma/C/LzmaDec.h @@ -1,19 +1,19 @@ /* LzmaDec.h -- LZMA Decoder -2020-03-19 : Igor Pavlov : Public domain */ +2023-04-02 : Igor Pavlov : Public domain */ -#ifndef __LZMA_DEC_H -#define __LZMA_DEC_H +#ifndef ZIP7_INC_LZMA_DEC_H +#define ZIP7_INC_LZMA_DEC_H #include "7zTypes.h" EXTERN_C_BEGIN -/* #define _LZMA_PROB32 */ -/* _LZMA_PROB32 can increase the speed on some CPUs, +/* #define Z7_LZMA_PROB32 */ +/* Z7_LZMA_PROB32 can increase the speed on some CPUs, but memory usage for CLzmaDec::probs will be doubled in that case */ typedef -#ifdef _LZMA_PROB32 +#ifdef Z7_LZMA_PROB32 UInt32 #else UInt16 @@ -25,7 +25,7 @@ typedef #define LZMA_PROPS_SIZE 5 -typedef struct _CLzmaProps +typedef struct { Byte lc; Byte lp; @@ -73,7 +73,8 @@ typedef struct Byte tempBuf[LZMA_REQUIRED_INPUT_MAX]; } CLzmaDec; -#define LzmaDec_Construct(p) { (p)->dic = NULL; (p)->probs = NULL; } +#define LzmaDec_CONSTRUCT(p) { (p)->dic = NULL; (p)->probs = NULL; } +#define LzmaDec_Construct(p) LzmaDec_CONSTRUCT(p) void LzmaDec_Init(CLzmaDec *p); diff --git a/libraries/lzma/C/LzmaEnc.c b/libraries/lzma/C/LzmaEnc.c index b04a7b7b7..6d13cac8c 100644 --- a/libraries/lzma/C/LzmaEnc.c +++ b/libraries/lzma/C/LzmaEnc.c @@ -1,5 +1,5 @@ /* LzmaEnc.c -- LZMA Encoder -2021-11-18: Igor Pavlov : Public domain */ +2023-04-13: Igor Pavlov : Public domain */ #include "Precomp.h" @@ -16,22 +16,22 @@ #include "LzmaEnc.h" #include "LzFind.h" -#ifndef _7ZIP_ST +#ifndef Z7_ST #include "LzFindMt.h" #endif /* the following LzmaEnc_* declarations is internal LZMA interface for LZMA2 encoder */ -SRes LzmaEnc_PrepareForLzma2(CLzmaEncHandle pp, ISeqInStream *inStream, UInt32 keepWindowSize, +SRes LzmaEnc_PrepareForLzma2(CLzmaEncHandle p, ISeqInStreamPtr inStream, UInt32 keepWindowSize, ISzAllocPtr alloc, ISzAllocPtr allocBig); -SRes LzmaEnc_MemPrepare(CLzmaEncHandle pp, const Byte *src, SizeT srcLen, +SRes LzmaEnc_MemPrepare(CLzmaEncHandle p, const Byte *src, SizeT srcLen, UInt32 keepWindowSize, ISzAllocPtr alloc, ISzAllocPtr allocBig); -SRes LzmaEnc_CodeOneMemBlock(CLzmaEncHandle pp, BoolInt reInit, +SRes LzmaEnc_CodeOneMemBlock(CLzmaEncHandle p, 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); +const Byte *LzmaEnc_GetCurBuf(CLzmaEncHandle p); +void LzmaEnc_Finish(CLzmaEncHandle p); +void LzmaEnc_SaveState(CLzmaEncHandle p); +void LzmaEnc_RestoreState(CLzmaEncHandle p); #ifdef SHOW_STAT static unsigned g_STAT_OFFSET = 0; @@ -40,8 +40,8 @@ static unsigned g_STAT_OFFSET = 0; /* for good normalization speed we still reserve 256 MB before 4 GB range */ #define kLzmaMaxHistorySize ((UInt32)15 << 28) -#define kNumTopBits 24 -#define kTopValue ((UInt32)1 << kNumTopBits) +// #define kNumTopBits 24 +#define kTopValue ((UInt32)1 << 24) #define kNumBitModelTotalBits 11 #define kBitModelTotal (1 << kNumBitModelTotalBits) @@ -60,6 +60,7 @@ void LzmaEncProps_Init(CLzmaEncProps *p) 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->numHashOutBits = 0; p->writeEndMark = 0; p->affinity = 0; } @@ -99,7 +100,7 @@ void LzmaEncProps_Normalize(CLzmaEncProps *p) if (p->numThreads < 0) p->numThreads = - #ifndef _7ZIP_ST + #ifndef Z7_ST ((p->btMode && p->algo) ? 2 : 1); #else 1; @@ -293,7 +294,7 @@ typedef struct #define kNumFullDistances (1 << (kEndPosModelIndex >> 1)) typedef -#ifdef _LZMA_PROB32 +#ifdef Z7_LZMA_PROB32 UInt32 #else UInt16 @@ -350,7 +351,7 @@ typedef struct Byte *buf; Byte *bufLim; Byte *bufBase; - ISeqOutStream *outStream; + ISeqOutStreamPtr outStream; UInt64 processed; SRes res; } CRangeEnc; @@ -383,7 +384,7 @@ typedef struct typedef UInt32 CProbPrice; -typedef struct +struct CLzmaEnc { void *matchFinderObj; IMatchFinder2 matchFinder; @@ -426,7 +427,7 @@ typedef struct UInt32 dictSize; SRes result; - #ifndef _7ZIP_ST + #ifndef Z7_ST BoolInt mtMode; // begin of CMatchFinderMt is used in LZ thread CMatchFinderMt matchFinderMt; @@ -439,7 +440,7 @@ typedef struct // we suppose that we have 8-bytes alignment after CMatchFinder - #ifndef _7ZIP_ST + #ifndef Z7_ST Byte pad[128]; #endif @@ -479,77 +480,59 @@ typedef struct CSaveState saveState; // BoolInt mf_Failure; - #ifndef _7ZIP_ST + #ifndef Z7_ST Byte pad2[128]; #endif -} CLzmaEnc; +}; #define MFB (p->matchFinderBase) /* -#ifndef _7ZIP_ST +#ifndef Z7_ST #define MFB (p->matchFinderMt.MatchFinder) #endif */ -#define COPY_ARR(dest, src, arr) memcpy(dest->arr, src->arr, sizeof(src->arr)); +// #define GET_CLzmaEnc_p CLzmaEnc *p = (CLzmaEnc*)(void *)p; +// #define GET_const_CLzmaEnc_p const CLzmaEnc *p = (const CLzmaEnc*)(const void *)p; -void LzmaEnc_SaveState(CLzmaEncHandle pp) +#define COPY_ARR(dest, src, arr) memcpy((dest)->arr, (src)->arr, sizeof((src)->arr)); + +#define COPY_LZMA_ENC_STATE(d, s, p) \ + (d)->state = (s)->state; \ + COPY_ARR(d, s, reps) \ + COPY_ARR(d, s, posAlignEncoder) \ + COPY_ARR(d, s, isRep) \ + COPY_ARR(d, s, isRepG0) \ + COPY_ARR(d, s, isRepG1) \ + COPY_ARR(d, s, isRepG2) \ + COPY_ARR(d, s, isMatch) \ + COPY_ARR(d, s, isRep0Long) \ + COPY_ARR(d, s, posSlotEncoder) \ + COPY_ARR(d, s, posEncoders) \ + (d)->lenProbs = (s)->lenProbs; \ + (d)->repLenProbs = (s)->repLenProbs; \ + memcpy((d)->litProbs, (s)->litProbs, ((UInt32)0x300 << (p)->lclp) * sizeof(CLzmaProb)); + +void LzmaEnc_SaveState(CLzmaEncHandle p) { - CLzmaEnc *p = (CLzmaEnc *)pp; - CSaveState *dest = &p->saveState; - - dest->state = p->state; - - dest->lenProbs = p->lenProbs; - dest->repLenProbs = p->repLenProbs; + // GET_CLzmaEnc_p + CSaveState *v = &p->saveState; + COPY_LZMA_ENC_STATE(v, p, p) +} - 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 p) +{ + // GET_CLzmaEnc_p + const CSaveState *v = &p->saveState; + COPY_LZMA_ENC_STATE(p, v, p) } -void LzmaEnc_RestoreState(CLzmaEncHandle pp) +Z7_NO_INLINE +SRes LzmaEnc_SetProps(CLzmaEncHandle p, const CLzmaEncProps *props2) { - 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; + // GET_CLzmaEnc_p CLzmaEncProps props = *props2; LzmaEncProps_Normalize(&props); @@ -585,6 +568,7 @@ SRes LzmaEnc_SetProps(CLzmaEncHandle pp, const CLzmaEncProps *props2) p->fastMode = (props.algo == 0); // p->_maxMode = True; MFB.btMode = (Byte)(props.btMode ? 1 : 0); + // MFB.btMode = (Byte)(props.btMode); { unsigned numHashBytes = 4; if (props.btMode) @@ -595,13 +579,15 @@ SRes LzmaEnc_SetProps(CLzmaEncHandle pp, const CLzmaEncProps *props2) if (props.numHashBytes >= 5) numHashBytes = 5; MFB.numHashBytes = numHashBytes; + // MFB.numHashBytes_Min = 2; + MFB.numHashOutBits = (Byte)props.numHashOutBits; } MFB.cutValue = props.mc; p->writeEndMark = (BoolInt)props.writeEndMark; - #ifndef _7ZIP_ST + #ifndef Z7_ST /* if (newMultiThread != _multiThread) { @@ -618,9 +604,9 @@ SRes LzmaEnc_SetProps(CLzmaEncHandle pp, const CLzmaEncProps *props2) } -void LzmaEnc_SetDataSize(CLzmaEncHandle pp, UInt64 expectedDataSiize) +void LzmaEnc_SetDataSize(CLzmaEncHandle p, UInt64 expectedDataSiize) { - CLzmaEnc *p = (CLzmaEnc *)pp; + // GET_CLzmaEnc_p MFB.expectedDataSize = expectedDataSiize; } @@ -684,7 +670,7 @@ static void RangeEnc_Init(CRangeEnc *p) p->res = SZ_OK; } -MY_NO_INLINE static void RangeEnc_FlushStream(CRangeEnc *p) +Z7_NO_INLINE static void RangeEnc_FlushStream(CRangeEnc *p) { const size_t num = (size_t)(p->buf - p->bufBase); if (p->res == SZ_OK) @@ -696,7 +682,7 @@ MY_NO_INLINE static void RangeEnc_FlushStream(CRangeEnc *p) p->buf = p->bufBase; } -MY_NO_INLINE static void MY_FAST_CALL RangeEnc_ShiftLow(CRangeEnc *p) +Z7_NO_INLINE static void Z7_FASTCALL RangeEnc_ShiftLow(CRangeEnc *p) { UInt32 low = (UInt32)p->low; unsigned high = (unsigned)(p->low >> 32); @@ -741,9 +727,9 @@ static void RangeEnc_FlushData(CRangeEnc *p) ttt = *(prob); \ newBound = (range >> kNumBitModelTotalBits) * ttt; -// #define _LZMA_ENC_USE_BRANCH +// #define Z7_LZMA_ENC_USE_BRANCH -#ifdef _LZMA_ENC_USE_BRANCH +#ifdef Z7_LZMA_ENC_USE_BRANCH #define RC_BIT(p, prob, bit) { \ RC_BIT_PRE(p, prob) \ @@ -811,7 +797,7 @@ static void LitEnc_Encode(CRangeEnc *p, CLzmaProb *probs, UInt32 sym) CLzmaProb *prob = probs + (sym >> 8); UInt32 bit = (sym >> 7) & 1; sym <<= 1; - RC_BIT(p, prob, bit); + RC_BIT(p, prob, bit) } while (sym < 0x10000); p->range = range; @@ -833,7 +819,7 @@ static void LitEnc_EncodeMatched(CRangeEnc *p, CLzmaProb *probs, UInt32 sym, UIn bit = (sym >> 7) & 1; sym <<= 1; offs &= ~(matchByte ^ sym); - RC_BIT(p, prob, bit); + RC_BIT(p, prob, bit) } while (sym < 0x10000); p->range = range; @@ -867,10 +853,10 @@ static void LzmaEnc_InitPriceTables(CProbPrice *ProbPrices) #define GET_PRICE(prob, bit) \ - p->ProbPrices[((prob) ^ (unsigned)(((-(int)(bit))) & (kBitModelTotal - 1))) >> kNumMoveReducingBits]; + p->ProbPrices[((prob) ^ (unsigned)(((-(int)(bit))) & (kBitModelTotal - 1))) >> kNumMoveReducingBits] #define GET_PRICEa(prob, bit) \ - ProbPrices[((prob) ^ (unsigned)((-((int)(bit))) & (kBitModelTotal - 1))) >> kNumMoveReducingBits]; + 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] @@ -921,7 +907,7 @@ static void RcTree_ReverseEncode(CRangeEnc *rc, CLzmaProb *probs, unsigned numBi unsigned bit = sym & 1; // RangeEnc_EncodeBit(rc, probs + m, bit); sym >>= 1; - RC_BIT(rc, probs + m, bit); + RC_BIT(rc, probs + m, bit) m = (m << 1) | bit; } while (--numBits); @@ -944,15 +930,15 @@ static void LenEnc_Encode(CLenEnc *p, CRangeEnc *rc, unsigned sym, unsigned posS UInt32 range, ttt, newBound; CLzmaProb *probs = p->low; range = rc->range; - RC_BIT_PRE(rc, probs); + RC_BIT_PRE(rc, probs) if (sym >= kLenNumLowSymbols) { - RC_BIT_1(rc, probs); + RC_BIT_1(rc, probs) probs += kLenNumLowSymbols; - RC_BIT_PRE(rc, probs); + RC_BIT_PRE(rc, probs) if (sym >= kLenNumLowSymbols * 2) { - RC_BIT_1(rc, probs); + 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); @@ -965,11 +951,11 @@ static void LenEnc_Encode(CLenEnc *p, CRangeEnc *rc, unsigned sym, unsigned posS { unsigned m; unsigned bit; - RC_BIT_0(rc, probs); + 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); + 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; } } @@ -990,7 +976,7 @@ static void SetPrices_3(const CLzmaProb *probs, UInt32 startPrice, UInt32 *price } -MY_NO_INLINE static void MY_FAST_CALL LenPriceEnc_UpdateTables( +Z7_NO_INLINE static void Z7_FASTCALL LenPriceEnc_UpdateTables( CLenPriceEnc *p, unsigned numPosStates, const CLenEnc *enc, @@ -1152,7 +1138,7 @@ static unsigned ReadMatchDistances(CLzmaEnc *p, unsigned *numPairsRes) + GET_PRICE_1(p->isRep[state]) \ + GET_PRICE_0(p->isRepG0[state]) -MY_FORCE_INLINE +Z7_FORCE_INLINE static UInt32 GetPrice_PureRep(const CLzmaEnc *p, unsigned repIndex, size_t state, size_t posState) { UInt32 price; @@ -1331,7 +1317,7 @@ static unsigned GetOptimum(CLzmaEnc *p, UInt32 position) LitEnc_GetPrice(probs, curByte, p->ProbPrices)); } - MakeAs_Lit(&p->opt[1]); + MakeAs_Lit(&p->opt[1]) matchPrice = GET_PRICE_1(p->isMatch[p->state][posState]); repMatchPrice = matchPrice + GET_PRICE_1(p->isRep[p->state]); @@ -1343,7 +1329,7 @@ static unsigned GetOptimum(CLzmaEnc *p, UInt32 position) if (shortRepPrice < p->opt[1].price) { p->opt[1].price = shortRepPrice; - MakeAs_ShortRep(&p->opt[1]); + MakeAs_ShortRep(&p->opt[1]) } if (last < 2) { @@ -1410,7 +1396,7 @@ static unsigned GetOptimum(CLzmaEnc *p, UInt32 position) else { unsigned slot; - GetPosSlot2(dist, slot); + GetPosSlot2(dist, slot) price += p->alignPrices[dist & kAlignMask]; price += p->posSlotPrices[lenToPosState][slot]; } @@ -1486,7 +1472,7 @@ static unsigned GetOptimum(CLzmaEnc *p, UInt32 position) unsigned delta = best - cur; if (delta != 0) { - MOVE_POS(p, delta); + MOVE_POS(p, delta) } } cur = best; @@ -1633,7 +1619,7 @@ static unsigned GetOptimum(CLzmaEnc *p, UInt32 position) { nextOpt->price = litPrice; nextOpt->len = 1; - MakeAs_Lit(nextOpt); + MakeAs_Lit(nextOpt) nextIsLit = True; } } @@ -1667,7 +1653,7 @@ static unsigned GetOptimum(CLzmaEnc *p, UInt32 position) { nextOpt->price = shortRepPrice; nextOpt->len = 1; - MakeAs_ShortRep(nextOpt); + MakeAs_ShortRep(nextOpt) nextIsLit = False; } } @@ -1871,7 +1857,7 @@ static unsigned GetOptimum(CLzmaEnc *p, UInt32 position) dist = MATCHES[(size_t)offs + 1]; // if (dist >= kNumFullDistances) - GetPosSlot2(dist, posSlot); + GetPosSlot2(dist, posSlot) for (len = /*2*/ startLen; ; len++) { @@ -1962,7 +1948,7 @@ static unsigned GetOptimum(CLzmaEnc *p, UInt32 position) break; dist = MATCHES[(size_t)offs + 1]; // if (dist >= kNumFullDistances) - GetPosSlot2(dist, posSlot); + GetPosSlot2(dist, posSlot) } } } @@ -2138,7 +2124,7 @@ static void WriteEndMarker(CLzmaEnc *p, unsigned posState) { UInt32 ttt, newBound; RC_BIT_PRE(p, probs + m) - RC_BIT_1(&p->rc, probs + m); + RC_BIT_1(&p->rc, probs + m) m = (m << 1) + 1; } while (m < (1 << kNumPosSlotBits)); @@ -2163,7 +2149,7 @@ static void WriteEndMarker(CLzmaEnc *p, unsigned posState) { UInt32 ttt, newBound; RC_BIT_PRE(p, probs + m) - RC_BIT_1(&p->rc, probs + m); + RC_BIT_1(&p->rc, probs + m) m = (m << 1) + 1; } while (m < kAlignTableSize); @@ -2179,7 +2165,7 @@ static SRes CheckErrors(CLzmaEnc *p) if (p->rc.res != SZ_OK) p->result = SZ_ERROR_WRITE; - #ifndef _7ZIP_ST + #ifndef Z7_ST if ( // p->mf_Failure || (p->mtMode && @@ -2187,7 +2173,7 @@ static SRes CheckErrors(CLzmaEnc *p) p->matchFinderMt.failure_LZ_BT)) ) { - p->result = MY_HRES_ERROR__INTERNAL_ERROR; + p->result = MY_HRES_ERROR_INTERNAL_ERROR; // printf("\nCheckErrors p->matchFinderMt.failureLZ\n"); } #endif @@ -2201,7 +2187,7 @@ static SRes CheckErrors(CLzmaEnc *p) } -MY_NO_INLINE static SRes Flush(CLzmaEnc *p, UInt32 nowPos) +Z7_NO_INLINE static SRes Flush(CLzmaEnc *p, UInt32 nowPos) { /* ReleaseMFStream(); */ p->finished = True; @@ -2213,7 +2199,7 @@ MY_NO_INLINE static SRes Flush(CLzmaEnc *p, UInt32 nowPos) } -MY_NO_INLINE static void FillAlignPrices(CLzmaEnc *p) +Z7_NO_INLINE static void FillAlignPrices(CLzmaEnc *p) { unsigned i; const CProbPrice *ProbPrices = p->ProbPrices; @@ -2237,7 +2223,7 @@ MY_NO_INLINE static void FillAlignPrices(CLzmaEnc *p) } -MY_NO_INLINE static void FillDistancesPrices(CLzmaEnc *p) +Z7_NO_INLINE static void FillDistancesPrices(CLzmaEnc *p) { // int y; for (y = 0; y < 100; y++) { @@ -2337,7 +2323,7 @@ static void LzmaEnc_Construct(CLzmaEnc *p) RangeEnc_Construct(&p->rc); MatchFinder_Construct(&MFB); - #ifndef _7ZIP_ST + #ifndef Z7_ST p->matchFinderMt.MatchFinder = &MFB; MatchFinderMt_Construct(&p->matchFinderMt); #endif @@ -2345,7 +2331,7 @@ static void LzmaEnc_Construct(CLzmaEnc *p) { CLzmaEncProps props; LzmaEncProps_Init(&props); - LzmaEnc_SetProps(p, &props); + LzmaEnc_SetProps((CLzmaEncHandle)(void *)p, &props); } #ifndef LZMA_LOG_BSR @@ -2376,7 +2362,7 @@ static void LzmaEnc_FreeLits(CLzmaEnc *p, ISzAllocPtr alloc) static void LzmaEnc_Destruct(CLzmaEnc *p, ISzAllocPtr alloc, ISzAllocPtr allocBig) { - #ifndef _7ZIP_ST + #ifndef Z7_ST MatchFinderMt_Destruct(&p->matchFinderMt, allocBig); #endif @@ -2387,21 +2373,22 @@ static void LzmaEnc_Destruct(CLzmaEnc *p, ISzAllocPtr alloc, ISzAllocPtr allocBi void LzmaEnc_Destroy(CLzmaEncHandle p, ISzAllocPtr alloc, ISzAllocPtr allocBig) { - LzmaEnc_Destruct((CLzmaEnc *)p, alloc, allocBig); + // GET_CLzmaEnc_p + LzmaEnc_Destruct(p, alloc, allocBig); ISzAlloc_Free(alloc, p); } -MY_NO_INLINE +Z7_NO_INLINE static SRes LzmaEnc_CodeOneBlock(CLzmaEnc *p, UInt32 maxPackSize, UInt32 maxUnpackSize) { UInt32 nowPos32, startPos32; if (p->needInit) { - #ifndef _7ZIP_ST + #ifndef Z7_ST if (p->mtMode) { - RINOK(MatchFinderMt_InitMt(&p->matchFinderMt)); + RINOK(MatchFinderMt_InitMt(&p->matchFinderMt)) } #endif p->matchFinder.Init(p->matchFinderObj); @@ -2410,7 +2397,7 @@ static SRes LzmaEnc_CodeOneBlock(CLzmaEnc *p, UInt32 maxPackSize, UInt32 maxUnpa if (p->finished) return p->result; - RINOK(CheckErrors(p)); + RINOK(CheckErrors(p)) nowPos32 = (UInt32)p->nowPos64; startPos32 = nowPos32; @@ -2473,7 +2460,7 @@ static SRes LzmaEnc_CodeOneBlock(CLzmaEnc *p, UInt32 maxPackSize, UInt32 maxUnpa const Byte *data; unsigned state; - RC_BIT_0(&p->rc, probs); + RC_BIT_0(&p->rc, probs) p->rc.range = range; data = p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - p->additionalOffset; probs = LIT_PROBS(nowPos32, *(data - 1)); @@ -2487,53 +2474,53 @@ static SRes LzmaEnc_CodeOneBlock(CLzmaEnc *p, UInt32 maxPackSize, UInt32 maxUnpa } else { - RC_BIT_1(&p->rc, probs); + 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); + 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); + 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); + RC_BIT_1_BASE(&p->rc, probs) } else { - RC_BIT_0_BASE(&p->rc, probs); + RC_BIT_0_BASE(&p->rc, probs) p->state = kShortRepNextStates[p->state]; } } else { - RC_BIT_1(&p->rc, probs); + 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); + RC_BIT_0_BASE(&p->rc, probs) dist = p->reps[1]; } else { - RC_BIT_1(&p->rc, probs); + 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); + RC_BIT_0_BASE(&p->rc, probs) dist = p->reps[2]; } else { - RC_BIT_1_BASE(&p->rc, probs); + RC_BIT_1_BASE(&p->rc, probs) dist = p->reps[3]; p->reps[3] = p->reps[2]; } @@ -2557,7 +2544,7 @@ static SRes LzmaEnc_CodeOneBlock(CLzmaEnc *p, UInt32 maxPackSize, UInt32 maxUnpa else { unsigned posSlot; - RC_BIT_0(&p->rc, probs); + RC_BIT_0(&p->rc, probs) p->rc.range = range; p->state = kMatchNextStates[p->state]; @@ -2571,7 +2558,7 @@ static SRes LzmaEnc_CodeOneBlock(CLzmaEnc *p, UInt32 maxPackSize, UInt32 maxUnpa p->reps[0] = dist + 1; p->matchPriceCount++; - GetPosSlot(dist, posSlot); + GetPosSlot(dist, posSlot) // RcTree_Encode_PosSlot(&p->rc, p->posSlotEncoder[GetLenToPosState(len)], posSlot); { UInt32 sym = (UInt32)posSlot + (1 << kNumPosSlotBits); @@ -2582,7 +2569,7 @@ static SRes LzmaEnc_CodeOneBlock(CLzmaEnc *p, UInt32 maxPackSize, UInt32 maxUnpa CLzmaProb *prob = probs + (sym >> kNumPosSlotBits); UInt32 bit = (sym >> (kNumPosSlotBits - 1)) & 1; sym <<= 1; - RC_BIT(&p->rc, prob, bit); + RC_BIT(&p->rc, prob, bit) } while (sym < (1 << kNumPosSlotBits * 2)); p->rc.range = range; @@ -2626,10 +2613,10 @@ static SRes LzmaEnc_CodeOneBlock(CLzmaEnc *p, UInt32 maxPackSize, UInt32 maxUnpa { 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); + 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++; } @@ -2704,7 +2691,7 @@ static SRes LzmaEnc_Alloc(CLzmaEnc *p, UInt32 keepWindowSize, ISzAllocPtr alloc, if (!RangeEnc_Alloc(&p->rc, alloc)) return SZ_ERROR_MEM; - #ifndef _7ZIP_ST + #ifndef Z7_ST p->mtMode = (p->multiThread && !p->fastMode && (MFB.btMode != 0)); #endif @@ -2748,15 +2735,14 @@ static SRes LzmaEnc_Alloc(CLzmaEnc *p, UInt32 keepWindowSize, ISzAllocPtr alloc, (numFastBytes + LZMA_MATCH_LEN_MAX + 1) */ - #ifndef _7ZIP_ST + #ifndef Z7_ST if (p->mtMode) { RINOK(MatchFinderMt_Create(&p->matchFinderMt, dictSize, beforeSize, p->numFastBytes, LZMA_MATCH_LEN_MAX + 1 /* 18.04 */ - , allocBig)); + , allocBig)) p->matchFinderObj = &p->matchFinderMt; - MFB.bigHash = (Byte)( - (p->dictSize > kBigHashDicLimit && MFB.hashMask >= 0xFFFFFF) ? 1 : 0); + MFB.bigHash = (Byte)(MFB.hashMask >= 0xFFFFFF ? 1 : 0); MatchFinderMt_CreateVTable(&p->matchFinderMt, &p->matchFinder); } else @@ -2872,59 +2858,53 @@ static SRes LzmaEnc_AllocAndInit(CLzmaEnc *p, UInt32 keepWindowSize, ISzAllocPtr p->finished = False; p->result = SZ_OK; - RINOK(LzmaEnc_Alloc(p, keepWindowSize, alloc, allocBig)); + p->nowPos64 = 0; + p->needInit = 1; + RINOK(LzmaEnc_Alloc(p, keepWindowSize, alloc, allocBig)) LzmaEnc_Init(p); LzmaEnc_InitPrices(p); - p->nowPos64 = 0; return SZ_OK; } -static SRes LzmaEnc_Prepare(CLzmaEncHandle pp, ISeqOutStream *outStream, ISeqInStream *inStream, +static SRes LzmaEnc_Prepare(CLzmaEncHandle p, + ISeqOutStreamPtr outStream, + ISeqInStreamPtr inStream, ISzAllocPtr alloc, ISzAllocPtr allocBig) { - CLzmaEnc *p = (CLzmaEnc *)pp; - MFB.stream = inStream; - p->needInit = 1; + // GET_CLzmaEnc_p + MatchFinder_SET_STREAM(&MFB, inStream) p->rc.outStream = outStream; return LzmaEnc_AllocAndInit(p, 0, alloc, allocBig); } -SRes LzmaEnc_PrepareForLzma2(CLzmaEncHandle pp, - ISeqInStream *inStream, UInt32 keepWindowSize, +SRes LzmaEnc_PrepareForLzma2(CLzmaEncHandle p, + ISeqInStreamPtr inStream, UInt32 keepWindowSize, ISzAllocPtr alloc, ISzAllocPtr allocBig) { - CLzmaEnc *p = (CLzmaEnc *)pp; - MFB.stream = inStream; - p->needInit = 1; + // GET_CLzmaEnc_p + MatchFinder_SET_STREAM(&MFB, inStream) return LzmaEnc_AllocAndInit(p, keepWindowSize, alloc, allocBig); } -static void LzmaEnc_SetInputBuf(CLzmaEnc *p, const Byte *src, SizeT srcLen) +SRes LzmaEnc_MemPrepare(CLzmaEncHandle p, + const Byte *src, SizeT srcLen, + UInt32 keepWindowSize, + ISzAllocPtr alloc, ISzAllocPtr allocBig) { - MFB.directInput = 1; - MFB.bufferBase = (Byte *)src; - MFB.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); + // GET_CLzmaEnc_p + MatchFinder_SET_DIRECT_INPUT_BUF(&MFB, src, srcLen) + LzmaEnc_SetDataSize(p, srcLen); return LzmaEnc_AllocAndInit(p, keepWindowSize, alloc, allocBig); } -void LzmaEnc_Finish(CLzmaEncHandle pp) +void LzmaEnc_Finish(CLzmaEncHandle p) { - #ifndef _7ZIP_ST - CLzmaEnc *p = (CLzmaEnc *)pp; + #ifndef Z7_ST + // GET_CLzmaEnc_p if (p->mtMode) MatchFinderMt_ReleaseStream(&p->matchFinderMt); #else - UNUSED_VAR(pp); + UNUSED_VAR(p) #endif } @@ -2933,13 +2913,13 @@ typedef struct { ISeqOutStream vt; Byte *data; - SizeT rem; + size_t rem; BoolInt overflow; } CLzmaEnc_SeqOutStreamBuf; -static size_t SeqOutStreamBuf_Write(const ISeqOutStream *pp, const void *data, size_t size) +static size_t SeqOutStreamBuf_Write(ISeqOutStreamPtr pp, const void *data, size_t size) { - CLzmaEnc_SeqOutStreamBuf *p = CONTAINER_FROM_VTBL(pp, CLzmaEnc_SeqOutStreamBuf, vt); + Z7_CONTAINER_FROM_VTBL_TO_DECL_VAR_pp_vt_p(CLzmaEnc_SeqOutStreamBuf) if (p->rem < size) { size = p->rem; @@ -2956,24 +2936,25 @@ static size_t SeqOutStreamBuf_Write(const ISeqOutStream *pp, const void *data, s /* -UInt32 LzmaEnc_GetNumAvailableBytes(CLzmaEncHandle pp) +UInt32 LzmaEnc_GetNumAvailableBytes(CLzmaEncHandle p) { - const CLzmaEnc *p = (CLzmaEnc *)pp; + GET_const_CLzmaEnc_p return p->matchFinder.GetNumAvailableBytes(p->matchFinderObj); } */ -const Byte *LzmaEnc_GetCurBuf(CLzmaEncHandle pp) +const Byte *LzmaEnc_GetCurBuf(CLzmaEncHandle p) { - const CLzmaEnc *p = (CLzmaEnc *)pp; + // GET_const_CLzmaEnc_p return p->matchFinder.GetPointerToCurrentPos(p->matchFinderObj) - p->additionalOffset; } -SRes LzmaEnc_CodeOneMemBlock(CLzmaEncHandle pp, BoolInt reInit, +// (desiredPackSize == 0) is not allowed +SRes LzmaEnc_CodeOneMemBlock(CLzmaEncHandle p, BoolInt reInit, Byte *dest, size_t *destLen, UInt32 desiredPackSize, UInt32 *unpackSize) { - CLzmaEnc *p = (CLzmaEnc *)pp; + // GET_CLzmaEnc_p UInt64 nowPos64; SRes res; CLzmaEnc_SeqOutStreamBuf outStream; @@ -2990,14 +2971,10 @@ SRes LzmaEnc_CodeOneMemBlock(CLzmaEncHandle pp, BoolInt reInit, 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; - + nowPos64 = p->nowPos64; + res = LzmaEnc_CodeOneBlock(p, desiredPackSize, *unpackSize); *unpackSize = (UInt32)(p->nowPos64 - nowPos64); @@ -3009,12 +2986,12 @@ SRes LzmaEnc_CodeOneMemBlock(CLzmaEncHandle pp, BoolInt reInit, } -MY_NO_INLINE -static SRes LzmaEnc_Encode2(CLzmaEnc *p, ICompressProgress *progress) +Z7_NO_INLINE +static SRes LzmaEnc_Encode2(CLzmaEnc *p, ICompressProgressPtr progress) { SRes res = SZ_OK; - #ifndef _7ZIP_ST + #ifndef Z7_ST Byte allocaDummy[0x300]; allocaDummy[0] = 0; allocaDummy[1] = allocaDummy[0]; @@ -3036,7 +3013,7 @@ static SRes LzmaEnc_Encode2(CLzmaEnc *p, ICompressProgress *progress) } } - LzmaEnc_Finish(p); + LzmaEnc_Finish((CLzmaEncHandle)(void *)p); /* if (res == SZ_OK && !Inline_MatchFinder_IsFinishedOK(&MFB)) @@ -3048,21 +3025,22 @@ static SRes LzmaEnc_Encode2(CLzmaEnc *p, ICompressProgress *progress) } -SRes LzmaEnc_Encode(CLzmaEncHandle pp, ISeqOutStream *outStream, ISeqInStream *inStream, ICompressProgress *progress, +SRes LzmaEnc_Encode(CLzmaEncHandle p, ISeqOutStreamPtr outStream, ISeqInStreamPtr inStream, ICompressProgressPtr progress, ISzAllocPtr alloc, ISzAllocPtr allocBig) { - RINOK(LzmaEnc_Prepare(pp, outStream, inStream, alloc, allocBig)); - return LzmaEnc_Encode2((CLzmaEnc *)pp, progress); + // GET_CLzmaEnc_p + RINOK(LzmaEnc_Prepare(p, outStream, inStream, alloc, allocBig)) + return LzmaEnc_Encode2(p, progress); } -SRes LzmaEnc_WriteProperties(CLzmaEncHandle pp, Byte *props, SizeT *size) +SRes LzmaEnc_WriteProperties(CLzmaEncHandle p, Byte *props, SizeT *size) { if (*size < LZMA_PROPS_SIZE) return SZ_ERROR_PARAM; *size = LZMA_PROPS_SIZE; { - const CLzmaEnc *p = (const CLzmaEnc *)pp; + // GET_CLzmaEnc_p const UInt32 dictSize = p->dictSize; UInt32 v; props[0] = (Byte)((p->pb * 5 + p->lp) * 9 + p->lc); @@ -3086,23 +3064,24 @@ SRes LzmaEnc_WriteProperties(CLzmaEncHandle pp, Byte *props, SizeT *size) while (v < dictSize); } - SetUi32(props + 1, v); + SetUi32(props + 1, v) return SZ_OK; } } -unsigned LzmaEnc_IsWriteEndMark(CLzmaEncHandle pp) +unsigned LzmaEnc_IsWriteEndMark(CLzmaEncHandle p) { - return (unsigned)((CLzmaEnc *)pp)->writeEndMark; + // GET_CLzmaEnc_p + return (unsigned)p->writeEndMark; } -SRes LzmaEnc_MemEncode(CLzmaEncHandle pp, Byte *dest, SizeT *destLen, const Byte *src, SizeT srcLen, - int writeEndMark, ICompressProgress *progress, ISzAllocPtr alloc, ISzAllocPtr allocBig) +SRes LzmaEnc_MemEncode(CLzmaEncHandle p, Byte *dest, SizeT *destLen, const Byte *src, SizeT srcLen, + int writeEndMark, ICompressProgressPtr progress, ISzAllocPtr alloc, ISzAllocPtr allocBig) { SRes res; - CLzmaEnc *p = (CLzmaEnc *)pp; + // GET_CLzmaEnc_p CLzmaEnc_SeqOutStreamBuf outStream; @@ -3114,7 +3093,7 @@ SRes LzmaEnc_MemEncode(CLzmaEncHandle pp, Byte *dest, SizeT *destLen, const Byte p->writeEndMark = writeEndMark; p->rc.outStream = &outStream.vt; - res = LzmaEnc_MemPrepare(pp, src, srcLen, 0, alloc, allocBig); + res = LzmaEnc_MemPrepare(p, src, srcLen, 0, alloc, allocBig); if (res == SZ_OK) { @@ -3123,7 +3102,7 @@ SRes LzmaEnc_MemEncode(CLzmaEncHandle pp, Byte *dest, SizeT *destLen, const Byte res = SZ_ERROR_FAIL; } - *destLen -= outStream.rem; + *destLen -= (SizeT)outStream.rem; if (outStream.overflow) return SZ_ERROR_OUTPUT_EOF; return res; @@ -3132,9 +3111,9 @@ SRes LzmaEnc_MemEncode(CLzmaEncHandle pp, Byte *dest, SizeT *destLen, const Byte 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) + ICompressProgressPtr progress, ISzAllocPtr alloc, ISzAllocPtr allocBig) { - CLzmaEnc *p = (CLzmaEnc *)LzmaEnc_Create(alloc); + CLzmaEncHandle p = LzmaEnc_Create(alloc); SRes res; if (!p) return SZ_ERROR_MEM; @@ -3154,10 +3133,10 @@ SRes LzmaEncode(Byte *dest, SizeT *destLen, const Byte *src, SizeT srcLen, /* -#ifndef _7ZIP_ST -void LzmaEnc_GetLzThreads(CLzmaEncHandle pp, HANDLE lz_threads[2]) +#ifndef Z7_ST +void LzmaEnc_GetLzThreads(CLzmaEncHandle p, HANDLE lz_threads[2]) { - const CLzmaEnc *p = (CLzmaEnc *)pp; + GET_const_CLzmaEnc_p lz_threads[0] = p->matchFinderMt.hashSync.thread; lz_threads[1] = p->matchFinderMt.btSync.thread; } diff --git a/libraries/lzma/C/LzmaEnc.h b/libraries/lzma/C/LzmaEnc.h index bc2ed5042..9f8039a10 100644 --- a/libraries/lzma/C/LzmaEnc.h +++ b/libraries/lzma/C/LzmaEnc.h @@ -1,8 +1,8 @@ /* LzmaEnc.h -- LZMA Encoder -2019-10-30 : Igor Pavlov : Public domain */ +2023-04-13 : Igor Pavlov : Public domain */ -#ifndef __LZMA_ENC_H -#define __LZMA_ENC_H +#ifndef ZIP7_INC_LZMA_ENC_H +#define ZIP7_INC_LZMA_ENC_H #include "7zTypes.h" @@ -10,7 +10,7 @@ EXTERN_C_BEGIN #define LZMA_PROPS_SIZE 5 -typedef struct _CLzmaEncProps +typedef struct { int level; /* 0 <= level <= 9 */ UInt32 dictSize; /* (1 << 12) <= dictSize <= (1 << 27) for 32-bit version @@ -23,10 +23,13 @@ typedef struct _CLzmaEncProps 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 */ + unsigned numHashOutBits; /* default = ? */ 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 */ + // int _pad; + UInt64 reduceSize; /* estimated size of data that will be compressed. default = (UInt64)(Int64)-1. Encoder uses this value to reduce dictionary size */ @@ -51,7 +54,9 @@ SRes: SZ_ERROR_THREAD - error in multithreading functions (only for Mt version) */ -typedef void * CLzmaEncHandle; +typedef struct CLzmaEnc CLzmaEnc; +typedef CLzmaEnc * CLzmaEncHandle; +// Z7_DECLARE_HANDLE(CLzmaEncHandle) CLzmaEncHandle LzmaEnc_Create(ISzAllocPtr alloc); void LzmaEnc_Destroy(CLzmaEncHandle p, ISzAllocPtr alloc, ISzAllocPtr allocBig); @@ -61,17 +66,17 @@ 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_Encode(CLzmaEncHandle p, ISeqOutStreamPtr outStream, ISeqInStreamPtr inStream, + ICompressProgressPtr 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); + int writeEndMark, ICompressProgressPtr 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); + ICompressProgressPtr progress, ISzAllocPtr alloc, ISzAllocPtr allocBig); EXTERN_C_END diff --git a/libraries/lzma/C/LzmaLib.c b/libraries/lzma/C/LzmaLib.c new file mode 100644 index 000000000..785e88487 --- /dev/null +++ b/libraries/lzma/C/LzmaLib.c @@ -0,0 +1,42 @@ +/* LzmaLib.c -- LZMA library wrapper +2023-04-02 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include "Alloc.h" +#include "LzmaDec.h" +#include "LzmaEnc.h" +#include "LzmaLib.h" + +Z7_STDAPI LzmaCompress(unsigned char *dest, size_t *destLen, const unsigned char *src, size_t srcLen, + unsigned char *outProps, size_t *outPropsSize, + int level, /* 0 <= level <= 9, default = 5 */ + unsigned dictSize, /* use (1 << N) or (3 << N). 4 KB < dictSize <= 128 MB */ + int lc, /* 0 <= lc <= 8, default = 3 */ + int lp, /* 0 <= lp <= 4, default = 0 */ + int pb, /* 0 <= pb <= 4, default = 2 */ + int fb, /* 5 <= fb <= 273, default = 32 */ + int numThreads /* 1 or 2, default = 2 */ +) +{ + CLzmaEncProps props; + LzmaEncProps_Init(&props); + props.level = level; + props.dictSize = dictSize; + props.lc = lc; + props.lp = lp; + props.pb = pb; + props.fb = fb; + props.numThreads = numThreads; + + return LzmaEncode(dest, destLen, src, srcLen, &props, outProps, outPropsSize, 0, + NULL, &g_Alloc, &g_Alloc); +} + + +Z7_STDAPI LzmaUncompress(unsigned char *dest, size_t *destLen, const unsigned char *src, size_t *srcLen, + const unsigned char *props, size_t propsSize) +{ + ELzmaStatus status; + return LzmaDecode(dest, destLen, src, srcLen, props, (unsigned)propsSize, LZMA_FINISH_ANY, &status, &g_Alloc); +} diff --git a/libraries/lzma/C/LzmaLib.h b/libraries/lzma/C/LzmaLib.h new file mode 100644 index 000000000..d7c0724de --- /dev/null +++ b/libraries/lzma/C/LzmaLib.h @@ -0,0 +1,138 @@ +/* LzmaLib.h -- LZMA library interface +2023-04-02 : Igor Pavlov : Public domain */ + +#ifndef ZIP7_INC_LZMA_LIB_H +#define ZIP7_INC_LZMA_LIB_H + +#include "7zTypes.h" + +EXTERN_C_BEGIN + +#define Z7_STDAPI int Z7_STDCALL + +#define LZMA_PROPS_SIZE 5 + +/* +RAM requirements for LZMA: + for compression: (dictSize * 11.5 + 6 MB) + state_size + for decompression: dictSize + state_size + state_size = (4 + (1.5 << (lc + lp))) KB + by default (lc=3, lp=0), state_size = 16 KB. + +LZMA properties (5 bytes) format + Offset Size Description + 0 1 lc, lp and pb in encoded form. + 1 4 dictSize (little endian). +*/ + +/* +LzmaCompress +------------ + +outPropsSize - + In: the pointer to the size of outProps buffer; *outPropsSize = LZMA_PROPS_SIZE = 5. + Out: the pointer to the size of written properties in outProps buffer; *outPropsSize = LZMA_PROPS_SIZE = 5. + + LZMA Encoder will use defult values for any parameter, if it is + -1 for any from: level, loc, lp, pb, fb, numThreads + 0 for dictSize + +level - compression level: 0 <= level <= 9; + + level dictSize algo fb + 0: 64 KB 0 32 + 1: 256 KB 0 32 + 2: 1 MB 0 32 + 3: 4 MB 0 32 + 4: 16 MB 0 32 + 5: 16 MB 1 32 + 6: 32 MB 1 32 + 7: 32 MB 1 64 + 8: 64 MB 1 64 + 9: 64 MB 1 64 + + The default value for "level" is 5. + + algo = 0 means fast method + algo = 1 means normal method + +dictSize - The dictionary size in bytes. The maximum value is + 128 MB = (1 << 27) bytes for 32-bit version + 1 GB = (1 << 30) bytes for 64-bit version + The default value is 16 MB = (1 << 24) bytes. + It's recommended to use the dictionary that is larger than 4 KB and + that can be calculated as (1 << N) or (3 << N) sizes. + +lc - The number of literal context bits (high bits of previous literal). + It can be in the range from 0 to 8. The default value is 3. + Sometimes lc=4 gives the gain for big files. + +lp - The number of literal pos bits (low bits of current position for literals). + It can be in the range from 0 to 4. The default value is 0. + The lp switch is intended for periodical data when the period is equal to 2^lp. + For example, for 32-bit (4 bytes) periodical data you can use lp=2. Often it's + better to set lc=0, if you change lp switch. + +pb - The number of pos bits (low bits of current position). + It can be in the range from 0 to 4. The default value is 2. + The pb switch is intended for periodical data when the period is equal 2^pb. + +fb - Word size (the number of fast bytes). + It can be in the range from 5 to 273. The default value is 32. + Usually, a big number gives a little bit better compression ratio and + slower compression process. + +numThreads - The number of thereads. 1 or 2. The default value is 2. + Fast mode (algo = 0) can use only 1 thread. + +In: + dest - output data buffer + destLen - output data buffer size + src - input data + srcLen - input data size +Out: + destLen - processed output size +Returns: + SZ_OK - OK + SZ_ERROR_MEM - Memory allocation error + SZ_ERROR_PARAM - Incorrect paramater + SZ_ERROR_OUTPUT_EOF - output buffer overflow + SZ_ERROR_THREAD - errors in multithreading functions (only for Mt version) +*/ + +Z7_STDAPI LzmaCompress(unsigned char *dest, size_t *destLen, const unsigned char *src, size_t srcLen, + unsigned char *outProps, size_t *outPropsSize, /* *outPropsSize must be = 5 */ + int level, /* 0 <= level <= 9, default = 5 */ + unsigned dictSize, /* 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 fb, /* 5 <= fb <= 273, default = 32 */ + int numThreads /* 1 or 2, default = 2 */ + ); + +/* +LzmaUncompress +-------------- +In: + dest - output data buffer + destLen - output data buffer size + src - input data + srcLen - input data size +Out: + destLen - processed output size + srcLen - processed input size +Returns: + SZ_OK - OK + SZ_ERROR_DATA - Data error + SZ_ERROR_MEM - Memory allocation arror + SZ_ERROR_UNSUPPORTED - Unsupported properties + SZ_ERROR_INPUT_EOF - it needs more bytes in input buffer (src) +*/ + +Z7_STDAPI LzmaUncompress(unsigned char *dest, size_t *destLen, const unsigned char *src, SizeT *srcLen, + const unsigned char *props, size_t propsSize); + +EXTERN_C_END + +#endif diff --git a/libraries/lzma/C/MtCoder.c b/libraries/lzma/C/MtCoder.c new file mode 100644 index 000000000..6f58abb2a --- /dev/null +++ b/libraries/lzma/C/MtCoder.c @@ -0,0 +1,571 @@ +/* MtCoder.c -- Multi-thread Coder +2023-04-13 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include "MtCoder.h" + +#ifndef Z7_ST + +static SRes MtProgressThunk_Progress(ICompressProgressPtr pp, UInt64 inSize, UInt64 outSize) +{ + Z7_CONTAINER_FROM_VTBL_TO_DECL_VAR_pp_vt_p(CMtProgressThunk) + UInt64 inSize2 = 0; + UInt64 outSize2 = 0; + if (inSize != (UInt64)(Int64)-1) + { + inSize2 = inSize - p->inSize; + p->inSize = inSize; + } + if (outSize != (UInt64)(Int64)-1) + { + outSize2 = outSize - p->outSize; + p->outSize = outSize; + } + return MtProgress_ProgressAdd(p->mtProgress, inSize2, outSize2); +} + + +void MtProgressThunk_CreateVTable(CMtProgressThunk *p) +{ + p->vt.Progress = MtProgressThunk_Progress; +} + + + +#define RINOK_THREAD(x) { if ((x) != 0) return SZ_ERROR_THREAD; } + + +static THREAD_FUNC_DECL ThreadFunc(void *pp); + + +static SRes MtCoderThread_CreateAndStart(CMtCoderThread *t) +{ + WRes wres = AutoResetEvent_OptCreate_And_Reset(&t->startEvent); + if (wres == 0) + { + t->stop = False; + if (!Thread_WasCreated(&t->thread)) + wres = Thread_Create(&t->thread, ThreadFunc, t); + if (wres == 0) + wres = Event_Set(&t->startEvent); + } + if (wres == 0) + return SZ_OK; + return MY_SRes_HRESULT_FROM_WRes(wres); +} + + +static void MtCoderThread_Destruct(CMtCoderThread *t) +{ + if (Thread_WasCreated(&t->thread)) + { + t->stop = 1; + Event_Set(&t->startEvent); + Thread_Wait_Close(&t->thread); + } + + Event_Close(&t->startEvent); + + if (t->inBuf) + { + ISzAlloc_Free(t->mtCoder->allocBig, t->inBuf); + t->inBuf = NULL; + } +} + + + + +/* + ThreadFunc2() returns: + SZ_OK - in all normal cases (even for stream error or memory allocation error) + SZ_ERROR_THREAD - in case of failure in system synch function +*/ + +static SRes ThreadFunc2(CMtCoderThread *t) +{ + CMtCoder *mtc = t->mtCoder; + + for (;;) + { + unsigned bi; + SRes res; + SRes res2; + BoolInt finished; + unsigned bufIndex; + size_t size; + const Byte *inData; + UInt64 readProcessed = 0; + + RINOK_THREAD(Event_Wait(&mtc->readEvent)) + + /* after Event_Wait(&mtc->readEvent) we must call Event_Set(&mtc->readEvent) in any case to unlock another threads */ + + if (mtc->stopReading) + { + return Event_Set(&mtc->readEvent) == 0 ? SZ_OK : SZ_ERROR_THREAD; + } + + res = MtProgress_GetError(&mtc->mtProgress); + + size = 0; + inData = NULL; + finished = True; + + if (res == SZ_OK) + { + size = mtc->blockSize; + if (mtc->inStream) + { + if (!t->inBuf) + { + t->inBuf = (Byte *)ISzAlloc_Alloc(mtc->allocBig, mtc->blockSize); + if (!t->inBuf) + res = SZ_ERROR_MEM; + } + if (res == SZ_OK) + { + res = SeqInStream_ReadMax(mtc->inStream, t->inBuf, &size); + readProcessed = mtc->readProcessed + size; + mtc->readProcessed = readProcessed; + } + if (res != SZ_OK) + { + mtc->readRes = res; + /* after reading error - we can stop encoding of previous blocks */ + MtProgress_SetError(&mtc->mtProgress, res); + } + else + finished = (size != mtc->blockSize); + } + else + { + size_t rem; + readProcessed = mtc->readProcessed; + rem = mtc->inDataSize - (size_t)readProcessed; + if (size > rem) + size = rem; + inData = mtc->inData + (size_t)readProcessed; + readProcessed += size; + mtc->readProcessed = readProcessed; + finished = (mtc->inDataSize == (size_t)readProcessed); + } + } + + /* we must get some block from blocksSemaphore before Event_Set(&mtc->readEvent) */ + + res2 = SZ_OK; + + if (Semaphore_Wait(&mtc->blocksSemaphore) != 0) + { + res2 = SZ_ERROR_THREAD; + if (res == SZ_OK) + { + res = res2; + // MtProgress_SetError(&mtc->mtProgress, res); + } + } + + bi = mtc->blockIndex; + + if (++mtc->blockIndex >= mtc->numBlocksMax) + mtc->blockIndex = 0; + + bufIndex = (unsigned)(int)-1; + + if (res == SZ_OK) + res = MtProgress_GetError(&mtc->mtProgress); + + if (res != SZ_OK) + finished = True; + + if (!finished) + { + if (mtc->numStartedThreads < mtc->numStartedThreadsLimit + && mtc->expectedDataSize != readProcessed) + { + res = MtCoderThread_CreateAndStart(&mtc->threads[mtc->numStartedThreads]); + if (res == SZ_OK) + mtc->numStartedThreads++; + else + { + MtProgress_SetError(&mtc->mtProgress, res); + finished = True; + } + } + } + + if (finished) + mtc->stopReading = True; + + RINOK_THREAD(Event_Set(&mtc->readEvent)) + + if (res2 != SZ_OK) + return res2; + + if (res == SZ_OK) + { + CriticalSection_Enter(&mtc->cs); + bufIndex = mtc->freeBlockHead; + mtc->freeBlockHead = mtc->freeBlockList[bufIndex]; + CriticalSection_Leave(&mtc->cs); + + res = mtc->mtCallback->Code(mtc->mtCallbackObject, t->index, bufIndex, + mtc->inStream ? t->inBuf : inData, size, finished); + + // MtProgress_Reinit(&mtc->mtProgress, t->index); + + if (res != SZ_OK) + MtProgress_SetError(&mtc->mtProgress, res); + } + + { + CMtCoderBlock *block = &mtc->blocks[bi]; + block->res = res; + block->bufIndex = bufIndex; + block->finished = finished; + } + + #ifdef MTCODER_USE_WRITE_THREAD + RINOK_THREAD(Event_Set(&mtc->writeEvents[bi])) + #else + { + unsigned wi; + { + CriticalSection_Enter(&mtc->cs); + wi = mtc->writeIndex; + if (wi == bi) + mtc->writeIndex = (unsigned)(int)-1; + else + mtc->ReadyBlocks[bi] = True; + CriticalSection_Leave(&mtc->cs); + } + + if (wi != bi) + { + if (res != SZ_OK || finished) + return 0; + continue; + } + + if (mtc->writeRes != SZ_OK) + res = mtc->writeRes; + + for (;;) + { + if (res == SZ_OK && bufIndex != (unsigned)(int)-1) + { + res = mtc->mtCallback->Write(mtc->mtCallbackObject, bufIndex); + if (res != SZ_OK) + { + mtc->writeRes = res; + MtProgress_SetError(&mtc->mtProgress, res); + } + } + + if (++wi >= mtc->numBlocksMax) + wi = 0; + { + BoolInt isReady; + + CriticalSection_Enter(&mtc->cs); + + if (bufIndex != (unsigned)(int)-1) + { + mtc->freeBlockList[bufIndex] = mtc->freeBlockHead; + mtc->freeBlockHead = bufIndex; + } + + isReady = mtc->ReadyBlocks[wi]; + + if (isReady) + mtc->ReadyBlocks[wi] = False; + else + mtc->writeIndex = wi; + + CriticalSection_Leave(&mtc->cs); + + RINOK_THREAD(Semaphore_Release1(&mtc->blocksSemaphore)) + + if (!isReady) + break; + } + + { + CMtCoderBlock *block = &mtc->blocks[wi]; + if (res == SZ_OK && block->res != SZ_OK) + res = block->res; + bufIndex = block->bufIndex; + finished = block->finished; + } + } + } + #endif + + if (finished || res != SZ_OK) + return 0; + } +} + + +static THREAD_FUNC_DECL ThreadFunc(void *pp) +{ + CMtCoderThread *t = (CMtCoderThread *)pp; + for (;;) + { + if (Event_Wait(&t->startEvent) != 0) + return (THREAD_FUNC_RET_TYPE)SZ_ERROR_THREAD; + if (t->stop) + return 0; + { + SRes res = ThreadFunc2(t); + CMtCoder *mtc = t->mtCoder; + if (res != SZ_OK) + { + MtProgress_SetError(&mtc->mtProgress, res); + } + + #ifndef MTCODER_USE_WRITE_THREAD + { + unsigned numFinished = (unsigned)InterlockedIncrement(&mtc->numFinishedThreads); + if (numFinished == mtc->numStartedThreads) + if (Event_Set(&mtc->finishedEvent) != 0) + return (THREAD_FUNC_RET_TYPE)SZ_ERROR_THREAD; + } + #endif + } + } +} + + + +void MtCoder_Construct(CMtCoder *p) +{ + unsigned i; + + p->blockSize = 0; + p->numThreadsMax = 0; + p->expectedDataSize = (UInt64)(Int64)-1; + + p->inStream = NULL; + p->inData = NULL; + p->inDataSize = 0; + + p->progress = NULL; + p->allocBig = NULL; + + p->mtCallback = NULL; + p->mtCallbackObject = NULL; + + p->allocatedBufsSize = 0; + + Event_Construct(&p->readEvent); + Semaphore_Construct(&p->blocksSemaphore); + + for (i = 0; i < MTCODER_THREADS_MAX; i++) + { + CMtCoderThread *t = &p->threads[i]; + t->mtCoder = p; + t->index = i; + t->inBuf = NULL; + t->stop = False; + Event_Construct(&t->startEvent); + Thread_CONSTRUCT(&t->thread) + } + + #ifdef MTCODER_USE_WRITE_THREAD + for (i = 0; i < MTCODER_BLOCKS_MAX; i++) + Event_Construct(&p->writeEvents[i]); + #else + Event_Construct(&p->finishedEvent); + #endif + + CriticalSection_Init(&p->cs); + CriticalSection_Init(&p->mtProgress.cs); +} + + + + +static void MtCoder_Free(CMtCoder *p) +{ + unsigned i; + + /* + p->stopReading = True; + if (Event_IsCreated(&p->readEvent)) + Event_Set(&p->readEvent); + */ + + for (i = 0; i < MTCODER_THREADS_MAX; i++) + MtCoderThread_Destruct(&p->threads[i]); + + Event_Close(&p->readEvent); + Semaphore_Close(&p->blocksSemaphore); + + #ifdef MTCODER_USE_WRITE_THREAD + for (i = 0; i < MTCODER_BLOCKS_MAX; i++) + Event_Close(&p->writeEvents[i]); + #else + Event_Close(&p->finishedEvent); + #endif +} + + +void MtCoder_Destruct(CMtCoder *p) +{ + MtCoder_Free(p); + + CriticalSection_Delete(&p->cs); + CriticalSection_Delete(&p->mtProgress.cs); +} + + +SRes MtCoder_Code(CMtCoder *p) +{ + unsigned numThreads = p->numThreadsMax; + unsigned numBlocksMax; + unsigned i; + SRes res = SZ_OK; + + if (numThreads > MTCODER_THREADS_MAX) + numThreads = MTCODER_THREADS_MAX; + numBlocksMax = MTCODER_GET_NUM_BLOCKS_FROM_THREADS(numThreads); + + if (p->blockSize < ((UInt32)1 << 26)) numBlocksMax++; + if (p->blockSize < ((UInt32)1 << 24)) numBlocksMax++; + if (p->blockSize < ((UInt32)1 << 22)) numBlocksMax++; + + if (numBlocksMax > MTCODER_BLOCKS_MAX) + numBlocksMax = MTCODER_BLOCKS_MAX; + + if (p->blockSize != p->allocatedBufsSize) + { + for (i = 0; i < MTCODER_THREADS_MAX; i++) + { + CMtCoderThread *t = &p->threads[i]; + if (t->inBuf) + { + ISzAlloc_Free(p->allocBig, t->inBuf); + t->inBuf = NULL; + } + } + p->allocatedBufsSize = p->blockSize; + } + + p->readRes = SZ_OK; + + MtProgress_Init(&p->mtProgress, p->progress); + + #ifdef MTCODER_USE_WRITE_THREAD + for (i = 0; i < numBlocksMax; i++) + { + RINOK_THREAD(AutoResetEvent_OptCreate_And_Reset(&p->writeEvents[i])) + } + #else + RINOK_THREAD(AutoResetEvent_OptCreate_And_Reset(&p->finishedEvent)) + #endif + + { + RINOK_THREAD(AutoResetEvent_OptCreate_And_Reset(&p->readEvent)) + RINOK_THREAD(Semaphore_OptCreateInit(&p->blocksSemaphore, numBlocksMax, numBlocksMax)) + } + + for (i = 0; i < MTCODER_BLOCKS_MAX - 1; i++) + p->freeBlockList[i] = i + 1; + p->freeBlockList[MTCODER_BLOCKS_MAX - 1] = (unsigned)(int)-1; + p->freeBlockHead = 0; + + p->readProcessed = 0; + p->blockIndex = 0; + p->numBlocksMax = numBlocksMax; + p->stopReading = False; + + #ifndef MTCODER_USE_WRITE_THREAD + p->writeIndex = 0; + p->writeRes = SZ_OK; + for (i = 0; i < MTCODER_BLOCKS_MAX; i++) + p->ReadyBlocks[i] = False; + p->numFinishedThreads = 0; + #endif + + p->numStartedThreadsLimit = numThreads; + p->numStartedThreads = 0; + + // for (i = 0; i < numThreads; i++) + { + CMtCoderThread *nextThread = &p->threads[p->numStartedThreads++]; + RINOK(MtCoderThread_CreateAndStart(nextThread)) + } + + RINOK_THREAD(Event_Set(&p->readEvent)) + + #ifdef MTCODER_USE_WRITE_THREAD + { + unsigned bi = 0; + + for (;; bi++) + { + if (bi >= numBlocksMax) + bi = 0; + + RINOK_THREAD(Event_Wait(&p->writeEvents[bi])) + + { + const CMtCoderBlock *block = &p->blocks[bi]; + unsigned bufIndex = block->bufIndex; + BoolInt finished = block->finished; + if (res == SZ_OK && block->res != SZ_OK) + res = block->res; + + if (bufIndex != (unsigned)(int)-1) + { + if (res == SZ_OK) + { + res = p->mtCallback->Write(p->mtCallbackObject, bufIndex); + if (res != SZ_OK) + MtProgress_SetError(&p->mtProgress, res); + } + + CriticalSection_Enter(&p->cs); + { + p->freeBlockList[bufIndex] = p->freeBlockHead; + p->freeBlockHead = bufIndex; + } + CriticalSection_Leave(&p->cs); + } + + RINOK_THREAD(Semaphore_Release1(&p->blocksSemaphore)) + + if (finished) + break; + } + } + } + #else + { + WRes wres = Event_Wait(&p->finishedEvent); + res = MY_SRes_HRESULT_FROM_WRes(wres); + } + #endif + + if (res == SZ_OK) + res = p->readRes; + + if (res == SZ_OK) + res = p->mtProgress.res; + + #ifndef MTCODER_USE_WRITE_THREAD + if (res == SZ_OK) + res = p->writeRes; + #endif + + if (res != SZ_OK) + MtCoder_Free(p); + return res; +} + +#endif + +#undef RINOK_THREAD diff --git a/libraries/lzma/C/MtCoder.h b/libraries/lzma/C/MtCoder.h new file mode 100644 index 000000000..1231d3c2a --- /dev/null +++ b/libraries/lzma/C/MtCoder.h @@ -0,0 +1,141 @@ +/* MtCoder.h -- Multi-thread Coder +2023-04-13 : Igor Pavlov : Public domain */ + +#ifndef ZIP7_INC_MT_CODER_H +#define ZIP7_INC_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 Z7_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 Z7_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; + + ISeqInStreamPtr inStream; + const Byte *inData; + size_t inDataSize; + + ICompressProgressPtr 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/libraries/lzma/C/MtDec.c b/libraries/lzma/C/MtDec.c new file mode 100644 index 000000000..782069926 --- /dev/null +++ b/libraries/lzma/C/MtDec.c @@ -0,0 +1,1114 @@ +/* MtDec.c -- Multi-thread Decoder +2023-04-02 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +// #define SHOW_DEBUG_INFO + +// #include +#include + +#ifdef SHOW_DEBUG_INFO +#include +#endif + +#include "MtDec.h" + +#ifndef Z7_ST + +#ifdef SHOW_DEBUG_INFO +#define PRF(x) x +#else +#define PRF(x) +#endif + +#define PRF_STR_INT(s, d) PRF(printf("\n" s " %d\n", (unsigned)d)) + +void MtProgress_Init(CMtProgress *p, ICompressProgressPtr progress) +{ + p->progress = progress; + p->res = SZ_OK; + p->totalInSize = 0; + p->totalOutSize = 0; +} + + +SRes MtProgress_Progress_ST(CMtProgress *p) +{ + if (p->res == SZ_OK && p->progress) + if (ICompressProgress_Progress(p->progress, p->totalInSize, p->totalOutSize) != SZ_OK) + p->res = SZ_ERROR_PROGRESS; + return p->res; +} + + +SRes MtProgress_ProgressAdd(CMtProgress *p, UInt64 inSize, UInt64 outSize) +{ + SRes res; + CriticalSection_Enter(&p->cs); + + p->totalInSize += inSize; + p->totalOutSize += outSize; + if (p->res == SZ_OK && p->progress) + if (ICompressProgress_Progress(p->progress, p->totalInSize, p->totalOutSize) != SZ_OK) + p->res = SZ_ERROR_PROGRESS; + res = p->res; + + CriticalSection_Leave(&p->cs); + return res; +} + + +SRes MtProgress_GetError(CMtProgress *p) +{ + SRes res; + CriticalSection_Enter(&p->cs); + res = p->res; + CriticalSection_Leave(&p->cs); + return res; +} + + +void MtProgress_SetError(CMtProgress *p, SRes res) +{ + CriticalSection_Enter(&p->cs); + if (p->res == SZ_OK) + p->res = res; + CriticalSection_Leave(&p->cs); +} + + +#define RINOK_THREAD(x) RINOK_WRes(x) + + +struct CMtDecBufLink_ +{ + struct CMtDecBufLink_ *next; + void *pad[3]; +}; + +typedef struct CMtDecBufLink_ CMtDecBufLink; + +#define MTDEC__LINK_DATA_OFFSET sizeof(CMtDecBufLink) +#define MTDEC__DATA_PTR_FROM_LINK(link) ((Byte *)(link) + MTDEC__LINK_DATA_OFFSET) + + + +static THREAD_FUNC_DECL MtDec_ThreadFunc(void *pp); + + +static WRes MtDecThread_CreateEvents(CMtDecThread *t) +{ + WRes wres = AutoResetEvent_OptCreate_And_Reset(&t->canWrite); + if (wres == 0) + { + wres = AutoResetEvent_OptCreate_And_Reset(&t->canRead); + if (wres == 0) + return SZ_OK; + } + return wres; +} + + +static SRes MtDecThread_CreateAndStart(CMtDecThread *t) +{ + WRes wres = MtDecThread_CreateEvents(t); + // wres = 17; // for test + if (wres == 0) + { + if (Thread_WasCreated(&t->thread)) + return SZ_OK; + wres = Thread_Create(&t->thread, MtDec_ThreadFunc, t); + if (wres == 0) + return SZ_OK; + } + return MY_SRes_HRESULT_FROM_WRes(wres); +} + + +void MtDecThread_FreeInBufs(CMtDecThread *t) +{ + if (t->inBuf) + { + void *link = t->inBuf; + t->inBuf = NULL; + do + { + void *next = ((CMtDecBufLink *)link)->next; + ISzAlloc_Free(t->mtDec->alloc, link); + link = next; + } + while (link); + } +} + + +static void MtDecThread_CloseThread(CMtDecThread *t) +{ + if (Thread_WasCreated(&t->thread)) + { + Event_Set(&t->canWrite); /* we can disable it. There are no threads waiting canWrite in normal cases */ + Event_Set(&t->canRead); + Thread_Wait_Close(&t->thread); + } + + Event_Close(&t->canRead); + Event_Close(&t->canWrite); +} + +static void MtDec_CloseThreads(CMtDec *p) +{ + unsigned i; + for (i = 0; i < MTDEC_THREADS_MAX; i++) + MtDecThread_CloseThread(&p->threads[i]); +} + +static void MtDecThread_Destruct(CMtDecThread *t) +{ + MtDecThread_CloseThread(t); + MtDecThread_FreeInBufs(t); +} + + + +static SRes MtDec_GetError_Spec(CMtDec *p, UInt64 interruptIndex, BoolInt *wasInterrupted) +{ + SRes res; + CriticalSection_Enter(&p->mtProgress.cs); + *wasInterrupted = (p->needInterrupt && interruptIndex > p->interruptIndex); + res = p->mtProgress.res; + CriticalSection_Leave(&p->mtProgress.cs); + return res; +} + +static SRes MtDec_Progress_GetError_Spec(CMtDec *p, UInt64 inSize, UInt64 outSize, UInt64 interruptIndex, BoolInt *wasInterrupted) +{ + SRes res; + CriticalSection_Enter(&p->mtProgress.cs); + + p->mtProgress.totalInSize += inSize; + p->mtProgress.totalOutSize += outSize; + if (p->mtProgress.res == SZ_OK && p->mtProgress.progress) + if (ICompressProgress_Progress(p->mtProgress.progress, p->mtProgress.totalInSize, p->mtProgress.totalOutSize) != SZ_OK) + p->mtProgress.res = SZ_ERROR_PROGRESS; + + *wasInterrupted = (p->needInterrupt && interruptIndex > p->interruptIndex); + res = p->mtProgress.res; + + CriticalSection_Leave(&p->mtProgress.cs); + + return res; +} + +static void MtDec_Interrupt(CMtDec *p, UInt64 interruptIndex) +{ + CriticalSection_Enter(&p->mtProgress.cs); + if (!p->needInterrupt || interruptIndex < p->interruptIndex) + { + p->interruptIndex = interruptIndex; + p->needInterrupt = True; + } + CriticalSection_Leave(&p->mtProgress.cs); +} + +Byte *MtDec_GetCrossBuff(CMtDec *p) +{ + Byte *cr = p->crossBlock; + if (!cr) + { + cr = (Byte *)ISzAlloc_Alloc(p->alloc, MTDEC__LINK_DATA_OFFSET + p->inBufSize); + if (!cr) + return NULL; + p->crossBlock = cr; + } + return MTDEC__DATA_PTR_FROM_LINK(cr); +} + + +/* + MtDec_ThreadFunc2() returns: + 0 - in all normal cases (even for stream error or memory allocation error) + (!= 0) - WRes error return by system threading function +*/ + +// #define MTDEC_ProgessStep (1 << 22) +#define MTDEC_ProgessStep (1 << 0) + +static WRes MtDec_ThreadFunc2(CMtDecThread *t) +{ + CMtDec *p = t->mtDec; + + PRF_STR_INT("MtDec_ThreadFunc2", t->index) + + // SetThreadAffinityMask(GetCurrentThread(), 1 << t->index); + + for (;;) + { + SRes res, codeRes; + BoolInt wasInterrupted, isAllocError, overflow, finish; + SRes threadingErrorSRes; + BoolInt needCode, needWrite, needContinue; + + size_t inDataSize_Start; + UInt64 inDataSize; + // UInt64 inDataSize_Full; + + UInt64 blockIndex; + + UInt64 inPrev = 0; + UInt64 outPrev = 0; + UInt64 inCodePos; + UInt64 outCodePos; + + Byte *afterEndData = NULL; + size_t afterEndData_Size = 0; + BoolInt afterEndData_IsCross = False; + + BoolInt canCreateNewThread = False; + // CMtDecCallbackInfo parse; + CMtDecThread *nextThread; + + PRF_STR_INT("=============== Event_Wait(&t->canRead)", t->index) + + RINOK_THREAD(Event_Wait(&t->canRead)) + if (p->exitThread) + return 0; + + PRF_STR_INT("after Event_Wait(&t->canRead)", t->index) + + // if (t->index == 3) return 19; // for test + + blockIndex = p->blockIndex++; + + // PRF(printf("\ncanRead\n")) + + res = MtDec_Progress_GetError_Spec(p, 0, 0, blockIndex, &wasInterrupted); + + finish = p->readWasFinished; + needCode = False; + needWrite = False; + isAllocError = False; + overflow = False; + + inDataSize_Start = 0; + inDataSize = 0; + // inDataSize_Full = 0; + + if (res == SZ_OK && !wasInterrupted) + { + // if (p->inStream) + { + CMtDecBufLink *prev = NULL; + CMtDecBufLink *link = (CMtDecBufLink *)t->inBuf; + size_t crossSize = p->crossEnd - p->crossStart; + + PRF(printf("\ncrossSize = %d\n", crossSize)); + + for (;;) + { + if (!link) + { + link = (CMtDecBufLink *)ISzAlloc_Alloc(p->alloc, MTDEC__LINK_DATA_OFFSET + p->inBufSize); + if (!link) + { + finish = True; + // p->allocError_for_Read_BlockIndex = blockIndex; + isAllocError = True; + break; + } + link->next = NULL; + if (prev) + { + // static unsigned g_num = 0; + // printf("\n%6d : %x", ++g_num, (unsigned)(size_t)((Byte *)link - (Byte *)prev)); + prev->next = link; + } + else + t->inBuf = (void *)link; + } + + { + Byte *data = MTDEC__DATA_PTR_FROM_LINK(link); + Byte *parseData = data; + size_t size; + + if (crossSize != 0) + { + inDataSize = crossSize; + // inDataSize_Full = inDataSize; + inDataSize_Start = crossSize; + size = crossSize; + parseData = MTDEC__DATA_PTR_FROM_LINK(p->crossBlock) + p->crossStart; + PRF(printf("\ncross : crossStart = %7d crossEnd = %7d finish = %1d", + (int)p->crossStart, (int)p->crossEnd, (int)finish)); + } + else + { + size = p->inBufSize; + + res = SeqInStream_ReadMax(p->inStream, data, &size); + + // size = 10; // test + + inDataSize += size; + // inDataSize_Full = inDataSize; + if (!prev) + inDataSize_Start = size; + + p->readProcessed += size; + finish = (size != p->inBufSize); + if (finish) + p->readWasFinished = True; + + // res = E_INVALIDARG; // test + + if (res != SZ_OK) + { + // PRF(printf("\nRead error = %d\n", res)) + // we want to decode all data before error + p->readRes = res; + // p->readError_BlockIndex = blockIndex; + p->readWasFinished = True; + finish = True; + res = SZ_OK; + // break; + } + + if (inDataSize - inPrev >= MTDEC_ProgessStep) + { + res = MtDec_Progress_GetError_Spec(p, 0, 0, blockIndex, &wasInterrupted); + if (res != SZ_OK || wasInterrupted) + break; + inPrev = inDataSize; + } + } + + { + CMtDecCallbackInfo parse; + + parse.startCall = (prev == NULL); + parse.src = parseData; + parse.srcSize = size; + parse.srcFinished = finish; + parse.canCreateNewThread = True; + + PRF(printf("\nParse size = %d\n", (unsigned)size)); + + p->mtCallback->Parse(p->mtCallbackObject, t->index, &parse); + + PRF(printf(" Parse processed = %d, state = %d \n", (unsigned)parse.srcSize, (unsigned)parse.state)); + + needWrite = True; + canCreateNewThread = parse.canCreateNewThread; + + // printf("\n\n%12I64u %12I64u", (UInt64)p->mtProgress.totalInSize, (UInt64)p->mtProgress.totalOutSize); + + if ( + // parseRes != SZ_OK || + // inDataSize - (size - parse.srcSize) > p->inBlockMax + // || + parse.state == MTDEC_PARSE_OVERFLOW + // || wasInterrupted + ) + { + // Overflow or Parse error - switch from MT decoding to ST decoding + finish = True; + overflow = True; + + { + PRF(printf("\n Overflow")); + // PRF(printf("\nisBlockFinished = %d", (unsigned)parse.blockWasFinished)); + PRF(printf("\n inDataSize = %d", (unsigned)inDataSize)); + } + + if (crossSize != 0) + memcpy(data, parseData, size); + p->crossStart = 0; + p->crossEnd = 0; + break; + } + + if (crossSize != 0) + { + memcpy(data, parseData, parse.srcSize); + p->crossStart += parse.srcSize; + } + + if (parse.state != MTDEC_PARSE_CONTINUE || finish) + { + // we don't need to parse in current thread anymore + + if (parse.state == MTDEC_PARSE_END) + finish = True; + + needCode = True; + // p->crossFinished = finish; + + if (parse.srcSize == size) + { + // full parsed - no cross transfer + p->crossStart = 0; + p->crossEnd = 0; + break; + } + + if (parse.state == MTDEC_PARSE_END) + { + afterEndData = parseData + parse.srcSize; + afterEndData_Size = size - parse.srcSize; + if (crossSize != 0) + afterEndData_IsCross = True; + // we reduce data size to required bytes (parsed only) + inDataSize -= afterEndData_Size; + if (!prev) + inDataSize_Start = parse.srcSize; + break; + } + + { + // partial parsed - need cross transfer + if (crossSize != 0) + inDataSize = parse.srcSize; // it's only parsed now + else + { + // partial parsed - is not in initial cross block - we need to copy new data to cross block + Byte *cr = MtDec_GetCrossBuff(p); + if (!cr) + { + { + PRF(printf("\ncross alloc error error\n")); + // res = SZ_ERROR_MEM; + finish = True; + // p->allocError_for_Read_BlockIndex = blockIndex; + isAllocError = True; + break; + } + } + + { + size_t crSize = size - parse.srcSize; + inDataSize -= crSize; + p->crossEnd = crSize; + p->crossStart = 0; + memcpy(cr, parseData + parse.srcSize, crSize); + } + } + + // inDataSize_Full = inDataSize; + if (!prev) + inDataSize_Start = parse.srcSize; // it's partial size (parsed only) + + finish = False; + break; + } + } + + if (parse.srcSize != size) + { + res = SZ_ERROR_FAIL; + PRF(printf("\nfinished error SZ_ERROR_FAIL = %d\n", res)); + break; + } + } + } + + prev = link; + link = link->next; + + if (crossSize != 0) + { + crossSize = 0; + p->crossStart = 0; + p->crossEnd = 0; + } + } + } + + if (res == SZ_OK) + res = MtDec_GetError_Spec(p, blockIndex, &wasInterrupted); + } + + codeRes = SZ_OK; + + if (res == SZ_OK && needCode && !wasInterrupted) + { + codeRes = p->mtCallback->PreCode(p->mtCallbackObject, t->index); + if (codeRes != SZ_OK) + { + needCode = False; + finish = True; + // SZ_ERROR_MEM is expected error here. + // if (codeRes == SZ_ERROR_MEM) - we will try single-thread decoding later. + // if (codeRes != SZ_ERROR_MEM) - we can stop decoding or try single-thread decoding. + } + } + + if (res != SZ_OK || wasInterrupted) + finish = True; + + nextThread = NULL; + threadingErrorSRes = SZ_OK; + + if (!finish) + { + if (p->numStartedThreads < p->numStartedThreads_Limit && canCreateNewThread) + { + SRes res2 = MtDecThread_CreateAndStart(&p->threads[p->numStartedThreads]); + if (res2 == SZ_OK) + { + // if (p->numStartedThreads % 1000 == 0) PRF(printf("\n numStartedThreads=%d\n", p->numStartedThreads)); + p->numStartedThreads++; + } + else + { + PRF(printf("\nERROR: numStartedThreads=%d\n", p->numStartedThreads)); + if (p->numStartedThreads == 1) + { + // if only one thread is possible, we leave muti-threading code + finish = True; + needCode = False; + threadingErrorSRes = res2; + } + else + p->numStartedThreads_Limit = p->numStartedThreads; + } + } + + if (!finish) + { + unsigned nextIndex = t->index + 1; + nextThread = &p->threads[nextIndex >= p->numStartedThreads ? 0 : nextIndex]; + RINOK_THREAD(Event_Set(&nextThread->canRead)) + // We have started executing for new iteration (with next thread) + // And that next thread now is responsible for possible exit from decoding (threading_code) + } + } + + // each call of Event_Set(&nextThread->canRead) must be followed by call of Event_Set(&nextThread->canWrite) + // if ( !finish ) we must call Event_Set(&nextThread->canWrite) in any case + // if ( finish ) we switch to single-thread mode and there are 2 ways at the end of current iteration (current block): + // - if (needContinue) after Write(&needContinue), we restore decoding with new iteration + // - otherwise we stop decoding and exit from MtDec_ThreadFunc2() + + // Don't change (finish) variable in the further code + + + // ---------- CODE ---------- + + inPrev = 0; + outPrev = 0; + inCodePos = 0; + outCodePos = 0; + + if (res == SZ_OK && needCode && codeRes == SZ_OK) + { + BoolInt isStartBlock = True; + CMtDecBufLink *link = (CMtDecBufLink *)t->inBuf; + + for (;;) + { + size_t inSize; + int stop; + + if (isStartBlock) + inSize = inDataSize_Start; + else + { + UInt64 rem = inDataSize - inCodePos; + inSize = p->inBufSize; + if (inSize > rem) + inSize = (size_t)rem; + } + + inCodePos += inSize; + stop = True; + + codeRes = p->mtCallback->Code(p->mtCallbackObject, t->index, + (const Byte *)MTDEC__DATA_PTR_FROM_LINK(link), inSize, + (inCodePos == inDataSize), // srcFinished + &inCodePos, &outCodePos, &stop); + + if (codeRes != SZ_OK) + { + PRF(printf("\nCode Interrupt error = %x\n", codeRes)); + // we interrupt only later blocks + MtDec_Interrupt(p, blockIndex); + break; + } + + if (stop || inCodePos == inDataSize) + break; + + { + const UInt64 inDelta = inCodePos - inPrev; + const UInt64 outDelta = outCodePos - outPrev; + if (inDelta >= MTDEC_ProgessStep || outDelta >= MTDEC_ProgessStep) + { + // Sleep(1); + res = MtDec_Progress_GetError_Spec(p, inDelta, outDelta, blockIndex, &wasInterrupted); + if (res != SZ_OK || wasInterrupted) + break; + inPrev = inCodePos; + outPrev = outCodePos; + } + } + + link = link->next; + isStartBlock = False; + } + } + + + // ---------- WRITE ---------- + + RINOK_THREAD(Event_Wait(&t->canWrite)) + + { + BoolInt isErrorMode = False; + BoolInt canRecode = True; + BoolInt needWriteToStream = needWrite; + + if (p->exitThread) return 0; // it's never executed in normal cases + + if (p->wasInterrupted) + wasInterrupted = True; + else + { + if (codeRes != SZ_OK) // || !needCode // check it !!! + { + p->wasInterrupted = True; + p->codeRes = codeRes; + if (codeRes == SZ_ERROR_MEM) + isAllocError = True; + } + + if (threadingErrorSRes) + { + p->wasInterrupted = True; + p->threadingErrorSRes = threadingErrorSRes; + needWriteToStream = False; + } + if (isAllocError) + { + p->wasInterrupted = True; + p->isAllocError = True; + needWriteToStream = False; + } + if (overflow) + { + p->wasInterrupted = True; + p->overflow = True; + needWriteToStream = False; + } + } + + if (needCode) + { + if (wasInterrupted) + { + inCodePos = 0; + outCodePos = 0; + } + { + const UInt64 inDelta = inCodePos - inPrev; + const UInt64 outDelta = outCodePos - outPrev; + // if (inDelta != 0 || outDelta != 0) + res = MtProgress_ProgressAdd(&p->mtProgress, inDelta, outDelta); + } + } + + needContinue = (!finish); + + // if (res == SZ_OK && needWrite && !wasInterrupted) + if (needWrite) + { + // p->inProcessed += inCodePos; + + PRF(printf("\n--Write afterSize = %d\n", (unsigned)afterEndData_Size)); + + res = p->mtCallback->Write(p->mtCallbackObject, t->index, + res == SZ_OK && needWriteToStream && !wasInterrupted, // needWrite + afterEndData, afterEndData_Size, afterEndData_IsCross, + &needContinue, + &canRecode); + + // res = SZ_ERROR_FAIL; // for test + + PRF(printf("\nAfter Write needContinue = %d\n", (unsigned)needContinue)); + PRF(printf("\nprocessed = %d\n", (unsigned)p->inProcessed)); + + if (res != SZ_OK) + { + PRF(printf("\nWrite error = %d\n", res)); + isErrorMode = True; + p->wasInterrupted = True; + } + if (res != SZ_OK + || (!needContinue && !finish)) + { + PRF(printf("\nWrite Interrupt error = %x\n", res)); + MtDec_Interrupt(p, blockIndex); + } + } + + if (canRecode) + if (!needCode + || res != SZ_OK + || p->wasInterrupted + || codeRes != SZ_OK + || wasInterrupted + || p->numFilledThreads != 0 + || isErrorMode) + { + if (p->numFilledThreads == 0) + p->filledThreadStart = t->index; + if (inDataSize != 0 || !finish) + { + t->inDataSize_Start = inDataSize_Start; + t->inDataSize = inDataSize; + p->numFilledThreads++; + } + PRF(printf("\np->numFilledThreads = %d\n", p->numFilledThreads)); + PRF(printf("p->filledThreadStart = %d\n", p->filledThreadStart)); + } + + if (!finish) + { + RINOK_THREAD(Event_Set(&nextThread->canWrite)) + } + else + { + if (needContinue) + { + // we restore decoding with new iteration + RINOK_THREAD(Event_Set(&p->threads[0].canWrite)) + } + else + { + // we exit from decoding + if (t->index == 0) + return SZ_OK; + p->exitThread = True; + } + RINOK_THREAD(Event_Set(&p->threads[0].canRead)) + } + } + } +} + +#ifdef _WIN32 +#define USE_ALLOCA +#endif + +#ifdef USE_ALLOCA +#ifdef _WIN32 +#include +#else +#include +#endif +#endif + + +static THREAD_FUNC_DECL MtDec_ThreadFunc1(void *pp) +{ + WRes res; + + CMtDecThread *t = (CMtDecThread *)pp; + CMtDec *p; + + // fprintf(stdout, "\n%d = %p\n", t->index, &t); + + res = MtDec_ThreadFunc2(t); + p = t->mtDec; + if (res == 0) + return (THREAD_FUNC_RET_TYPE)(UINT_PTR)p->exitThreadWRes; + { + // it's unexpected situation for some threading function error + if (p->exitThreadWRes == 0) + p->exitThreadWRes = res; + PRF(printf("\nthread exit error = %d\n", res)); + p->exitThread = True; + Event_Set(&p->threads[0].canRead); + Event_Set(&p->threads[0].canWrite); + MtProgress_SetError(&p->mtProgress, MY_SRes_HRESULT_FROM_WRes(res)); + } + return (THREAD_FUNC_RET_TYPE)(UINT_PTR)res; +} + +static Z7_NO_INLINE THREAD_FUNC_DECL MtDec_ThreadFunc(void *pp) +{ + #ifdef USE_ALLOCA + CMtDecThread *t = (CMtDecThread *)pp; + // fprintf(stderr, "\n%d = %p - before", t->index, &t); + t->allocaPtr = alloca(t->index * 128); + #endif + return MtDec_ThreadFunc1(pp); +} + + +int MtDec_PrepareRead(CMtDec *p) +{ + if (p->crossBlock && p->crossStart == p->crossEnd) + { + ISzAlloc_Free(p->alloc, p->crossBlock); + p->crossBlock = NULL; + } + + { + unsigned i; + for (i = 0; i < MTDEC_THREADS_MAX; i++) + if (i > p->numStartedThreads + || p->numFilledThreads <= + (i >= p->filledThreadStart ? + i - p->filledThreadStart : + i + p->numStartedThreads - p->filledThreadStart)) + MtDecThread_FreeInBufs(&p->threads[i]); + } + + return (p->numFilledThreads != 0) || (p->crossStart != p->crossEnd); +} + + +const Byte *MtDec_Read(CMtDec *p, size_t *inLim) +{ + while (p->numFilledThreads != 0) + { + CMtDecThread *t = &p->threads[p->filledThreadStart]; + + if (*inLim != 0) + { + { + void *link = t->inBuf; + void *next = ((CMtDecBufLink *)link)->next; + ISzAlloc_Free(p->alloc, link); + t->inBuf = next; + } + + if (t->inDataSize == 0) + { + MtDecThread_FreeInBufs(t); + if (--p->numFilledThreads == 0) + break; + if (++p->filledThreadStart == p->numStartedThreads) + p->filledThreadStart = 0; + t = &p->threads[p->filledThreadStart]; + } + } + + { + size_t lim = t->inDataSize_Start; + if (lim != 0) + t->inDataSize_Start = 0; + else + { + UInt64 rem = t->inDataSize; + lim = p->inBufSize; + if (lim > rem) + lim = (size_t)rem; + } + t->inDataSize -= lim; + *inLim = lim; + return (const Byte *)MTDEC__DATA_PTR_FROM_LINK(t->inBuf); + } + } + + { + size_t crossSize = p->crossEnd - p->crossStart; + if (crossSize != 0) + { + const Byte *data = MTDEC__DATA_PTR_FROM_LINK(p->crossBlock) + p->crossStart; + *inLim = crossSize; + p->crossStart = 0; + p->crossEnd = 0; + return data; + } + *inLim = 0; + if (p->crossBlock) + { + ISzAlloc_Free(p->alloc, p->crossBlock); + p->crossBlock = NULL; + } + return NULL; + } +} + + +void MtDec_Construct(CMtDec *p) +{ + unsigned i; + + p->inBufSize = (size_t)1 << 18; + + p->numThreadsMax = 0; + + p->inStream = NULL; + + // p->inData = NULL; + // p->inDataSize = 0; + + p->crossBlock = NULL; + p->crossStart = 0; + p->crossEnd = 0; + + p->numFilledThreads = 0; + + p->progress = NULL; + p->alloc = NULL; + + p->mtCallback = NULL; + p->mtCallbackObject = NULL; + + p->allocatedBufsSize = 0; + + for (i = 0; i < MTDEC_THREADS_MAX; i++) + { + CMtDecThread *t = &p->threads[i]; + t->mtDec = p; + t->index = i; + t->inBuf = NULL; + Event_Construct(&t->canRead); + Event_Construct(&t->canWrite); + Thread_CONSTRUCT(&t->thread) + } + + // Event_Construct(&p->finishedEvent); + + CriticalSection_Init(&p->mtProgress.cs); +} + + +static void MtDec_Free(CMtDec *p) +{ + unsigned i; + + p->exitThread = True; + + for (i = 0; i < MTDEC_THREADS_MAX; i++) + MtDecThread_Destruct(&p->threads[i]); + + // Event_Close(&p->finishedEvent); + + if (p->crossBlock) + { + ISzAlloc_Free(p->alloc, p->crossBlock); + p->crossBlock = NULL; + } +} + + +void MtDec_Destruct(CMtDec *p) +{ + MtDec_Free(p); + + CriticalSection_Delete(&p->mtProgress.cs); +} + + +SRes MtDec_Code(CMtDec *p) +{ + unsigned i; + + p->inProcessed = 0; + + p->blockIndex = 1; // it must be larger than not_defined index (0) + p->isAllocError = False; + p->overflow = False; + p->threadingErrorSRes = SZ_OK; + + p->needContinue = True; + + p->readWasFinished = False; + p->needInterrupt = False; + p->interruptIndex = (UInt64)(Int64)-1; + + p->readProcessed = 0; + p->readRes = SZ_OK; + p->codeRes = SZ_OK; + p->wasInterrupted = False; + + p->crossStart = 0; + p->crossEnd = 0; + + p->filledThreadStart = 0; + p->numFilledThreads = 0; + + { + unsigned numThreads = p->numThreadsMax; + if (numThreads > MTDEC_THREADS_MAX) + numThreads = MTDEC_THREADS_MAX; + p->numStartedThreads_Limit = numThreads; + p->numStartedThreads = 0; + } + + if (p->inBufSize != p->allocatedBufsSize) + { + for (i = 0; i < MTDEC_THREADS_MAX; i++) + { + CMtDecThread *t = &p->threads[i]; + if (t->inBuf) + MtDecThread_FreeInBufs(t); + } + if (p->crossBlock) + { + ISzAlloc_Free(p->alloc, p->crossBlock); + p->crossBlock = NULL; + } + + p->allocatedBufsSize = p->inBufSize; + } + + MtProgress_Init(&p->mtProgress, p->progress); + + // RINOK_THREAD(AutoResetEvent_OptCreate_And_Reset(&p->finishedEvent)) + p->exitThread = False; + p->exitThreadWRes = 0; + + { + WRes wres; + SRes sres; + CMtDecThread *nextThread = &p->threads[p->numStartedThreads++]; + // wres = MtDecThread_CreateAndStart(nextThread); + wres = MtDecThread_CreateEvents(nextThread); + if (wres == 0) { wres = Event_Set(&nextThread->canWrite); + if (wres == 0) { wres = Event_Set(&nextThread->canRead); + if (wres == 0) { THREAD_FUNC_RET_TYPE res = MtDec_ThreadFunc(nextThread); + wres = (WRes)(UINT_PTR)res; + if (wres != 0) + { + p->needContinue = False; + MtDec_CloseThreads(p); + }}}} + + // wres = 17; // for test + // wres = Event_Wait(&p->finishedEvent); + + sres = MY_SRes_HRESULT_FROM_WRes(wres); + + if (sres != 0) + p->threadingErrorSRes = sres; + + if ( + // wres == 0 + // wres != 0 + // || p->mtc.codeRes == SZ_ERROR_MEM + p->isAllocError + || p->threadingErrorSRes != SZ_OK + || p->overflow) + { + // p->needContinue = True; + } + else + p->needContinue = False; + + if (p->needContinue) + return SZ_OK; + + // if (sres != SZ_OK) + return sres; + // return SZ_ERROR_FAIL; + } +} + +#endif + +#undef PRF diff --git a/libraries/lzma/C/MtDec.h b/libraries/lzma/C/MtDec.h new file mode 100644 index 000000000..c28e8d9ac --- /dev/null +++ b/libraries/lzma/C/MtDec.h @@ -0,0 +1,202 @@ +/* MtDec.h -- Multi-thread Decoder +2023-04-02 : Igor Pavlov : Public domain */ + +#ifndef ZIP7_INC_MT_DEC_H +#define ZIP7_INC_MT_DEC_H + +#include "7zTypes.h" + +#ifndef Z7_ST +#include "Threads.h" +#endif + +EXTERN_C_BEGIN + +#ifndef Z7_ST + +#ifndef Z7_ST + #define MTDEC_THREADS_MAX 32 +#else + #define MTDEC_THREADS_MAX 1 +#endif + + +typedef struct +{ + ICompressProgressPtr progress; + SRes res; + UInt64 totalInSize; + UInt64 totalOutSize; + CCriticalSection cs; +} CMtProgress; + +void MtProgress_Init(CMtProgress *p, ICompressProgressPtr 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, BoolInt isCross, + // int srcFinished, + BoolInt *needContinue, + BoolInt *canRecode); + +} IMtDecCallback2; + + + +typedef struct CMtDec_ +{ + /* input variables */ + + size_t inBufSize; /* size of input block */ + unsigned numThreadsMax; + // size_t inBlockMax; + unsigned numThreadsMax_2; + + ISeqInStreamPtr inStream; + // const Byte *inData; + // size_t inDataSize; + + ICompressProgressPtr progress; + ISzAllocPtr alloc; + + IMtDecCallback2 *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 Z7_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/libraries/lzma/C/Ppmd.h b/libraries/lzma/C/Ppmd.h index b19879208..66b26266f 100644 --- a/libraries/lzma/C/Ppmd.h +++ b/libraries/lzma/C/Ppmd.h @@ -1,9 +1,9 @@ /* Ppmd.h -- PPMD codec common code -2021-04-13 : Igor Pavlov : Public domain +2023-03-05 : Igor Pavlov : Public domain This code is based on PPMd var.H (2001): Dmitry Shkarin : Public domain */ -#ifndef __PPMD_H -#define __PPMD_H +#ifndef ZIP7_INC_PPMD_H +#define ZIP7_INC_PPMD_H #include "CpuArch.h" @@ -48,8 +48,10 @@ typedef struct Byte Count; /* Count to next change of Shift */ } CPpmd_See; -#define Ppmd_See_Update(p) if ((p)->Shift < PPMD_PERIOD_BITS && --(p)->Count == 0) \ - { (p)->Summ = (UInt16)((p)->Summ << 1); (p)->Count = (Byte)(3 << (p)->Shift++); } +#define Ppmd_See_UPDATE(p) \ + { if ((p)->Shift < PPMD_PERIOD_BITS && --(p)->Count == 0) \ + { (p)->Summ = (UInt16)((p)->Summ << 1); \ + (p)->Count = (Byte)(3 << (p)->Shift++); }} typedef struct diff --git a/libraries/lzma/C/Ppmd7.c b/libraries/lzma/C/Ppmd7.c index cf401cb37..6e1307e2d 100644 --- a/libraries/lzma/C/Ppmd7.c +++ b/libraries/lzma/C/Ppmd7.c @@ -1,5 +1,5 @@ /* Ppmd7.c -- PPMdH codec -2021-04-13 : Igor Pavlov : Public domain +2023-04-02 : Igor Pavlov : Public domain This code is based on PPMd var.H (2001): Dmitry Shkarin : Public domain */ #include "Precomp.h" @@ -14,7 +14,7 @@ This code is based on PPMd var.H (2001): Dmitry Shkarin : Public domain */ MY_ALIGN(16) static const Byte PPMD7_kExpEscape[16] = { 25, 14, 9, 7, 5, 5, 4, 4, 4, 3, 3, 3, 2, 2, 2, 2 }; MY_ALIGN(16) -static const UInt16 kInitBinEsc[] = { 0x3CDD, 0x1F3F, 0x59BF, 0x48F3, 0x64A1, 0x5ABC, 0x6632, 0x6051}; +static const UInt16 PPMD7_kInitBinEsc[] = { 0x3CDD, 0x1F3F, 0x59BF, 0x48F3, 0x64A1, 0x5ABC, 0x6632, 0x6051}; #define MAX_FREQ 124 #define UNIT_SIZE 12 @@ -33,7 +33,7 @@ static const UInt16 kInitBinEsc[] = { 0x3CDD, 0x1F3F, 0x59BF, 0x48F3, 0x64A1, 0x #define ONE_STATE(ctx) Ppmd7Context_OneState(ctx) #define SUFFIX(ctx) CTX((ctx)->Suffix) -typedef CPpmd7_Context * CTX_PTR; +typedef CPpmd7_Context * PPMD7_CTX_PTR; struct CPpmd7_Node_; @@ -107,14 +107,14 @@ BoolInt Ppmd7_Alloc(CPpmd7 *p, UInt32 size, ISzAllocPtr alloc) // ---------- Internal Memory Allocator ---------- /* We can use CPpmd7_Node in list of free units (as in Ppmd8) - But we still need one additional list walk pass in GlueFreeBlocks(). - So we use simple CPpmd_Void_Ref instead of CPpmd7_Node in InsertNode() / RemoveNode() + But we still need one additional list walk pass in Ppmd7_GlueFreeBlocks(). + So we use simple CPpmd_Void_Ref instead of CPpmd7_Node in Ppmd7_InsertNode() / Ppmd7_RemoveNode() */ #define EMPTY_NODE 0 -static void InsertNode(CPpmd7 *p, void *node, unsigned indx) +static void Ppmd7_InsertNode(CPpmd7 *p, void *node, unsigned indx) { *((CPpmd_Void_Ref *)node) = p->FreeList[indx]; // ((CPpmd7_Node *)node)->Next = (CPpmd7_Node_Ref)p->FreeList[indx]; @@ -124,7 +124,7 @@ static void InsertNode(CPpmd7 *p, void *node, unsigned indx) } -static void *RemoveNode(CPpmd7 *p, unsigned indx) +static void *Ppmd7_RemoveNode(CPpmd7 *p, unsigned indx) { CPpmd_Void_Ref *node = (CPpmd_Void_Ref *)Ppmd7_GetPtr(p, p->FreeList[indx]); p->FreeList[indx] = *node; @@ -134,32 +134,32 @@ static void *RemoveNode(CPpmd7 *p, unsigned indx) } -static void SplitBlock(CPpmd7 *p, void *ptr, unsigned oldIndx, unsigned newIndx) +static void Ppmd7_SplitBlock(CPpmd7 *p, void *ptr, unsigned oldIndx, unsigned newIndx) { unsigned i, nu = I2U(oldIndx) - I2U(newIndx); ptr = (Byte *)ptr + U2B(I2U(newIndx)); if (I2U(i = U2I(nu)) != nu) { unsigned k = I2U(--i); - InsertNode(p, ((Byte *)ptr) + U2B(k), nu - k - 1); + Ppmd7_InsertNode(p, ((Byte *)ptr) + U2B(k), nu - k - 1); } - InsertNode(p, ptr, i); + Ppmd7_InsertNode(p, ptr, i); } /* we use CPpmd7_Node_Union union to solve XLC -O2 strict pointer aliasing problem */ -typedef union _CPpmd7_Node_Union +typedef union { CPpmd7_Node Node; CPpmd7_Node_Ref NextRef; } CPpmd7_Node_Union; -/* Original PPmdH (Ppmd7) code uses doubly linked list in GlueFreeBlocks() +/* Original PPmdH (Ppmd7) code uses doubly linked list in Ppmd7_GlueFreeBlocks() we use single linked list similar to Ppmd8 code */ -static void GlueFreeBlocks(CPpmd7 *p) +static void Ppmd7_GlueFreeBlocks(CPpmd7 *p) { /* we use first UInt16 field of 12-bytes UNITs as record type stamp @@ -239,27 +239,27 @@ static void GlueFreeBlocks(CPpmd7 *p) if (nu == 0) continue; for (; nu > 128; nu -= 128, node += 128) - InsertNode(p, node, PPMD_NUM_INDEXES - 1); + Ppmd7_InsertNode(p, node, PPMD_NUM_INDEXES - 1); if (I2U(i = U2I(nu)) != nu) { unsigned k = I2U(--i); - InsertNode(p, node + k, (unsigned)nu - k - 1); + Ppmd7_InsertNode(p, node + k, (unsigned)nu - k - 1); } - InsertNode(p, node, i); + Ppmd7_InsertNode(p, node, i); } } -MY_NO_INLINE -static void *AllocUnitsRare(CPpmd7 *p, unsigned indx) +Z7_NO_INLINE +static void *Ppmd7_AllocUnitsRare(CPpmd7 *p, unsigned indx) { unsigned i; if (p->GlueCount == 0) { - GlueFreeBlocks(p); + Ppmd7_GlueFreeBlocks(p); if (p->FreeList[indx] != 0) - return RemoveNode(p, indx); + return Ppmd7_RemoveNode(p, indx); } i = indx; @@ -277,17 +277,17 @@ static void *AllocUnitsRare(CPpmd7 *p, unsigned indx) while (p->FreeList[i] == 0); { - void *block = RemoveNode(p, i); - SplitBlock(p, block, i, indx); + void *block = Ppmd7_RemoveNode(p, i); + Ppmd7_SplitBlock(p, block, i, indx); return block; } } -static void *AllocUnits(CPpmd7 *p, unsigned indx) +static void *Ppmd7_AllocUnits(CPpmd7 *p, unsigned indx) { if (p->FreeList[indx] != 0) - return RemoveNode(p, indx); + return Ppmd7_RemoveNode(p, indx); { UInt32 numBytes = U2B(I2U(indx)); Byte *lo = p->LoUnit; @@ -297,11 +297,11 @@ static void *AllocUnits(CPpmd7 *p, unsigned indx) return lo; } } - return AllocUnitsRare(p, indx); + return Ppmd7_AllocUnitsRare(p, indx); } -#define MyMem12Cpy(dest, src, num) \ +#define MEM_12_CPY(dest, src, num) \ { UInt32 *d = (UInt32 *)dest; const UInt32 *z = (const UInt32 *)src; UInt32 n = num; \ do { d[0] = z[0]; d[1] = z[1]; d[2] = z[2]; z += 3; d += 3; } while (--n); } @@ -315,12 +315,12 @@ static void *ShrinkUnits(CPpmd7 *p, void *oldPtr, unsigned oldNU, unsigned newNU return oldPtr; if (p->FreeList[i1] != 0) { - void *ptr = RemoveNode(p, i1); - MyMem12Cpy(ptr, oldPtr, newNU); - InsertNode(p, oldPtr, i0); + void *ptr = Ppmd7_RemoveNode(p, i1); + MEM_12_CPY(ptr, oldPtr, newNU) + Ppmd7_InsertNode(p, oldPtr, i0); return ptr; } - SplitBlock(p, oldPtr, i0, i1); + Ppmd7_SplitBlock(p, oldPtr, i0, i1); return oldPtr; } */ @@ -329,14 +329,14 @@ static void *ShrinkUnits(CPpmd7 *p, void *oldPtr, unsigned oldNU, unsigned newNU #define SUCCESSOR(p) Ppmd_GET_SUCCESSOR(p) static void SetSuccessor(CPpmd_State *p, CPpmd_Void_Ref v) { - Ppmd_SET_SUCCESSOR(p, v); + Ppmd_SET_SUCCESSOR(p, v) } -MY_NO_INLINE +Z7_NO_INLINE static -void RestartModel(CPpmd7 *p) +void Ppmd7_RestartModel(CPpmd7 *p) { unsigned i, k; @@ -352,8 +352,8 @@ void RestartModel(CPpmd7 *p) p->PrevSuccess = 0; { - CPpmd7_Context *mc = (CTX_PTR)(void *)(p->HiUnit -= UNIT_SIZE); /* AllocContext(p); */ - CPpmd_State *s = (CPpmd_State *)p->LoUnit; /* AllocUnits(p, PPMD_NUM_INDEXES - 1); */ + CPpmd7_Context *mc = (PPMD7_CTX_PTR)(void *)(p->HiUnit -= UNIT_SIZE); /* AllocContext(p); */ + CPpmd_State *s = (CPpmd_State *)p->LoUnit; /* Ppmd7_AllocUnits(p, PPMD_NUM_INDEXES - 1); */ p->LoUnit += U2B(256 / 2); p->MaxContext = p->MinContext = mc; @@ -391,7 +391,7 @@ void RestartModel(CPpmd7 *p) { unsigned m; UInt16 *dest = p->BinSumm[i] + k; - UInt16 val = (UInt16)(PPMD_BIN_SCALE - kInitBinEsc[k] / (i + 2)); + const UInt16 val = (UInt16)(PPMD_BIN_SCALE - PPMD7_kInitBinEsc[k] / (i + 2)); for (m = 0; m < 64; m += 8) dest[m] = val; } @@ -423,13 +423,13 @@ void Ppmd7_Init(CPpmd7 *p, unsigned maxOrder) { p->MaxOrder = maxOrder; - RestartModel(p); + Ppmd7_RestartModel(p); } /* - CreateSuccessors() + Ppmd7_CreateSuccessors() It's called when (FoundState->Successor) is RAW-Successor, that is the link to position in Raw text. So we create Context records and write the links to @@ -445,10 +445,10 @@ void Ppmd7_Init(CPpmd7 *p, unsigned maxOrder) also it can return pointer to real context of same order, */ -MY_NO_INLINE -static CTX_PTR CreateSuccessors(CPpmd7 *p) +Z7_NO_INLINE +static PPMD7_CTX_PTR Ppmd7_CreateSuccessors(CPpmd7 *p) { - CTX_PTR c = p->MinContext; + PPMD7_CTX_PTR c = p->MinContext; CPpmd_Byte_Ref upBranch = (CPpmd_Byte_Ref)SUCCESSOR(p->FoundState); Byte newSym, newFreq; unsigned numPs = 0; @@ -522,15 +522,15 @@ static CTX_PTR CreateSuccessors(CPpmd7 *p) do { - CTX_PTR c1; + PPMD7_CTX_PTR c1; /* = AllocContext(p); */ if (p->HiUnit != p->LoUnit) - c1 = (CTX_PTR)(void *)(p->HiUnit -= UNIT_SIZE); + c1 = (PPMD7_CTX_PTR)(void *)(p->HiUnit -= UNIT_SIZE); else if (p->FreeList[0] != 0) - c1 = (CTX_PTR)RemoveNode(p, 0); + c1 = (PPMD7_CTX_PTR)Ppmd7_RemoveNode(p, 0); else { - c1 = (CTX_PTR)AllocUnitsRare(p, 0); + c1 = (PPMD7_CTX_PTR)Ppmd7_AllocUnitsRare(p, 0); if (!c1) return NULL; } @@ -550,16 +550,16 @@ static CTX_PTR CreateSuccessors(CPpmd7 *p) -#define SwapStates(s) \ +#define SWAP_STATES(s) \ { CPpmd_State tmp = s[0]; s[0] = s[-1]; s[-1] = tmp; } void Ppmd7_UpdateModel(CPpmd7 *p); -MY_NO_INLINE +Z7_NO_INLINE void Ppmd7_UpdateModel(CPpmd7 *p) { CPpmd_Void_Ref maxSuccessor, minSuccessor; - CTX_PTR c, mc; + PPMD7_CTX_PTR c, mc; unsigned s0, ns; @@ -592,7 +592,7 @@ void Ppmd7_UpdateModel(CPpmd7 *p) if (s[0].Freq >= s[-1].Freq) { - SwapStates(s); + SWAP_STATES(s) s--; } } @@ -610,10 +610,10 @@ void Ppmd7_UpdateModel(CPpmd7 *p) { /* MAX ORDER context */ /* (FoundState->Successor) is RAW-Successor. */ - p->MaxContext = p->MinContext = CreateSuccessors(p); + p->MaxContext = p->MinContext = Ppmd7_CreateSuccessors(p); if (!p->MinContext) { - RestartModel(p); + Ppmd7_RestartModel(p); return; } SetSuccessor(p->FoundState, REF(p->MinContext)); @@ -629,7 +629,7 @@ void Ppmd7_UpdateModel(CPpmd7 *p) p->Text = text; if (text >= p->UnitsStart) { - RestartModel(p); + Ppmd7_RestartModel(p); return; } maxSuccessor = REF(text); @@ -645,10 +645,10 @@ void Ppmd7_UpdateModel(CPpmd7 *p) if (minSuccessor <= maxSuccessor) { // minSuccessor is RAW-Successor. So we will create real contexts records: - CTX_PTR cs = CreateSuccessors(p); + PPMD7_CTX_PTR cs = Ppmd7_CreateSuccessors(p); if (!cs) { - RestartModel(p); + Ppmd7_RestartModel(p); return; } minSuccessor = REF(cs); @@ -715,16 +715,16 @@ void Ppmd7_UpdateModel(CPpmd7 *p) unsigned i = U2I(oldNU); if (i != U2I((size_t)oldNU + 1)) { - void *ptr = AllocUnits(p, i + 1); + void *ptr = Ppmd7_AllocUnits(p, i + 1); void *oldPtr; if (!ptr) { - RestartModel(p); + Ppmd7_RestartModel(p); return; } oldPtr = STATS(c); - MyMem12Cpy(ptr, oldPtr, oldNU); - InsertNode(p, oldPtr, i); + MEM_12_CPY(ptr, oldPtr, oldNU) + Ppmd7_InsertNode(p, oldPtr, i); c->Union4.Stats = STATS_REF(ptr); } } @@ -739,10 +739,10 @@ void Ppmd7_UpdateModel(CPpmd7 *p) else { // instead of One-symbol context we create 2-symbol context - CPpmd_State *s = (CPpmd_State*)AllocUnits(p, 0); + CPpmd_State *s = (CPpmd_State*)Ppmd7_AllocUnits(p, 0); if (!s) { - RestartModel(p); + Ppmd7_RestartModel(p); return; } { @@ -795,8 +795,8 @@ void Ppmd7_UpdateModel(CPpmd7 *p) -MY_NO_INLINE -static void Rescale(CPpmd7 *p) +Z7_NO_INLINE +static void Ppmd7_Rescale(CPpmd7 *p) { unsigned i, adder, sumFreq, escFreq; CPpmd_State *stats = STATS(p->MinContext); @@ -885,7 +885,7 @@ static void Rescale(CPpmd7 *p) *s = *stats; s->Freq = (Byte)freq; // (freq <= 260 / 4) p->FoundState = s; - InsertNode(p, stats, U2I(n0)); + Ppmd7_InsertNode(p, stats, U2I(n0)); return; } @@ -899,13 +899,13 @@ static void Rescale(CPpmd7 *p) { if (p->FreeList[i1] != 0) { - void *ptr = RemoveNode(p, i1); + void *ptr = Ppmd7_RemoveNode(p, i1); p->MinContext->Union4.Stats = STATS_REF(ptr); - MyMem12Cpy(ptr, (const void *)stats, n1); - InsertNode(p, stats, i0); + MEM_12_CPY(ptr, (const void *)stats, n1) + Ppmd7_InsertNode(p, stats, i0); } else - SplitBlock(p, stats, i0, i1); + Ppmd7_SplitBlock(p, stats, i0, i1); } } } @@ -948,9 +948,9 @@ CPpmd_See *Ppmd7_MakeEscFreq(CPpmd7 *p, unsigned numMasked, UInt32 *escFreq) } -static void NextContext(CPpmd7 *p) +static void Ppmd7_NextContext(CPpmd7 *p) { - CTX_PTR c = CTX(SUCCESSOR(p->FoundState)); + PPMD7_CTX_PTR c = CTX(SUCCESSOR(p->FoundState)); if (p->OrderFall == 0 && (const Byte *)c > p->Text) p->MaxContext = p->MinContext = c; else @@ -967,12 +967,12 @@ void Ppmd7_Update1(CPpmd7 *p) s->Freq = (Byte)freq; if (freq > s[-1].Freq) { - SwapStates(s); + SWAP_STATES(s) p->FoundState = --s; if (freq > MAX_FREQ) - Rescale(p); + Ppmd7_Rescale(p); } - NextContext(p); + Ppmd7_NextContext(p); } @@ -988,8 +988,8 @@ void Ppmd7_Update1_0(CPpmd7 *p) freq += 4; s->Freq = (Byte)freq; if (freq > MAX_FREQ) - Rescale(p); - NextContext(p); + Ppmd7_Rescale(p); + Ppmd7_NextContext(p); } @@ -1000,7 +1000,7 @@ void Ppmd7_UpdateBin(CPpmd7 *p) p->FoundState->Freq = (Byte)(freq + (freq < 128)); p->PrevSuccess = 1; p->RunLength++; - NextContext(p); + Ppmd7_NextContext(p); } */ @@ -1013,7 +1013,7 @@ void Ppmd7_Update2(CPpmd7 *p) p->MinContext->Union2.SummFreq = (UInt16)(p->MinContext->Union2.SummFreq + 4); s->Freq = (Byte)freq; if (freq > MAX_FREQ) - Rescale(p); + Ppmd7_Rescale(p); Ppmd7_UpdateModel(p); } @@ -1042,8 +1042,8 @@ Last UNIT of array at offset (Size - 12) is root order-0 CPpmd7_Context record. The code can free UNITs memory blocks that were allocated to store CPpmd_State vectors. The code doesn't free UNITs allocated for CPpmd7_Context records. -The code calls RestartModel(), when there is no free memory for allocation. -And RestartModel() changes the state to orignal start state, with full free block. +The code calls Ppmd7_RestartModel(), when there is no free memory for allocation. +And Ppmd7_RestartModel() changes the state to orignal start state, with full free block. The code allocates UNITs with the following order: @@ -1051,14 +1051,14 @@ The code allocates UNITs with the following order: Allocation of 1 UNIT for Context record - from free space (HiUnit) down to (LoUnit) - from FreeList[0] - - AllocUnitsRare() + - Ppmd7_AllocUnitsRare() -AllocUnits() for CPpmd_State vectors: +Ppmd7_AllocUnits() for CPpmd_State vectors: - from FreeList[i] - from free space (LoUnit) up to (HiUnit) - - AllocUnitsRare() + - Ppmd7_AllocUnitsRare() -AllocUnitsRare() +Ppmd7_AllocUnitsRare() - if (GlueCount == 0) { Glue lists, GlueCount = 255, allocate from FreeList[i]] } - loop for all higher sized FreeList[...] lists @@ -1093,8 +1093,8 @@ The PPMd code tries to fulfill the condition: We have (Sum(Stats[].Freq) <= 256 * 124), because of (MAX_FREQ = 124) So (4 = 128 - 124) is average reserve for Escape_Freq for each symbol. If (CPpmd_State::Freq) is not aligned for 4, the reserve can be 5, 6 or 7. -SummFreq and Escape_Freq can be changed in Rescale() and *Update*() functions. -Rescale() can remove symbols only from max-order contexts. So Escape_Freq can increase after multiple calls of Rescale() for +SummFreq and Escape_Freq can be changed in Ppmd7_Rescale() and *Update*() functions. +Ppmd7_Rescale() can remove symbols only from max-order contexts. So Escape_Freq can increase after multiple calls of Ppmd7_Rescale() for max-order context. When the PPMd code still break (Total <= RC::Range) condition in range coder, @@ -1102,3 +1102,21 @@ we have two ways to resolve that problem: 1) we can report error, if we want to keep compatibility with original PPMd code that has no fix for such cases. 2) we can reduce (Total) value to (RC::Range) by reducing (Escape_Freq) part of (Total) value. */ + +#undef MAX_FREQ +#undef UNIT_SIZE +#undef U2B +#undef U2I +#undef I2U +#undef I2U_UInt16 +#undef REF +#undef STATS_REF +#undef CTX +#undef STATS +#undef ONE_STATE +#undef SUFFIX +#undef NODE +#undef EMPTY_NODE +#undef MEM_12_CPY +#undef SUCCESSOR +#undef SWAP_STATES diff --git a/libraries/lzma/C/Ppmd7.h b/libraries/lzma/C/Ppmd7.h index d31809aeb..d9eb326d6 100644 --- a/libraries/lzma/C/Ppmd7.h +++ b/libraries/lzma/C/Ppmd7.h @@ -1,11 +1,11 @@ /* Ppmd7.h -- Ppmd7 (PPMdH) compression codec -2021-04-13 : Igor Pavlov : Public domain +2023-04-02 : Igor Pavlov : Public domain This code is based on: PPMd var.H (2001): Dmitry Shkarin : Public domain */ -#ifndef __PPMD7_H -#define __PPMD7_H +#ifndef ZIP7_INC_PPMD7_H +#define ZIP7_INC_PPMD7_H #include "Ppmd.h" @@ -55,7 +55,7 @@ typedef struct UInt32 Range; UInt32 Code; UInt32 Low; - IByteIn *Stream; + IByteInPtr Stream; } CPpmd7_RangeDec; @@ -66,7 +66,7 @@ typedef struct // Byte _dummy_[3]; UInt64 Low; UInt64 CacheSize; - IByteOut *Stream; + IByteOutPtr Stream; } CPpmd7z_RangeEnc; diff --git a/libraries/lzma/C/Ppmd7Dec.c b/libraries/lzma/C/Ppmd7Dec.c index 55d74ff9d..832382828 100644 --- a/libraries/lzma/C/Ppmd7Dec.c +++ b/libraries/lzma/C/Ppmd7Dec.c @@ -1,5 +1,5 @@ /* Ppmd7Dec.c -- Ppmd7z (PPMdH with 7z Range Coder) Decoder -2021-04-13 : Igor Pavlov : Public domain +2023-04-02 : Igor Pavlov : Public domain This code is based on: PPMd var.H (2001): Dmitry Shkarin : Public domain */ @@ -8,7 +8,7 @@ This code is based on: #include "Ppmd7.h" -#define kTopValue (1 << 24) +#define kTopValue ((UInt32)1 << 24) #define READ_BYTE(p) IByteIn_Read((p)->Stream) @@ -37,9 +37,9 @@ BoolInt Ppmd7z_RangeDec_Init(CPpmd7_RangeDec *p) #define R (&p->rc.dec) -MY_FORCE_INLINE -// MY_NO_INLINE -static void RangeDec_Decode(CPpmd7 *p, UInt32 start, UInt32 size) +Z7_FORCE_INLINE +// Z7_NO_INLINE +static void Ppmd7z_RD_Decode(CPpmd7 *p, UInt32 start, UInt32 size) { @@ -48,18 +48,18 @@ static void RangeDec_Decode(CPpmd7 *p, UInt32 start, UInt32 size) RC_NORM_LOCAL(R) } -#define RC_Decode(start, size) RangeDec_Decode(p, start, size); -#define RC_DecodeFinal(start, size) RC_Decode(start, size) RC_NORM_REMOTE(R) -#define RC_GetThreshold(total) (R->Code / (R->Range /= (total))) +#define RC_Decode(start, size) Ppmd7z_RD_Decode(p, start, size); +#define RC_DecodeFinal(start, size) RC_Decode(start, size) RC_NORM_REMOTE(R) +#define RC_GetThreshold(total) (R->Code / (R->Range /= (total))) #define CTX(ref) ((CPpmd7_Context *)Ppmd7_GetContext(p, ref)) -typedef CPpmd7_Context * CTX_PTR; +// typedef CPpmd7_Context * CTX_PTR; #define SUCCESSOR(p) Ppmd_GET_SUCCESSOR(p) void Ppmd7_UpdateModel(CPpmd7 *p); #define MASK(sym) ((unsigned char *)charMask)[sym] -// MY_FORCE_INLINE +// Z7_FORCE_INLINE // static int Ppmd7z_DecodeSymbol(CPpmd7 *p) { @@ -70,7 +70,7 @@ int Ppmd7z_DecodeSymbol(CPpmd7 *p) CPpmd_State *s = Ppmd7_GetStats(p, p->MinContext); unsigned i; UInt32 count, hiCnt; - UInt32 summFreq = p->MinContext->Union2.SummFreq; + const UInt32 summFreq = p->MinContext->Union2.SummFreq; @@ -81,7 +81,7 @@ int Ppmd7z_DecodeSymbol(CPpmd7 *p) if ((Int32)(count -= s->Freq) < 0) { Byte sym; - RC_DecodeFinal(0, s->Freq); + RC_DecodeFinal(0, s->Freq) p->FoundState = s; sym = s->Symbol; Ppmd7_Update1_0(p); @@ -96,7 +96,7 @@ int Ppmd7z_DecodeSymbol(CPpmd7 *p) if ((Int32)(count -= (++s)->Freq) < 0) { Byte sym; - RC_DecodeFinal((hiCnt - count) - s->Freq, s->Freq); + RC_DecodeFinal((hiCnt - count) - s->Freq, s->Freq) p->FoundState = s; sym = s->Symbol; Ppmd7_Update1(p); @@ -109,10 +109,10 @@ int Ppmd7z_DecodeSymbol(CPpmd7 *p) return PPMD7_SYM_ERROR; hiCnt -= count; - RC_Decode(hiCnt, summFreq - hiCnt); + RC_Decode(hiCnt, summFreq - hiCnt) p->HiBitsFlag = PPMD7_HiBitsFlag_3(p->FoundState->Symbol); - PPMD_SetAllBitsIn256Bytes(charMask); + PPMD_SetAllBitsIn256Bytes(charMask) // i = p->MinContext->NumStats - 1; // do { MASK((--s)->Symbol) = 0; } while (--i); { @@ -152,7 +152,7 @@ int Ppmd7z_DecodeSymbol(CPpmd7 *p) // Ppmd7_UpdateBin(p); { unsigned freq = s->Freq; - CTX_PTR c = CTX(SUCCESSOR(s)); + CPpmd7_Context *c = CTX(SUCCESSOR(s)); sym = s->Symbol; p->FoundState = s; p->PrevSuccess = 1; @@ -176,7 +176,7 @@ int Ppmd7z_DecodeSymbol(CPpmd7 *p) R->Range -= size0; RC_NORM_LOCAL(R) - PPMD_SetAllBitsIn256Bytes(charMask); + PPMD_SetAllBitsIn256Bytes(charMask) MASK(Ppmd7Context_OneState(p->MinContext)->Symbol) = 0; p->PrevSuccess = 0; } @@ -245,13 +245,13 @@ int Ppmd7z_DecodeSymbol(CPpmd7 *p) { count -= s->Freq & (unsigned)(MASK((s)->Symbol)); s++; if ((Int32)count < 0) break; // count -= s->Freq & (unsigned)(MASK((s)->Symbol)); s++; if ((Int32)count < 0) break; - }; + } } s--; - RC_DecodeFinal((hiCnt - count) - s->Freq, s->Freq); + RC_DecodeFinal((hiCnt - count) - s->Freq, s->Freq) // new (see->Summ) value can overflow over 16-bits in some rare cases - Ppmd_See_Update(see); + Ppmd_See_UPDATE(see) p->FoundState = s; sym = s->Symbol; Ppmd7_Update2(p); @@ -261,7 +261,7 @@ int Ppmd7z_DecodeSymbol(CPpmd7 *p) if (count >= freqSum) return PPMD7_SYM_ERROR; - RC_Decode(hiCnt, freqSum - hiCnt); + RC_Decode(hiCnt, freqSum - hiCnt) // We increase (see->Summ) for sum of Freqs of all non_Masked symbols. // new (see->Summ) value can overflow over 16-bits in some rare cases @@ -295,3 +295,18 @@ Byte *Ppmd7z_DecodeSymbols(CPpmd7 *p, Byte *buf, const Byte *lim) return buf; } */ + +#undef kTopValue +#undef READ_BYTE +#undef RC_NORM_BASE +#undef RC_NORM_1 +#undef RC_NORM +#undef RC_NORM_LOCAL +#undef RC_NORM_REMOTE +#undef R +#undef RC_Decode +#undef RC_DecodeFinal +#undef RC_GetThreshold +#undef CTX +#undef SUCCESSOR +#undef MASK diff --git a/libraries/lzma/C/Ppmd7Enc.c b/libraries/lzma/C/Ppmd7Enc.c new file mode 100644 index 000000000..41106bab4 --- /dev/null +++ b/libraries/lzma/C/Ppmd7Enc.c @@ -0,0 +1,338 @@ +/* Ppmd7Enc.c -- Ppmd7z (PPMdH with 7z Range Coder) Encoder +2023-04-02 : Igor Pavlov : Public domain +This code is based on: + PPMd var.H (2001): Dmitry Shkarin : Public domain */ + + +#include "Precomp.h" + +#include "Ppmd7.h" + +#define kTopValue ((UInt32)1 << 24) + +#define R (&p->rc.enc) + +void Ppmd7z_Init_RangeEnc(CPpmd7 *p) +{ + R->Low = 0; + R->Range = 0xFFFFFFFF; + R->Cache = 0; + R->CacheSize = 1; +} + +Z7_NO_INLINE +static void Ppmd7z_RangeEnc_ShiftLow(CPpmd7 *p) +{ + if ((UInt32)R->Low < (UInt32)0xFF000000 || (unsigned)(R->Low >> 32) != 0) + { + Byte temp = R->Cache; + do + { + IByteOut_Write(R->Stream, (Byte)(temp + (Byte)(R->Low >> 32))); + temp = 0xFF; + } + while (--R->CacheSize != 0); + R->Cache = (Byte)((UInt32)R->Low >> 24); + } + R->CacheSize++; + R->Low = (UInt32)((UInt32)R->Low << 8); +} + +#define RC_NORM_BASE(p) if (R->Range < kTopValue) { R->Range <<= 8; Ppmd7z_RangeEnc_ShiftLow(p); +#define RC_NORM_1(p) RC_NORM_BASE(p) } +#define RC_NORM(p) RC_NORM_BASE(p) RC_NORM_BASE(p) }} + +// we must use only one type of Normalization from two: LOCAL or REMOTE +#define RC_NORM_LOCAL(p) // RC_NORM(p) +#define RC_NORM_REMOTE(p) RC_NORM(p) + +/* +#define Ppmd7z_RangeEnc_Encode(p, start, _size_) \ + { UInt32 size = _size_; \ + R->Low += start * R->Range; \ + R->Range *= size; \ + RC_NORM_LOCAL(p); } +*/ + +Z7_FORCE_INLINE +// Z7_NO_INLINE +static void Ppmd7z_RangeEnc_Encode(CPpmd7 *p, UInt32 start, UInt32 size) +{ + R->Low += start * R->Range; + R->Range *= size; + RC_NORM_LOCAL(p) +} + +void Ppmd7z_Flush_RangeEnc(CPpmd7 *p) +{ + unsigned i; + for (i = 0; i < 5; i++) + Ppmd7z_RangeEnc_ShiftLow(p); +} + + + +#define RC_Encode(start, size) Ppmd7z_RangeEnc_Encode(p, start, size); +#define RC_EncodeFinal(start, size) RC_Encode(start, size) RC_NORM_REMOTE(p) + +#define CTX(ref) ((CPpmd7_Context *)Ppmd7_GetContext(p, ref)) +#define SUFFIX(ctx) CTX((ctx)->Suffix) +// typedef CPpmd7_Context * CTX_PTR; +#define SUCCESSOR(p) Ppmd_GET_SUCCESSOR(p) + +void Ppmd7_UpdateModel(CPpmd7 *p); + +#define MASK(sym) ((unsigned char *)charMask)[sym] + +Z7_FORCE_INLINE +static +void Ppmd7z_EncodeSymbol(CPpmd7 *p, int symbol) +{ + size_t charMask[256 / sizeof(size_t)]; + + if (p->MinContext->NumStats != 1) + { + CPpmd_State *s = Ppmd7_GetStats(p, p->MinContext); + UInt32 sum; + unsigned i; + + + + + R->Range /= p->MinContext->Union2.SummFreq; + + if (s->Symbol == symbol) + { + // R->Range /= p->MinContext->Union2.SummFreq; + RC_EncodeFinal(0, s->Freq) + p->FoundState = s; + Ppmd7_Update1_0(p); + return; + } + p->PrevSuccess = 0; + sum = s->Freq; + i = (unsigned)p->MinContext->NumStats - 1; + do + { + if ((++s)->Symbol == symbol) + { + // R->Range /= p->MinContext->Union2.SummFreq; + RC_EncodeFinal(sum, s->Freq) + p->FoundState = s; + Ppmd7_Update1(p); + return; + } + sum += s->Freq; + } + while (--i); + + // R->Range /= p->MinContext->Union2.SummFreq; + RC_Encode(sum, p->MinContext->Union2.SummFreq - sum) + + p->HiBitsFlag = PPMD7_HiBitsFlag_3(p->FoundState->Symbol); + PPMD_SetAllBitsIn256Bytes(charMask) + // MASK(s->Symbol) = 0; + // i = p->MinContext->NumStats - 1; + // do { MASK((--s)->Symbol) = 0; } while (--i); + { + CPpmd_State *s2 = Ppmd7_GetStats(p, p->MinContext); + MASK(s->Symbol) = 0; + do + { + unsigned sym0 = s2[0].Symbol; + unsigned sym1 = s2[1].Symbol; + s2 += 2; + MASK(sym0) = 0; + MASK(sym1) = 0; + } + while (s2 < s); + } + } + else + { + UInt16 *prob = Ppmd7_GetBinSumm(p); + CPpmd_State *s = Ppmd7Context_OneState(p->MinContext); + UInt32 pr = *prob; + const UInt32 bound = (R->Range >> 14) * pr; + pr = PPMD_UPDATE_PROB_1(pr); + if (s->Symbol == symbol) + { + *prob = (UInt16)(pr + (1 << PPMD_INT_BITS)); + // RangeEnc_EncodeBit_0(p, bound); + R->Range = bound; + RC_NORM_1(p) + + // p->FoundState = s; + // Ppmd7_UpdateBin(p); + { + const unsigned freq = s->Freq; + CPpmd7_Context *c = CTX(SUCCESSOR(s)); + p->FoundState = s; + p->PrevSuccess = 1; + p->RunLength++; + s->Freq = (Byte)(freq + (freq < 128)); + // NextContext(p); + if (p->OrderFall == 0 && (const Byte *)c > p->Text) + p->MaxContext = p->MinContext = c; + else + Ppmd7_UpdateModel(p); + } + return; + } + + *prob = (UInt16)pr; + p->InitEsc = p->ExpEscape[pr >> 10]; + // RangeEnc_EncodeBit_1(p, bound); + R->Low += bound; + R->Range -= bound; + RC_NORM_LOCAL(p) + + PPMD_SetAllBitsIn256Bytes(charMask) + MASK(s->Symbol) = 0; + p->PrevSuccess = 0; + } + + for (;;) + { + CPpmd_See *see; + CPpmd_State *s; + UInt32 sum, escFreq; + CPpmd7_Context *mc; + unsigned i, numMasked; + + RC_NORM_REMOTE(p) + + mc = p->MinContext; + numMasked = mc->NumStats; + + do + { + p->OrderFall++; + if (!mc->Suffix) + return; /* EndMarker (symbol = -1) */ + mc = Ppmd7_GetContext(p, mc->Suffix); + i = mc->NumStats; + } + while (i == numMasked); + + p->MinContext = mc; + + // see = Ppmd7_MakeEscFreq(p, numMasked, &escFreq); + { + if (i != 256) + { + unsigned nonMasked = i - numMasked; + see = p->See[(unsigned)p->NS2Indx[(size_t)nonMasked - 1]] + + p->HiBitsFlag + + (nonMasked < (unsigned)SUFFIX(mc)->NumStats - i) + + 2 * (unsigned)(mc->Union2.SummFreq < 11 * i) + + 4 * (unsigned)(numMasked > nonMasked); + { + // if (see->Summ) field is larger than 16-bit, we need only low 16 bits of Summ + unsigned summ = (UInt16)see->Summ; // & 0xFFFF + unsigned r = (summ >> see->Shift); + see->Summ = (UInt16)(summ - r); + escFreq = r + (r == 0); + } + } + else + { + see = &p->DummySee; + escFreq = 1; + } + } + + s = Ppmd7_GetStats(p, mc); + sum = 0; + // i = mc->NumStats; + + do + { + const unsigned cur = s->Symbol; + if ((int)cur == symbol) + { + const UInt32 low = sum; + const UInt32 freq = s->Freq; + unsigned num2; + + Ppmd_See_UPDATE(see) + p->FoundState = s; + sum += escFreq; + + num2 = i / 2; + i &= 1; + sum += freq & (0 - (UInt32)i); + if (num2 != 0) + { + s += i; + for (;;) + { + unsigned sym0 = s[0].Symbol; + unsigned sym1 = s[1].Symbol; + s += 2; + sum += (s[-2].Freq & (unsigned)(MASK(sym0))); + sum += (s[-1].Freq & (unsigned)(MASK(sym1))); + if (--num2 == 0) + break; + } + } + + + R->Range /= sum; + RC_EncodeFinal(low, freq) + Ppmd7_Update2(p); + return; + } + sum += (s->Freq & (unsigned)(MASK(cur))); + s++; + } + while (--i); + + { + const UInt32 total = sum + escFreq; + see->Summ = (UInt16)(see->Summ + total); + + R->Range /= total; + RC_Encode(sum, escFreq) + } + + { + const CPpmd_State *s2 = Ppmd7_GetStats(p, p->MinContext); + s--; + MASK(s->Symbol) = 0; + do + { + const unsigned sym0 = s2[0].Symbol; + const unsigned sym1 = s2[1].Symbol; + s2 += 2; + MASK(sym0) = 0; + MASK(sym1) = 0; + } + while (s2 < s); + } + } +} + + +void Ppmd7z_EncodeSymbols(CPpmd7 *p, const Byte *buf, const Byte *lim) +{ + for (; buf < lim; buf++) + { + Ppmd7z_EncodeSymbol(p, *buf); + } +} + +#undef kTopValue +#undef WRITE_BYTE +#undef RC_NORM_BASE +#undef RC_NORM_1 +#undef RC_NORM +#undef RC_NORM_LOCAL +#undef RC_NORM_REMOTE +#undef R +#undef RC_Encode +#undef RC_EncodeFinal +#undef SUFFIX +#undef CTX +#undef SUCCESSOR +#undef MASK diff --git a/libraries/lzma/C/Precomp.h b/libraries/lzma/C/Precomp.h index e8ff8b40e..69afb2ffd 100644 --- a/libraries/lzma/C/Precomp.h +++ b/libraries/lzma/C/Precomp.h @@ -1,8 +1,8 @@ /* Precomp.h -- StdAfx -2013-11-12 : Igor Pavlov : Public domain */ +2023-04-02 : Igor Pavlov : Public domain */ -#ifndef __7Z_PRECOMP_H -#define __7Z_PRECOMP_H +#ifndef ZIP7_INC_PRECOMP_H +#define ZIP7_INC_PRECOMP_H #include "Compiler.h" /* #include "7zTypes.h" */ diff --git a/libraries/lzma/C/RotateDefs.h b/libraries/lzma/C/RotateDefs.h new file mode 100644 index 000000000..c16b4f8e6 --- /dev/null +++ b/libraries/lzma/C/RotateDefs.h @@ -0,0 +1,50 @@ +/* RotateDefs.h -- Rotate functions +2023-06-18 : Igor Pavlov : Public domain */ + +#ifndef ZIP7_INC_ROTATE_DEFS_H +#define ZIP7_INC_ROTATE_DEFS_H + +#ifdef _MSC_VER + +#include + +/* don't use _rotl with old 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)) + +#if (_MSC_VER >= 1300) +#define Z7_ROTL64(x, n) _rotl64((x), (n)) +#define Z7_ROTR64(x, n) _rotr64((x), (n)) +#else +#define Z7_ROTL64(x, n) (((x) << (n)) | ((x) >> (64 - (n)))) +#define Z7_ROTR64(x, n) (((x) >> (n)) | ((x) << (64 - (n)))) +#endif + +#else + +/* new compilers can translate these macros to fast commands. */ + +#if defined(__clang__) && (__clang_major__ >= 4) \ + || defined(__GNUC__) && (__GNUC__ >= 5) +/* GCC 4.9.0 and clang 3.5 can recognize more correct version: */ +#define rotlFixed(x, n) (((x) << (n)) | ((x) >> (-(n) & 31))) +#define rotrFixed(x, n) (((x) >> (n)) | ((x) << (-(n) & 31))) +#define Z7_ROTL64(x, n) (((x) << (n)) | ((x) >> (-(n) & 63))) +#define Z7_ROTR64(x, n) (((x) >> (n)) | ((x) << (-(n) & 63))) +#else +/* for old GCC / clang: */ +#define rotlFixed(x, n) (((x) << (n)) | ((x) >> (32 - (n)))) +#define rotrFixed(x, n) (((x) >> (n)) | ((x) << (32 - (n)))) +#define Z7_ROTL64(x, n) (((x) << (n)) | ((x) >> (64 - (n)))) +#define Z7_ROTR64(x, n) (((x) >> (n)) | ((x) << (64 - (n)))) +#endif + +#endif + +#endif diff --git a/libraries/lzma/C/Sha256.c b/libraries/lzma/C/Sha256.c new file mode 100644 index 000000000..018cf6f4b --- /dev/null +++ b/libraries/lzma/C/Sha256.c @@ -0,0 +1,516 @@ +/* Sha256.c -- SHA-256 Hash +2023-04-02 : Igor Pavlov : Public domain +This code is based on public domain code from Wei Dai's Crypto++ library. */ + +#include "Precomp.h" + +#include + +#include "CpuArch.h" +#include "RotateDefs.h" +#include "Sha256.h" + +#if defined(_MSC_VER) && (_MSC_VER < 1900) +// #define USE_MY_MM +#endif + +#ifdef MY_CPU_X86_OR_AMD64 + #ifdef _MSC_VER + #if _MSC_VER >= 1200 + #define Z7_COMPILER_SHA256_SUPPORTED + #endif + #elif defined(__clang__) + #if (__clang_major__ >= 8) // fix that check + #define Z7_COMPILER_SHA256_SUPPORTED + #endif + #elif defined(__GNUC__) + #if (__GNUC__ >= 8) // fix that check + #define Z7_COMPILER_SHA256_SUPPORTED + #endif + #elif defined(__INTEL_COMPILER) + #if (__INTEL_COMPILER >= 1800) // fix that check + #define Z7_COMPILER_SHA256_SUPPORTED + #endif + #endif +#elif defined(MY_CPU_ARM_OR_ARM64) + #ifdef _MSC_VER + #if _MSC_VER >= 1910 + #define Z7_COMPILER_SHA256_SUPPORTED + #endif + #elif defined(__clang__) + #if (__clang_major__ >= 8) // fix that check + #define Z7_COMPILER_SHA256_SUPPORTED + #endif + #elif defined(__GNUC__) + #if (__GNUC__ >= 6) // fix that check + #define Z7_COMPILER_SHA256_SUPPORTED + #endif + #endif +#endif + +void Z7_FASTCALL Sha256_UpdateBlocks(UInt32 state[8], const Byte *data, size_t numBlocks); + +#ifdef Z7_COMPILER_SHA256_SUPPORTED + void Z7_FASTCALL Sha256_UpdateBlocks_HW(UInt32 state[8], const Byte *data, size_t numBlocks); + + static SHA256_FUNC_UPDATE_BLOCKS g_SHA256_FUNC_UPDATE_BLOCKS = Sha256_UpdateBlocks; + static SHA256_FUNC_UPDATE_BLOCKS g_SHA256_FUNC_UPDATE_BLOCKS_HW; + + #define SHA256_UPDATE_BLOCKS(p) p->func_UpdateBlocks +#else + #define SHA256_UPDATE_BLOCKS(p) Sha256_UpdateBlocks +#endif + + +BoolInt Sha256_SetFunction(CSha256 *p, unsigned algo) +{ + SHA256_FUNC_UPDATE_BLOCKS func = Sha256_UpdateBlocks; + + #ifdef Z7_COMPILER_SHA256_SUPPORTED + if (algo != SHA256_ALGO_SW) + { + if (algo == SHA256_ALGO_DEFAULT) + func = g_SHA256_FUNC_UPDATE_BLOCKS; + else + { + if (algo != SHA256_ALGO_HW) + return False; + func = g_SHA256_FUNC_UPDATE_BLOCKS_HW; + if (!func) + return False; + } + } + #else + if (algo > 1) + return False; + #endif + + p->func_UpdateBlocks = func; + return True; +} + + +/* define it for speed optimization */ + +#ifdef Z7_SFX + #define STEP_PRE 1 + #define STEP_MAIN 1 +#else + #define STEP_PRE 2 + #define STEP_MAIN 4 + // #define Z7_SHA256_UNROLL +#endif + +#undef Z7_SHA256_BIG_W +#if STEP_MAIN != 16 + #define Z7_SHA256_BIG_W +#endif + + + + +void Sha256_InitState(CSha256 *p) +{ + p->count = 0; + 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; +} + +void Sha256_Init(CSha256 *p) +{ + p->func_UpdateBlocks = + #ifdef Z7_COMPILER_SHA256_SUPPORTED + g_SHA256_FUNC_UPDATE_BLOCKS; + #else + NULL; + #endif + Sha256_InitState(p); +} + +#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 Ch(x,y,z) (z^(x&(y^z))) +#define Maj(x,y,z) ((x&y)|(z&(x|y))) + + +#define W_PRE(i) (W[(i) + (size_t)(j)] = GetBe32(data + ((size_t)(j) + i) * 4)) + +#define blk2_main(j, i) s1(w(j, (i)-2)) + w(j, (i)-7) + s0(w(j, (i)-15)) + +#ifdef Z7_SHA256_BIG_W + // we use +i instead of +(i) to change the order to solve CLANG compiler warning for signed/unsigned. + #define w(j, i) W[(size_t)(j) + i] + #define blk2(j, i) (w(j, i) = w(j, (i)-16) + blk2_main(j, i)) +#else + #if STEP_MAIN == 16 + #define w(j, i) W[(i) & 15] + #else + #define w(j, i) W[((size_t)(j) + (i)) & 15] + #endif + #define blk2(j, i) (w(j, i) += blk2_main(j, i)) +#endif + +#define W_MAIN(i) blk2(j, i) + + +#define T1(wx, i) \ + tmp = h + S1(e) + Ch(e,f,g) + K[(i)+(size_t)(j)] + wx(i); \ + h = g; \ + g = f; \ + f = e; \ + e = d + tmp; \ + tmp += S0(a) + Maj(a, b, c); \ + d = c; \ + c = b; \ + b = a; \ + a = tmp; \ + +#define R1_PRE(i) T1( W_PRE, i) +#define R1_MAIN(i) T1( W_MAIN, i) + +#if (!defined(Z7_SHA256_UNROLL) || STEP_MAIN < 8) && (STEP_MAIN >= 4) +#define R2_MAIN(i) \ + R1_MAIN(i) \ + R1_MAIN(i + 1) \ + +#endif + + + +#if defined(Z7_SHA256_UNROLL) && STEP_MAIN >= 8 + +#define T4( a,b,c,d,e,f,g,h, wx, i) \ + h += S1(e) + Ch(e,f,g) + K[(i)+(size_t)(j)] + wx(i); \ + tmp = h; \ + h += d; \ + d = tmp + S0(a) + Maj(a, b, c); \ + +#define R4( wx, i) \ + T4 ( a,b,c,d,e,f,g,h, wx, (i )); \ + T4 ( d,a,b,c,h,e,f,g, wx, (i+1)); \ + T4 ( c,d,a,b,g,h,e,f, wx, (i+2)); \ + T4 ( b,c,d,a,f,g,h,e, wx, (i+3)); \ + +#define R4_PRE(i) R4( W_PRE, i) +#define R4_MAIN(i) R4( W_MAIN, i) + + +#define T8( a,b,c,d,e,f,g,h, wx, i) \ + h += S1(e) + Ch(e,f,g) + K[(i)+(size_t)(j)] + wx(i); \ + d += h; \ + h += S0(a) + Maj(a, b, c); \ + +#define R8( wx, i) \ + T8 ( a,b,c,d,e,f,g,h, wx, i ); \ + T8 ( h,a,b,c,d,e,f,g, wx, i+1); \ + T8 ( g,h,a,b,c,d,e,f, wx, i+2); \ + T8 ( f,g,h,a,b,c,d,e, wx, i+3); \ + T8 ( e,f,g,h,a,b,c,d, wx, i+4); \ + T8 ( d,e,f,g,h,a,b,c, wx, i+5); \ + T8 ( c,d,e,f,g,h,a,b, wx, i+6); \ + T8 ( b,c,d,e,f,g,h,a, wx, i+7); \ + +#define R8_PRE(i) R8( W_PRE, i) +#define R8_MAIN(i) R8( W_MAIN, i) + +#endif + +void Z7_FASTCALL Sha256_UpdateBlocks_HW(UInt32 state[8], const Byte *data, size_t numBlocks); + +// static +extern MY_ALIGN(64) +const UInt32 SHA256_K_ARRAY[64]; + +MY_ALIGN(64) +const UInt32 SHA256_K_ARRAY[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 +}; + +#define K SHA256_K_ARRAY + + +Z7_NO_INLINE +void Z7_FASTCALL Sha256_UpdateBlocks(UInt32 state[8], const Byte *data, size_t numBlocks) +{ + UInt32 W + #ifdef Z7_SHA256_BIG_W + [64]; + #else + [16]; + #endif + + unsigned j; + + UInt32 a,b,c,d,e,f,g,h; + + #if !defined(Z7_SHA256_UNROLL) || (STEP_MAIN <= 4) || (STEP_PRE <= 4) + UInt32 tmp; + #endif + + a = state[0]; + b = state[1]; + c = state[2]; + d = state[3]; + e = state[4]; + f = state[5]; + g = state[6]; + h = state[7]; + + while (numBlocks) + { + + for (j = 0; j < 16; j += STEP_PRE) + { + #if STEP_PRE > 4 + + #if STEP_PRE < 8 + R4_PRE(0); + #else + R8_PRE(0); + #if STEP_PRE == 16 + R8_PRE(8); + #endif + #endif + + #else + + R1_PRE(0) + #if STEP_PRE >= 2 + R1_PRE(1) + #if STEP_PRE >= 4 + R1_PRE(2) + R1_PRE(3) + #endif + #endif + + #endif + } + + for (j = 16; j < 64; j += STEP_MAIN) + { + #if defined(Z7_SHA256_UNROLL) && STEP_MAIN >= 8 + + #if STEP_MAIN < 8 + R4_MAIN(0) + #else + R8_MAIN(0) + #if STEP_MAIN == 16 + R8_MAIN(8) + #endif + #endif + + #else + + R1_MAIN(0) + #if STEP_MAIN >= 2 + R1_MAIN(1) + #if STEP_MAIN >= 4 + R2_MAIN(2) + #if STEP_MAIN >= 8 + R2_MAIN(4) + R2_MAIN(6) + #if STEP_MAIN >= 16 + R2_MAIN(8) + R2_MAIN(10) + R2_MAIN(12) + R2_MAIN(14) + #endif + #endif + #endif + #endif + #endif + } + + a += state[0]; state[0] = a; + b += state[1]; state[1] = b; + c += state[2]; state[2] = c; + d += state[3]; state[3] = d; + e += state[4]; state[4] = e; + f += state[5]; state[5] = f; + g += state[6]; state[6] = g; + h += state[7]; state[7] = h; + + data += 64; + numBlocks--; + } + + /* Wipe variables */ + /* memset(W, 0, sizeof(W)); */ +} + +#undef S0 +#undef S1 +#undef s0 +#undef s1 +#undef K + +#define Sha256_UpdateBlock(p) SHA256_UPDATE_BLOCKS(p)(p->state, p->buffer, 1) + +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; + } + + if (pos != 0) + { + size -= num; + memcpy(p->buffer + pos, data, num); + data += num; + Sha256_UpdateBlock(p); + } + } + { + size_t numBlocks = size >> 6; + SHA256_UPDATE_BLOCKS(p)(p->state, data, numBlocks); + size &= 0x3F; + if (size == 0) + return; + data += (numBlocks << 6); + memcpy(p->buffer, data, size); + } +} + + +void Sha256_Final(CSha256 *p, Byte *digest) +{ + unsigned pos = (unsigned)p->count & 0x3F; + unsigned i; + + p->buffer[pos++] = 0x80; + + if (pos > (64 - 8)) + { + while (pos != 64) { p->buffer[pos++] = 0; } + // memset(&p->buf.buffer[pos], 0, 64 - pos); + Sha256_UpdateBlock(p); + pos = 0; + } + + /* + if (pos & 3) + { + p->buffer[pos] = 0; + p->buffer[pos + 1] = 0; + p->buffer[pos + 2] = 0; + pos += 3; + pos &= ~3; + } + { + for (; pos < 64 - 8; pos += 4) + *(UInt32 *)(&p->buffer[pos]) = 0; + } + */ + + memset(&p->buffer[pos], 0, (64 - 8) - pos); + + { + UInt64 numBits = (p->count << 3); + SetBe32(p->buffer + 64 - 8, (UInt32)(numBits >> 32)) + SetBe32(p->buffer + 64 - 4, (UInt32)(numBits)) + } + + Sha256_UpdateBlock(p); + + for (i = 0; i < 8; i += 2) + { + UInt32 v0 = p->state[i]; + UInt32 v1 = p->state[(size_t)i + 1]; + SetBe32(digest , v0) + SetBe32(digest + 4, v1) + digest += 8; + } + + Sha256_InitState(p); +} + + +void Sha256Prepare(void) +{ + #ifdef Z7_COMPILER_SHA256_SUPPORTED + SHA256_FUNC_UPDATE_BLOCKS f, f_hw; + f = Sha256_UpdateBlocks; + f_hw = NULL; + #ifdef MY_CPU_X86_OR_AMD64 + #ifndef USE_MY_MM + if (CPU_IsSupported_SHA() + && CPU_IsSupported_SSSE3() + // && CPU_IsSupported_SSE41() + ) + #endif + #else + if (CPU_IsSupported_SHA2()) + #endif + { + // printf("\n========== HW SHA256 ======== \n"); + f = f_hw = Sha256_UpdateBlocks_HW; + } + g_SHA256_FUNC_UPDATE_BLOCKS = f; + g_SHA256_FUNC_UPDATE_BLOCKS_HW = f_hw; + #endif +} + +#undef S0 +#undef S1 +#undef s0 +#undef s1 +#undef Ch +#undef Maj +#undef W_MAIN +#undef W_PRE +#undef w +#undef blk2_main +#undef blk2 +#undef T1 +#undef T4 +#undef T8 +#undef R1_PRE +#undef R1_MAIN +#undef R2_MAIN +#undef R4 +#undef R4_PRE +#undef R4_MAIN +#undef R8 +#undef R8_PRE +#undef R8_MAIN +#undef STEP_PRE +#undef STEP_MAIN +#undef Z7_SHA256_BIG_W +#undef Z7_SHA256_UNROLL +#undef Z7_COMPILER_SHA256_SUPPORTED diff --git a/libraries/lzma/C/Sha256.h b/libraries/lzma/C/Sha256.h new file mode 100644 index 000000000..9e0422320 --- /dev/null +++ b/libraries/lzma/C/Sha256.h @@ -0,0 +1,76 @@ +/* Sha256.h -- SHA-256 Hash +2023-04-02 : Igor Pavlov : Public domain */ + +#ifndef ZIP7_INC_SHA256_H +#define ZIP7_INC_SHA256_H + +#include "7zTypes.h" + +EXTERN_C_BEGIN + +#define SHA256_NUM_BLOCK_WORDS 16 +#define SHA256_NUM_DIGEST_WORDS 8 + +#define SHA256_BLOCK_SIZE (SHA256_NUM_BLOCK_WORDS * 4) +#define SHA256_DIGEST_SIZE (SHA256_NUM_DIGEST_WORDS * 4) + +typedef void (Z7_FASTCALL *SHA256_FUNC_UPDATE_BLOCKS)(UInt32 state[8], const Byte *data, size_t numBlocks); + +/* + if (the system supports different SHA256 code implementations) + { + (CSha256::func_UpdateBlocks) will be used + (CSha256::func_UpdateBlocks) can be set by + Sha256_Init() - to default (fastest) + Sha256_SetFunction() - to any algo + } + else + { + (CSha256::func_UpdateBlocks) is ignored. + } +*/ + +typedef struct +{ + SHA256_FUNC_UPDATE_BLOCKS func_UpdateBlocks; + UInt64 count; + UInt64 _pad_2[2]; + UInt32 state[SHA256_NUM_DIGEST_WORDS]; + + Byte buffer[SHA256_BLOCK_SIZE]; +} CSha256; + + +#define SHA256_ALGO_DEFAULT 0 +#define SHA256_ALGO_SW 1 +#define SHA256_ALGO_HW 2 + +/* +Sha256_SetFunction() +return: + 0 - (algo) value is not supported, and func_UpdateBlocks was not changed + 1 - func_UpdateBlocks was set according (algo) value. +*/ + +BoolInt Sha256_SetFunction(CSha256 *p, unsigned algo); + +void Sha256_InitState(CSha256 *p); +void Sha256_Init(CSha256 *p); +void Sha256_Update(CSha256 *p, const Byte *data, size_t size); +void Sha256_Final(CSha256 *p, Byte *digest); + + + + +// void Z7_FASTCALL Sha256_UpdateBlocks(UInt32 state[8], const Byte *data, size_t numBlocks); + +/* +call Sha256Prepare() once at program start. +It prepares all supported implementations, and detects the fastest implementation. +*/ + +void Sha256Prepare(void); + +EXTERN_C_END + +#endif diff --git a/libraries/lzma/C/Sha256Opt.c b/libraries/lzma/C/Sha256Opt.c new file mode 100644 index 000000000..e4465e3e7 --- /dev/null +++ b/libraries/lzma/C/Sha256Opt.c @@ -0,0 +1,386 @@ +/* Sha256Opt.c -- SHA-256 optimized code for SHA-256 hardware instructions +2023-04-02 : Igor Pavlov : Public domain */ + +#include "Precomp.h" +#include "Compiler.h" +#include "CpuArch.h" + +#if defined(_MSC_VER) +#if (_MSC_VER < 1900) && (_MSC_VER >= 1200) +// #define USE_MY_MM +#endif +#endif + +#ifdef MY_CPU_X86_OR_AMD64 + #if defined(__INTEL_COMPILER) && (__INTEL_COMPILER >= 1600) // fix that check + #define USE_HW_SHA + #elif defined(Z7_LLVM_CLANG_VERSION) && (Z7_LLVM_CLANG_VERSION >= 30800) \ + || defined(Z7_APPLE_CLANG_VERSION) && (Z7_APPLE_CLANG_VERSION >= 50100) \ + || defined(Z7_GCC_VERSION) && (Z7_GCC_VERSION >= 40900) + #define USE_HW_SHA + #if !defined(_INTEL_COMPILER) + // icc defines __GNUC__, but icc doesn't support __attribute__(__target__) + #if !defined(__SHA__) || !defined(__SSSE3__) + #define ATTRIB_SHA __attribute__((__target__("sha,ssse3"))) + #endif + #endif + #elif defined(_MSC_VER) + #ifdef USE_MY_MM + #define USE_VER_MIN 1300 + #else + #define USE_VER_MIN 1900 + #endif + #if (_MSC_VER >= USE_VER_MIN) + #define USE_HW_SHA + #endif + #endif +// #endif // MY_CPU_X86_OR_AMD64 + +#ifdef USE_HW_SHA + +// #pragma message("Sha256 HW") + +// sse/sse2/ssse3: +#include +// sha*: +#include + +#if defined (__clang__) && defined(_MSC_VER) + // #if !defined(__SSSE3__) + // #endif + #if !defined(__SHA__) + #include + #endif +#else + +#ifdef USE_MY_MM +#include "My_mm.h" +#endif + +#endif + +/* +SHA256 uses: +SSE2: + _mm_loadu_si128 + _mm_storeu_si128 + _mm_set_epi32 + _mm_add_epi32 + _mm_shuffle_epi32 / pshufd + + + +SSSE3: + _mm_shuffle_epi8 / pshufb + _mm_alignr_epi8 +SHA: + _mm_sha256* +*/ + +// K array must be aligned for 16-bytes at least. +// The compiler can look align attribute and selects +// movdqu - for code without align attribute +// movdqa - for code with align attribute +extern +MY_ALIGN(64) +const UInt32 SHA256_K_ARRAY[64]; + +#define K SHA256_K_ARRAY + + +#define ADD_EPI32(dest, src) dest = _mm_add_epi32(dest, src); +#define SHA256_MSG1(dest, src) dest = _mm_sha256msg1_epu32(dest, src); +#define SHA25G_MSG2(dest, src) dest = _mm_sha256msg2_epu32(dest, src); + + +#define LOAD_SHUFFLE(m, k) \ + m = _mm_loadu_si128((const __m128i *)(const void *)(data + (k) * 16)); \ + m = _mm_shuffle_epi8(m, mask); \ + +#define SM1(g0, g1, g2, g3) \ + SHA256_MSG1(g3, g0); \ + +#define SM2(g0, g1, g2, g3) \ + tmp = _mm_alignr_epi8(g1, g0, 4); \ + ADD_EPI32(g2, tmp) \ + SHA25G_MSG2(g2, g1); \ + +// #define LS0(k, g0, g1, g2, g3) LOAD_SHUFFLE(g0, k) +// #define LS1(k, g0, g1, g2, g3) LOAD_SHUFFLE(g1, k+1) + + +#define NNN(g0, g1, g2, g3) + + +#define RND2(t0, t1) \ + t0 = _mm_sha256rnds2_epu32(t0, t1, msg); + +#define RND2_0(m, k) \ + msg = _mm_add_epi32(m, *(const __m128i *) (const void *) &K[(k) * 4]); \ + RND2(state0, state1); \ + msg = _mm_shuffle_epi32(msg, 0x0E); \ + + +#define RND2_1 \ + RND2(state1, state0); \ + + +// We use scheme with 3 rounds ahead for SHA256_MSG1 / 2 rounds ahead for SHA256_MSG2 + +#define R4(k, g0, g1, g2, g3, OP0, OP1) \ + RND2_0(g0, k) \ + OP0(g0, g1, g2, g3) \ + RND2_1 \ + OP1(g0, g1, g2, g3) \ + +#define R16(k, OP0, OP1, OP2, OP3, OP4, OP5, OP6, OP7) \ + R4 ( (k)*4+0, m0,m1,m2,m3, OP0, OP1 ) \ + R4 ( (k)*4+1, m1,m2,m3,m0, OP2, OP3 ) \ + R4 ( (k)*4+2, m2,m3,m0,m1, OP4, OP5 ) \ + R4 ( (k)*4+3, m3,m0,m1,m2, OP6, OP7 ) \ + +#define PREPARE_STATE \ + tmp = _mm_shuffle_epi32(state0, 0x1B); /* abcd */ \ + state0 = _mm_shuffle_epi32(state1, 0x1B); /* efgh */ \ + state1 = state0; \ + state0 = _mm_unpacklo_epi64(state0, tmp); /* cdgh */ \ + state1 = _mm_unpackhi_epi64(state1, tmp); /* abef */ \ + + +void Z7_FASTCALL Sha256_UpdateBlocks_HW(UInt32 state[8], const Byte *data, size_t numBlocks); +#ifdef ATTRIB_SHA +ATTRIB_SHA +#endif +void Z7_FASTCALL Sha256_UpdateBlocks_HW(UInt32 state[8], const Byte *data, size_t numBlocks) +{ + const __m128i mask = _mm_set_epi32(0x0c0d0e0f, 0x08090a0b, 0x04050607, 0x00010203); + __m128i tmp; + __m128i state0, state1; + + if (numBlocks == 0) + return; + + state0 = _mm_loadu_si128((const __m128i *) (const void *) &state[0]); + state1 = _mm_loadu_si128((const __m128i *) (const void *) &state[4]); + + PREPARE_STATE + + do + { + __m128i state0_save, state1_save; + __m128i m0, m1, m2, m3; + __m128i msg; + // #define msg tmp + + state0_save = state0; + state1_save = state1; + + LOAD_SHUFFLE (m0, 0) + LOAD_SHUFFLE (m1, 1) + LOAD_SHUFFLE (m2, 2) + LOAD_SHUFFLE (m3, 3) + + + + R16 ( 0, NNN, NNN, SM1, NNN, SM1, SM2, SM1, SM2 ) + R16 ( 1, SM1, SM2, SM1, SM2, SM1, SM2, SM1, SM2 ) + R16 ( 2, SM1, SM2, SM1, SM2, SM1, SM2, SM1, SM2 ) + R16 ( 3, SM1, SM2, NNN, SM2, NNN, NNN, NNN, NNN ) + + ADD_EPI32(state0, state0_save) + ADD_EPI32(state1, state1_save) + + data += 64; + } + while (--numBlocks); + + PREPARE_STATE + + _mm_storeu_si128((__m128i *) (void *) &state[0], state0); + _mm_storeu_si128((__m128i *) (void *) &state[4], state1); +} + +#endif // USE_HW_SHA + +#elif defined(MY_CPU_ARM_OR_ARM64) + + #if defined(__clang__) + #if (__clang_major__ >= 8) // fix that check + #define USE_HW_SHA + #endif + #elif defined(__GNUC__) + #if (__GNUC__ >= 6) // fix that check + #define USE_HW_SHA + #endif + #elif defined(_MSC_VER) + #if _MSC_VER >= 1910 + #define USE_HW_SHA + #endif + #endif + +#ifdef USE_HW_SHA + +// #pragma message("=== Sha256 HW === ") + +#if defined(__clang__) || defined(__GNUC__) + #ifdef MY_CPU_ARM64 + #define ATTRIB_SHA __attribute__((__target__("+crypto"))) + #else + #define ATTRIB_SHA __attribute__((__target__("fpu=crypto-neon-fp-armv8"))) + #endif +#else + // _MSC_VER + // for arm32 + #define _ARM_USE_NEW_NEON_INTRINSICS +#endif + +#if defined(_MSC_VER) && defined(MY_CPU_ARM64) +#include +#else +#include +#endif + +typedef uint32x4_t v128; +// typedef __n128 v128; // MSVC + +#ifdef MY_CPU_BE + #define MY_rev32_for_LE(x) +#else + #define MY_rev32_for_LE(x) x = vreinterpretq_u32_u8(vrev32q_u8(vreinterpretq_u8_u32(x))) +#endif + +#define LOAD_128(_p) (*(const v128 *)(const void *)(_p)) +#define STORE_128(_p, _v) *(v128 *)(void *)(_p) = (_v) + +#define LOAD_SHUFFLE(m, k) \ + m = LOAD_128((data + (k) * 16)); \ + MY_rev32_for_LE(m); \ + +// K array must be aligned for 16-bytes at least. +extern +MY_ALIGN(64) +const UInt32 SHA256_K_ARRAY[64]; + +#define K SHA256_K_ARRAY + + +#define SHA256_SU0(dest, src) dest = vsha256su0q_u32(dest, src); +#define SHA25G_SU1(dest, src2, src3) dest = vsha256su1q_u32(dest, src2, src3); + +#define SM1(g0, g1, g2, g3) SHA256_SU0(g3, g0) +#define SM2(g0, g1, g2, g3) SHA25G_SU1(g2, g0, g1) +#define NNN(g0, g1, g2, g3) + + +#define R4(k, g0, g1, g2, g3, OP0, OP1) \ + msg = vaddq_u32(g0, *(const v128 *) (const void *) &K[(k) * 4]); \ + tmp = state0; \ + state0 = vsha256hq_u32( state0, state1, msg ); \ + state1 = vsha256h2q_u32( state1, tmp, msg ); \ + OP0(g0, g1, g2, g3); \ + OP1(g0, g1, g2, g3); \ + + +#define R16(k, OP0, OP1, OP2, OP3, OP4, OP5, OP6, OP7) \ + R4 ( (k)*4+0, m0, m1, m2, m3, OP0, OP1 ) \ + R4 ( (k)*4+1, m1, m2, m3, m0, OP2, OP3 ) \ + R4 ( (k)*4+2, m2, m3, m0, m1, OP4, OP5 ) \ + R4 ( (k)*4+3, m3, m0, m1, m2, OP6, OP7 ) \ + + +void Z7_FASTCALL Sha256_UpdateBlocks_HW(UInt32 state[8], const Byte *data, size_t numBlocks); +#ifdef ATTRIB_SHA +ATTRIB_SHA +#endif +void Z7_FASTCALL Sha256_UpdateBlocks_HW(UInt32 state[8], const Byte *data, size_t numBlocks) +{ + v128 state0, state1; + + if (numBlocks == 0) + return; + + state0 = LOAD_128(&state[0]); + state1 = LOAD_128(&state[4]); + + do + { + v128 state0_save, state1_save; + v128 m0, m1, m2, m3; + v128 msg, tmp; + + state0_save = state0; + state1_save = state1; + + LOAD_SHUFFLE (m0, 0) + LOAD_SHUFFLE (m1, 1) + LOAD_SHUFFLE (m2, 2) + LOAD_SHUFFLE (m3, 3) + + R16 ( 0, NNN, NNN, SM1, NNN, SM1, SM2, SM1, SM2 ); + R16 ( 1, SM1, SM2, SM1, SM2, SM1, SM2, SM1, SM2 ); + R16 ( 2, SM1, SM2, SM1, SM2, SM1, SM2, SM1, SM2 ); + R16 ( 3, SM1, SM2, NNN, SM2, NNN, NNN, NNN, NNN ); + + state0 = vaddq_u32(state0, state0_save); + state1 = vaddq_u32(state1, state1_save); + + data += 64; + } + while (--numBlocks); + + STORE_128(&state[0], state0); + STORE_128(&state[4], state1); +} + +#endif // USE_HW_SHA + +#endif // MY_CPU_ARM_OR_ARM64 + + +#ifndef USE_HW_SHA + +// #error Stop_Compiling_UNSUPPORTED_SHA +// #include + +// #include "Sha256.h" +void Z7_FASTCALL Sha256_UpdateBlocks(UInt32 state[8], const Byte *data, size_t numBlocks); + +#pragma message("Sha256 HW-SW stub was used") + +void Z7_FASTCALL Sha256_UpdateBlocks_HW(UInt32 state[8], const Byte *data, size_t numBlocks); +void Z7_FASTCALL Sha256_UpdateBlocks_HW(UInt32 state[8], const Byte *data, size_t numBlocks) +{ + Sha256_UpdateBlocks(state, data, numBlocks); + /* + UNUSED_VAR(state); + UNUSED_VAR(data); + UNUSED_VAR(numBlocks); + exit(1); + return; + */ +} + +#endif + + + +#undef K +#undef RND2 +#undef RND2_0 +#undef RND2_1 + +#undef MY_rev32_for_LE +#undef NNN +#undef LOAD_128 +#undef STORE_128 +#undef LOAD_SHUFFLE +#undef SM1 +#undef SM2 + +#undef NNN +#undef R4 +#undef R16 +#undef PREPARE_STATE +#undef USE_HW_SHA +#undef ATTRIB_SHA +#undef USE_VER_MIN diff --git a/libraries/lzma/C/Sort.c b/libraries/lzma/C/Sort.c new file mode 100644 index 000000000..e1097e380 --- /dev/null +++ b/libraries/lzma/C/Sort.c @@ -0,0 +1,141 @@ +/* Sort.c -- Sort functions +2014-04-05 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include "Sort.h" + +#define HeapSortDown(p, k, size, temp) \ + { for (;;) { \ + size_t s = (k << 1); \ + if (s > size) break; \ + if (s < size && p[s + 1] > p[s]) s++; \ + if (temp >= p[s]) break; \ + p[k] = p[s]; k = s; \ + } p[k] = temp; } + +void HeapSort(UInt32 *p, size_t size) +{ + if (size <= 1) + return; + p--; + { + size_t i = size / 2; + do + { + UInt32 temp = p[i]; + size_t k = i; + HeapSortDown(p, k, size, temp) + } + while (--i != 0); + } + /* + do + { + size_t k = 1; + UInt32 temp = p[size]; + p[size--] = p[1]; + HeapSortDown(p, k, size, temp) + } + while (size > 1); + */ + while (size > 3) + { + UInt32 temp = p[size]; + size_t k = (p[3] > p[2]) ? 3 : 2; + p[size--] = p[1]; + p[1] = p[k]; + HeapSortDown(p, k, size, temp) + } + { + UInt32 temp = p[size]; + p[size] = p[1]; + if (size > 2 && p[2] < temp) + { + p[1] = p[2]; + p[2] = temp; + } + else + p[1] = temp; + } +} + +void HeapSort64(UInt64 *p, size_t size) +{ + if (size <= 1) + return; + p--; + { + size_t i = size / 2; + do + { + UInt64 temp = p[i]; + size_t k = i; + HeapSortDown(p, k, size, temp) + } + while (--i != 0); + } + /* + do + { + size_t k = 1; + UInt64 temp = p[size]; + p[size--] = p[1]; + HeapSortDown(p, k, size, temp) + } + while (size > 1); + */ + while (size > 3) + { + UInt64 temp = p[size]; + size_t k = (p[3] > p[2]) ? 3 : 2; + p[size--] = p[1]; + p[1] = p[k]; + HeapSortDown(p, k, size, temp) + } + { + UInt64 temp = p[size]; + p[size] = p[1]; + if (size > 2 && p[2] < temp) + { + p[1] = p[2]; + p[2] = temp; + } + else + p[1] = temp; + } +} + +/* +#define HeapSortRefDown(p, vals, n, size, temp) \ + { size_t k = n; UInt32 val = vals[temp]; for (;;) { \ + size_t s = (k << 1); \ + if (s > size) break; \ + if (s < size && vals[p[s + 1]] > vals[p[s]]) s++; \ + if (val >= vals[p[s]]) break; \ + p[k] = p[s]; k = s; \ + } p[k] = temp; } + +void HeapSortRef(UInt32 *p, UInt32 *vals, size_t size) +{ + if (size <= 1) + return; + p--; + { + size_t i = size / 2; + do + { + UInt32 temp = p[i]; + HeapSortRefDown(p, vals, i, size, temp); + } + while (--i != 0); + } + do + { + UInt32 temp = p[size]; + p[size--] = p[1]; + HeapSortRefDown(p, vals, 1, size, temp); + } + while (size > 1); +} +*/ diff --git a/libraries/lzma/C/Sort.h b/libraries/lzma/C/Sort.h new file mode 100644 index 000000000..1817b652f --- /dev/null +++ b/libraries/lzma/C/Sort.h @@ -0,0 +1,18 @@ +/* Sort.h -- Sort functions +2023-03-05 : Igor Pavlov : Public domain */ + +#ifndef ZIP7_INC_SORT_H +#define ZIP7_INC_SORT_H + +#include "7zTypes.h" + +EXTERN_C_BEGIN + +void HeapSort(UInt32 *p, size_t size); +void HeapSort64(UInt64 *p, size_t size); + +/* void HeapSortRef(UInt32 *p, UInt32 *vals, size_t size); */ + +EXTERN_C_END + +#endif diff --git a/libraries/lzma/C/SwapBytes.c b/libraries/lzma/C/SwapBytes.c new file mode 100644 index 000000000..7901bbaa8 --- /dev/null +++ b/libraries/lzma/C/SwapBytes.c @@ -0,0 +1,800 @@ +/* SwapBytes.c -- Byte Swap conversion filter +2023-04-07 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include "Compiler.h" +#include "CpuArch.h" +#include "RotateDefs.h" +#include "SwapBytes.h" + +typedef UInt16 CSwapUInt16; +typedef UInt32 CSwapUInt32; + +// #define k_SwapBytes_Mode_BASE 0 + +#ifdef MY_CPU_X86_OR_AMD64 + +#define k_SwapBytes_Mode_SSE2 1 +#define k_SwapBytes_Mode_SSSE3 2 +#define k_SwapBytes_Mode_AVX2 3 + + // #if defined(__INTEL_COMPILER) && (__INTEL_COMPILER >= 1900) + #if defined(__clang__) && (__clang_major__ >= 4) \ + || defined(Z7_GCC_VERSION) && (Z7_GCC_VERSION >= 40701) + #define k_SwapBytes_Mode_MAX k_SwapBytes_Mode_AVX2 + #define SWAP_ATTRIB_SSE2 __attribute__((__target__("sse2"))) + #define SWAP_ATTRIB_SSSE3 __attribute__((__target__("ssse3"))) + #define SWAP_ATTRIB_AVX2 __attribute__((__target__("avx2"))) + #elif defined(_MSC_VER) + #if (_MSC_VER == 1900) + #pragma warning(disable : 4752) // found Intel(R) Advanced Vector Extensions; consider using /arch:AVX + #endif + #if (_MSC_VER >= 1900) + #define k_SwapBytes_Mode_MAX k_SwapBytes_Mode_AVX2 + #elif (_MSC_VER >= 1500) // (VS2008) + #define k_SwapBytes_Mode_MAX k_SwapBytes_Mode_SSSE3 + #elif (_MSC_VER >= 1310) // (VS2003) + #define k_SwapBytes_Mode_MAX k_SwapBytes_Mode_SSE2 + #endif + #endif // _MSC_VER + +/* +// for debug +#ifdef k_SwapBytes_Mode_MAX +#undef k_SwapBytes_Mode_MAX +#endif +*/ + +#ifndef k_SwapBytes_Mode_MAX +#define k_SwapBytes_Mode_MAX 0 +#endif + +#if (k_SwapBytes_Mode_MAX != 0) && defined(MY_CPU_AMD64) + #define k_SwapBytes_Mode_MIN k_SwapBytes_Mode_SSE2 +#else + #define k_SwapBytes_Mode_MIN 0 +#endif + +#if (k_SwapBytes_Mode_MAX >= k_SwapBytes_Mode_AVX2) + #define USE_SWAP_AVX2 +#endif +#if (k_SwapBytes_Mode_MAX >= k_SwapBytes_Mode_SSSE3) + #define USE_SWAP_SSSE3 +#endif +#if (k_SwapBytes_Mode_MAX >= k_SwapBytes_Mode_SSE2) + #define USE_SWAP_128 +#endif + +#if k_SwapBytes_Mode_MAX <= k_SwapBytes_Mode_MIN || !defined(USE_SWAP_128) +#define FORCE_SWAP_MODE +#endif + + +#ifdef USE_SWAP_128 +/* + MMX + SSE + SSE2 + SSE3 + SSSE3 + SSE4.1 + SSE4.2 + SSE4A + AES + AVX, AVX2, FMA +*/ + +#include // sse2 +// typedef __m128i v128; + +#define SWAP2_128(i) { \ + const __m128i v = *(const __m128i *)(const void *)(items + (i) * 8); \ + *( __m128i *)( void *)(items + (i) * 8) = \ + _mm_or_si128( \ + _mm_slli_epi16(v, 8), \ + _mm_srli_epi16(v, 8)); } +// _mm_or_si128() has more ports to execute than _mm_add_epi16(). + +static +#ifdef SWAP_ATTRIB_SSE2 +SWAP_ATTRIB_SSE2 +#endif +void +Z7_FASTCALL +SwapBytes2_128(CSwapUInt16 *items, const CSwapUInt16 *lim) +{ + Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE + do + { + SWAP2_128(0) SWAP2_128(1) items += 2 * 8; + SWAP2_128(0) SWAP2_128(1) items += 2 * 8; + } + while (items != lim); +} + +/* +// sse2 +#define SWAP4_128_pack(i) { \ + __m128i v = *(const __m128i *)(const void *)(items + (i) * 4); \ + __m128i v0 = _mm_unpacklo_epi8(v, mask); \ + __m128i v1 = _mm_unpackhi_epi8(v, mask); \ + v0 = _mm_shufflelo_epi16(v0, 0x1b); \ + v1 = _mm_shufflelo_epi16(v1, 0x1b); \ + v0 = _mm_shufflehi_epi16(v0, 0x1b); \ + v1 = _mm_shufflehi_epi16(v1, 0x1b); \ + *(__m128i *)(void *)(items + (i) * 4) = _mm_packus_epi16(v0, v1); } + +static +#ifdef SWAP_ATTRIB_SSE2 +SWAP_ATTRIB_SSE2 +#endif +void +Z7_FASTCALL +SwapBytes4_128_pack(CSwapUInt32 *items, const CSwapUInt32 *lim) +{ + const __m128i mask = _mm_setzero_si128(); + // const __m128i mask = _mm_set_epi16(0, 0, 0, 0, 0, 0, 0, 0); + Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE + do + { + SWAP4_128_pack(0); items += 1 * 4; + // SWAP4_128_pack(0); SWAP4_128_pack(1); items += 2 * 4; + } + while (items != lim); +} + +// sse2 +#define SWAP4_128_shift(i) { \ + __m128i v = *(const __m128i *)(const void *)(items + (i) * 4); \ + __m128i v2; \ + v2 = _mm_or_si128( \ + _mm_slli_si128(_mm_and_si128(v, mask), 1), \ + _mm_and_si128(_mm_srli_si128(v, 1), mask)); \ + v = _mm_or_si128( \ + _mm_slli_epi32(v, 24), \ + _mm_srli_epi32(v, 24)); \ + *(__m128i *)(void *)(items + (i) * 4) = _mm_or_si128(v2, v); } + +static +#ifdef SWAP_ATTRIB_SSE2 +SWAP_ATTRIB_SSE2 +#endif +void +Z7_FASTCALL +SwapBytes4_128_shift(CSwapUInt32 *items, const CSwapUInt32 *lim) +{ + #define M1 0xff00 + const __m128i mask = _mm_set_epi32(M1, M1, M1, M1); + Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE + do + { + // SWAP4_128_shift(0) SWAP4_128_shift(1) items += 2 * 4; + // SWAP4_128_shift(0) SWAP4_128_shift(1) items += 2 * 4; + SWAP4_128_shift(0); items += 1 * 4; + } + while (items != lim); +} +*/ + + +#if defined(USE_SWAP_SSSE3) || defined(USE_SWAP_AVX2) + +#define SWAP_SHUF_REV_SEQ_2_VALS(v) (v)+1, (v) +#define SWAP_SHUF_REV_SEQ_4_VALS(v) (v)+3, (v)+2, (v)+1, (v) + +#define SWAP2_SHUF_MASK_16_BYTES \ + SWAP_SHUF_REV_SEQ_2_VALS (0 * 2), \ + SWAP_SHUF_REV_SEQ_2_VALS (1 * 2), \ + SWAP_SHUF_REV_SEQ_2_VALS (2 * 2), \ + SWAP_SHUF_REV_SEQ_2_VALS (3 * 2), \ + SWAP_SHUF_REV_SEQ_2_VALS (4 * 2), \ + SWAP_SHUF_REV_SEQ_2_VALS (5 * 2), \ + SWAP_SHUF_REV_SEQ_2_VALS (6 * 2), \ + SWAP_SHUF_REV_SEQ_2_VALS (7 * 2) + +#define SWAP4_SHUF_MASK_16_BYTES \ + SWAP_SHUF_REV_SEQ_4_VALS (0 * 4), \ + SWAP_SHUF_REV_SEQ_4_VALS (1 * 4), \ + SWAP_SHUF_REV_SEQ_4_VALS (2 * 4), \ + SWAP_SHUF_REV_SEQ_4_VALS (3 * 4) + +#if defined(USE_SWAP_AVX2) +/* if we use 256_BIT_INIT_MASK, each static array mask will be larger for 16 bytes */ +// #define SWAP_USE_256_BIT_INIT_MASK +#endif + +#if defined(SWAP_USE_256_BIT_INIT_MASK) && defined(USE_SWAP_AVX2) +#define SWAP_MASK_INIT_SIZE 32 +#else +#define SWAP_MASK_INIT_SIZE 16 +#endif + +MY_ALIGN(SWAP_MASK_INIT_SIZE) +static const Byte k_ShufMask_Swap2[] = +{ + SWAP2_SHUF_MASK_16_BYTES + #if SWAP_MASK_INIT_SIZE > 16 + , SWAP2_SHUF_MASK_16_BYTES + #endif +}; + +MY_ALIGN(SWAP_MASK_INIT_SIZE) +static const Byte k_ShufMask_Swap4[] = +{ + SWAP4_SHUF_MASK_16_BYTES + #if SWAP_MASK_INIT_SIZE > 16 + , SWAP4_SHUF_MASK_16_BYTES + #endif +}; + + +#ifdef USE_SWAP_SSSE3 + +#include // ssse3 + +#define SHUF_128(i) *(items + (i)) = \ + _mm_shuffle_epi8(*(items + (i)), mask); // SSSE3 + +// Z7_NO_INLINE +static +#ifdef SWAP_ATTRIB_SSSE3 +SWAP_ATTRIB_SSSE3 +#endif +Z7_ATTRIB_NO_VECTORIZE +void +Z7_FASTCALL +ShufBytes_128(void *items8, const void *lim8, const void *mask128_ptr) +{ + __m128i *items = (__m128i *)items8; + const __m128i *lim = (const __m128i *)lim8; + // const __m128i mask = _mm_set_epi8(SHUF_SWAP2_MASK_16_VALS); + // const __m128i mask = _mm_set_epi8(SHUF_SWAP4_MASK_16_VALS); + // const __m128i mask = _mm_load_si128((const __m128i *)(const void *)&(k_ShufMask_Swap4[0])); + // const __m128i mask = _mm_load_si128((const __m128i *)(const void *)&(k_ShufMask_Swap4[0])); + // const __m128i mask = *(const __m128i *)(const void *)&(k_ShufMask_Swap4[0]); + const __m128i mask = *(const __m128i *)mask128_ptr; + Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE + do + { + SHUF_128(0) SHUF_128(1) items += 2; + SHUF_128(0) SHUF_128(1) items += 2; + } + while (items != lim); +} + +#endif // USE_SWAP_SSSE3 + + + +#ifdef USE_SWAP_AVX2 + +#include // avx, avx2 +#if defined(__clang__) +#include +#include +#endif + +#define SHUF_256(i) *(items + (i)) = \ + _mm256_shuffle_epi8(*(items + (i)), mask); // AVX2 + +// Z7_NO_INLINE +static +#ifdef SWAP_ATTRIB_AVX2 +SWAP_ATTRIB_AVX2 +#endif +Z7_ATTRIB_NO_VECTORIZE +void +Z7_FASTCALL +ShufBytes_256(void *items8, const void *lim8, const void *mask128_ptr) +{ + __m256i *items = (__m256i *)items8; + const __m256i *lim = (const __m256i *)lim8; + /* + UNUSED_VAR(mask128_ptr) + __m256i mask = + for Swap4: _mm256_setr_epi8(SWAP4_SHUF_MASK_16_BYTES, SWAP4_SHUF_MASK_16_BYTES); + for Swap2: _mm256_setr_epi8(SWAP2_SHUF_MASK_16_BYTES, SWAP2_SHUF_MASK_16_BYTES); + */ + const __m256i mask = + #if SWAP_MASK_INIT_SIZE > 16 + *(const __m256i *)(const void *)mask128_ptr; + #else + /* msvc: broadcastsi128() version reserves the stack for no reason + msvc 19.29-: _mm256_insertf128_si256() / _mm256_set_m128i)) versions use non-avx movdqu xmm0,XMMWORD PTR [r8] + msvc 19.30+ (VS2022): replaces _mm256_set_m128i(m,m) to vbroadcastf128(m) as we want + */ + // _mm256_broadcastsi128_si256(*mask128_ptr); + /* + #define MY_mm256_set_m128i(hi, lo) _mm256_insertf128_si256(_mm256_castsi128_si256(lo), (hi), 1) + MY_mm256_set_m128i + */ + _mm256_set_m128i( + *(const __m128i *)mask128_ptr, + *(const __m128i *)mask128_ptr); + #endif + + Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE + do + { + SHUF_256(0) SHUF_256(1) items += 2; + SHUF_256(0) SHUF_256(1) items += 2; + } + while (items != lim); +} + +#endif // USE_SWAP_AVX2 +#endif // USE_SWAP_SSSE3 || USE_SWAP_AVX2 +#endif // USE_SWAP_128 + + + +// compile message "NEON intrinsics not available with the soft-float ABI" +#elif defined(MY_CPU_ARM_OR_ARM64) || \ + (defined(__ARM_ARCH) && (__ARM_ARCH >= 7)) +// #elif defined(MY_CPU_ARM64) + + #if defined(__clang__) && (__clang_major__ >= 8) \ + || defined(__GNUC__) && (__GNUC__ >= 8) + #if (defined(__ARM_ARCH) && (__ARM_ARCH >= 7)) \ + || defined(MY_CPU_ARM64) + #define USE_SWAP_128 + #endif + #ifdef MY_CPU_ARM64 + // #define SWAP_ATTRIB_NEON __attribute__((__target__(""))) + #else + // #define SWAP_ATTRIB_NEON __attribute__((__target__("fpu=crypto-neon-fp-armv8"))) + #endif + #elif defined(_MSC_VER) + #if (_MSC_VER >= 1910) + #define USE_SWAP_128 + #endif + #endif + + #if defined(_MSC_VER) && defined(MY_CPU_ARM64) + #include + #else + #include + #endif + +#ifndef USE_SWAP_128 + #define FORCE_SWAP_MODE +#else + +#ifdef MY_CPU_ARM64 + // for debug : comment it + #define FORCE_SWAP_MODE +#else + #define k_SwapBytes_Mode_NEON 1 +#endif +// typedef uint8x16_t v128; +#define SWAP2_128(i) *(uint8x16_t *) (void *)(items + (i) * 8) = \ + vrev16q_u8(*(const uint8x16_t *)(const void *)(items + (i) * 8)); +#define SWAP4_128(i) *(uint8x16_t *) (void *)(items + (i) * 4) = \ + vrev32q_u8(*(const uint8x16_t *)(const void *)(items + (i) * 4)); + +// Z7_NO_INLINE +static +#ifdef SWAP_ATTRIB_NEON +SWAP_ATTRIB_NEON +#endif +Z7_ATTRIB_NO_VECTORIZE +void +Z7_FASTCALL +SwapBytes2_128(CSwapUInt16 *items, const CSwapUInt16 *lim) +{ + Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE + do + { + SWAP2_128(0) SWAP2_128(1) items += 2 * 8; + SWAP2_128(0) SWAP2_128(1) items += 2 * 8; + } + while (items != lim); +} + +// Z7_NO_INLINE +static +#ifdef SWAP_ATTRIB_NEON +SWAP_ATTRIB_NEON +#endif +Z7_ATTRIB_NO_VECTORIZE +void +Z7_FASTCALL +SwapBytes4_128(CSwapUInt32 *items, const CSwapUInt32 *lim) +{ + Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE + do + { + SWAP4_128(0) SWAP4_128(1) items += 2 * 4; + SWAP4_128(0) SWAP4_128(1) items += 2 * 4; + } + while (items != lim); +} + +#endif // USE_SWAP_128 + +#else // MY_CPU_ARM_OR_ARM64 +#define FORCE_SWAP_MODE +#endif // MY_CPU_ARM_OR_ARM64 + + + + + + +#if defined(Z7_MSC_VER_ORIGINAL) && defined(MY_CPU_X86) + /* _byteswap_ushort() in MSVC x86 32-bit works via slow { mov dh, al; mov dl, ah } + So we use own versions of byteswap function */ + #if (_MSC_VER < 1400 ) // old MSVC-X86 without _rotr16() support + #define SWAP2_16(i) { UInt32 v = items[i]; v += (v << 16); v >>= 8; items[i] = (CSwapUInt16)v; } + #else // is new MSVC-X86 with fast _rotr16() + #include + #define SWAP2_16(i) { items[i] = _rotr16(items[i], 8); } + #endif +#else // is not MSVC-X86 + #define SWAP2_16(i) { CSwapUInt16 v = items[i]; items[i] = Z7_BSWAP16(v); } +#endif // MSVC-X86 + +#if defined(Z7_CPU_FAST_BSWAP_SUPPORTED) + #define SWAP4_32(i) { CSwapUInt32 v = items[i]; items[i] = Z7_BSWAP32(v); } +#else + #define SWAP4_32(i) \ + { UInt32 v = items[i]; \ + v = ((v & 0xff00ff) << 8) + ((v >> 8) & 0xff00ff); \ + v = rotlFixed(v, 16); \ + items[i] = v; } +#endif + + + + +#if defined(FORCE_SWAP_MODE) && defined(USE_SWAP_128) + #define DEFAULT_Swap2 SwapBytes2_128 + #if !defined(MY_CPU_X86_OR_AMD64) + #define DEFAULT_Swap4 SwapBytes4_128 + #endif +#endif + +#if !defined(DEFAULT_Swap2) || !defined(DEFAULT_Swap4) + +#define SWAP_BASE_FUNCS_PREFIXES \ +Z7_FORCE_INLINE \ +static \ +Z7_ATTRIB_NO_VECTOR \ +void Z7_FASTCALL + + +#ifdef MY_CPU_64BIT + +#if defined(MY_CPU_ARM64) \ + && defined(__ARM_ARCH) && (__ARM_ARCH >= 8) \ + && ( (defined(__GNUC__) && (__GNUC__ >= 4)) \ + || (defined(__clang__) && (__clang_major__ >= 4))) + + #define SWAP2_64_VAR(v) asm ("rev16 %x0,%x0" : "+r" (v)); + #define SWAP4_64_VAR(v) asm ("rev32 %x0,%x0" : "+r" (v)); + +#else // is not ARM64-GNU + +#if !defined(MY_CPU_X86_OR_AMD64) || (k_SwapBytes_Mode_MIN == 0) || !defined(USE_SWAP_128) + #define SWAP2_64_VAR(v) \ + v = ( 0x00ff00ff00ff00ff & (v >> 8)) \ + + ((0x00ff00ff00ff00ff & v) << 8); + /* plus gives faster code in MSVC */ +#endif + +#ifdef Z7_CPU_FAST_BSWAP_SUPPORTED + #define SWAP4_64_VAR(v) \ + v = Z7_BSWAP64(v); \ + v = Z7_ROTL64(v, 32); +#else + #define SWAP4_64_VAR(v) \ + v = ( 0x000000ff000000ff & (v >> 24)) \ + + ((0x000000ff000000ff & v) << 24 ) \ + + ( 0x0000ff000000ff00 & (v >> 8)) \ + + ((0x0000ff000000ff00 & v) << 8 ) \ + ; +#endif + +#endif // ARM64-GNU + + +#ifdef SWAP2_64_VAR + +#define SWAP2_64(i) { \ + UInt64 v = *(const UInt64 *)(const void *)(items + (i) * 4); \ + SWAP2_64_VAR(v) \ + *(UInt64 *)(void *)(items + (i) * 4) = v; } + +SWAP_BASE_FUNCS_PREFIXES +SwapBytes2_64(CSwapUInt16 *items, const CSwapUInt16 *lim) +{ + Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE + do + { + SWAP2_64(0) SWAP2_64(1) items += 2 * 4; + SWAP2_64(0) SWAP2_64(1) items += 2 * 4; + } + while (items != lim); +} + + #define DEFAULT_Swap2 SwapBytes2_64 + #if !defined(FORCE_SWAP_MODE) + #define SWAP2_DEFAULT_MODE 0 + #endif +#else // !defined(SWAP2_64_VAR) + #define DEFAULT_Swap2 SwapBytes2_128 + #if !defined(FORCE_SWAP_MODE) + #define SWAP2_DEFAULT_MODE 1 + #endif +#endif // SWAP2_64_VAR + + +#define SWAP4_64(i) { \ + UInt64 v = *(const UInt64 *)(const void *)(items + (i) * 2); \ + SWAP4_64_VAR(v) \ + *(UInt64 *)(void *)(items + (i) * 2) = v; } + +SWAP_BASE_FUNCS_PREFIXES +SwapBytes4_64(CSwapUInt32 *items, const CSwapUInt32 *lim) +{ + Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE + do + { + SWAP4_64(0) SWAP4_64(1) items += 2 * 2; + SWAP4_64(0) SWAP4_64(1) items += 2 * 2; + } + while (items != lim); +} + +#define DEFAULT_Swap4 SwapBytes4_64 + +#else // is not 64BIT + + +#if defined(MY_CPU_ARM_OR_ARM64) \ + && defined(__ARM_ARCH) && (__ARM_ARCH >= 6) \ + && ( (defined(__GNUC__) && (__GNUC__ >= 4)) \ + || (defined(__clang__) && (__clang_major__ >= 4))) + +#ifdef MY_CPU_64BIT + #define SWAP2_32_VAR(v) asm ("rev16 %w0,%w0" : "+r" (v)); +#else + #define SWAP2_32_VAR(v) asm ("rev16 %0,%0" : "+r" (v)); // for clang/gcc + // asm ("rev16 %r0,%r0" : "+r" (a)); // for gcc +#endif + +#elif defined(_MSC_VER) && (_MSC_VER < 1300) && defined(MY_CPU_X86) \ + || !defined(Z7_CPU_FAST_BSWAP_SUPPORTED) \ + || !defined(Z7_CPU_FAST_ROTATE_SUPPORTED) + // old msvc doesn't support _byteswap_ulong() + #define SWAP2_32_VAR(v) \ + v = ((v & 0xff00ff) << 8) + ((v >> 8) & 0xff00ff); + +#else // is not ARM and is not old-MSVC-X86 and fast BSWAP/ROTATE are supported + #define SWAP2_32_VAR(v) \ + v = Z7_BSWAP32(v); \ + v = rotlFixed(v, 16); + +#endif // GNU-ARM* + +#define SWAP2_32(i) { \ + UInt32 v = *(const UInt32 *)(const void *)(items + (i) * 2); \ + SWAP2_32_VAR(v); \ + *(UInt32 *)(void *)(items + (i) * 2) = v; } + + +SWAP_BASE_FUNCS_PREFIXES +SwapBytes2_32(CSwapUInt16 *items, const CSwapUInt16 *lim) +{ + Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE + do + { + SWAP2_32(0) SWAP2_32(1) items += 2 * 2; + SWAP2_32(0) SWAP2_32(1) items += 2 * 2; + } + while (items != lim); +} + + +SWAP_BASE_FUNCS_PREFIXES +SwapBytes4_32(CSwapUInt32 *items, const CSwapUInt32 *lim) +{ + Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE + do + { + SWAP4_32(0) SWAP4_32(1) items += 2; + SWAP4_32(0) SWAP4_32(1) items += 2; + } + while (items != lim); +} + +#define DEFAULT_Swap2 SwapBytes2_32 +#define DEFAULT_Swap4 SwapBytes4_32 +#if !defined(FORCE_SWAP_MODE) + #define SWAP2_DEFAULT_MODE 0 +#endif + +#endif // MY_CPU_64BIT +#endif // if !defined(DEFAULT_Swap2) || !defined(DEFAULT_Swap4) + + + +#if !defined(FORCE_SWAP_MODE) +static unsigned g_SwapBytes_Mode; +#endif + +/* size of largest unrolled loop iteration: 128 bytes = 4 * 32 bytes (AVX). */ +#define SWAP_ITERATION_BLOCK_SIZE_MAX (1 << 7) + +// 32 bytes for (AVX) or 2 * 16-bytes for NEON. +#define SWAP_VECTOR_ALIGN_SIZE (1 << 5) + +Z7_NO_INLINE +void z7_SwapBytes2(CSwapUInt16 *items, size_t numItems) +{ + Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE + for (; numItems != 0 && ((unsigned)(ptrdiff_t)items & (SWAP_VECTOR_ALIGN_SIZE - 1)) != 0; numItems--) + { + SWAP2_16(0) + items++; + } + { + const size_t k_Align_Mask = SWAP_ITERATION_BLOCK_SIZE_MAX / sizeof(CSwapUInt16) - 1; + size_t numItems2 = numItems; + CSwapUInt16 *lim; + numItems &= k_Align_Mask; + numItems2 &= ~(size_t)k_Align_Mask; + lim = items + numItems2; + if (numItems2 != 0) + { + #if !defined(FORCE_SWAP_MODE) + #ifdef MY_CPU_X86_OR_AMD64 + #ifdef USE_SWAP_AVX2 + if (g_SwapBytes_Mode > k_SwapBytes_Mode_SSSE3) + ShufBytes_256((__m256i *)(void *)items, + (const __m256i *)(const void *)lim, + (const __m128i *)(const void *)&(k_ShufMask_Swap2[0])); + else + #endif + #ifdef USE_SWAP_SSSE3 + if (g_SwapBytes_Mode >= k_SwapBytes_Mode_SSSE3) + ShufBytes_128((__m128i *)(void *)items, + (const __m128i *)(const void *)lim, + (const __m128i *)(const void *)&(k_ShufMask_Swap2[0])); + else + #endif + #endif // MY_CPU_X86_OR_AMD64 + #if SWAP2_DEFAULT_MODE == 0 + if (g_SwapBytes_Mode != 0) + SwapBytes2_128(items, lim); + else + #endif + #endif // FORCE_SWAP_MODE + DEFAULT_Swap2(items, lim); + } + items = lim; + } + Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE + for (; numItems != 0; numItems--) + { + SWAP2_16(0) + items++; + } +} + + +Z7_NO_INLINE +void z7_SwapBytes4(CSwapUInt32 *items, size_t numItems) +{ + Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE + for (; numItems != 0 && ((unsigned)(ptrdiff_t)items & (SWAP_VECTOR_ALIGN_SIZE - 1)) != 0; numItems--) + { + SWAP4_32(0) + items++; + } + { + const size_t k_Align_Mask = SWAP_ITERATION_BLOCK_SIZE_MAX / sizeof(CSwapUInt32) - 1; + size_t numItems2 = numItems; + CSwapUInt32 *lim; + numItems &= k_Align_Mask; + numItems2 &= ~(size_t)k_Align_Mask; + lim = items + numItems2; + if (numItems2 != 0) + { + #if !defined(FORCE_SWAP_MODE) + #ifdef MY_CPU_X86_OR_AMD64 + #ifdef USE_SWAP_AVX2 + if (g_SwapBytes_Mode > k_SwapBytes_Mode_SSSE3) + ShufBytes_256((__m256i *)(void *)items, + (const __m256i *)(const void *)lim, + (const __m128i *)(const void *)&(k_ShufMask_Swap4[0])); + else + #endif + #ifdef USE_SWAP_SSSE3 + if (g_SwapBytes_Mode >= k_SwapBytes_Mode_SSSE3) + ShufBytes_128((__m128i *)(void *)items, + (const __m128i *)(const void *)lim, + (const __m128i *)(const void *)&(k_ShufMask_Swap4[0])); + else + #endif + #else // MY_CPU_X86_OR_AMD64 + + if (g_SwapBytes_Mode != 0) + SwapBytes4_128(items, lim); + else + #endif // MY_CPU_X86_OR_AMD64 + #endif // FORCE_SWAP_MODE + DEFAULT_Swap4(items, lim); + } + items = lim; + } + Z7_PRAGMA_OPT_DISABLE_LOOP_UNROLL_VECTORIZE + for (; numItems != 0; numItems--) + { + SWAP4_32(0) + items++; + } +} + + +// #define SHOW_HW_STATUS + +#ifdef SHOW_HW_STATUS +#include +#define PRF(x) x +#else +#define PRF(x) +#endif + +void z7_SwapBytesPrepare(void) +{ +#ifndef FORCE_SWAP_MODE + unsigned mode = 0; // k_SwapBytes_Mode_BASE; + +#ifdef MY_CPU_ARM_OR_ARM64 + { + if (CPU_IsSupported_NEON()) + { + // #pragma message ("=== SwapBytes NEON") + PRF(printf("\n=== SwapBytes NEON\n");) + mode = k_SwapBytes_Mode_NEON; + } + } +#else // MY_CPU_ARM_OR_ARM64 + { + #ifdef USE_SWAP_AVX2 + if (CPU_IsSupported_AVX2()) + { + // #pragma message ("=== SwapBytes AVX2") + PRF(printf("\n=== SwapBytes AVX2\n");) + mode = k_SwapBytes_Mode_AVX2; + } + else + #endif + #ifdef USE_SWAP_SSSE3 + if (CPU_IsSupported_SSSE3()) + { + // #pragma message ("=== SwapBytes SSSE3") + PRF(printf("\n=== SwapBytes SSSE3\n");) + mode = k_SwapBytes_Mode_SSSE3; + } + else + #endif + #if !defined(MY_CPU_AMD64) + if (CPU_IsSupported_SSE2()) + #endif + { + // #pragma message ("=== SwapBytes SSE2") + PRF(printf("\n=== SwapBytes SSE2\n");) + mode = k_SwapBytes_Mode_SSE2; + } + } +#endif // MY_CPU_ARM_OR_ARM64 + g_SwapBytes_Mode = mode; + // g_SwapBytes_Mode = 0; // for debug +#endif // FORCE_SWAP_MODE + PRF(printf("\n=== SwapBytesPrepare\n");) +} + +#undef PRF diff --git a/libraries/lzma/C/SwapBytes.h b/libraries/lzma/C/SwapBytes.h new file mode 100644 index 000000000..d44246738 --- /dev/null +++ b/libraries/lzma/C/SwapBytes.h @@ -0,0 +1,17 @@ +/* SwapBytes.h -- Byte Swap conversion filter +2023-04-02 : Igor Pavlov : Public domain */ + +#ifndef ZIP7_INC_SWAP_BYTES_H +#define ZIP7_INC_SWAP_BYTES_H + +#include "7zTypes.h" + +EXTERN_C_BEGIN + +void z7_SwapBytes2(UInt16 *data, size_t numItems); +void z7_SwapBytes4(UInt32 *data, size_t numItems); +void z7_SwapBytesPrepare(void); + +EXTERN_C_END + +#endif diff --git a/libraries/lzma/C/Threads.c b/libraries/lzma/C/Threads.c index 58eb90ffa..cf52bd303 100644 --- a/libraries/lzma/C/Threads.c +++ b/libraries/lzma/C/Threads.c @@ -1,5 +1,5 @@ /* Threads.c -- multithreading library -2021-12-21 : Igor Pavlov : Public domain */ +2023-03-04 : Igor Pavlov : Public domain */ #include "Precomp.h" @@ -11,9 +11,9 @@ #include "Threads.h" -static WRes GetError() +static WRes GetError(void) { - DWORD res = GetLastError(); + const DWORD res = GetLastError(); return res ? (WRes)res : 1; } @@ -173,6 +173,9 @@ WRes CriticalSection_Init(CCriticalSection *p) Windows XP, 2003 : can raise a STATUS_NO_MEMORY exception Windows Vista+ : no exceptions */ #ifdef _MSC_VER + #ifdef __clang__ + #pragma GCC diagnostic ignored "-Wlanguage-extension-token" + #endif __try #endif { @@ -193,18 +196,26 @@ WRes CriticalSection_Init(CCriticalSection *p) // ---------- POSIX ---------- #ifndef __APPLE__ -#ifndef _7ZIP_AFFINITY_DISABLE +#ifndef Z7_AFFINITY_DISABLE // _GNU_SOURCE can be required for pthread_setaffinity_np() / CPU_ZERO / CPU_SET +// clang < 3.6 : unknown warning group '-Wreserved-id-macro' +// clang 3.6 - 12.01 : gives warning "macro name is a reserved identifier" +// clang >= 13 : do not give warning +#if !defined(_GNU_SOURCE) + #if defined(__clang__) && (__clang_major__ >= 4) && (__clang_major__ <= 12) + #pragma GCC diagnostic ignored "-Wreserved-id-macro" + #endif #define _GNU_SOURCE -#endif -#endif +#endif // !defined(_GNU_SOURCE) +#endif // Z7_AFFINITY_DISABLE +#endif // __APPLE__ #include "Threads.h" #include #include #include -#ifdef _7ZIP_AFFINITY_SUPPORTED +#ifdef Z7_AFFINITY_SUPPORTED // #include #endif @@ -212,15 +223,12 @@ WRes CriticalSection_Init(CCriticalSection *p) // #include // #define PRF(p) p #define PRF(p) - -#define Print(s) PRF(printf("\n%s\n", s)) - -// #include +#define Print(s) PRF(printf("\n%s\n", s);) WRes Thread_Create_With_CpuSet(CThread *p, THREAD_FUNC_TYPE func, LPVOID param, const CCpuSet *cpuSet) { // new thread in Posix probably inherits affinity from parrent thread - Print("Thread_Create_With_CpuSet"); + Print("Thread_Create_With_CpuSet") pthread_attr_t attr; int ret; @@ -228,7 +236,7 @@ WRes Thread_Create_With_CpuSet(CThread *p, THREAD_FUNC_TYPE func, LPVOID param, p->_created = 0; - RINOK(pthread_attr_init(&attr)); + RINOK(pthread_attr_init(&attr)) ret = pthread_attr_setdetachstate(&attr, PTHREAD_CREATE_JOINABLE); @@ -236,7 +244,7 @@ WRes Thread_Create_With_CpuSet(CThread *p, THREAD_FUNC_TYPE func, LPVOID param, { if (cpuSet) { - #ifdef _7ZIP_AFFINITY_SUPPORTED + #ifdef Z7_AFFINITY_SUPPORTED /* printf("\n affinity :"); @@ -292,7 +300,7 @@ WRes Thread_Create(CThread *p, THREAD_FUNC_TYPE func, LPVOID param) WRes Thread_Create_With_Affinity(CThread *p, THREAD_FUNC_TYPE func, LPVOID param, CAffinityMask affinity) { - Print("Thread_Create_WithAffinity"); + Print("Thread_Create_WithAffinity") CCpuSet cs; unsigned i; CpuSet_Zero(&cs); @@ -312,7 +320,7 @@ WRes Thread_Create_With_Affinity(CThread *p, THREAD_FUNC_TYPE func, LPVOID param WRes Thread_Close(CThread *p) { - // Print("Thread_Close"); + // Print("Thread_Close") int ret; if (!p->_created) return 0; @@ -326,7 +334,7 @@ WRes Thread_Close(CThread *p) WRes Thread_Wait_Close(CThread *p) { - // Print("Thread_Wait_Close"); + // Print("Thread_Wait_Close") void *thread_return; int ret; if (!p->_created) @@ -343,8 +351,8 @@ WRes Thread_Wait_Close(CThread *p) static WRes Event_Create(CEvent *p, int manualReset, int signaled) { - RINOK(pthread_mutex_init(&p->_mutex, NULL)); - RINOK(pthread_cond_init(&p->_cond, NULL)); + RINOK(pthread_mutex_init(&p->_mutex, NULL)) + RINOK(pthread_cond_init(&p->_cond, NULL)) p->_manual_reset = manualReset; p->_state = (signaled ? True : False); p->_created = 1; @@ -363,7 +371,7 @@ WRes AutoResetEvent_CreateNotSignaled(CAutoResetEvent *p) WRes Event_Set(CEvent *p) { - RINOK(pthread_mutex_lock(&p->_mutex)); + RINOK(pthread_mutex_lock(&p->_mutex)) p->_state = True; int res1 = pthread_cond_broadcast(&p->_cond); int res2 = pthread_mutex_unlock(&p->_mutex); @@ -372,14 +380,14 @@ WRes Event_Set(CEvent *p) WRes Event_Reset(CEvent *p) { - RINOK(pthread_mutex_lock(&p->_mutex)); + RINOK(pthread_mutex_lock(&p->_mutex)) p->_state = False; return pthread_mutex_unlock(&p->_mutex); } WRes Event_Wait(CEvent *p) { - RINOK(pthread_mutex_lock(&p->_mutex)); + RINOK(pthread_mutex_lock(&p->_mutex)) while (p->_state == False) { // ETIMEDOUT @@ -411,8 +419,8 @@ WRes Semaphore_Create(CSemaphore *p, UInt32 initCount, UInt32 maxCount) { if (initCount > maxCount || maxCount < 1) return EINVAL; - RINOK(pthread_mutex_init(&p->_mutex, NULL)); - RINOK(pthread_cond_init(&p->_cond, NULL)); + RINOK(pthread_mutex_init(&p->_mutex, NULL)) + RINOK(pthread_cond_init(&p->_cond, NULL)) p->_count = initCount; p->_maxCount = maxCount; p->_created = 1; @@ -448,7 +456,7 @@ WRes Semaphore_ReleaseN(CSemaphore *p, UInt32 releaseCount) if (releaseCount < 1) return EINVAL; - RINOK(pthread_mutex_lock(&p->_mutex)); + RINOK(pthread_mutex_lock(&p->_mutex)) newCount = p->_count + releaseCount; if (newCount > p->_maxCount) @@ -458,13 +466,13 @@ WRes Semaphore_ReleaseN(CSemaphore *p, UInt32 releaseCount) p->_count = newCount; ret = pthread_cond_broadcast(&p->_cond); } - RINOK(pthread_mutex_unlock(&p->_mutex)); + RINOK(pthread_mutex_unlock(&p->_mutex)) return ret; } WRes Semaphore_Wait(CSemaphore *p) { - RINOK(pthread_mutex_lock(&p->_mutex)); + RINOK(pthread_mutex_lock(&p->_mutex)) while (p->_count < 1) { pthread_cond_wait(&p->_cond, &p->_mutex); @@ -489,7 +497,7 @@ WRes Semaphore_Close(CSemaphore *p) WRes CriticalSection_Init(CCriticalSection *p) { - // Print("CriticalSection_Init"); + // Print("CriticalSection_Init") if (!p) return EINTR; return pthread_mutex_init(&p->_mutex, NULL); @@ -497,7 +505,7 @@ WRes CriticalSection_Init(CCriticalSection *p) void CriticalSection_Enter(CCriticalSection *p) { - // Print("CriticalSection_Enter"); + // Print("CriticalSection_Enter") if (p) { // int ret = @@ -507,7 +515,7 @@ void CriticalSection_Enter(CCriticalSection *p) void CriticalSection_Leave(CCriticalSection *p) { - // Print("CriticalSection_Leave"); + // Print("CriticalSection_Leave") if (p) { // int ret = @@ -517,7 +525,7 @@ void CriticalSection_Leave(CCriticalSection *p) void CriticalSection_Delete(CCriticalSection *p) { - // Print("CriticalSection_Delete"); + // Print("CriticalSection_Delete") if (p) { // int ret = @@ -527,14 +535,28 @@ void CriticalSection_Delete(CCriticalSection *p) LONG InterlockedIncrement(LONG volatile *addend) { - // Print("InterlockedIncrement"); + // Print("InterlockedIncrement") #ifdef USE_HACK_UNSAFE_ATOMIC LONG val = *addend + 1; *addend = val; return val; #else + + #if defined(__clang__) && (__clang_major__ >= 8) + #pragma GCC diagnostic ignored "-Watomic-implicit-seq-cst" + #endif return __sync_add_and_fetch(addend, 1); #endif } #endif // _WIN32 + +WRes AutoResetEvent_OptCreate_And_Reset(CAutoResetEvent *p) +{ + if (Event_IsCreated(p)) + return Event_Reset(p); + return AutoResetEvent_CreateNotSignaled(p); +} + +#undef PRF +#undef Print diff --git a/libraries/lzma/C/Threads.h b/libraries/lzma/C/Threads.h index 89ecb92be..4028464a3 100644 --- a/libraries/lzma/C/Threads.h +++ b/libraries/lzma/C/Threads.h @@ -1,18 +1,19 @@ /* Threads.h -- multithreading library -2021-12-21 : Igor Pavlov : Public domain */ +2023-04-02 : Igor Pavlov : Public domain */ -#ifndef __7Z_THREADS_H -#define __7Z_THREADS_H +#ifndef ZIP7_INC_THREADS_H +#define ZIP7_INC_THREADS_H #ifdef _WIN32 -#include +#include "7zWindows.h" + #else #if defined(__linux__) #if !defined(__APPLE__) && !defined(_AIX) && !defined(__ANDROID__) -#ifndef _7ZIP_AFFINITY_DISABLE -#define _7ZIP_AFFINITY_SUPPORTED -// #pragma message(" ==== _7ZIP_AFFINITY_SUPPORTED") +#ifndef Z7_AFFINITY_DISABLE +#define Z7_AFFINITY_SUPPORTED +// #pragma message(" ==== Z7_AFFINITY_SUPPORTED") // #define _GNU_SOURCE #endif #endif @@ -33,7 +34,7 @@ WRes Handle_WaitObject(HANDLE h); typedef HANDLE CThread; -#define Thread_Construct(p) { *(p) = NULL; } +#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)) @@ -52,42 +53,46 @@ typedef #endif THREAD_FUNC_RET_TYPE; +#define THREAD_FUNC_RET_ZERO 0 + typedef DWORD_PTR CAffinityMask; typedef DWORD_PTR CCpuSet; -#define CpuSet_Zero(p) { *(p) = 0; } -#define CpuSet_Set(p, cpu) { *(p) |= ((DWORD_PTR)1 << (cpu)); } +#define CpuSet_Zero(p) *(p) = (0) +#define CpuSet_Set(p, cpu) *(p) |= ((DWORD_PTR)1 << (cpu)) #else // _WIN32 -typedef struct _CThread +typedef struct { pthread_t _tid; int _created; } CThread; -#define Thread_Construct(p) { (p)->_tid = 0; (p)->_created = 0; } -#define Thread_WasCreated(p) ((p)->_created != 0) +#define Thread_CONSTRUCT(p) { (p)->_tid = 0; (p)->_created = 0; } +#define Thread_WasCreated(p) ((p)->_created != 0) WRes Thread_Close(CThread *p); // #define Thread_Wait Thread_Wait_Close typedef void * THREAD_FUNC_RET_TYPE; +#define THREAD_FUNC_RET_ZERO NULL + typedef UInt64 CAffinityMask; -#ifdef _7ZIP_AFFINITY_SUPPORTED +#ifdef Z7_AFFINITY_SUPPORTED typedef cpu_set_t CCpuSet; -#define CpuSet_Zero(p) CPU_ZERO(p) -#define CpuSet_Set(p, cpu) CPU_SET(cpu, p) -#define CpuSet_IsSet(p, cpu) CPU_ISSET(cpu, p) +#define CpuSet_Zero(p) CPU_ZERO(p) +#define CpuSet_Set(p, cpu) CPU_SET(cpu, p) +#define CpuSet_IsSet(p, cpu) CPU_ISSET(cpu, p) #else typedef UInt64 CCpuSet; -#define CpuSet_Zero(p) { *(p) = 0; } -#define CpuSet_Set(p, cpu) { *(p) |= ((UInt64)1 << (cpu)); } -#define CpuSet_IsSet(p, cpu) ((*(p) & ((UInt64)1 << (cpu))) != 0) +#define CpuSet_Zero(p) *(p) = (0) +#define CpuSet_Set(p, cpu) *(p) |= ((UInt64)1 << (cpu)) +#define CpuSet_IsSet(p, cpu) ((*(p) & ((UInt64)1 << (cpu))) != 0) #endif @@ -95,7 +100,7 @@ typedef UInt64 CCpuSet; #endif // _WIN32 -#define THREAD_FUNC_CALL_TYPE MY_STD_CALL +#define THREAD_FUNC_CALL_TYPE Z7_STDCALL #if defined(_WIN32) && defined(__GNUC__) /* GCC compiler for x86 32-bit uses the rule: @@ -187,6 +192,7 @@ WRes ManualResetEvent_Create(CManualResetEvent *p, int signaled); WRes ManualResetEvent_CreateNotSignaled(CManualResetEvent *p); WRes AutoResetEvent_Create(CAutoResetEvent *p, int signaled); WRes AutoResetEvent_CreateNotSignaled(CAutoResetEvent *p); + WRes Event_Set(CEvent *p); WRes Event_Reset(CEvent *p); WRes Event_Wait(CEvent *p); @@ -227,6 +233,8 @@ LONG InterlockedIncrement(LONG volatile *addend); #endif // _WIN32 +WRes AutoResetEvent_OptCreate_And_Reset(CAutoResetEvent *p); + EXTERN_C_END #endif diff --git a/libraries/lzma/C/Xz.c b/libraries/lzma/C/Xz.c new file mode 100644 index 000000000..4ad071060 --- /dev/null +++ b/libraries/lzma/C/Xz.c @@ -0,0 +1,90 @@ +/* Xz.c - Xz +2023-04-02 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include "7zCrc.h" +#include "CpuArch.h" +#include "Xz.h" +#include "XzCrc64.h" + +const Byte XZ_SIG[XZ_SIG_SIZE] = { 0xFD, '7', 'z', 'X', 'Z', 0 }; +/* const Byte XZ_FOOTER_SIG[XZ_FOOTER_SIG_SIZE] = { 'Y', 'Z' }; */ + +unsigned Xz_WriteVarInt(Byte *buf, UInt64 v) +{ + unsigned i = 0; + do + { + buf[i++] = (Byte)((v & 0x7F) | 0x80); + v >>= 7; + } + while (v != 0); + buf[(size_t)i - 1] &= 0x7F; + return i; +} + +void Xz_Construct(CXzStream *p) +{ + p->numBlocks = 0; + p->blocks = NULL; + p->flags = 0; +} + +void Xz_Free(CXzStream *p, ISzAllocPtr alloc) +{ + ISzAlloc_Free(alloc, p->blocks); + p->numBlocks = 0; + p->blocks = NULL; +} + +unsigned XzFlags_GetCheckSize(CXzStreamFlags f) +{ + unsigned t = XzFlags_GetCheckType(f); + return (t == 0) ? 0 : ((unsigned)4 << ((t - 1) / 3)); +} + +void XzCheck_Init(CXzCheck *p, unsigned mode) +{ + p->mode = mode; + switch (mode) + { + case XZ_CHECK_CRC32: p->crc = CRC_INIT_VAL; break; + case XZ_CHECK_CRC64: p->crc64 = CRC64_INIT_VAL; break; + case XZ_CHECK_SHA256: Sha256_Init(&p->sha); break; + } +} + +void XzCheck_Update(CXzCheck *p, const void *data, size_t size) +{ + switch (p->mode) + { + case XZ_CHECK_CRC32: p->crc = CrcUpdate(p->crc, data, size); break; + case XZ_CHECK_CRC64: p->crc64 = Crc64Update(p->crc64, data, size); break; + case XZ_CHECK_SHA256: Sha256_Update(&p->sha, (const Byte *)data, size); break; + } +} + +int XzCheck_Final(CXzCheck *p, Byte *digest) +{ + switch (p->mode) + { + case XZ_CHECK_CRC32: + SetUi32(digest, CRC_GET_DIGEST(p->crc)) + break; + case XZ_CHECK_CRC64: + { + int i; + UInt64 v = CRC64_GET_DIGEST(p->crc64); + for (i = 0; i < 8; i++, v >>= 8) + digest[i] = (Byte)(v & 0xFF); + break; + } + case XZ_CHECK_SHA256: + Sha256_Final(&p->sha, digest); + break; + default: + return 0; + } + return 1; +} diff --git a/libraries/lzma/C/Xz.h b/libraries/lzma/C/Xz.h new file mode 100644 index 000000000..d5001f6ca --- /dev/null +++ b/libraries/lzma/C/Xz.h @@ -0,0 +1,535 @@ +/* Xz.h - Xz interface +2023-04-13 : Igor Pavlov : Public domain */ + +#ifndef ZIP7_INC_XZ_H +#define ZIP7_INC_XZ_H + +#include "Sha256.h" +#include "Delta.h" + +EXTERN_C_BEGIN + +#define XZ_ID_Subblock 1 +#define XZ_ID_Delta 3 +#define XZ_ID_X86 4 +#define XZ_ID_PPC 5 +#define XZ_ID_IA64 6 +#define XZ_ID_ARM 7 +#define XZ_ID_ARMT 8 +#define XZ_ID_SPARC 9 +#define XZ_ID_ARM64 0xa +#define XZ_ID_LZMA2 0x21 + +unsigned Xz_ReadVarInt(const Byte *p, size_t maxSize, UInt64 *value); +unsigned Xz_WriteVarInt(Byte *buf, UInt64 v); + +/* ---------- xz block ---------- */ + +#define XZ_BLOCK_HEADER_SIZE_MAX 1024 + +#define XZ_NUM_FILTERS_MAX 4 +#define XZ_BF_NUM_FILTERS_MASK 3 +#define XZ_BF_PACK_SIZE (1 << 6) +#define XZ_BF_UNPACK_SIZE (1 << 7) + +#define XZ_FILTER_PROPS_SIZE_MAX 20 + +typedef struct +{ + UInt64 id; + UInt32 propsSize; + Byte props[XZ_FILTER_PROPS_SIZE_MAX]; +} CXzFilter; + +typedef struct +{ + UInt64 packSize; + UInt64 unpackSize; + Byte flags; + CXzFilter filters[XZ_NUM_FILTERS_MAX]; +} CXzBlock; + +#define XzBlock_GetNumFilters(p) (((unsigned)(p)->flags & XZ_BF_NUM_FILTERS_MASK) + 1) +#define XzBlock_HasPackSize(p) (((p)->flags & XZ_BF_PACK_SIZE) != 0) +#define XzBlock_HasUnpackSize(p) (((p)->flags & XZ_BF_UNPACK_SIZE) != 0) +#define XzBlock_HasUnsupportedFlags(p) (((p)->flags & ~(XZ_BF_NUM_FILTERS_MASK | XZ_BF_PACK_SIZE | XZ_BF_UNPACK_SIZE)) != 0) + +SRes XzBlock_Parse(CXzBlock *p, const Byte *header); +SRes XzBlock_ReadHeader(CXzBlock *p, ISeqInStreamPtr inStream, BoolInt *isIndex, UInt32 *headerSizeRes); + +/* ---------- xz stream ---------- */ + +#define XZ_SIG_SIZE 6 +#define XZ_FOOTER_SIG_SIZE 2 + +extern const Byte XZ_SIG[XZ_SIG_SIZE]; + +/* +extern const Byte XZ_FOOTER_SIG[XZ_FOOTER_SIG_SIZE]; +*/ + +#define XZ_FOOTER_SIG_0 'Y' +#define XZ_FOOTER_SIG_1 'Z' + +#define XZ_STREAM_FLAGS_SIZE 2 +#define XZ_STREAM_CRC_SIZE 4 + +#define XZ_STREAM_HEADER_SIZE (XZ_SIG_SIZE + XZ_STREAM_FLAGS_SIZE + XZ_STREAM_CRC_SIZE) +#define XZ_STREAM_FOOTER_SIZE (XZ_FOOTER_SIG_SIZE + XZ_STREAM_FLAGS_SIZE + XZ_STREAM_CRC_SIZE + 4) + +#define XZ_CHECK_MASK 0xF +#define XZ_CHECK_NO 0 +#define XZ_CHECK_CRC32 1 +#define XZ_CHECK_CRC64 4 +#define XZ_CHECK_SHA256 10 + +typedef struct +{ + unsigned mode; + UInt32 crc; + UInt64 crc64; + CSha256 sha; +} CXzCheck; + +void XzCheck_Init(CXzCheck *p, unsigned mode); +void XzCheck_Update(CXzCheck *p, const void *data, size_t size); +int XzCheck_Final(CXzCheck *p, Byte *digest); + +typedef UInt16 CXzStreamFlags; + +#define XzFlags_IsSupported(f) ((f) <= XZ_CHECK_MASK) +#define XzFlags_GetCheckType(f) ((f) & XZ_CHECK_MASK) +#define XzFlags_HasDataCrc32(f) (Xz_GetCheckType(f) == XZ_CHECK_CRC32) +unsigned XzFlags_GetCheckSize(CXzStreamFlags f); + +SRes Xz_ParseHeader(CXzStreamFlags *p, const Byte *buf); +SRes Xz_ReadHeader(CXzStreamFlags *p, ISeqInStreamPtr inStream); + +typedef struct +{ + UInt64 unpackSize; + UInt64 totalSize; +} CXzBlockSizes; + +typedef struct +{ + CXzStreamFlags flags; + // Byte _pad[6]; + size_t numBlocks; + CXzBlockSizes *blocks; + UInt64 startOffset; +} CXzStream; + +void Xz_Construct(CXzStream *p); +void Xz_Free(CXzStream *p, ISzAllocPtr alloc); + +#define XZ_SIZE_OVERFLOW ((UInt64)(Int64)-1) + +UInt64 Xz_GetUnpackSize(const CXzStream *p); +UInt64 Xz_GetPackSize(const CXzStream *p); + +typedef struct +{ + size_t num; + size_t numAllocated; + CXzStream *streams; +} CXzs; + +void Xzs_Construct(CXzs *p); +void Xzs_Free(CXzs *p, ISzAllocPtr alloc); +SRes Xzs_ReadBackward(CXzs *p, ILookInStreamPtr inStream, Int64 *startOffset, ICompressProgressPtr progress, ISzAllocPtr alloc); + +UInt64 Xzs_GetNumBlocks(const CXzs *p); +UInt64 Xzs_GetUnpackSize(const CXzs *p); + + +// ECoderStatus values are identical to ELzmaStatus values of LZMA2 decoder + +typedef enum +{ + CODER_STATUS_NOT_SPECIFIED, /* use main error code instead */ + CODER_STATUS_FINISHED_WITH_MARK, /* stream was finished with end mark. */ + CODER_STATUS_NOT_FINISHED, /* stream was not finished */ + CODER_STATUS_NEEDS_MORE_INPUT /* you must provide more input bytes */ +} ECoderStatus; + + +// ECoderFinishMode values are identical to ELzmaFinishMode + +typedef enum +{ + CODER_FINISH_ANY, /* finish at any point */ + CODER_FINISH_END /* block must be finished at the end */ +} ECoderFinishMode; + + +typedef struct +{ + void *p; // state object; + void (*Free)(void *p, ISzAllocPtr alloc); + SRes (*SetProps)(void *p, const Byte *props, size_t propSize, ISzAllocPtr alloc); + void (*Init)(void *p); + SRes (*Code2)(void *p, Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen, + int srcWasFinished, ECoderFinishMode finishMode, + // int *wasFinished, + ECoderStatus *status); + SizeT (*Filter)(void *p, Byte *data, SizeT size); +} IStateCoder; + + +typedef struct +{ + UInt32 methodId; + UInt32 delta; + UInt32 ip; + UInt32 X86_State; + Byte delta_State[DELTA_STATE_SIZE]; +} CXzBcFilterStateBase; + +typedef SizeT (*Xz_Func_BcFilterStateBase_Filter)(CXzBcFilterStateBase *p, Byte *data, SizeT size); + +SRes Xz_StateCoder_Bc_SetFromMethod_Func(IStateCoder *p, UInt64 id, + Xz_Func_BcFilterStateBase_Filter func, ISzAllocPtr alloc); + + +#define MIXCODER_NUM_FILTERS_MAX 4 + +typedef struct +{ + ISzAllocPtr alloc; + Byte *buf; + unsigned numCoders; + + Byte *outBuf; + size_t outBufSize; + size_t outWritten; // is equal to lzmaDecoder.dicPos (in outBuf mode) + BoolInt wasFinished; + SRes res; + ECoderStatus status; + // BoolInt SingleBufMode; + + int finished[MIXCODER_NUM_FILTERS_MAX - 1]; + size_t pos[MIXCODER_NUM_FILTERS_MAX - 1]; + size_t size[MIXCODER_NUM_FILTERS_MAX - 1]; + UInt64 ids[MIXCODER_NUM_FILTERS_MAX]; + SRes results[MIXCODER_NUM_FILTERS_MAX]; + IStateCoder coders[MIXCODER_NUM_FILTERS_MAX]; +} CMixCoder; + + +typedef enum +{ + XZ_STATE_STREAM_HEADER, + XZ_STATE_STREAM_INDEX, + XZ_STATE_STREAM_INDEX_CRC, + XZ_STATE_STREAM_FOOTER, + XZ_STATE_STREAM_PADDING, + XZ_STATE_BLOCK_HEADER, + XZ_STATE_BLOCK, + XZ_STATE_BLOCK_FOOTER +} EXzState; + + +typedef struct +{ + EXzState state; + UInt32 pos; + unsigned alignPos; + unsigned indexPreSize; + + CXzStreamFlags streamFlags; + + UInt32 blockHeaderSize; + UInt64 packSize; + UInt64 unpackSize; + + UInt64 numBlocks; // number of finished blocks in current stream + UInt64 indexSize; + UInt64 indexPos; + UInt64 padSize; + + UInt64 numStartedStreams; + UInt64 numFinishedStreams; + UInt64 numTotalBlocks; + + UInt32 crc; + CMixCoder decoder; + CXzBlock block; + CXzCheck check; + CSha256 sha; + + BoolInt parseMode; + BoolInt headerParsedOk; + BoolInt decodeToStreamSignature; + unsigned decodeOnlyOneBlock; + + Byte *outBuf; + size_t outBufSize; + size_t outDataWritten; // the size of data in (outBuf) that were fully unpacked + + Byte shaDigest[SHA256_DIGEST_SIZE]; + Byte buf[XZ_BLOCK_HEADER_SIZE_MAX]; +} CXzUnpacker; + +/* alloc : aligned for cache line allocation is better */ +void XzUnpacker_Construct(CXzUnpacker *p, ISzAllocPtr alloc); +void XzUnpacker_Init(CXzUnpacker *p); +void XzUnpacker_SetOutBuf(CXzUnpacker *p, Byte *outBuf, size_t outBufSize); +void XzUnpacker_Free(CXzUnpacker *p); + +/* + XzUnpacker + The sequence for decoding functions: + { + XzUnpacker_Construct() + [Decoding_Calls] + XzUnpacker_Free() + } + + [Decoding_Calls] + + There are 3 types of interfaces for [Decoding_Calls] calls: + + Interface-1 : Partial output buffers: + { + XzUnpacker_Init() + for() + { + XzUnpacker_Code(); + } + XzUnpacker_IsStreamWasFinished() + } + + Interface-2 : Direct output buffer: + Use it, if you know exact size of decoded data, and you need + whole xz unpacked data in one output buffer. + xz unpacker doesn't allocate additional buffer for lzma2 dictionary in that mode. + { + XzUnpacker_Init() + XzUnpacker_SetOutBufMode(); // to set output buffer and size + for() + { + XzUnpacker_Code(); // (dest = NULL) in XzUnpacker_Code() + } + XzUnpacker_IsStreamWasFinished() + } + + Interface-3 : Direct output buffer : One call full decoding + It unpacks whole input buffer to output buffer in one call. + It uses Interface-2 internally. + { + XzUnpacker_CodeFull() + XzUnpacker_IsStreamWasFinished() + } +*/ + +/* +finishMode: + It has meaning only if the decoding reaches output limit (*destLen). + CODER_FINISH_ANY - use smallest number of input bytes + CODER_FINISH_END - read EndOfStream marker after decoding + +Returns: + SZ_OK + status: + CODER_STATUS_NOT_FINISHED, + CODER_STATUS_NEEDS_MORE_INPUT - the decoder can return it in two cases: + 1) it needs more input data to finish current xz stream + 2) xz stream was finished successfully. But the decoder supports multiple + concatented xz streams. So it expects more input data for new xz streams. + Call XzUnpacker_IsStreamWasFinished() to check that latest xz stream was finished successfully. + + SZ_ERROR_MEM - Memory allocation error + SZ_ERROR_DATA - Data error + SZ_ERROR_UNSUPPORTED - Unsupported method or method properties + SZ_ERROR_CRC - CRC error + // SZ_ERROR_INPUT_EOF - It needs more bytes in input buffer (src). + + SZ_ERROR_NO_ARCHIVE - the error with xz Stream Header with one of the following reasons: + - xz Stream Signature failure + - CRC32 of xz Stream Header is failed + - The size of Stream padding is not multiple of four bytes. + It's possible to get that error, if xz stream was finished and the stream + contains some another data. In that case you can call XzUnpacker_GetExtraSize() + function to get real size of xz stream. +*/ + + +SRes XzUnpacker_Code(CXzUnpacker *p, Byte *dest, SizeT *destLen, + const Byte *src, SizeT *srcLen, int srcFinished, + ECoderFinishMode finishMode, ECoderStatus *status); + +SRes XzUnpacker_CodeFull(CXzUnpacker *p, Byte *dest, SizeT *destLen, + const Byte *src, SizeT *srcLen, + ECoderFinishMode finishMode, ECoderStatus *status); + +/* +If you decode full xz stream(s), then you can call XzUnpacker_IsStreamWasFinished() +after successful XzUnpacker_CodeFull() or after last call of XzUnpacker_Code(). +*/ + +BoolInt XzUnpacker_IsStreamWasFinished(const CXzUnpacker *p); + +/* +XzUnpacker_GetExtraSize() returns then number of unconfirmed bytes, + if it's in (XZ_STATE_STREAM_HEADER) state or in (XZ_STATE_STREAM_PADDING) state. +These bytes can be some data after xz archive, or +it can be start of new xz stream. + +Call XzUnpacker_GetExtraSize() after XzUnpacker_Code() function to detect real size of +xz stream in two cases, if XzUnpacker_Code() returns: + res == SZ_OK && status == CODER_STATUS_NEEDS_MORE_INPUT + res == SZ_ERROR_NO_ARCHIVE +*/ + +UInt64 XzUnpacker_GetExtraSize(const CXzUnpacker *p); + + +/* + for random block decoding: + XzUnpacker_Init(); + set CXzUnpacker::streamFlags + XzUnpacker_PrepareToRandomBlockDecoding() + loop + { + XzUnpacker_Code() + XzUnpacker_IsBlockFinished() + } +*/ + +void XzUnpacker_PrepareToRandomBlockDecoding(CXzUnpacker *p); +BoolInt XzUnpacker_IsBlockFinished(const CXzUnpacker *p); + +#define XzUnpacker_GetPackSizeForIndex(p) ((p)->packSize + (p)->blockHeaderSize + XzFlags_GetCheckSize((p)->streamFlags)) + + + + + + +/* ---- Single-Thread and Multi-Thread xz Decoding with Input/Output Streams ---- */ + +/* + if (CXzDecMtProps::numThreads > 1), the decoder can try to use + Multi-Threading. The decoder analyses xz block header, and if + there are pack size and unpack size values stored in xz block header, + the decoder reads compressed data of block to internal buffers, + and then it can start parallel decoding, if there are another blocks. + The decoder can switch back to Single-Thread decoding after some conditions. + + The sequence of calls for xz decoding with in/out Streams: + { + XzDecMt_Create() + XzDecMtProps_Init(XzDecMtProps) to set default values of properties + // then you can change some XzDecMtProps parameters with required values + // here you can set the number of threads and (memUseMax) - the maximum + Memory usage for multithreading decoding. + for() + { + XzDecMt_Decode() // one call per one file + } + XzDecMt_Destroy() + } +*/ + + +typedef struct +{ + size_t inBufSize_ST; // size of input buffer for Single-Thread decoding + size_t outStep_ST; // size of output buffer for Single-Thread decoding + BoolInt ignoreErrors; // if set to 1, the decoder can ignore some errors and it skips broken parts of data. + + #ifndef Z7_ST + unsigned numThreads; // the number of threads for Multi-Thread decoding. if (umThreads == 1) it will use Single-thread decoding + size_t inBufSize_MT; // size of small input data buffers for Multi-Thread decoding. Big number of such small buffers can be created + size_t memUseMax; // the limit of total memory usage for Multi-Thread decoding. + // it's recommended to set (memUseMax) manually to value that is smaller of total size of RAM in computer. + #endif +} CXzDecMtProps; + +void XzDecMtProps_Init(CXzDecMtProps *p); + +typedef struct CXzDecMt CXzDecMt; +typedef CXzDecMt * CXzDecMtHandle; +// Z7_DECLARE_HANDLE(CXzDecMtHandle) + +/* + alloc : XzDecMt uses CAlignOffsetAlloc internally for addresses allocated by (alloc). + allocMid : for big allocations, aligned allocation is better +*/ + +CXzDecMtHandle XzDecMt_Create(ISzAllocPtr alloc, ISzAllocPtr allocMid); +void XzDecMt_Destroy(CXzDecMtHandle p); + + +typedef struct +{ + Byte UnpackSize_Defined; + Byte NumStreams_Defined; + Byte NumBlocks_Defined; + + Byte DataAfterEnd; // there are some additional data after good xz streams, and that data is not new xz stream. + Byte DecodingTruncated; // Decoding was Truncated, we need only partial output data + + UInt64 InSize; // pack size processed. That value doesn't include the data after + // end of xz stream, if that data was not correct + UInt64 OutSize; + + UInt64 NumStreams; + UInt64 NumBlocks; + + SRes DecodeRes; // the error code of xz streams data decoding + SRes ReadRes; // error code from ISeqInStream:Read() + SRes ProgressRes; // error code from ICompressProgress:Progress() + + SRes CombinedRes; // Combined result error code that shows main rusult + // = S_OK, if there is no error. + // but check also (DataAfterEnd) that can show additional minor errors. + + SRes CombinedRes_Type; // = SZ_ERROR_READ, if error from ISeqInStream + // = SZ_ERROR_PROGRESS, if error from ICompressProgress + // = SZ_ERROR_WRITE, if error from ISeqOutStream + // = SZ_ERROR_* codes for decoding +} CXzStatInfo; + +void XzStatInfo_Clear(CXzStatInfo *p); + +/* + +XzDecMt_Decode() +SRes: it's combined decoding result. It also is equal to stat->CombinedRes. + + SZ_OK - no error + check also output value in (stat->DataAfterEnd) + that can show additional possible error + + SZ_ERROR_MEM - Memory allocation error + SZ_ERROR_NO_ARCHIVE - is not xz archive + SZ_ERROR_ARCHIVE - Headers error + SZ_ERROR_DATA - Data Error + SZ_ERROR_UNSUPPORTED - Unsupported method or method properties + SZ_ERROR_CRC - CRC Error + SZ_ERROR_INPUT_EOF - it needs more input data + SZ_ERROR_WRITE - ISeqOutStream error + (SZ_ERROR_READ) - ISeqInStream errors + (SZ_ERROR_PROGRESS) - ICompressProgress errors + // SZ_ERROR_THREAD - error in multi-threading functions + MY_SRes_HRESULT_FROM_WRes(WRes_error) - error in multi-threading function +*/ + +SRes XzDecMt_Decode(CXzDecMtHandle p, + const CXzDecMtProps *props, + const UInt64 *outDataSize, // NULL means undefined + int finishMode, // 0 - partial unpacking is allowed, 1 - xz stream(s) must be finished + ISeqOutStreamPtr outStream, + // Byte *outBuf, size_t *outBufSize, + ISeqInStreamPtr inStream, + // const Byte *inData, size_t inDataSize, + CXzStatInfo *stat, // out: decoding results and statistics + int *isMT, // out: 0 means that ST (Single-Thread) version was used + // 1 means that MT (Multi-Thread) version was used + ICompressProgressPtr progress); + +EXTERN_C_END + +#endif diff --git a/libraries/lzma/C/XzCrc64.c b/libraries/lzma/C/XzCrc64.c new file mode 100644 index 000000000..c2fad6cda --- /dev/null +++ b/libraries/lzma/C/XzCrc64.c @@ -0,0 +1,80 @@ +/* XzCrc64.c -- CRC64 calculation +2023-04-02 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include "XzCrc64.h" +#include "CpuArch.h" + +#define kCrc64Poly UINT64_CONST(0xC96C5795D7870F42) + +#ifdef MY_CPU_LE + #define CRC64_NUM_TABLES 4 +#else + #define CRC64_NUM_TABLES 5 + + UInt64 Z7_FASTCALL XzCrc64UpdateT1_BeT4(UInt64 v, const void *data, size_t size, const UInt64 *table); +#endif + +#ifndef MY_CPU_BE + UInt64 Z7_FASTCALL XzCrc64UpdateT4(UInt64 v, const void *data, size_t size, const UInt64 *table); +#endif + +typedef UInt64 (Z7_FASTCALL *CRC64_FUNC)(UInt64 v, const void *data, size_t size, const UInt64 *table); + +static CRC64_FUNC g_Crc64Update; +UInt64 g_Crc64Table[256 * CRC64_NUM_TABLES]; + +UInt64 Z7_FASTCALL Crc64Update(UInt64 v, const void *data, size_t size) +{ + return g_Crc64Update(v, data, size, g_Crc64Table); +} + +UInt64 Z7_FASTCALL Crc64Calc(const void *data, size_t size) +{ + return g_Crc64Update(CRC64_INIT_VAL, data, size, g_Crc64Table) ^ CRC64_INIT_VAL; +} + +void Z7_FASTCALL Crc64GenerateTable(void) +{ + UInt32 i; + for (i = 0; i < 256; i++) + { + UInt64 r = i; + unsigned j; + for (j = 0; j < 8; j++) + r = (r >> 1) ^ (kCrc64Poly & ((UInt64)0 - (r & 1))); + g_Crc64Table[i] = r; + } + for (i = 256; i < 256 * CRC64_NUM_TABLES; i++) + { + const UInt64 r = g_Crc64Table[(size_t)i - 256]; + g_Crc64Table[i] = g_Crc64Table[r & 0xFF] ^ (r >> 8); + } + + #ifdef MY_CPU_LE + + g_Crc64Update = XzCrc64UpdateT4; + + #else + { + #ifndef MY_CPU_BE + UInt32 k = 1; + if (*(const Byte *)&k == 1) + g_Crc64Update = XzCrc64UpdateT4; + else + #endif + { + for (i = 256 * CRC64_NUM_TABLES - 1; i >= 256; i--) + { + const UInt64 x = g_Crc64Table[(size_t)i - 256]; + g_Crc64Table[i] = Z7_BSWAP64(x); + } + g_Crc64Update = XzCrc64UpdateT1_BeT4; + } + } + #endif +} + +#undef kCrc64Poly +#undef CRC64_NUM_TABLES diff --git a/libraries/lzma/C/XzCrc64.h b/libraries/lzma/C/XzCrc64.h new file mode 100644 index 000000000..ca46869a6 --- /dev/null +++ b/libraries/lzma/C/XzCrc64.h @@ -0,0 +1,26 @@ +/* XzCrc64.h -- CRC64 calculation +2023-04-02 : Igor Pavlov : Public domain */ + +#ifndef ZIP7_INC_XZ_CRC64_H +#define ZIP7_INC_XZ_CRC64_H + +#include + +#include "7zTypes.h" + +EXTERN_C_BEGIN + +extern UInt64 g_Crc64Table[]; + +void Z7_FASTCALL Crc64GenerateTable(void); + +#define CRC64_INIT_VAL UINT64_CONST(0xFFFFFFFFFFFFFFFF) +#define CRC64_GET_DIGEST(crc) ((crc) ^ CRC64_INIT_VAL) +#define CRC64_UPDATE_BYTE(crc, b) (g_Crc64Table[((crc) ^ (b)) & 0xFF] ^ ((crc) >> 8)) + +UInt64 Z7_FASTCALL Crc64Update(UInt64 crc, const void *data, size_t size); +UInt64 Z7_FASTCALL Crc64Calc(const void *data, size_t size); + +EXTERN_C_END + +#endif diff --git a/libraries/lzma/C/XzCrc64Opt.c b/libraries/lzma/C/XzCrc64Opt.c new file mode 100644 index 000000000..d03374c00 --- /dev/null +++ b/libraries/lzma/C/XzCrc64Opt.c @@ -0,0 +1,61 @@ +/* XzCrc64Opt.c -- CRC64 calculation +2023-04-02 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include "CpuArch.h" + +#ifndef MY_CPU_BE + +#define CRC64_UPDATE_BYTE_2(crc, b) (table[((crc) ^ (b)) & 0xFF] ^ ((crc) >> 8)) + +UInt64 Z7_FASTCALL XzCrc64UpdateT4(UInt64 v, const void *data, size_t size, const UInt64 *table); +UInt64 Z7_FASTCALL XzCrc64UpdateT4(UInt64 v, const void *data, size_t size, const UInt64 *table) +{ + const Byte *p = (const Byte *)data; + for (; size > 0 && ((unsigned)(ptrdiff_t)p & 3) != 0; size--, p++) + v = CRC64_UPDATE_BYTE_2(v, *p); + for (; size >= 4; size -= 4, p += 4) + { + const UInt32 d = (UInt32)v ^ *(const UInt32 *)(const void *)p; + v = (v >> 32) + ^ (table + 0x300)[((d ) & 0xFF)] + ^ (table + 0x200)[((d >> 8) & 0xFF)] + ^ (table + 0x100)[((d >> 16) & 0xFF)] + ^ (table + 0x000)[((d >> 24))]; + } + for (; size > 0; size--, p++) + v = CRC64_UPDATE_BYTE_2(v, *p); + return v; +} + +#endif + + +#ifndef MY_CPU_LE + +#define CRC64_UPDATE_BYTE_2_BE(crc, b) (table[(Byte)((crc) >> 56) ^ (b)] ^ ((crc) << 8)) + +UInt64 Z7_FASTCALL XzCrc64UpdateT1_BeT4(UInt64 v, const void *data, size_t size, const UInt64 *table); +UInt64 Z7_FASTCALL XzCrc64UpdateT1_BeT4(UInt64 v, const void *data, size_t size, const UInt64 *table) +{ + const Byte *p = (const Byte *)data; + table += 0x100; + v = Z7_BSWAP64(v); + for (; size > 0 && ((unsigned)(ptrdiff_t)p & 3) != 0; size--, p++) + v = CRC64_UPDATE_BYTE_2_BE(v, *p); + for (; size >= 4; size -= 4, p += 4) + { + const UInt32 d = (UInt32)(v >> 32) ^ *(const UInt32 *)(const void *)p; + v = (v << 32) + ^ (table + 0x000)[((d ) & 0xFF)] + ^ (table + 0x100)[((d >> 8) & 0xFF)] + ^ (table + 0x200)[((d >> 16) & 0xFF)] + ^ (table + 0x300)[((d >> 24))]; + } + for (; size > 0; size--, p++) + v = CRC64_UPDATE_BYTE_2_BE(v, *p); + return Z7_BSWAP64(v); +} + +#endif diff --git a/libraries/lzma/C/XzDec.c b/libraries/lzma/C/XzDec.c new file mode 100644 index 000000000..a5f703966 --- /dev/null +++ b/libraries/lzma/C/XzDec.c @@ -0,0 +1,2875 @@ +/* XzDec.c -- Xz Decode +2023-04-13 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +// #include + +// #define XZ_DUMP + +/* #define XZ_DUMP */ + +#ifdef XZ_DUMP +#include +#endif + +// #define SHOW_DEBUG_INFO + +#ifdef SHOW_DEBUG_INFO +#include +#endif + +#ifdef SHOW_DEBUG_INFO +#define PRF(x) x +#else +#define PRF(x) +#endif + +#define PRF_STR(s) PRF(printf("\n" s "\n")) +#define PRF_STR_INT(s, d) PRF(printf("\n" s " %d\n", (unsigned)d)) + +#include +#include + +#include "7zCrc.h" +#include "Alloc.h" +#include "Bra.h" +#include "CpuArch.h" +#include "Delta.h" +#include "Lzma2Dec.h" + +// #define USE_SUBBLOCK + +#ifdef USE_SUBBLOCK +#include "Bcj3Dec.c" +#include "SbDec.h" +#endif + +#include "Xz.h" + +#define XZ_CHECK_SIZE_MAX 64 + +#define CODER_BUF_SIZE ((size_t)1 << 17) + +unsigned Xz_ReadVarInt(const Byte *p, size_t maxSize, UInt64 *value) +{ + unsigned i, limit; + *value = 0; + limit = (maxSize > 9) ? 9 : (unsigned)maxSize; + + for (i = 0; i < limit;) + { + Byte b = p[i]; + *value |= (UInt64)(b & 0x7F) << (7 * i++); + if ((b & 0x80) == 0) + return (b == 0 && i != 1) ? 0 : i; + } + return 0; +} + + +/* ---------- XzBcFilterState ---------- */ + +#define BRA_BUF_SIZE (1 << 14) + +typedef struct +{ + size_t bufPos; + size_t bufConv; + size_t bufTotal; + Byte *buf; // must be aligned for 4 bytes + Xz_Func_BcFilterStateBase_Filter filter_func; + // int encodeMode; + CXzBcFilterStateBase base; + // Byte buf[BRA_BUF_SIZE]; +} CXzBcFilterState; + + +static void XzBcFilterState_Free(void *pp, ISzAllocPtr alloc) +{ + if (pp) + { + CXzBcFilterState *p = ((CXzBcFilterState *)pp); + ISzAlloc_Free(alloc, p->buf); + ISzAlloc_Free(alloc, pp); + } +} + + +static SRes XzBcFilterState_SetProps(void *pp, const Byte *props, size_t propSize, ISzAllocPtr alloc) +{ + CXzBcFilterStateBase *p = &((CXzBcFilterState *)pp)->base; + UNUSED_VAR(alloc) + p->ip = 0; + if (p->methodId == XZ_ID_Delta) + { + if (propSize != 1) + return SZ_ERROR_UNSUPPORTED; + p->delta = (unsigned)props[0] + 1; + } + else + { + if (propSize == 4) + { + UInt32 v = GetUi32(props); + switch (p->methodId) + { + case XZ_ID_PPC: + case XZ_ID_ARM: + case XZ_ID_SPARC: + case XZ_ID_ARM64: + if ((v & 3) != 0) + return SZ_ERROR_UNSUPPORTED; + break; + case XZ_ID_ARMT: + if ((v & 1) != 0) + return SZ_ERROR_UNSUPPORTED; + break; + case XZ_ID_IA64: + if ((v & 0xF) != 0) + return SZ_ERROR_UNSUPPORTED; + break; + } + p->ip = v; + } + else if (propSize != 0) + return SZ_ERROR_UNSUPPORTED; + } + return SZ_OK; +} + + +static void XzBcFilterState_Init(void *pp) +{ + CXzBcFilterState *p = ((CXzBcFilterState *)pp); + p->bufPos = p->bufConv = p->bufTotal = 0; + p->base.X86_State = Z7_BRANCH_CONV_ST_X86_STATE_INIT_VAL; + if (p->base.methodId == XZ_ID_Delta) + Delta_Init(p->base.delta_State); +} + + +static const z7_Func_BranchConv g_Funcs_BranchConv_RISC_Dec[] = +{ + Z7_BRANCH_CONV_DEC(PPC), + Z7_BRANCH_CONV_DEC(IA64), + Z7_BRANCH_CONV_DEC(ARM), + Z7_BRANCH_CONV_DEC(ARMT), + Z7_BRANCH_CONV_DEC(SPARC), + Z7_BRANCH_CONV_DEC(ARM64) +}; + +static SizeT XzBcFilterStateBase_Filter_Dec(CXzBcFilterStateBase *p, Byte *data, SizeT size) +{ + switch (p->methodId) + { + case XZ_ID_Delta: + Delta_Decode(p->delta_State, p->delta, data, size); + break; + case XZ_ID_X86: + size = (SizeT)(z7_BranchConvSt_X86_Dec(data, size, p->ip, &p->X86_State) - data); + break; + default: + if (p->methodId >= XZ_ID_PPC) + { + const UInt32 i = p->methodId - XZ_ID_PPC; + if (i < Z7_ARRAY_SIZE(g_Funcs_BranchConv_RISC_Dec)) + size = (SizeT)(g_Funcs_BranchConv_RISC_Dec[i](data, size, p->ip) - data); + } + break; + } + p->ip += (UInt32)size; + return size; +} + + +static SizeT XzBcFilterState_Filter(void *pp, Byte *data, SizeT size) +{ + CXzBcFilterState *p = ((CXzBcFilterState *)pp); + return p->filter_func(&p->base, data, size); +} + + +static SRes XzBcFilterState_Code2(void *pp, + Byte *dest, SizeT *destLen, + const Byte *src, SizeT *srcLen, int srcWasFinished, + ECoderFinishMode finishMode, + // int *wasFinished + ECoderStatus *status) +{ + CXzBcFilterState *p = ((CXzBcFilterState *)pp); + SizeT destRem = *destLen; + SizeT srcRem = *srcLen; + UNUSED_VAR(finishMode) + + *destLen = 0; + *srcLen = 0; + // *wasFinished = False; + *status = CODER_STATUS_NOT_FINISHED; + + while (destRem != 0) + { + { + size_t size = p->bufConv - p->bufPos; + if (size) + { + if (size > destRem) + size = destRem; + memcpy(dest, p->buf + p->bufPos, size); + p->bufPos += size; + *destLen += size; + dest += size; + destRem -= size; + continue; + } + } + + p->bufTotal -= p->bufPos; + memmove(p->buf, p->buf + p->bufPos, p->bufTotal); + p->bufPos = 0; + p->bufConv = 0; + { + size_t size = BRA_BUF_SIZE - p->bufTotal; + if (size > srcRem) + size = srcRem; + memcpy(p->buf + p->bufTotal, src, size); + *srcLen += size; + src += size; + srcRem -= size; + p->bufTotal += size; + } + if (p->bufTotal == 0) + break; + + p->bufConv = p->filter_func(&p->base, p->buf, p->bufTotal); + + if (p->bufConv == 0) + { + if (!srcWasFinished) + break; + p->bufConv = p->bufTotal; + } + } + + if (p->bufTotal == p->bufPos && srcRem == 0 && srcWasFinished) + { + *status = CODER_STATUS_FINISHED_WITH_MARK; + // *wasFinished = 1; + } + + return SZ_OK; +} + + +#define XZ_IS_SUPPORTED_FILTER_ID(id) \ + ((id) >= XZ_ID_Delta && (id) <= XZ_ID_ARM64) + +SRes Xz_StateCoder_Bc_SetFromMethod_Func(IStateCoder *p, UInt64 id, + Xz_Func_BcFilterStateBase_Filter func, ISzAllocPtr alloc) +{ + CXzBcFilterState *decoder; + if (!XZ_IS_SUPPORTED_FILTER_ID(id)) + return SZ_ERROR_UNSUPPORTED; + decoder = (CXzBcFilterState *)p->p; + if (!decoder) + { + decoder = (CXzBcFilterState *)ISzAlloc_Alloc(alloc, sizeof(CXzBcFilterState)); + if (!decoder) + return SZ_ERROR_MEM; + decoder->buf = ISzAlloc_Alloc(alloc, BRA_BUF_SIZE); + if (!decoder->buf) + { + ISzAlloc_Free(alloc, decoder); + return SZ_ERROR_MEM; + } + p->p = decoder; + p->Free = XzBcFilterState_Free; + p->SetProps = XzBcFilterState_SetProps; + p->Init = XzBcFilterState_Init; + p->Code2 = XzBcFilterState_Code2; + p->Filter = XzBcFilterState_Filter; + decoder->filter_func = func; + } + decoder->base.methodId = (UInt32)id; + // decoder->encodeMode = encodeMode; + return SZ_OK; +} + + + +/* ---------- SbState ---------- */ + +#ifdef USE_SUBBLOCK + +static void SbState_Free(void *pp, ISzAllocPtr alloc) +{ + CSbDec *p = (CSbDec *)pp; + SbDec_Free(p); + ISzAlloc_Free(alloc, pp); +} + +static SRes SbState_SetProps(void *pp, const Byte *props, size_t propSize, ISzAllocPtr alloc) +{ + UNUSED_VAR(pp) + UNUSED_VAR(props) + UNUSED_VAR(alloc) + return (propSize == 0) ? SZ_OK : SZ_ERROR_UNSUPPORTED; +} + +static void SbState_Init(void *pp) +{ + SbDec_Init((CSbDec *)pp); +} + +static SRes SbState_Code2(void *pp, Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen, + int srcWasFinished, ECoderFinishMode finishMode, + // int *wasFinished + ECoderStatus *status) +{ + CSbDec *p = (CSbDec *)pp; + SRes res; + UNUSED_VAR(srcWasFinished) + p->dest = dest; + p->destLen = *destLen; + p->src = src; + p->srcLen = *srcLen; + p->finish = finishMode; /* change it */ + res = SbDec_Decode((CSbDec *)pp); + *destLen -= p->destLen; + *srcLen -= p->srcLen; + // *wasFinished = (*destLen == 0 && *srcLen == 0); /* change it */ + *status = (*destLen == 0 && *srcLen == 0) ? + CODER_STATUS_FINISHED_WITH_MARK : + CODER_STATUS_NOT_FINISHED; + return res; +} + +static SRes SbState_SetFromMethod(IStateCoder *p, ISzAllocPtr alloc) +{ + CSbDec *decoder = (CSbDec *)p->p; + if (!decoder) + { + decoder = (CSbDec *)ISzAlloc_Alloc(alloc, sizeof(CSbDec)); + if (!decoder) + return SZ_ERROR_MEM; + p->p = decoder; + p->Free = SbState_Free; + p->SetProps = SbState_SetProps; + p->Init = SbState_Init; + p->Code2 = SbState_Code2; + p->Filter = NULL; + } + SbDec_Construct(decoder); + SbDec_SetAlloc(decoder, alloc); + return SZ_OK; +} + +#endif + + + +/* ---------- Lzma2 ---------- */ + +typedef struct +{ + CLzma2Dec decoder; + BoolInt outBufMode; +} CLzma2Dec_Spec; + + +static void Lzma2State_Free(void *pp, ISzAllocPtr alloc) +{ + CLzma2Dec_Spec *p = (CLzma2Dec_Spec *)pp; + if (p->outBufMode) + Lzma2Dec_FreeProbs(&p->decoder, alloc); + else + Lzma2Dec_Free(&p->decoder, alloc); + ISzAlloc_Free(alloc, pp); +} + +static SRes Lzma2State_SetProps(void *pp, const Byte *props, size_t propSize, ISzAllocPtr alloc) +{ + if (propSize != 1) + return SZ_ERROR_UNSUPPORTED; + { + CLzma2Dec_Spec *p = (CLzma2Dec_Spec *)pp; + if (p->outBufMode) + return Lzma2Dec_AllocateProbs(&p->decoder, props[0], alloc); + else + return Lzma2Dec_Allocate(&p->decoder, props[0], alloc); + } +} + +static void Lzma2State_Init(void *pp) +{ + Lzma2Dec_Init(&((CLzma2Dec_Spec *)pp)->decoder); +} + + +/* + if (outBufMode), then (dest) is not used. Use NULL. + Data is unpacked to (spec->decoder.decoder.dic) output buffer. +*/ + +static SRes Lzma2State_Code2(void *pp, Byte *dest, SizeT *destLen, const Byte *src, SizeT *srcLen, + int srcWasFinished, ECoderFinishMode finishMode, + // int *wasFinished, + ECoderStatus *status) +{ + CLzma2Dec_Spec *spec = (CLzma2Dec_Spec *)pp; + ELzmaStatus status2; + /* ELzmaFinishMode fm = (finishMode == LZMA_FINISH_ANY) ? LZMA_FINISH_ANY : LZMA_FINISH_END; */ + SRes res; + UNUSED_VAR(srcWasFinished) + if (spec->outBufMode) + { + SizeT dicPos = spec->decoder.decoder.dicPos; + SizeT dicLimit = dicPos + *destLen; + res = Lzma2Dec_DecodeToDic(&spec->decoder, dicLimit, src, srcLen, (ELzmaFinishMode)finishMode, &status2); + *destLen = spec->decoder.decoder.dicPos - dicPos; + } + else + res = Lzma2Dec_DecodeToBuf(&spec->decoder, dest, destLen, src, srcLen, (ELzmaFinishMode)finishMode, &status2); + // *wasFinished = (status2 == LZMA_STATUS_FINISHED_WITH_MARK); + // ECoderStatus values are identical to ELzmaStatus values of LZMA2 decoder + *status = (ECoderStatus)status2; + return res; +} + + +static SRes Lzma2State_SetFromMethod(IStateCoder *p, Byte *outBuf, size_t outBufSize, ISzAllocPtr alloc) +{ + CLzma2Dec_Spec *spec = (CLzma2Dec_Spec *)p->p; + if (!spec) + { + spec = (CLzma2Dec_Spec *)ISzAlloc_Alloc(alloc, sizeof(CLzma2Dec_Spec)); + if (!spec) + return SZ_ERROR_MEM; + p->p = spec; + p->Free = Lzma2State_Free; + p->SetProps = Lzma2State_SetProps; + p->Init = Lzma2State_Init; + p->Code2 = Lzma2State_Code2; + p->Filter = NULL; + Lzma2Dec_CONSTRUCT(&spec->decoder) + } + spec->outBufMode = False; + if (outBuf) + { + spec->outBufMode = True; + spec->decoder.decoder.dic = outBuf; + spec->decoder.decoder.dicBufSize = outBufSize; + } + return SZ_OK; +} + + +static SRes Lzma2State_ResetOutBuf(IStateCoder *p, Byte *outBuf, size_t outBufSize) +{ + CLzma2Dec_Spec *spec = (CLzma2Dec_Spec *)p->p; + if ((spec->outBufMode && !outBuf) || (!spec->outBufMode && outBuf)) + return SZ_ERROR_FAIL; + if (outBuf) + { + spec->decoder.decoder.dic = outBuf; + spec->decoder.decoder.dicBufSize = outBufSize; + } + return SZ_OK; +} + + + +static void MixCoder_Construct(CMixCoder *p, ISzAllocPtr alloc) +{ + unsigned i; + p->alloc = alloc; + p->buf = NULL; + p->numCoders = 0; + + p->outBufSize = 0; + p->outBuf = NULL; + // p->SingleBufMode = False; + + for (i = 0; i < MIXCODER_NUM_FILTERS_MAX; i++) + p->coders[i].p = NULL; +} + + +static void MixCoder_Free(CMixCoder *p) +{ + unsigned i; + p->numCoders = 0; + for (i = 0; i < MIXCODER_NUM_FILTERS_MAX; i++) + { + IStateCoder *sc = &p->coders[i]; + if (sc->p) + { + sc->Free(sc->p, p->alloc); + sc->p = NULL; + } + } + if (p->buf) + { + ISzAlloc_Free(p->alloc, p->buf); + p->buf = NULL; /* 9.31: the BUG was fixed */ + } +} + +static void MixCoder_Init(CMixCoder *p) +{ + unsigned i; + for (i = 0; i < MIXCODER_NUM_FILTERS_MAX - 1; i++) + { + p->size[i] = 0; + p->pos[i] = 0; + p->finished[i] = 0; + } + for (i = 0; i < p->numCoders; i++) + { + IStateCoder *coder = &p->coders[i]; + coder->Init(coder->p); + p->results[i] = SZ_OK; + } + p->outWritten = 0; + p->wasFinished = False; + p->res = SZ_OK; + p->status = CODER_STATUS_NOT_SPECIFIED; +} + + +static SRes MixCoder_SetFromMethod(CMixCoder *p, unsigned coderIndex, UInt64 methodId, Byte *outBuf, size_t outBufSize) +{ + IStateCoder *sc = &p->coders[coderIndex]; + p->ids[coderIndex] = methodId; + switch (methodId) + { + case XZ_ID_LZMA2: return Lzma2State_SetFromMethod(sc, outBuf, outBufSize, p->alloc); + #ifdef USE_SUBBLOCK + case XZ_ID_Subblock: return SbState_SetFromMethod(sc, p->alloc); + #endif + } + if (coderIndex == 0) + return SZ_ERROR_UNSUPPORTED; + return Xz_StateCoder_Bc_SetFromMethod_Func(sc, methodId, + XzBcFilterStateBase_Filter_Dec, p->alloc); +} + + +static SRes MixCoder_ResetFromMethod(CMixCoder *p, unsigned coderIndex, UInt64 methodId, Byte *outBuf, size_t outBufSize) +{ + IStateCoder *sc = &p->coders[coderIndex]; + switch (methodId) + { + case XZ_ID_LZMA2: return Lzma2State_ResetOutBuf(sc, outBuf, outBufSize); + } + return SZ_ERROR_UNSUPPORTED; +} + + + +/* + if (destFinish) - then unpack data block is finished at (*destLen) position, + and we can return data that were not processed by filter + +output (status) can be : + CODER_STATUS_NOT_FINISHED + CODER_STATUS_FINISHED_WITH_MARK + CODER_STATUS_NEEDS_MORE_INPUT - not implemented still +*/ + +static SRes MixCoder_Code(CMixCoder *p, + Byte *dest, SizeT *destLen, int destFinish, + const Byte *src, SizeT *srcLen, int srcWasFinished, + ECoderFinishMode finishMode) +{ + SizeT destLenOrig = *destLen; + SizeT srcLenOrig = *srcLen; + + *destLen = 0; + *srcLen = 0; + + if (p->wasFinished) + return p->res; + + p->status = CODER_STATUS_NOT_FINISHED; + + // if (p->SingleBufMode) + if (p->outBuf) + { + SRes res; + SizeT destLen2, srcLen2; + int wasFinished; + + PRF_STR("------- MixCoder Single ----------") + + srcLen2 = srcLenOrig; + destLen2 = destLenOrig; + + { + IStateCoder *coder = &p->coders[0]; + res = coder->Code2(coder->p, NULL, &destLen2, src, &srcLen2, srcWasFinished, finishMode, + // &wasFinished, + &p->status); + wasFinished = (p->status == CODER_STATUS_FINISHED_WITH_MARK); + } + + p->res = res; + + /* + if (wasFinished) + p->status = CODER_STATUS_FINISHED_WITH_MARK; + else + { + if (res == SZ_OK) + if (destLen2 != destLenOrig) + p->status = CODER_STATUS_NEEDS_MORE_INPUT; + } + */ + + + *srcLen = srcLen2; + src += srcLen2; + p->outWritten += destLen2; + + if (res != SZ_OK || srcWasFinished || wasFinished) + p->wasFinished = True; + + if (p->numCoders == 1) + *destLen = destLen2; + else if (p->wasFinished) + { + unsigned i; + size_t processed = p->outWritten; + + for (i = 1; i < p->numCoders; i++) + { + IStateCoder *coder = &p->coders[i]; + processed = coder->Filter(coder->p, p->outBuf, processed); + if (wasFinished || (destFinish && p->outWritten == destLenOrig)) + processed = p->outWritten; + PRF_STR_INT("filter", i) + } + *destLen = processed; + } + return res; + } + + PRF_STR("standard mix") + + if (p->numCoders != 1) + { + if (!p->buf) + { + p->buf = (Byte *)ISzAlloc_Alloc(p->alloc, CODER_BUF_SIZE * (MIXCODER_NUM_FILTERS_MAX - 1)); + if (!p->buf) + return SZ_ERROR_MEM; + } + + finishMode = CODER_FINISH_ANY; + } + + for (;;) + { + BoolInt processed = False; + BoolInt allFinished = True; + SRes resMain = SZ_OK; + unsigned i; + + p->status = CODER_STATUS_NOT_FINISHED; + /* + if (p->numCoders == 1 && *destLen == destLenOrig && finishMode == LZMA_FINISH_ANY) + break; + */ + + for (i = 0; i < p->numCoders; i++) + { + SRes res; + IStateCoder *coder = &p->coders[i]; + Byte *dest2; + SizeT destLen2, srcLen2; // destLen2_Orig; + const Byte *src2; + int srcFinished2; + int encodingWasFinished; + ECoderStatus status2; + + if (i == 0) + { + src2 = src; + srcLen2 = srcLenOrig - *srcLen; + srcFinished2 = srcWasFinished; + } + else + { + size_t k = i - 1; + src2 = p->buf + (CODER_BUF_SIZE * k) + p->pos[k]; + srcLen2 = p->size[k] - p->pos[k]; + srcFinished2 = p->finished[k]; + } + + if (i == p->numCoders - 1) + { + dest2 = dest; + destLen2 = destLenOrig - *destLen; + } + else + { + if (p->pos[i] != p->size[i]) + continue; + dest2 = p->buf + (CODER_BUF_SIZE * i); + destLen2 = CODER_BUF_SIZE; + } + + // destLen2_Orig = destLen2; + + if (p->results[i] != SZ_OK) + { + if (resMain == SZ_OK) + resMain = p->results[i]; + continue; + } + + res = coder->Code2(coder->p, + dest2, &destLen2, + src2, &srcLen2, srcFinished2, + finishMode, + // &encodingWasFinished, + &status2); + + if (res != SZ_OK) + { + p->results[i] = res; + if (resMain == SZ_OK) + resMain = res; + } + + encodingWasFinished = (status2 == CODER_STATUS_FINISHED_WITH_MARK); + + if (!encodingWasFinished) + { + allFinished = False; + if (p->numCoders == 1 && res == SZ_OK) + p->status = status2; + } + + if (i == 0) + { + *srcLen += srcLen2; + src += srcLen2; + } + else + p->pos[(size_t)i - 1] += srcLen2; + + if (i == p->numCoders - 1) + { + *destLen += destLen2; + dest += destLen2; + } + else + { + p->size[i] = destLen2; + p->pos[i] = 0; + p->finished[i] = encodingWasFinished; + } + + if (destLen2 != 0 || srcLen2 != 0) + processed = True; + } + + if (!processed) + { + if (allFinished) + p->status = CODER_STATUS_FINISHED_WITH_MARK; + return resMain; + } + } +} + + +SRes Xz_ParseHeader(CXzStreamFlags *p, const Byte *buf) +{ + *p = (CXzStreamFlags)GetBe16(buf + XZ_SIG_SIZE); + if (CrcCalc(buf + XZ_SIG_SIZE, XZ_STREAM_FLAGS_SIZE) != + GetUi32(buf + XZ_SIG_SIZE + XZ_STREAM_FLAGS_SIZE)) + return SZ_ERROR_NO_ARCHIVE; + return XzFlags_IsSupported(*p) ? SZ_OK : SZ_ERROR_UNSUPPORTED; +} + +static BoolInt Xz_CheckFooter(CXzStreamFlags flags, UInt64 indexSize, const Byte *buf) +{ + return indexSize == (((UInt64)GetUi32(buf + 4) + 1) << 2) + && GetUi32(buf) == CrcCalc(buf + 4, 6) + && flags == GetBe16(buf + 8) + && buf[10] == XZ_FOOTER_SIG_0 + && buf[11] == XZ_FOOTER_SIG_1; +} + +#define READ_VARINT_AND_CHECK(buf, pos, size, res) \ + { unsigned s = Xz_ReadVarInt(buf + pos, size - pos, res); \ + if (s == 0) return SZ_ERROR_ARCHIVE; \ + pos += s; } + + +static BoolInt XzBlock_AreSupportedFilters(const CXzBlock *p) +{ + const unsigned numFilters = XzBlock_GetNumFilters(p) - 1; + unsigned i; + { + const CXzFilter *f = &p->filters[numFilters]; + if (f->id != XZ_ID_LZMA2 || f->propsSize != 1 || f->props[0] > 40) + return False; + } + + for (i = 0; i < numFilters; i++) + { + const CXzFilter *f = &p->filters[i]; + if (f->id == XZ_ID_Delta) + { + if (f->propsSize != 1) + return False; + } + else if (!XZ_IS_SUPPORTED_FILTER_ID(f->id) + || (f->propsSize != 0 && f->propsSize != 4)) + return False; + } + return True; +} + + +SRes XzBlock_Parse(CXzBlock *p, const Byte *header) +{ + unsigned pos; + unsigned numFilters, i; + unsigned headerSize = (unsigned)header[0] << 2; + + /* (headerSize != 0) : another code checks */ + + if (CrcCalc(header, headerSize) != GetUi32(header + headerSize)) + return SZ_ERROR_ARCHIVE; + + pos = 1; + p->flags = header[pos++]; + + p->packSize = (UInt64)(Int64)-1; + if (XzBlock_HasPackSize(p)) + { + READ_VARINT_AND_CHECK(header, pos, headerSize, &p->packSize) + if (p->packSize == 0 || p->packSize + headerSize >= (UInt64)1 << 63) + return SZ_ERROR_ARCHIVE; + } + + p->unpackSize = (UInt64)(Int64)-1; + if (XzBlock_HasUnpackSize(p)) + { + READ_VARINT_AND_CHECK(header, pos, headerSize, &p->unpackSize) + } + + numFilters = XzBlock_GetNumFilters(p); + for (i = 0; i < numFilters; i++) + { + CXzFilter *filter = p->filters + i; + UInt64 size; + READ_VARINT_AND_CHECK(header, pos, headerSize, &filter->id) + READ_VARINT_AND_CHECK(header, pos, headerSize, &size) + if (size > headerSize - pos || size > XZ_FILTER_PROPS_SIZE_MAX) + return SZ_ERROR_ARCHIVE; + filter->propsSize = (UInt32)size; + memcpy(filter->props, header + pos, (size_t)size); + pos += (unsigned)size; + + #ifdef XZ_DUMP + printf("\nf[%u] = %2X: ", i, (unsigned)filter->id); + { + unsigned i; + for (i = 0; i < size; i++) + printf(" %2X", filter->props[i]); + } + #endif + } + + if (XzBlock_HasUnsupportedFlags(p)) + return SZ_ERROR_UNSUPPORTED; + + while (pos < headerSize) + if (header[pos++] != 0) + return SZ_ERROR_ARCHIVE; + return SZ_OK; +} + + + + +static SRes XzDecMix_Init(CMixCoder *p, const CXzBlock *block, Byte *outBuf, size_t outBufSize) +{ + unsigned i; + BoolInt needReInit = True; + unsigned numFilters = XzBlock_GetNumFilters(block); + + if (numFilters == p->numCoders && ((p->outBuf && outBuf) || (!p->outBuf && !outBuf))) + { + needReInit = False; + for (i = 0; i < numFilters; i++) + if (p->ids[i] != block->filters[numFilters - 1 - i].id) + { + needReInit = True; + break; + } + } + + // p->SingleBufMode = (outBuf != NULL); + p->outBuf = outBuf; + p->outBufSize = outBufSize; + + // p->SingleBufMode = False; + // outBuf = NULL; + + if (needReInit) + { + MixCoder_Free(p); + for (i = 0; i < numFilters; i++) + { + RINOK(MixCoder_SetFromMethod(p, i, block->filters[numFilters - 1 - i].id, outBuf, outBufSize)) + } + p->numCoders = numFilters; + } + else + { + RINOK(MixCoder_ResetFromMethod(p, 0, block->filters[numFilters - 1].id, outBuf, outBufSize)) + } + + for (i = 0; i < numFilters; i++) + { + const CXzFilter *f = &block->filters[numFilters - 1 - i]; + IStateCoder *sc = &p->coders[i]; + RINOK(sc->SetProps(sc->p, f->props, f->propsSize, p->alloc)) + } + + MixCoder_Init(p); + return SZ_OK; +} + + + +void XzUnpacker_Init(CXzUnpacker *p) +{ + p->state = XZ_STATE_STREAM_HEADER; + p->pos = 0; + p->numStartedStreams = 0; + p->numFinishedStreams = 0; + p->numTotalBlocks = 0; + p->padSize = 0; + p->decodeOnlyOneBlock = 0; + + p->parseMode = False; + p->decodeToStreamSignature = False; + + // p->outBuf = NULL; + // p->outBufSize = 0; + p->outDataWritten = 0; +} + + +void XzUnpacker_SetOutBuf(CXzUnpacker *p, Byte *outBuf, size_t outBufSize) +{ + p->outBuf = outBuf; + p->outBufSize = outBufSize; +} + + +void XzUnpacker_Construct(CXzUnpacker *p, ISzAllocPtr alloc) +{ + MixCoder_Construct(&p->decoder, alloc); + p->outBuf = NULL; + p->outBufSize = 0; + XzUnpacker_Init(p); +} + + +void XzUnpacker_Free(CXzUnpacker *p) +{ + MixCoder_Free(&p->decoder); +} + + +void XzUnpacker_PrepareToRandomBlockDecoding(CXzUnpacker *p) +{ + p->indexSize = 0; + p->numBlocks = 0; + Sha256_Init(&p->sha); + p->state = XZ_STATE_BLOCK_HEADER; + p->pos = 0; + p->decodeOnlyOneBlock = 1; +} + + +static void XzUnpacker_UpdateIndex(CXzUnpacker *p, UInt64 packSize, UInt64 unpackSize) +{ + Byte temp[32]; + unsigned num = Xz_WriteVarInt(temp, packSize); + num += Xz_WriteVarInt(temp + num, unpackSize); + Sha256_Update(&p->sha, temp, num); + p->indexSize += num; + p->numBlocks++; +} + + + +SRes XzUnpacker_Code(CXzUnpacker *p, Byte *dest, SizeT *destLen, + const Byte *src, SizeT *srcLen, int srcFinished, + ECoderFinishMode finishMode, ECoderStatus *status) +{ + SizeT destLenOrig = *destLen; + SizeT srcLenOrig = *srcLen; + *destLen = 0; + *srcLen = 0; + *status = CODER_STATUS_NOT_SPECIFIED; + + for (;;) + { + SizeT srcRem; + + if (p->state == XZ_STATE_BLOCK) + { + SizeT destLen2 = destLenOrig - *destLen; + SizeT srcLen2 = srcLenOrig - *srcLen; + SRes res; + + ECoderFinishMode finishMode2 = finishMode; + BoolInt srcFinished2 = srcFinished; + BoolInt destFinish = False; + + if (p->block.packSize != (UInt64)(Int64)-1) + { + UInt64 rem = p->block.packSize - p->packSize; + if (srcLen2 >= rem) + { + srcFinished2 = True; + srcLen2 = (SizeT)rem; + } + if (rem == 0 && p->block.unpackSize == p->unpackSize) + return SZ_ERROR_DATA; + } + + if (p->block.unpackSize != (UInt64)(Int64)-1) + { + UInt64 rem = p->block.unpackSize - p->unpackSize; + if (destLen2 >= rem) + { + destFinish = True; + finishMode2 = CODER_FINISH_END; + destLen2 = (SizeT)rem; + } + } + + /* + if (srcLen2 == 0 && destLen2 == 0) + { + *status = CODER_STATUS_NOT_FINISHED; + return SZ_OK; + } + */ + + { + res = MixCoder_Code(&p->decoder, + (p->outBuf ? NULL : dest), &destLen2, destFinish, + src, &srcLen2, srcFinished2, + finishMode2); + + *status = p->decoder.status; + XzCheck_Update(&p->check, (p->outBuf ? p->outBuf + p->outDataWritten : dest), destLen2); + if (!p->outBuf) + dest += destLen2; + p->outDataWritten += destLen2; + } + + (*srcLen) += srcLen2; + src += srcLen2; + p->packSize += srcLen2; + (*destLen) += destLen2; + p->unpackSize += destLen2; + + RINOK(res) + + if (*status != CODER_STATUS_FINISHED_WITH_MARK) + { + if (p->block.packSize == p->packSize + && *status == CODER_STATUS_NEEDS_MORE_INPUT) + { + PRF_STR("CODER_STATUS_NEEDS_MORE_INPUT") + *status = CODER_STATUS_NOT_SPECIFIED; + return SZ_ERROR_DATA; + } + + return SZ_OK; + } + { + XzUnpacker_UpdateIndex(p, XzUnpacker_GetPackSizeForIndex(p), p->unpackSize); + p->state = XZ_STATE_BLOCK_FOOTER; + p->pos = 0; + p->alignPos = 0; + *status = CODER_STATUS_NOT_SPECIFIED; + + if ((p->block.packSize != (UInt64)(Int64)-1 && p->block.packSize != p->packSize) + || (p->block.unpackSize != (UInt64)(Int64)-1 && p->block.unpackSize != p->unpackSize)) + { + PRF_STR("ERROR: block.size mismatch") + return SZ_ERROR_DATA; + } + } + // continue; + } + + srcRem = srcLenOrig - *srcLen; + + // XZ_STATE_BLOCK_FOOTER can transit to XZ_STATE_BLOCK_HEADER without input bytes + if (srcRem == 0 && p->state != XZ_STATE_BLOCK_FOOTER) + { + *status = CODER_STATUS_NEEDS_MORE_INPUT; + return SZ_OK; + } + + switch (p->state) + { + case XZ_STATE_STREAM_HEADER: + { + if (p->pos < XZ_STREAM_HEADER_SIZE) + { + if (p->pos < XZ_SIG_SIZE && *src != XZ_SIG[p->pos]) + return SZ_ERROR_NO_ARCHIVE; + if (p->decodeToStreamSignature) + return SZ_OK; + p->buf[p->pos++] = *src++; + (*srcLen)++; + } + else + { + RINOK(Xz_ParseHeader(&p->streamFlags, p->buf)) + p->numStartedStreams++; + p->indexSize = 0; + p->numBlocks = 0; + Sha256_Init(&p->sha); + p->state = XZ_STATE_BLOCK_HEADER; + p->pos = 0; + } + break; + } + + case XZ_STATE_BLOCK_HEADER: + { + if (p->pos == 0) + { + p->buf[p->pos++] = *src++; + (*srcLen)++; + if (p->buf[0] == 0) + { + if (p->decodeOnlyOneBlock) + return SZ_ERROR_DATA; + p->indexPreSize = 1 + Xz_WriteVarInt(p->buf + 1, p->numBlocks); + p->indexPos = p->indexPreSize; + p->indexSize += p->indexPreSize; + Sha256_Final(&p->sha, p->shaDigest); + Sha256_Init(&p->sha); + p->crc = CrcUpdate(CRC_INIT_VAL, p->buf, p->indexPreSize); + p->state = XZ_STATE_STREAM_INDEX; + break; + } + p->blockHeaderSize = ((UInt32)p->buf[0] << 2) + 4; + break; + } + + if (p->pos != p->blockHeaderSize) + { + UInt32 cur = p->blockHeaderSize - p->pos; + if (cur > srcRem) + cur = (UInt32)srcRem; + memcpy(p->buf + p->pos, src, cur); + p->pos += cur; + (*srcLen) += cur; + src += cur; + } + else + { + RINOK(XzBlock_Parse(&p->block, p->buf)) + if (!XzBlock_AreSupportedFilters(&p->block)) + return SZ_ERROR_UNSUPPORTED; + p->numTotalBlocks++; + p->state = XZ_STATE_BLOCK; + p->packSize = 0; + p->unpackSize = 0; + XzCheck_Init(&p->check, XzFlags_GetCheckType(p->streamFlags)); + if (p->parseMode) + { + p->headerParsedOk = True; + return SZ_OK; + } + RINOK(XzDecMix_Init(&p->decoder, &p->block, p->outBuf, p->outBufSize)) + } + break; + } + + case XZ_STATE_BLOCK_FOOTER: + { + if ((((unsigned)p->packSize + p->alignPos) & 3) != 0) + { + if (srcRem == 0) + { + *status = CODER_STATUS_NEEDS_MORE_INPUT; + return SZ_OK; + } + (*srcLen)++; + p->alignPos++; + if (*src++ != 0) + return SZ_ERROR_CRC; + } + else + { + UInt32 checkSize = XzFlags_GetCheckSize(p->streamFlags); + UInt32 cur = checkSize - p->pos; + if (cur != 0) + { + if (srcRem == 0) + { + *status = CODER_STATUS_NEEDS_MORE_INPUT; + return SZ_OK; + } + if (cur > srcRem) + cur = (UInt32)srcRem; + memcpy(p->buf + p->pos, src, cur); + p->pos += cur; + (*srcLen) += cur; + src += cur; + if (checkSize != p->pos) + break; + } + { + Byte digest[XZ_CHECK_SIZE_MAX]; + p->state = XZ_STATE_BLOCK_HEADER; + p->pos = 0; + if (XzCheck_Final(&p->check, digest) && memcmp(digest, p->buf, checkSize) != 0) + return SZ_ERROR_CRC; + if (p->decodeOnlyOneBlock) + { + *status = CODER_STATUS_FINISHED_WITH_MARK; + return SZ_OK; + } + } + } + break; + } + + case XZ_STATE_STREAM_INDEX: + { + if (p->pos < p->indexPreSize) + { + (*srcLen)++; + if (*src++ != p->buf[p->pos++]) + return SZ_ERROR_CRC; + } + else + { + if (p->indexPos < p->indexSize) + { + UInt64 cur = p->indexSize - p->indexPos; + if (srcRem > cur) + srcRem = (SizeT)cur; + p->crc = CrcUpdate(p->crc, src, srcRem); + Sha256_Update(&p->sha, src, srcRem); + (*srcLen) += srcRem; + src += srcRem; + p->indexPos += srcRem; + } + else if ((p->indexPos & 3) != 0) + { + Byte b = *src++; + p->crc = CRC_UPDATE_BYTE(p->crc, b); + (*srcLen)++; + p->indexPos++; + p->indexSize++; + if (b != 0) + return SZ_ERROR_CRC; + } + else + { + Byte digest[SHA256_DIGEST_SIZE]; + p->state = XZ_STATE_STREAM_INDEX_CRC; + p->indexSize += 4; + p->pos = 0; + Sha256_Final(&p->sha, digest); + if (memcmp(digest, p->shaDigest, SHA256_DIGEST_SIZE) != 0) + return SZ_ERROR_CRC; + } + } + break; + } + + case XZ_STATE_STREAM_INDEX_CRC: + { + if (p->pos < 4) + { + (*srcLen)++; + p->buf[p->pos++] = *src++; + } + else + { + const Byte *ptr = p->buf; + p->state = XZ_STATE_STREAM_FOOTER; + p->pos = 0; + if (CRC_GET_DIGEST(p->crc) != GetUi32(ptr)) + return SZ_ERROR_CRC; + } + break; + } + + case XZ_STATE_STREAM_FOOTER: + { + UInt32 cur = XZ_STREAM_FOOTER_SIZE - p->pos; + if (cur > srcRem) + cur = (UInt32)srcRem; + memcpy(p->buf + p->pos, src, cur); + p->pos += cur; + (*srcLen) += cur; + src += cur; + if (p->pos == XZ_STREAM_FOOTER_SIZE) + { + p->state = XZ_STATE_STREAM_PADDING; + p->numFinishedStreams++; + p->padSize = 0; + if (!Xz_CheckFooter(p->streamFlags, p->indexSize, p->buf)) + return SZ_ERROR_CRC; + } + break; + } + + case XZ_STATE_STREAM_PADDING: + { + if (*src != 0) + { + if (((UInt32)p->padSize & 3) != 0) + return SZ_ERROR_NO_ARCHIVE; + p->pos = 0; + p->state = XZ_STATE_STREAM_HEADER; + } + else + { + (*srcLen)++; + src++; + p->padSize++; + } + break; + } + + case XZ_STATE_BLOCK: break; /* to disable GCC warning */ + } + } + /* + if (p->state == XZ_STATE_FINISHED) + *status = CODER_STATUS_FINISHED_WITH_MARK; + return SZ_OK; + */ +} + + +SRes XzUnpacker_CodeFull(CXzUnpacker *p, Byte *dest, SizeT *destLen, + const Byte *src, SizeT *srcLen, + ECoderFinishMode finishMode, ECoderStatus *status) +{ + XzUnpacker_Init(p); + XzUnpacker_SetOutBuf(p, dest, *destLen); + + return XzUnpacker_Code(p, + NULL, destLen, + src, srcLen, True, + finishMode, status); +} + + +BoolInt XzUnpacker_IsBlockFinished(const CXzUnpacker *p) +{ + return (p->state == XZ_STATE_BLOCK_HEADER) && (p->pos == 0); +} + +BoolInt XzUnpacker_IsStreamWasFinished(const CXzUnpacker *p) +{ + return (p->state == XZ_STATE_STREAM_PADDING) && (((UInt32)p->padSize & 3) == 0); +} + +UInt64 XzUnpacker_GetExtraSize(const CXzUnpacker *p) +{ + UInt64 num = 0; + if (p->state == XZ_STATE_STREAM_PADDING) + num = p->padSize; + else if (p->state == XZ_STATE_STREAM_HEADER) + num = p->padSize + p->pos; + return num; +} + + + + + + + + + + + + + + + + + + + + + +#ifndef Z7_ST +#include "MtDec.h" +#endif + + +void XzDecMtProps_Init(CXzDecMtProps *p) +{ + p->inBufSize_ST = 1 << 18; + p->outStep_ST = 1 << 20; + p->ignoreErrors = False; + + #ifndef Z7_ST + p->numThreads = 1; + p->inBufSize_MT = 1 << 18; + p->memUseMax = sizeof(size_t) << 28; + #endif +} + + + +#ifndef Z7_ST + +/* ---------- CXzDecMtThread ---------- */ + +typedef struct +{ + Byte *outBuf; + size_t outBufSize; + size_t outPreSize; + size_t inPreSize; + size_t inPreHeaderSize; + size_t blockPackSize_for_Index; // including block header and checksum. + size_t blockPackTotal; // including stream header, block header and checksum. + size_t inCodeSize; + size_t outCodeSize; + ECoderStatus status; + SRes codeRes; + BoolInt skipMode; + // BoolInt finishedWithMark; + EMtDecParseState parseState; + BoolInt parsing_Truncated; + BoolInt atBlockHeader; + CXzStreamFlags streamFlags; + // UInt64 numFinishedStreams + UInt64 numStreams; + UInt64 numTotalBlocks; + UInt64 numBlocks; + + BoolInt dec_created; + CXzUnpacker dec; + + Byte mtPad[1 << 7]; +} CXzDecMtThread; + +#endif + + +/* ---------- CXzDecMt ---------- */ + +struct CXzDecMt +{ + CAlignOffsetAlloc alignOffsetAlloc; + ISzAllocPtr allocMid; + + CXzDecMtProps props; + size_t unpackBlockMaxSize; + + ISeqInStreamPtr inStream; + ISeqOutStreamPtr outStream; + ICompressProgressPtr progress; + + BoolInt finishMode; + BoolInt outSize_Defined; + UInt64 outSize; + + UInt64 outProcessed; + UInt64 inProcessed; + UInt64 readProcessed; + BoolInt readWasFinished; + SRes readRes; + SRes writeRes; + + Byte *outBuf; + size_t outBufSize; + Byte *inBuf; + size_t inBufSize; + + CXzUnpacker dec; + + ECoderStatus status; + SRes codeRes; + + #ifndef Z7_ST + BoolInt mainDecoderWasCalled; + // int statErrorDefined; + int finishedDecoderIndex; + + // global values that are used in Parse stage + CXzStreamFlags streamFlags; + // UInt64 numFinishedStreams + UInt64 numStreams; + UInt64 numTotalBlocks; + UInt64 numBlocks; + + // UInt64 numBadBlocks; + SRes mainErrorCode; // it's set to error code, if the size Code() output doesn't patch the size from Parsing stage + // it can be = SZ_ERROR_INPUT_EOF + // it can be = SZ_ERROR_DATA, in some another cases + BoolInt isBlockHeaderState_Parse; + BoolInt isBlockHeaderState_Write; + UInt64 outProcessed_Parse; + BoolInt parsing_Truncated; + + BoolInt mtc_WasConstructed; + CMtDec mtc; + CXzDecMtThread coders[MTDEC_THREADS_MAX]; + #endif +}; + + + +CXzDecMtHandle XzDecMt_Create(ISzAllocPtr alloc, ISzAllocPtr allocMid) +{ + CXzDecMt *p = (CXzDecMt *)ISzAlloc_Alloc(alloc, sizeof(CXzDecMt)); + if (!p) + return NULL; + + AlignOffsetAlloc_CreateVTable(&p->alignOffsetAlloc); + p->alignOffsetAlloc.baseAlloc = alloc; + p->alignOffsetAlloc.numAlignBits = 7; + p->alignOffsetAlloc.offset = 0; + + p->allocMid = allocMid; + + p->outBuf = NULL; + p->outBufSize = 0; + p->inBuf = NULL; + p->inBufSize = 0; + + XzUnpacker_Construct(&p->dec, &p->alignOffsetAlloc.vt); + + p->unpackBlockMaxSize = 0; + + XzDecMtProps_Init(&p->props); + + #ifndef Z7_ST + p->mtc_WasConstructed = False; + { + unsigned i; + for (i = 0; i < MTDEC_THREADS_MAX; i++) + { + CXzDecMtThread *coder = &p->coders[i]; + coder->dec_created = False; + coder->outBuf = NULL; + coder->outBufSize = 0; + } + } + #endif + + return (CXzDecMtHandle)p; +} + + +#ifndef Z7_ST + +static void XzDecMt_FreeOutBufs(CXzDecMt *p) +{ + unsigned i; + for (i = 0; i < MTDEC_THREADS_MAX; i++) + { + CXzDecMtThread *coder = &p->coders[i]; + if (coder->outBuf) + { + ISzAlloc_Free(p->allocMid, coder->outBuf); + coder->outBuf = NULL; + coder->outBufSize = 0; + } + } + p->unpackBlockMaxSize = 0; +} + +#endif + + + +static void XzDecMt_FreeSt(CXzDecMt *p) +{ + XzUnpacker_Free(&p->dec); + + if (p->outBuf) + { + ISzAlloc_Free(p->allocMid, p->outBuf); + p->outBuf = NULL; + } + p->outBufSize = 0; + + if (p->inBuf) + { + ISzAlloc_Free(p->allocMid, p->inBuf); + p->inBuf = NULL; + } + p->inBufSize = 0; +} + + +// #define GET_CXzDecMt_p CXzDecMt *p = pp; + +void XzDecMt_Destroy(CXzDecMtHandle p) +{ + // GET_CXzDecMt_p + + XzDecMt_FreeSt(p); + + #ifndef Z7_ST + + if (p->mtc_WasConstructed) + { + MtDec_Destruct(&p->mtc); + p->mtc_WasConstructed = False; + } + { + unsigned i; + for (i = 0; i < MTDEC_THREADS_MAX; i++) + { + CXzDecMtThread *t = &p->coders[i]; + if (t->dec_created) + { + // we don't need to free dict here + XzUnpacker_Free(&t->dec); + t->dec_created = False; + } + } + } + XzDecMt_FreeOutBufs(p); + + #endif + + ISzAlloc_Free(p->alignOffsetAlloc.baseAlloc, p); +} + + + +#ifndef Z7_ST + +static void XzDecMt_Callback_Parse(void *obj, unsigned coderIndex, CMtDecCallbackInfo *cc) +{ + CXzDecMt *me = (CXzDecMt *)obj; + CXzDecMtThread *coder = &me->coders[coderIndex]; + size_t srcSize = cc->srcSize; + + cc->srcSize = 0; + cc->outPos = 0; + cc->state = MTDEC_PARSE_CONTINUE; + + cc->canCreateNewThread = True; + + if (cc->startCall) + { + coder->outPreSize = 0; + coder->inPreSize = 0; + coder->inPreHeaderSize = 0; + coder->parseState = MTDEC_PARSE_CONTINUE; + coder->parsing_Truncated = False; + coder->skipMode = False; + coder->codeRes = SZ_OK; + coder->status = CODER_STATUS_NOT_SPECIFIED; + coder->inCodeSize = 0; + coder->outCodeSize = 0; + + coder->numStreams = me->numStreams; + coder->numTotalBlocks = me->numTotalBlocks; + coder->numBlocks = me->numBlocks; + + if (!coder->dec_created) + { + XzUnpacker_Construct(&coder->dec, &me->alignOffsetAlloc.vt); + coder->dec_created = True; + } + + XzUnpacker_Init(&coder->dec); + + if (me->isBlockHeaderState_Parse) + { + coder->dec.streamFlags = me->streamFlags; + coder->atBlockHeader = True; + XzUnpacker_PrepareToRandomBlockDecoding(&coder->dec); + } + else + { + coder->atBlockHeader = False; + me->isBlockHeaderState_Parse = True; + } + + coder->dec.numStartedStreams = me->numStreams; + coder->dec.numTotalBlocks = me->numTotalBlocks; + coder->dec.numBlocks = me->numBlocks; + } + + while (!coder->skipMode) + { + ECoderStatus status; + SRes res; + size_t srcSize2 = srcSize; + size_t destSize = (size_t)0 - 1; + + coder->dec.parseMode = True; + coder->dec.headerParsedOk = False; + + PRF_STR_INT("Parse", srcSize2) + + res = XzUnpacker_Code(&coder->dec, + NULL, &destSize, + cc->src, &srcSize2, cc->srcFinished, + CODER_FINISH_END, &status); + + // PRF(printf(" res = %d, srcSize2 = %d", res, (unsigned)srcSize2)); + + coder->codeRes = res; + coder->status = status; + cc->srcSize += srcSize2; + srcSize -= srcSize2; + coder->inPreHeaderSize += srcSize2; + coder->inPreSize = coder->inPreHeaderSize; + + if (res != SZ_OK) + { + cc->state = + coder->parseState = MTDEC_PARSE_END; + /* + if (res == SZ_ERROR_MEM) + return res; + return SZ_OK; + */ + return; // res; + } + + if (coder->dec.headerParsedOk) + { + const CXzBlock *block = &coder->dec.block; + if (XzBlock_HasUnpackSize(block) + // && block->unpackSize <= me->props.outBlockMax + && XzBlock_HasPackSize(block)) + { + { + if (block->unpackSize * 2 * me->mtc.numStartedThreads > me->props.memUseMax) + { + cc->state = MTDEC_PARSE_OVERFLOW; + return; // SZ_OK; + } + } + { + UInt64 packSize = block->packSize; + UInt64 packSizeAligned = packSize + ((0 - (unsigned)packSize) & 3); + UInt32 checkSize = XzFlags_GetCheckSize(coder->dec.streamFlags); + UInt64 blockPackSum = coder->inPreSize + packSizeAligned + checkSize; + // if (blockPackSum <= me->props.inBlockMax) + // unpackBlockMaxSize + { + coder->blockPackSize_for_Index = (size_t)(coder->dec.blockHeaderSize + packSize + checkSize); + coder->blockPackTotal = (size_t)blockPackSum; + coder->outPreSize = (size_t)block->unpackSize; + coder->streamFlags = coder->dec.streamFlags; + me->streamFlags = coder->dec.streamFlags; + coder->skipMode = True; + break; + } + } + } + } + else + // if (coder->inPreSize <= me->props.inBlockMax) + { + if (!cc->srcFinished) + return; // SZ_OK; + cc->state = + coder->parseState = MTDEC_PARSE_END; + return; // SZ_OK; + } + cc->state = MTDEC_PARSE_OVERFLOW; + return; // SZ_OK; + } + + // ---------- skipMode ---------- + { + UInt64 rem = coder->blockPackTotal - coder->inPreSize; + size_t cur = srcSize; + if (cur > rem) + cur = (size_t)rem; + cc->srcSize += cur; + coder->inPreSize += cur; + srcSize -= cur; + + if (coder->inPreSize == coder->blockPackTotal) + { + if (srcSize == 0) + { + if (!cc->srcFinished) + return; // SZ_OK; + cc->state = MTDEC_PARSE_END; + } + else if ((cc->src)[cc->srcSize] == 0) // we check control byte of next block + cc->state = MTDEC_PARSE_END; + else + { + cc->state = MTDEC_PARSE_NEW; + + { + size_t blockMax = me->unpackBlockMaxSize; + if (blockMax < coder->outPreSize) + blockMax = coder->outPreSize; + { + UInt64 required = (UInt64)blockMax * (me->mtc.numStartedThreads + 1) * 2; + if (me->props.memUseMax < required) + cc->canCreateNewThread = False; + } + } + + if (me->outSize_Defined) + { + // next block can be zero size + const UInt64 rem2 = me->outSize - me->outProcessed_Parse; + if (rem2 < coder->outPreSize) + { + coder->parsing_Truncated = True; + cc->state = MTDEC_PARSE_END; + } + me->outProcessed_Parse += coder->outPreSize; + } + } + } + else if (cc->srcFinished) + cc->state = MTDEC_PARSE_END; + else + return; // SZ_OK; + + coder->parseState = cc->state; + cc->outPos = coder->outPreSize; + + me->numStreams = coder->dec.numStartedStreams; + me->numTotalBlocks = coder->dec.numTotalBlocks; + me->numBlocks = coder->dec.numBlocks + 1; + return; // SZ_OK; + } +} + + +static SRes XzDecMt_Callback_PreCode(void *pp, unsigned coderIndex) +{ + CXzDecMt *me = (CXzDecMt *)pp; + CXzDecMtThread *coder = &me->coders[coderIndex]; + Byte *dest; + + if (!coder->dec.headerParsedOk) + return SZ_OK; + + dest = coder->outBuf; + + if (!dest || coder->outBufSize < coder->outPreSize) + { + if (dest) + { + ISzAlloc_Free(me->allocMid, dest); + coder->outBuf = NULL; + coder->outBufSize = 0; + } + { + size_t outPreSize = coder->outPreSize; + if (outPreSize == 0) + outPreSize = 1; + dest = (Byte *)ISzAlloc_Alloc(me->allocMid, outPreSize); + } + if (!dest) + return SZ_ERROR_MEM; + coder->outBuf = dest; + coder->outBufSize = coder->outPreSize; + + if (coder->outBufSize > me->unpackBlockMaxSize) + me->unpackBlockMaxSize = coder->outBufSize; + } + + // return SZ_ERROR_MEM; + + XzUnpacker_SetOutBuf(&coder->dec, coder->outBuf, coder->outBufSize); + + { + SRes res = XzDecMix_Init(&coder->dec.decoder, &coder->dec.block, coder->outBuf, coder->outBufSize); + // res = SZ_ERROR_UNSUPPORTED; // to test + coder->codeRes = res; + if (res != SZ_OK) + { + // if (res == SZ_ERROR_MEM) return res; + if (me->props.ignoreErrors && res != SZ_ERROR_MEM) + return SZ_OK; + return res; + } + } + + return SZ_OK; +} + + +static SRes XzDecMt_Callback_Code(void *pp, unsigned coderIndex, + const Byte *src, size_t srcSize, int srcFinished, + // int finished, int blockFinished, + UInt64 *inCodePos, UInt64 *outCodePos, int *stop) +{ + CXzDecMt *me = (CXzDecMt *)pp; + CXzDecMtThread *coder = &me->coders[coderIndex]; + + *inCodePos = coder->inCodeSize; + *outCodePos = coder->outCodeSize; + *stop = True; + + if (srcSize > coder->inPreSize - coder->inCodeSize) + return SZ_ERROR_FAIL; + + if (coder->inCodeSize < coder->inPreHeaderSize) + { + size_t step = coder->inPreHeaderSize - coder->inCodeSize; + if (step > srcSize) + step = srcSize; + src += step; + srcSize -= step; + coder->inCodeSize += step; + *inCodePos = coder->inCodeSize; + if (coder->inCodeSize < coder->inPreHeaderSize) + { + *stop = False; + return SZ_OK; + } + } + + if (!coder->dec.headerParsedOk) + return SZ_OK; + if (!coder->outBuf) + return SZ_OK; + + if (coder->codeRes == SZ_OK) + { + ECoderStatus status; + SRes res; + size_t srcProcessed = srcSize; + size_t outSizeCur = coder->outPreSize - coder->dec.outDataWritten; + + // PRF(printf("\nCallback_Code: Code %d %d\n", (unsigned)srcSize, (unsigned)outSizeCur)); + + res = XzUnpacker_Code(&coder->dec, + NULL, &outSizeCur, + src, &srcProcessed, srcFinished, + // coder->finishedWithMark ? CODER_FINISH_END : CODER_FINISH_ANY, + CODER_FINISH_END, + &status); + + // PRF(printf(" res = %d, srcSize2 = %d, outSizeCur = %d", res, (unsigned)srcProcessed, (unsigned)outSizeCur)); + + coder->codeRes = res; + coder->status = status; + coder->inCodeSize += srcProcessed; + coder->outCodeSize = coder->dec.outDataWritten; + *inCodePos = coder->inCodeSize; + *outCodePos = coder->outCodeSize; + + if (res == SZ_OK) + { + if (srcProcessed == srcSize) + *stop = False; + return SZ_OK; + } + } + + if (me->props.ignoreErrors && coder->codeRes != SZ_ERROR_MEM) + { + *inCodePos = coder->inPreSize; + *outCodePos = coder->outPreSize; + return SZ_OK; + } + return coder->codeRes; +} + + +#define XZDECMT_STREAM_WRITE_STEP (1 << 24) + +static SRes XzDecMt_Callback_Write(void *pp, unsigned coderIndex, + BoolInt needWriteToStream, + const Byte *src, size_t srcSize, BoolInt isCross, + // int srcFinished, + BoolInt *needContinue, + BoolInt *canRecode) +{ + CXzDecMt *me = (CXzDecMt *)pp; + const CXzDecMtThread *coder = &me->coders[coderIndex]; + + // PRF(printf("\nWrite processed = %d srcSize = %d\n", (unsigned)me->mtc.inProcessed, (unsigned)srcSize)); + + *needContinue = False; + *canRecode = True; + + if (!needWriteToStream) + return SZ_OK; + + if (!coder->dec.headerParsedOk || !coder->outBuf) + { + if (me->finishedDecoderIndex < 0) + me->finishedDecoderIndex = (int)coderIndex; + return SZ_OK; + } + + if (me->finishedDecoderIndex >= 0) + return SZ_OK; + + me->mtc.inProcessed += coder->inCodeSize; + + *canRecode = False; + + { + SRes res; + size_t size = coder->outCodeSize; + Byte *data = coder->outBuf; + + // we use in me->dec: sha, numBlocks, indexSize + + if (!me->isBlockHeaderState_Write) + { + XzUnpacker_PrepareToRandomBlockDecoding(&me->dec); + me->dec.decodeOnlyOneBlock = False; + me->dec.numStartedStreams = coder->dec.numStartedStreams; + me->dec.streamFlags = coder->streamFlags; + + me->isBlockHeaderState_Write = True; + } + + me->dec.numTotalBlocks = coder->dec.numTotalBlocks; + XzUnpacker_UpdateIndex(&me->dec, coder->blockPackSize_for_Index, coder->outPreSize); + + if (coder->outPreSize != size) + { + if (me->props.ignoreErrors) + { + memset(data + size, 0, coder->outPreSize - size); + size = coder->outPreSize; + } + // me->numBadBlocks++; + if (me->mainErrorCode == SZ_OK) + { + if ((int)coder->status == LZMA_STATUS_NEEDS_MORE_INPUT) + me->mainErrorCode = SZ_ERROR_INPUT_EOF; + else + me->mainErrorCode = SZ_ERROR_DATA; + } + } + + if (me->writeRes != SZ_OK) + return me->writeRes; + + res = SZ_OK; + { + if (me->outSize_Defined) + { + const UInt64 rem = me->outSize - me->outProcessed; + if (size > rem) + size = (SizeT)rem; + } + + for (;;) + { + size_t cur = size; + size_t written; + if (cur > XZDECMT_STREAM_WRITE_STEP) + cur = XZDECMT_STREAM_WRITE_STEP; + + written = ISeqOutStream_Write(me->outStream, data, cur); + + // PRF(printf("\nWritten ask = %d written = %d\n", (unsigned)cur, (unsigned)written)); + + me->outProcessed += written; + if (written != cur) + { + me->writeRes = SZ_ERROR_WRITE; + res = me->writeRes; + break; + } + data += cur; + size -= cur; + // PRF_STR_INT("Written size =", size) + if (size == 0) + break; + res = MtProgress_ProgressAdd(&me->mtc.mtProgress, 0, 0); + if (res != SZ_OK) + break; + } + } + + if (coder->codeRes != SZ_OK) + if (!me->props.ignoreErrors) + { + me->finishedDecoderIndex = (int)coderIndex; + return res; + } + + RINOK(res) + + if (coder->inPreSize != coder->inCodeSize + || coder->blockPackTotal != coder->inCodeSize) + { + me->finishedDecoderIndex = (int)coderIndex; + return SZ_OK; + } + + if (coder->parseState != MTDEC_PARSE_END) + { + *needContinue = True; + return SZ_OK; + } + } + + // (coder->state == MTDEC_PARSE_END) means that there are no other working threads + // so we can use mtc variables without lock + + PRF_STR_INT("Write MTDEC_PARSE_END", me->mtc.inProcessed) + + me->mtc.mtProgress.totalInSize = me->mtc.inProcessed; + { + CXzUnpacker *dec = &me->dec; + + PRF_STR_INT("PostSingle", srcSize) + + { + size_t srcProcessed = srcSize; + ECoderStatus status; + size_t outSizeCur = 0; + SRes res; + + // dec->decodeOnlyOneBlock = False; + dec->decodeToStreamSignature = True; + + me->mainDecoderWasCalled = True; + + if (coder->parsing_Truncated) + { + me->parsing_Truncated = True; + return SZ_OK; + } + + /* + We have processed all xz-blocks of stream, + And xz unpacker is at XZ_STATE_BLOCK_HEADER state, where + (src) is a pointer to xz-Index structure. + We finish reading of current xz-Stream, including Zero padding after xz-Stream. + We exit, if we reach extra byte (first byte of new-Stream or another data). + But we don't update input stream pointer for that new extra byte. + If extra byte is not correct first byte of xz-signature, + we have SZ_ERROR_NO_ARCHIVE error here. + */ + + res = XzUnpacker_Code(dec, + NULL, &outSizeCur, + src, &srcProcessed, + me->mtc.readWasFinished, // srcFinished + CODER_FINISH_END, // CODER_FINISH_ANY, + &status); + + // res = SZ_ERROR_ARCHIVE; // for failure test + + me->status = status; + me->codeRes = res; + + if (isCross) + me->mtc.crossStart += srcProcessed; + + me->mtc.inProcessed += srcProcessed; + me->mtc.mtProgress.totalInSize = me->mtc.inProcessed; + + srcSize -= srcProcessed; + src += srcProcessed; + + if (res != SZ_OK) + { + return SZ_OK; + // return res; + } + + if (dec->state == XZ_STATE_STREAM_HEADER) + { + *needContinue = True; + me->isBlockHeaderState_Parse = False; + me->isBlockHeaderState_Write = False; + + if (!isCross) + { + Byte *crossBuf = MtDec_GetCrossBuff(&me->mtc); + if (!crossBuf) + return SZ_ERROR_MEM; + if (srcSize != 0) + memcpy(crossBuf, src, srcSize); + me->mtc.crossStart = 0; + me->mtc.crossEnd = srcSize; + } + + PRF_STR_INT("XZ_STATE_STREAM_HEADER crossEnd = ", (unsigned)me->mtc.crossEnd) + + return SZ_OK; + } + + if (status != CODER_STATUS_NEEDS_MORE_INPUT || srcSize != 0) + { + return SZ_ERROR_FAIL; + } + + if (me->mtc.readWasFinished) + { + return SZ_OK; + } + } + + { + size_t inPos; + size_t inLim; + // const Byte *inData; + UInt64 inProgressPrev = me->mtc.inProcessed; + + // XzDecMt_Prepare_InBuf_ST(p); + Byte *crossBuf = MtDec_GetCrossBuff(&me->mtc); + if (!crossBuf) + return SZ_ERROR_MEM; + + inPos = 0; + inLim = 0; + + // inData = crossBuf; + + for (;;) + { + SizeT inProcessed; + SizeT outProcessed; + ECoderStatus status; + SRes res; + + if (inPos == inLim) + { + if (!me->mtc.readWasFinished) + { + inPos = 0; + inLim = me->mtc.inBufSize; + me->mtc.readRes = ISeqInStream_Read(me->inStream, (void *)crossBuf, &inLim); + me->mtc.readProcessed += inLim; + if (inLim == 0 || me->mtc.readRes != SZ_OK) + me->mtc.readWasFinished = True; + } + } + + inProcessed = inLim - inPos; + outProcessed = 0; + + res = XzUnpacker_Code(dec, + NULL, &outProcessed, + crossBuf + inPos, &inProcessed, + (inProcessed == 0), // srcFinished + CODER_FINISH_END, &status); + + me->codeRes = res; + me->status = status; + inPos += inProcessed; + me->mtc.inProcessed += inProcessed; + me->mtc.mtProgress.totalInSize = me->mtc.inProcessed; + + if (res != SZ_OK) + { + return SZ_OK; + // return res; + } + + if (dec->state == XZ_STATE_STREAM_HEADER) + { + *needContinue = True; + me->mtc.crossStart = inPos; + me->mtc.crossEnd = inLim; + me->isBlockHeaderState_Parse = False; + me->isBlockHeaderState_Write = False; + return SZ_OK; + } + + if (status != CODER_STATUS_NEEDS_MORE_INPUT) + return SZ_ERROR_FAIL; + + if (me->mtc.progress) + { + UInt64 inDelta = me->mtc.inProcessed - inProgressPrev; + if (inDelta >= (1 << 22)) + { + RINOK(MtProgress_Progress_ST(&me->mtc.mtProgress)) + inProgressPrev = me->mtc.inProcessed; + } + } + if (me->mtc.readWasFinished) + return SZ_OK; + } + } + } +} + + +#endif + + + +void XzStatInfo_Clear(CXzStatInfo *p) +{ + p->InSize = 0; + p->OutSize = 0; + + p->NumStreams = 0; + p->NumBlocks = 0; + + p->UnpackSize_Defined = False; + + p->NumStreams_Defined = False; + p->NumBlocks_Defined = False; + + p->DataAfterEnd = False; + p->DecodingTruncated = False; + + p->DecodeRes = SZ_OK; + p->ReadRes = SZ_OK; + p->ProgressRes = SZ_OK; + + p->CombinedRes = SZ_OK; + p->CombinedRes_Type = SZ_OK; +} + + + +/* + XzDecMt_Decode_ST() can return SZ_OK or the following errors + - SZ_ERROR_MEM for memory allocation error + - error from XzUnpacker_Code() function + - SZ_ERROR_WRITE for ISeqOutStream::Write(). stat->CombinedRes_Type = SZ_ERROR_WRITE in that case + - ICompressProgress::Progress() error, stat->CombinedRes_Type = SZ_ERROR_PROGRESS. + But XzDecMt_Decode_ST() doesn't return ISeqInStream::Read() errors. + ISeqInStream::Read() result is set to p->readRes. + also it can set stat->CombinedRes_Type to SZ_ERROR_WRITE or SZ_ERROR_PROGRESS. +*/ + +static SRes XzDecMt_Decode_ST(CXzDecMt *p + #ifndef Z7_ST + , BoolInt tMode + #endif + , CXzStatInfo *stat) +{ + size_t outPos; + size_t inPos, inLim; + const Byte *inData; + UInt64 inPrev, outPrev; + + CXzUnpacker *dec; + + #ifndef Z7_ST + if (tMode) + { + XzDecMt_FreeOutBufs(p); + tMode = MtDec_PrepareRead(&p->mtc); + } + #endif + + if (!p->outBuf || p->outBufSize != p->props.outStep_ST) + { + ISzAlloc_Free(p->allocMid, p->outBuf); + p->outBufSize = 0; + p->outBuf = (Byte *)ISzAlloc_Alloc(p->allocMid, p->props.outStep_ST); + if (!p->outBuf) + return SZ_ERROR_MEM; + p->outBufSize = p->props.outStep_ST; + } + + if (!p->inBuf || p->inBufSize != p->props.inBufSize_ST) + { + ISzAlloc_Free(p->allocMid, p->inBuf); + p->inBufSize = 0; + p->inBuf = (Byte *)ISzAlloc_Alloc(p->allocMid, p->props.inBufSize_ST); + if (!p->inBuf) + return SZ_ERROR_MEM; + p->inBufSize = p->props.inBufSize_ST; + } + + dec = &p->dec; + dec->decodeToStreamSignature = False; + // dec->decodeOnlyOneBlock = False; + + XzUnpacker_SetOutBuf(dec, NULL, 0); + + inPrev = p->inProcessed; + outPrev = p->outProcessed; + + inPos = 0; + inLim = 0; + inData = NULL; + outPos = 0; + + for (;;) + { + SizeT outSize; + BoolInt finished; + ECoderFinishMode finishMode; + SizeT inProcessed; + ECoderStatus status; + SRes res; + + SizeT outProcessed; + + + + if (inPos == inLim) + { + #ifndef Z7_ST + if (tMode) + { + inData = MtDec_Read(&p->mtc, &inLim); + inPos = 0; + if (inData) + continue; + tMode = False; + inLim = 0; + } + #endif + + if (!p->readWasFinished) + { + inPos = 0; + inLim = p->inBufSize; + inData = p->inBuf; + p->readRes = ISeqInStream_Read(p->inStream, (void *)p->inBuf, &inLim); + p->readProcessed += inLim; + if (inLim == 0 || p->readRes != SZ_OK) + p->readWasFinished = True; + } + } + + outSize = p->props.outStep_ST - outPos; + + finishMode = CODER_FINISH_ANY; + if (p->outSize_Defined) + { + const UInt64 rem = p->outSize - p->outProcessed; + if (outSize >= rem) + { + outSize = (SizeT)rem; + if (p->finishMode) + finishMode = CODER_FINISH_END; + } + } + + inProcessed = inLim - inPos; + outProcessed = outSize; + + res = XzUnpacker_Code(dec, p->outBuf + outPos, &outProcessed, + inData + inPos, &inProcessed, + (inPos == inLim), // srcFinished + finishMode, &status); + + p->codeRes = res; + p->status = status; + + inPos += inProcessed; + outPos += outProcessed; + p->inProcessed += inProcessed; + p->outProcessed += outProcessed; + + finished = ((inProcessed == 0 && outProcessed == 0) || res != SZ_OK); + + if (finished || outProcessed >= outSize) + if (outPos != 0) + { + const size_t written = ISeqOutStream_Write(p->outStream, p->outBuf, outPos); + // p->outProcessed += written; // 21.01: BUG fixed + if (written != outPos) + { + stat->CombinedRes_Type = SZ_ERROR_WRITE; + return SZ_ERROR_WRITE; + } + outPos = 0; + } + + if (p->progress && res == SZ_OK) + { + if (p->inProcessed - inPrev >= (1 << 22) || + p->outProcessed - outPrev >= (1 << 22)) + { + res = ICompressProgress_Progress(p->progress, p->inProcessed, p->outProcessed); + if (res != SZ_OK) + { + stat->CombinedRes_Type = SZ_ERROR_PROGRESS; + stat->ProgressRes = res; + return res; + } + inPrev = p->inProcessed; + outPrev = p->outProcessed; + } + } + + if (finished) + { + // p->codeRes is preliminary error from XzUnpacker_Code. + // and it can be corrected later as final result + // so we return SZ_OK here instead of (res); + return SZ_OK; + // return res; + } + } +} + + + +/* +XzStatInfo_SetStat() transforms + CXzUnpacker return code and status to combined CXzStatInfo results. + it can convert SZ_OK to SZ_ERROR_INPUT_EOF + it can convert SZ_ERROR_NO_ARCHIVE to SZ_OK and (DataAfterEnd = 1) +*/ + +static void XzStatInfo_SetStat(const CXzUnpacker *dec, + int finishMode, + // UInt64 readProcessed, + UInt64 inProcessed, + SRes res, // it's result from CXzUnpacker unpacker + ECoderStatus status, + BoolInt decodingTruncated, + CXzStatInfo *stat) +{ + UInt64 extraSize; + + stat->DecodingTruncated = (Byte)(decodingTruncated ? 1 : 0); + stat->InSize = inProcessed; + stat->NumStreams = dec->numStartedStreams; + stat->NumBlocks = dec->numTotalBlocks; + + stat->UnpackSize_Defined = True; + stat->NumStreams_Defined = True; + stat->NumBlocks_Defined = True; + + extraSize = XzUnpacker_GetExtraSize(dec); + + if (res == SZ_OK) + { + if (status == CODER_STATUS_NEEDS_MORE_INPUT) + { + // CODER_STATUS_NEEDS_MORE_INPUT is expected status for correct xz streams + // any extra data is part of correct data + extraSize = 0; + // if xz stream was not finished, then we need more data + if (!XzUnpacker_IsStreamWasFinished(dec)) + res = SZ_ERROR_INPUT_EOF; + } + else + { + // CODER_STATUS_FINISHED_WITH_MARK is not possible for multi stream xz decoding + // so he we have (status == CODER_STATUS_NOT_FINISHED) + // if (status != CODER_STATUS_FINISHED_WITH_MARK) + if (!decodingTruncated || finishMode) + res = SZ_ERROR_DATA; + } + } + else if (res == SZ_ERROR_NO_ARCHIVE) + { + /* + SZ_ERROR_NO_ARCHIVE is possible for 2 states: + XZ_STATE_STREAM_HEADER - if bad signature or bad CRC + XZ_STATE_STREAM_PADDING - if non-zero padding data + extraSize and inProcessed don't include "bad" byte + */ + // if (inProcessed == extraSize), there was no any good xz stream header, and we keep error + if (inProcessed != extraSize) // if there were good xz streams before error + { + // if (extraSize != 0 || readProcessed != inProcessed) + { + // he we suppose that all xz streams were finsihed OK, and we have + // some extra data after all streams + stat->DataAfterEnd = True; + res = SZ_OK; + } + } + } + + if (stat->DecodeRes == SZ_OK) + stat->DecodeRes = res; + + stat->InSize -= extraSize; +} + + + +SRes XzDecMt_Decode(CXzDecMtHandle p, + const CXzDecMtProps *props, + const UInt64 *outDataSize, int finishMode, + ISeqOutStreamPtr outStream, + // Byte *outBuf, size_t *outBufSize, + ISeqInStreamPtr inStream, + // const Byte *inData, size_t inDataSize, + CXzStatInfo *stat, + int *isMT, + ICompressProgressPtr progress) +{ + // GET_CXzDecMt_p + #ifndef Z7_ST + BoolInt tMode; + #endif + + XzStatInfo_Clear(stat); + + p->props = *props; + + p->inStream = inStream; + p->outStream = outStream; + p->progress = progress; + // p->stat = stat; + + p->outSize = 0; + p->outSize_Defined = False; + if (outDataSize) + { + p->outSize_Defined = True; + p->outSize = *outDataSize; + } + + p->finishMode = finishMode; + + // p->outSize = 457; p->outSize_Defined = True; p->finishMode = False; // for test + + p->writeRes = SZ_OK; + p->outProcessed = 0; + p->inProcessed = 0; + p->readProcessed = 0; + p->readWasFinished = False; + p->readRes = SZ_OK; + + p->codeRes = SZ_OK; + p->status = CODER_STATUS_NOT_SPECIFIED; + + XzUnpacker_Init(&p->dec); + + *isMT = False; + + /* + p->outBuf = NULL; + p->outBufSize = 0; + if (!outStream) + { + p->outBuf = outBuf; + p->outBufSize = *outBufSize; + *outBufSize = 0; + } + */ + + + #ifndef Z7_ST + + p->isBlockHeaderState_Parse = False; + p->isBlockHeaderState_Write = False; + // p->numBadBlocks = 0; + p->mainErrorCode = SZ_OK; + p->mainDecoderWasCalled = False; + + tMode = False; + + if (p->props.numThreads > 1) + { + IMtDecCallback2 vt; + BoolInt needContinue; + SRes res; + // we just free ST buffers here + // but we still keep state variables, that was set in XzUnpacker_Init() + XzDecMt_FreeSt(p); + + p->outProcessed_Parse = 0; + p->parsing_Truncated = False; + + p->numStreams = 0; + p->numTotalBlocks = 0; + p->numBlocks = 0; + p->finishedDecoderIndex = -1; + + if (!p->mtc_WasConstructed) + { + p->mtc_WasConstructed = True; + MtDec_Construct(&p->mtc); + } + + p->mtc.mtCallback = &vt; + p->mtc.mtCallbackObject = p; + + p->mtc.progress = progress; + p->mtc.inStream = inStream; + p->mtc.alloc = &p->alignOffsetAlloc.vt; + // p->mtc.inData = inData; + // p->mtc.inDataSize = inDataSize; + p->mtc.inBufSize = p->props.inBufSize_MT; + // p->mtc.inBlockMax = p->props.inBlockMax; + p->mtc.numThreadsMax = p->props.numThreads; + + *isMT = True; + + vt.Parse = XzDecMt_Callback_Parse; + vt.PreCode = XzDecMt_Callback_PreCode; + vt.Code = XzDecMt_Callback_Code; + vt.Write = XzDecMt_Callback_Write; + + + res = MtDec_Code(&p->mtc); + + + stat->InSize = p->mtc.inProcessed; + + p->inProcessed = p->mtc.inProcessed; + p->readRes = p->mtc.readRes; + p->readWasFinished = p->mtc.readWasFinished; + p->readProcessed = p->mtc.readProcessed; + + tMode = True; + needContinue = False; + + if (res == SZ_OK) + { + if (p->mtc.mtProgress.res != SZ_OK) + { + res = p->mtc.mtProgress.res; + stat->ProgressRes = res; + stat->CombinedRes_Type = SZ_ERROR_PROGRESS; + } + else + needContinue = p->mtc.needContinue; + } + + if (!needContinue) + { + { + SRes codeRes; + BoolInt truncated = False; + ECoderStatus status; + const CXzUnpacker *dec; + + stat->OutSize = p->outProcessed; + + if (p->finishedDecoderIndex >= 0) + { + const CXzDecMtThread *coder = &p->coders[(unsigned)p->finishedDecoderIndex]; + codeRes = coder->codeRes; + dec = &coder->dec; + status = coder->status; + } + else if (p->mainDecoderWasCalled) + { + codeRes = p->codeRes; + dec = &p->dec; + status = p->status; + truncated = p->parsing_Truncated; + } + else + return SZ_ERROR_FAIL; + + if (p->mainErrorCode != SZ_OK) + stat->DecodeRes = p->mainErrorCode; + + XzStatInfo_SetStat(dec, p->finishMode, + // p->mtc.readProcessed, + p->mtc.inProcessed, + codeRes, status, + truncated, + stat); + } + + if (res == SZ_OK) + { + stat->ReadRes = p->mtc.readRes; + + if (p->writeRes != SZ_OK) + { + res = p->writeRes; + stat->CombinedRes_Type = SZ_ERROR_WRITE; + } + else if (p->mtc.readRes != SZ_OK + // && p->mtc.inProcessed == p->mtc.readProcessed + && stat->DecodeRes == SZ_ERROR_INPUT_EOF) + { + res = p->mtc.readRes; + stat->CombinedRes_Type = SZ_ERROR_READ; + } + else if (stat->DecodeRes != SZ_OK) + res = stat->DecodeRes; + } + + stat->CombinedRes = res; + if (stat->CombinedRes_Type == SZ_OK) + stat->CombinedRes_Type = res; + return res; + } + + PRF_STR("----- decoding ST -----") + } + + #endif + + + *isMT = False; + + { + SRes res = XzDecMt_Decode_ST(p + #ifndef Z7_ST + , tMode + #endif + , stat + ); + + #ifndef Z7_ST + // we must set error code from MT decoding at first + if (p->mainErrorCode != SZ_OK) + stat->DecodeRes = p->mainErrorCode; + #endif + + XzStatInfo_SetStat(&p->dec, + p->finishMode, + // p->readProcessed, + p->inProcessed, + p->codeRes, p->status, + False, // truncated + stat); + + stat->ReadRes = p->readRes; + + if (res == SZ_OK) + { + if (p->readRes != SZ_OK + // && p->inProcessed == p->readProcessed + && stat->DecodeRes == SZ_ERROR_INPUT_EOF) + { + // we set read error as combined error, only if that error was the reason + // of decoding problem + res = p->readRes; + stat->CombinedRes_Type = SZ_ERROR_READ; + } + else if (stat->DecodeRes != SZ_OK) + res = stat->DecodeRes; + } + + stat->CombinedRes = res; + if (stat->CombinedRes_Type == SZ_OK) + stat->CombinedRes_Type = res; + return res; + } +} + +#undef PRF +#undef PRF_STR +#undef PRF_STR_INT_2 diff --git a/libraries/lzma/C/XzEnc.c b/libraries/lzma/C/XzEnc.c new file mode 100644 index 000000000..22408e261 --- /dev/null +++ b/libraries/lzma/C/XzEnc.c @@ -0,0 +1,1362 @@ +/* XzEnc.c -- Xz Encode +2023-04-13 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include +#include + +#include "7zCrc.h" +#include "Bra.h" +#include "CpuArch.h" + +#ifdef USE_SUBBLOCK +#include "Bcj3Enc.c" +#include "SbFind.c" +#include "SbEnc.c" +#endif + +#include "XzEnc.h" + +// #define Z7_ST + +#ifndef Z7_ST +#include "MtCoder.h" +#else +#define MTCODER_THREADS_MAX 1 +#define MTCODER_BLOCKS_MAX 1 +#endif + +#define XZ_GET_PAD_SIZE(dataSize) ((4 - ((unsigned)(dataSize) & 3)) & 3) + +/* max pack size for LZMA2 block + check-64bytrs: */ +#define XZ_GET_MAX_BLOCK_PACK_SIZE(unpackSize) ((unpackSize) + ((unpackSize) >> 10) + 16 + 64) + +#define XZ_GET_ESTIMATED_BLOCK_TOTAL_PACK_SIZE(unpackSize) (XZ_BLOCK_HEADER_SIZE_MAX + XZ_GET_MAX_BLOCK_PACK_SIZE(unpackSize)) + + +// #define XzBlock_ClearFlags(p) (p)->flags = 0; +#define XzBlock_ClearFlags_SetNumFilters(p, n) (p)->flags = (Byte)((n) - 1); +#define XzBlock_SetHasPackSize(p) (p)->flags |= XZ_BF_PACK_SIZE; +#define XzBlock_SetHasUnpackSize(p) (p)->flags |= XZ_BF_UNPACK_SIZE; + + +static SRes WriteBytes(ISeqOutStreamPtr s, const void *buf, size_t size) +{ + return (ISeqOutStream_Write(s, buf, size) == size) ? SZ_OK : SZ_ERROR_WRITE; +} + +static SRes WriteBytes_UpdateCrc(ISeqOutStreamPtr s, const void *buf, size_t size, UInt32 *crc) +{ + *crc = CrcUpdate(*crc, buf, size); + return WriteBytes(s, buf, size); +} + + +static SRes Xz_WriteHeader(CXzStreamFlags f, ISeqOutStreamPtr s) +{ + UInt32 crc; + Byte header[XZ_STREAM_HEADER_SIZE]; + memcpy(header, XZ_SIG, XZ_SIG_SIZE); + header[XZ_SIG_SIZE] = (Byte)(f >> 8); + header[XZ_SIG_SIZE + 1] = (Byte)(f & 0xFF); + crc = CrcCalc(header + XZ_SIG_SIZE, XZ_STREAM_FLAGS_SIZE); + SetUi32(header + XZ_SIG_SIZE + XZ_STREAM_FLAGS_SIZE, crc) + return WriteBytes(s, header, XZ_STREAM_HEADER_SIZE); +} + + +static SRes XzBlock_WriteHeader(const CXzBlock *p, ISeqOutStreamPtr s) +{ + Byte header[XZ_BLOCK_HEADER_SIZE_MAX]; + + unsigned pos = 1; + unsigned numFilters, i; + header[pos++] = p->flags; + + if (XzBlock_HasPackSize(p)) pos += Xz_WriteVarInt(header + pos, p->packSize); + if (XzBlock_HasUnpackSize(p)) pos += Xz_WriteVarInt(header + pos, p->unpackSize); + numFilters = XzBlock_GetNumFilters(p); + + for (i = 0; i < numFilters; i++) + { + const CXzFilter *f = &p->filters[i]; + pos += Xz_WriteVarInt(header + pos, f->id); + pos += Xz_WriteVarInt(header + pos, f->propsSize); + memcpy(header + pos, f->props, f->propsSize); + pos += f->propsSize; + } + + while ((pos & 3) != 0) + header[pos++] = 0; + + header[0] = (Byte)(pos >> 2); + SetUi32(header + pos, CrcCalc(header, pos)) + return WriteBytes(s, header, pos + 4); +} + + + + +typedef struct +{ + size_t numBlocks; + size_t size; + size_t allocated; + Byte *blocks; +} CXzEncIndex; + + +static void XzEncIndex_Construct(CXzEncIndex *p) +{ + p->numBlocks = 0; + p->size = 0; + p->allocated = 0; + p->blocks = NULL; +} + +static void XzEncIndex_Init(CXzEncIndex *p) +{ + p->numBlocks = 0; + p->size = 0; +} + +static void XzEncIndex_Free(CXzEncIndex *p, ISzAllocPtr alloc) +{ + if (p->blocks) + { + ISzAlloc_Free(alloc, p->blocks); + p->blocks = NULL; + } + p->numBlocks = 0; + p->size = 0; + p->allocated = 0; +} + + +static SRes XzEncIndex_ReAlloc(CXzEncIndex *p, size_t newSize, ISzAllocPtr alloc) +{ + Byte *blocks = (Byte *)ISzAlloc_Alloc(alloc, newSize); + if (!blocks) + return SZ_ERROR_MEM; + if (p->size != 0) + memcpy(blocks, p->blocks, p->size); + if (p->blocks) + ISzAlloc_Free(alloc, p->blocks); + p->blocks = blocks; + p->allocated = newSize; + return SZ_OK; +} + + +static SRes XzEncIndex_PreAlloc(CXzEncIndex *p, UInt64 numBlocks, UInt64 unpackSize, UInt64 totalSize, ISzAllocPtr alloc) +{ + UInt64 pos; + { + Byte buf[32]; + unsigned pos2 = Xz_WriteVarInt(buf, totalSize); + pos2 += Xz_WriteVarInt(buf + pos2, unpackSize); + pos = numBlocks * pos2; + } + + if (pos <= p->allocated - p->size) + return SZ_OK; + { + UInt64 newSize64 = p->size + pos; + size_t newSize = (size_t)newSize64; + if (newSize != newSize64) + return SZ_ERROR_MEM; + return XzEncIndex_ReAlloc(p, newSize, alloc); + } +} + + +static SRes XzEncIndex_AddIndexRecord(CXzEncIndex *p, UInt64 unpackSize, UInt64 totalSize, ISzAllocPtr alloc) +{ + Byte buf[32]; + unsigned pos = Xz_WriteVarInt(buf, totalSize); + pos += Xz_WriteVarInt(buf + pos, unpackSize); + + if (pos > p->allocated - p->size) + { + size_t newSize = p->allocated * 2 + 16 * 2; + if (newSize < p->size + pos) + return SZ_ERROR_MEM; + RINOK(XzEncIndex_ReAlloc(p, newSize, alloc)) + } + memcpy(p->blocks + p->size, buf, pos); + p->size += pos; + p->numBlocks++; + return SZ_OK; +} + + +static SRes XzEncIndex_WriteFooter(const CXzEncIndex *p, CXzStreamFlags flags, ISeqOutStreamPtr s) +{ + Byte buf[32]; + UInt64 globalPos; + UInt32 crc = CRC_INIT_VAL; + unsigned pos = 1 + Xz_WriteVarInt(buf + 1, p->numBlocks); + + globalPos = pos; + buf[0] = 0; + RINOK(WriteBytes_UpdateCrc(s, buf, pos, &crc)) + RINOK(WriteBytes_UpdateCrc(s, p->blocks, p->size, &crc)) + globalPos += p->size; + + pos = XZ_GET_PAD_SIZE(globalPos); + buf[1] = 0; + buf[2] = 0; + buf[3] = 0; + globalPos += pos; + + crc = CrcUpdate(crc, buf + 4 - pos, pos); + SetUi32(buf + 4, CRC_GET_DIGEST(crc)) + + SetUi32(buf + 8 + 4, (UInt32)(globalPos >> 2)) + buf[8 + 8] = (Byte)(flags >> 8); + buf[8 + 9] = (Byte)(flags & 0xFF); + SetUi32(buf + 8, CrcCalc(buf + 8 + 4, 6)) + buf[8 + 10] = XZ_FOOTER_SIG_0; + buf[8 + 11] = XZ_FOOTER_SIG_1; + + return WriteBytes(s, buf + 4 - pos, pos + 4 + 12); +} + + + +/* ---------- CSeqCheckInStream ---------- */ + +typedef struct +{ + ISeqInStream vt; + ISeqInStreamPtr realStream; + const Byte *data; + UInt64 limit; + UInt64 processed; + int realStreamFinished; + CXzCheck check; +} CSeqCheckInStream; + +static void SeqCheckInStream_Init(CSeqCheckInStream *p, unsigned checkMode) +{ + p->limit = (UInt64)(Int64)-1; + p->processed = 0; + p->realStreamFinished = 0; + XzCheck_Init(&p->check, checkMode); +} + +static void SeqCheckInStream_GetDigest(CSeqCheckInStream *p, Byte *digest) +{ + XzCheck_Final(&p->check, digest); +} + +static SRes SeqCheckInStream_Read(ISeqInStreamPtr pp, void *data, size_t *size) +{ + Z7_CONTAINER_FROM_VTBL_TO_DECL_VAR_pp_vt_p(CSeqCheckInStream) + 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) + { + if (p->realStream) + { + res = ISeqInStream_Read(p->realStream, data, &size2); + p->realStreamFinished = (size2 == 0) ? 1 : 0; + } + else + memcpy(data, p->data + (size_t)p->processed, size2); + XzCheck_Update(&p->check, data, size2); + p->processed += size2; + } + *size = size2; + return res; +} + + +/* ---------- CSeqSizeOutStream ---------- */ + +typedef struct +{ + ISeqOutStream vt; + ISeqOutStreamPtr realStream; + Byte *outBuf; + size_t outBufLimit; + UInt64 processed; +} CSeqSizeOutStream; + +static size_t SeqSizeOutStream_Write(ISeqOutStreamPtr pp, const void *data, size_t size) +{ + Z7_CONTAINER_FROM_VTBL_TO_DECL_VAR_pp_vt_p(CSeqSizeOutStream) + if (p->realStream) + size = ISeqOutStream_Write(p->realStream, data, size); + else + { + if (size > p->outBufLimit - (size_t)p->processed) + return 0; + memcpy(p->outBuf + (size_t)p->processed, data, size); + } + p->processed += size; + return size; +} + + +/* ---------- CSeqInFilter ---------- */ + +#define FILTER_BUF_SIZE (1 << 20) + +typedef struct +{ + ISeqInStream vt; + ISeqInStreamPtr realStream; + IStateCoder StateCoder; + Byte *buf; + size_t curPos; + size_t endPos; + int srcWasFinished; +} CSeqInFilter; + + +static const z7_Func_BranchConv g_Funcs_BranchConv_RISC_Enc[] = +{ + Z7_BRANCH_CONV_ENC(PPC), + Z7_BRANCH_CONV_ENC(IA64), + Z7_BRANCH_CONV_ENC(ARM), + Z7_BRANCH_CONV_ENC(ARMT), + Z7_BRANCH_CONV_ENC(SPARC), + Z7_BRANCH_CONV_ENC(ARM64) +}; + +static SizeT XzBcFilterStateBase_Filter_Enc(CXzBcFilterStateBase *p, Byte *data, SizeT size) +{ + switch (p->methodId) + { + case XZ_ID_Delta: + Delta_Encode(p->delta_State, p->delta, data, size); + break; + case XZ_ID_X86: + size = (SizeT)(z7_BranchConvSt_X86_Enc(data, size, p->ip, &p->X86_State) - data); + break; + default: + if (p->methodId >= XZ_ID_PPC) + { + const UInt32 i = p->methodId - XZ_ID_PPC; + if (i < Z7_ARRAY_SIZE(g_Funcs_BranchConv_RISC_Enc)) + size = (SizeT)(g_Funcs_BranchConv_RISC_Enc[i](data, size, p->ip) - data); + } + break; + } + p->ip += (UInt32)size; + return size; +} + + +static SRes SeqInFilter_Init(CSeqInFilter *p, const CXzFilter *props, ISzAllocPtr alloc) +{ + if (!p->buf) + { + p->buf = (Byte *)ISzAlloc_Alloc(alloc, FILTER_BUF_SIZE); + if (!p->buf) + return SZ_ERROR_MEM; + } + p->curPos = p->endPos = 0; + p->srcWasFinished = 0; + RINOK(Xz_StateCoder_Bc_SetFromMethod_Func(&p->StateCoder, props->id, XzBcFilterStateBase_Filter_Enc, alloc)) + RINOK(p->StateCoder.SetProps(p->StateCoder.p, props->props, props->propsSize, alloc)) + p->StateCoder.Init(p->StateCoder.p); + return SZ_OK; +} + + +static SRes SeqInFilter_Read(ISeqInStreamPtr pp, void *data, size_t *size) +{ + Z7_CONTAINER_FROM_VTBL_TO_DECL_VAR_pp_vt_p(CSeqInFilter) + const size_t sizeOriginal = *size; + if (sizeOriginal == 0) + return SZ_OK; + *size = 0; + + for (;;) + { + if (!p->srcWasFinished && p->curPos == p->endPos) + { + p->curPos = 0; + p->endPos = FILTER_BUF_SIZE; + RINOK(ISeqInStream_Read(p->realStream, p->buf, &p->endPos)) + if (p->endPos == 0) + p->srcWasFinished = 1; + } + { + SizeT srcLen = p->endPos - p->curPos; + ECoderStatus status; + SRes res; + *size = sizeOriginal; + res = p->StateCoder.Code2(p->StateCoder.p, + (Byte *)data, size, + p->buf + p->curPos, &srcLen, + p->srcWasFinished, CODER_FINISH_ANY, + &status); + p->curPos += srcLen; + if (*size != 0 || srcLen == 0 || res != SZ_OK) + return res; + } + } +} + +static void SeqInFilter_Construct(CSeqInFilter *p) +{ + p->buf = NULL; + p->StateCoder.p = NULL; + p->vt.Read = SeqInFilter_Read; +} + +static void SeqInFilter_Free(CSeqInFilter *p, ISzAllocPtr alloc) +{ + if (p->StateCoder.p) + { + p->StateCoder.Free(p->StateCoder.p, alloc); + p->StateCoder.p = NULL; + } + if (p->buf) + { + ISzAlloc_Free(alloc, p->buf); + p->buf = NULL; + } +} + + +/* ---------- CSbEncInStream ---------- */ + +#ifdef USE_SUBBLOCK + +typedef struct +{ + ISeqInStream vt; + ISeqInStreamPtr inStream; + CSbEnc enc; +} CSbEncInStream; + +static SRes SbEncInStream_Read(ISeqInStreamPtr pp, void *data, size_t *size) +{ + CSbEncInStream *p = Z7_CONTAINER_FROM_VTBL(pp, CSbEncInStream, vt); + size_t sizeOriginal = *size; + if (sizeOriginal == 0) + return SZ_OK; + + for (;;) + { + if (p->enc.needRead && !p->enc.readWasFinished) + { + size_t processed = p->enc.needReadSizeMax; + RINOK(p->inStream->Read(p->inStream, p->enc.buf + p->enc.readPos, &processed)) + p->enc.readPos += processed; + if (processed == 0) + { + p->enc.readWasFinished = True; + p->enc.isFinalFinished = True; + } + p->enc.needRead = False; + } + + *size = sizeOriginal; + RINOK(SbEnc_Read(&p->enc, data, size)) + if (*size != 0 || !p->enc.needRead) + return SZ_OK; + } +} + +void SbEncInStream_Construct(CSbEncInStream *p, ISzAllocPtr alloc) +{ + SbEnc_Construct(&p->enc, alloc); + p->vt.Read = SbEncInStream_Read; +} + +SRes SbEncInStream_Init(CSbEncInStream *p) +{ + return SbEnc_Init(&p->enc); +} + +void SbEncInStream_Free(CSbEncInStream *p) +{ + SbEnc_Free(&p->enc); +} + +#endif + + + +/* ---------- CXzProps ---------- */ + + +void XzFilterProps_Init(CXzFilterProps *p) +{ + p->id = 0; + p->delta = 0; + p->ip = 0; + p->ipDefined = False; +} + +void XzProps_Init(CXzProps *p) +{ + p->checkId = XZ_CHECK_CRC32; + p->blockSize = XZ_PROPS_BLOCK_SIZE_AUTO; + p->numBlockThreads_Reduced = -1; + p->numBlockThreads_Max = -1; + p->numTotalThreads = -1; + p->reduceSize = (UInt64)(Int64)-1; + p->forceWriteSizesInHeader = 0; + // p->forceWriteSizesInHeader = 1; + + XzFilterProps_Init(&p->filterProps); + Lzma2EncProps_Init(&p->lzma2Props); +} + + +static void XzEncProps_Normalize_Fixed(CXzProps *p) +{ + UInt64 fileSize; + int t1, t1n, t2, t2r, t3; + { + CLzma2EncProps tp = p->lzma2Props; + if (tp.numTotalThreads <= 0) + tp.numTotalThreads = p->numTotalThreads; + Lzma2EncProps_Normalize(&tp); + t1n = tp.numTotalThreads; + } + + t1 = p->lzma2Props.numTotalThreads; + 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->lzma2Props.numTotalThreads = t1; + + t2r = t2; + + fileSize = p->reduceSize; + + if ((p->blockSize < fileSize || fileSize == (UInt64)(Int64)-1)) + p->lzma2Props.lzmaProps.reduceSize = p->blockSize; + + Lzma2EncProps_Normalize(&p->lzma2Props); + + t1 = p->lzma2Props.numTotalThreads; + + { + if (t2 > 1 && fileSize != (UInt64)(Int64)-1) + { + UInt64 numBlocks = fileSize / p->blockSize; + if (numBlocks * p->blockSize != fileSize) + numBlocks++; + if (numBlocks < (unsigned)t2) + { + t2r = (int)numBlocks; + if (t2r == 0) + t2r = 1; + t3 = t1 * t2r; + } + } + } + + p->numBlockThreads_Max = t2; + p->numBlockThreads_Reduced = t2r; + p->numTotalThreads = t3; +} + + +static void XzProps_Normalize(CXzProps *p) +{ + /* we normalize xzProps properties, but we normalize only some of CXzProps::lzma2Props properties. + Lzma2Enc_SetProps() will normalize lzma2Props later. */ + + if (p->blockSize == XZ_PROPS_BLOCK_SIZE_SOLID) + { + p->lzma2Props.lzmaProps.reduceSize = p->reduceSize; + p->numBlockThreads_Reduced = 1; + p->numBlockThreads_Max = 1; + if (p->lzma2Props.numTotalThreads <= 0) + p->lzma2Props.numTotalThreads = p->numTotalThreads; + return; + } + else + { + CLzma2EncProps *lzma2 = &p->lzma2Props; + if (p->blockSize == LZMA2_ENC_PROPS_BLOCK_SIZE_AUTO) + { + // xz-auto + p->lzma2Props.lzmaProps.reduceSize = p->reduceSize; + + if (lzma2->blockSize == LZMA2_ENC_PROPS_BLOCK_SIZE_SOLID) + { + // if (xz-auto && lzma2-solid) - we use solid for both + p->blockSize = XZ_PROPS_BLOCK_SIZE_SOLID; + p->numBlockThreads_Reduced = 1; + p->numBlockThreads_Max = 1; + if (p->lzma2Props.numTotalThreads <= 0) + p->lzma2Props.numTotalThreads = p->numTotalThreads; + } + else + { + // if (xz-auto && (lzma2-auto || lzma2-fixed_) + // we calculate block size for lzma2 and use that block size for xz, lzma2 uses single-chunk per block + CLzma2EncProps tp = p->lzma2Props; + if (tp.numTotalThreads <= 0) + tp.numTotalThreads = p->numTotalThreads; + + Lzma2EncProps_Normalize(&tp); + + p->blockSize = tp.blockSize; // fixed or solid + p->numBlockThreads_Reduced = tp.numBlockThreads_Reduced; + p->numBlockThreads_Max = tp.numBlockThreads_Max; + if (lzma2->blockSize == LZMA2_ENC_PROPS_BLOCK_SIZE_AUTO) + lzma2->blockSize = tp.blockSize; // fixed or solid, LZMA2_ENC_PROPS_BLOCK_SIZE_SOLID + if (lzma2->lzmaProps.reduceSize > tp.blockSize && tp.blockSize != LZMA2_ENC_PROPS_BLOCK_SIZE_SOLID) + lzma2->lzmaProps.reduceSize = tp.blockSize; + lzma2->numBlockThreads_Reduced = 1; + lzma2->numBlockThreads_Max = 1; + return; + } + } + else + { + // xz-fixed + // we can use xz::reduceSize or xz::blockSize as base for lzmaProps::reduceSize + + p->lzma2Props.lzmaProps.reduceSize = p->reduceSize; + { + UInt64 r = p->reduceSize; + if (r > p->blockSize || r == (UInt64)(Int64)-1) + r = p->blockSize; + lzma2->lzmaProps.reduceSize = r; + } + if (lzma2->blockSize == LZMA2_ENC_PROPS_BLOCK_SIZE_AUTO) + lzma2->blockSize = LZMA2_ENC_PROPS_BLOCK_SIZE_SOLID; + else if (lzma2->blockSize > p->blockSize && lzma2->blockSize != LZMA2_ENC_PROPS_BLOCK_SIZE_SOLID) + lzma2->blockSize = p->blockSize; + + XzEncProps_Normalize_Fixed(p); + } + } +} + + +/* ---------- CLzma2WithFilters ---------- */ + +typedef struct +{ + CLzma2EncHandle lzma2; + CSeqInFilter filter; + + #ifdef USE_SUBBLOCK + CSbEncInStream sb; + #endif +} CLzma2WithFilters; + + +static void Lzma2WithFilters_Construct(CLzma2WithFilters *p) +{ + p->lzma2 = NULL; + SeqInFilter_Construct(&p->filter); + + #ifdef USE_SUBBLOCK + SbEncInStream_Construct(&p->sb, alloc); + #endif +} + + +static SRes Lzma2WithFilters_Create(CLzma2WithFilters *p, ISzAllocPtr alloc, ISzAllocPtr bigAlloc) +{ + if (!p->lzma2) + { + p->lzma2 = Lzma2Enc_Create(alloc, bigAlloc); + if (!p->lzma2) + return SZ_ERROR_MEM; + } + return SZ_OK; +} + + +static void Lzma2WithFilters_Free(CLzma2WithFilters *p, ISzAllocPtr alloc) +{ + #ifdef USE_SUBBLOCK + SbEncInStream_Free(&p->sb); + #endif + + SeqInFilter_Free(&p->filter, alloc); + if (p->lzma2) + { + Lzma2Enc_Destroy(p->lzma2); + p->lzma2 = NULL; + } +} + + +typedef struct +{ + UInt64 unpackSize; + UInt64 totalSize; + size_t headerSize; +} CXzEncBlockInfo; + + +static SRes Xz_CompressBlock( + CLzma2WithFilters *lzmaf, + + ISeqOutStreamPtr outStream, + Byte *outBufHeader, + Byte *outBufData, size_t outBufDataLimit, + + ISeqInStreamPtr inStream, + // UInt64 expectedSize, + const Byte *inBuf, // used if (!inStream) + size_t inBufSize, // used if (!inStream), it's block size, props->blockSize is ignored + + const CXzProps *props, + ICompressProgressPtr progress, + int *inStreamFinished, /* only for inStream version */ + CXzEncBlockInfo *blockSizes, + ISzAllocPtr alloc, + ISzAllocPtr allocBig) +{ + CSeqCheckInStream checkInStream; + CSeqSizeOutStream seqSizeOutStream; + CXzBlock block; + unsigned filterIndex = 0; + CXzFilter *filter = NULL; + const CXzFilterProps *fp = &props->filterProps; + if (fp->id == 0) + fp = NULL; + + *inStreamFinished = False; + + RINOK(Lzma2WithFilters_Create(lzmaf, alloc, allocBig)) + + RINOK(Lzma2Enc_SetProps(lzmaf->lzma2, &props->lzma2Props)) + + // XzBlock_ClearFlags(&block) + XzBlock_ClearFlags_SetNumFilters(&block, 1 + (fp ? 1 : 0)) + + if (fp) + { + filter = &block.filters[filterIndex++]; + filter->id = fp->id; + filter->propsSize = 0; + + if (fp->id == XZ_ID_Delta) + { + filter->props[0] = (Byte)(fp->delta - 1); + filter->propsSize = 1; + } + else if (fp->ipDefined) + { + Byte *ptr = filter->props; + SetUi32(ptr, fp->ip) + filter->propsSize = 4; + } + } + + { + CXzFilter *f = &block.filters[filterIndex++]; + f->id = XZ_ID_LZMA2; + f->propsSize = 1; + f->props[0] = Lzma2Enc_WriteProperties(lzmaf->lzma2); + } + + seqSizeOutStream.vt.Write = SeqSizeOutStream_Write; + seqSizeOutStream.realStream = outStream; + seqSizeOutStream.outBuf = outBufData; + seqSizeOutStream.outBufLimit = outBufDataLimit; + seqSizeOutStream.processed = 0; + + /* + if (expectedSize != (UInt64)(Int64)-1) + { + block.unpackSize = expectedSize; + if (props->blockSize != (UInt64)(Int64)-1) + if (expectedSize > props->blockSize) + block.unpackSize = props->blockSize; + XzBlock_SetHasUnpackSize(&block) + } + */ + + if (outStream) + { + RINOK(XzBlock_WriteHeader(&block, &seqSizeOutStream.vt)) + } + + checkInStream.vt.Read = SeqCheckInStream_Read; + SeqCheckInStream_Init(&checkInStream, props->checkId); + + checkInStream.realStream = inStream; + checkInStream.data = inBuf; + checkInStream.limit = props->blockSize; + if (!inStream) + checkInStream.limit = inBufSize; + + if (fp) + { + #ifdef USE_SUBBLOCK + if (fp->id == XZ_ID_Subblock) + { + lzmaf->sb.inStream = &checkInStream.vt; + RINOK(SbEncInStream_Init(&lzmaf->sb)) + } + else + #endif + { + lzmaf->filter.realStream = &checkInStream.vt; + RINOK(SeqInFilter_Init(&lzmaf->filter, filter, alloc)) + } + } + + { + SRes res; + Byte *outBuf = NULL; + size_t outSize = 0; + BoolInt useStream = (fp || inStream); + // useStream = True; + + if (!useStream) + { + XzCheck_Update(&checkInStream.check, inBuf, inBufSize); + checkInStream.processed = inBufSize; + } + + if (!outStream) + { + outBuf = seqSizeOutStream.outBuf; // + (size_t)seqSizeOutStream.processed; + outSize = seqSizeOutStream.outBufLimit; // - (size_t)seqSizeOutStream.processed; + } + + res = Lzma2Enc_Encode2(lzmaf->lzma2, + outBuf ? NULL : &seqSizeOutStream.vt, + outBuf, + outBuf ? &outSize : NULL, + + useStream ? + (fp ? + ( + #ifdef USE_SUBBLOCK + (fp->id == XZ_ID_Subblock) ? &lzmaf->sb.vt: + #endif + &lzmaf->filter.vt) : + &checkInStream.vt) : NULL, + + useStream ? NULL : inBuf, + useStream ? 0 : inBufSize, + + progress); + + if (outBuf) + seqSizeOutStream.processed += outSize; + + RINOK(res) + blockSizes->unpackSize = checkInStream.processed; + } + { + Byte buf[4 + 64]; + unsigned padSize = XZ_GET_PAD_SIZE(seqSizeOutStream.processed); + UInt64 packSize = seqSizeOutStream.processed; + + buf[0] = 0; + buf[1] = 0; + buf[2] = 0; + buf[3] = 0; + + SeqCheckInStream_GetDigest(&checkInStream, buf + 4); + RINOK(WriteBytes(&seqSizeOutStream.vt, buf + (4 - padSize), padSize + XzFlags_GetCheckSize((CXzStreamFlags)props->checkId))) + + blockSizes->totalSize = seqSizeOutStream.processed - padSize; + + if (!outStream) + { + seqSizeOutStream.outBuf = outBufHeader; + seqSizeOutStream.outBufLimit = XZ_BLOCK_HEADER_SIZE_MAX; + seqSizeOutStream.processed = 0; + + block.unpackSize = blockSizes->unpackSize; + XzBlock_SetHasUnpackSize(&block) + + block.packSize = packSize; + XzBlock_SetHasPackSize(&block) + + RINOK(XzBlock_WriteHeader(&block, &seqSizeOutStream.vt)) + + blockSizes->headerSize = (size_t)seqSizeOutStream.processed; + blockSizes->totalSize += seqSizeOutStream.processed; + } + } + + if (inStream) + *inStreamFinished = checkInStream.realStreamFinished; + else + { + *inStreamFinished = False; + if (checkInStream.processed != inBufSize) + return SZ_ERROR_FAIL; + } + + return SZ_OK; +} + + + +typedef struct +{ + ICompressProgress vt; + ICompressProgressPtr progress; + UInt64 inOffset; + UInt64 outOffset; +} CCompressProgress_XzEncOffset; + + +static SRes CompressProgress_XzEncOffset_Progress(ICompressProgressPtr pp, UInt64 inSize, UInt64 outSize) +{ + const CCompressProgress_XzEncOffset *p = Z7_CONTAINER_FROM_VTBL_CONST(pp, CCompressProgress_XzEncOffset, vt); + inSize += p->inOffset; + outSize += p->outOffset; + return ICompressProgress_Progress(p->progress, inSize, outSize); +} + + + + +struct CXzEnc +{ + ISzAllocPtr alloc; + ISzAllocPtr allocBig; + + CXzProps xzProps; + UInt64 expectedDataSize; + + CXzEncIndex xzIndex; + + CLzma2WithFilters lzmaf_Items[MTCODER_THREADS_MAX]; + + size_t outBufSize; /* size of allocated outBufs[i] */ + Byte *outBufs[MTCODER_BLOCKS_MAX]; + + #ifndef Z7_ST + unsigned checkType; + ISeqOutStreamPtr outStream; + BoolInt mtCoder_WasConstructed; + CMtCoder mtCoder; + CXzEncBlockInfo EncBlocks[MTCODER_BLOCKS_MAX]; + #endif +}; + + +static void XzEnc_Construct(CXzEnc *p) +{ + unsigned i; + + XzEncIndex_Construct(&p->xzIndex); + + for (i = 0; i < MTCODER_THREADS_MAX; i++) + Lzma2WithFilters_Construct(&p->lzmaf_Items[i]); + + #ifndef Z7_ST + p->mtCoder_WasConstructed = False; + { + for (i = 0; i < MTCODER_BLOCKS_MAX; i++) + p->outBufs[i] = NULL; + p->outBufSize = 0; + } + #endif +} + + +static void XzEnc_FreeOutBufs(CXzEnc *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; +} + + +static void XzEnc_Free(CXzEnc *p, ISzAllocPtr alloc) +{ + unsigned i; + + XzEncIndex_Free(&p->xzIndex, alloc); + + for (i = 0; i < MTCODER_THREADS_MAX; i++) + Lzma2WithFilters_Free(&p->lzmaf_Items[i], alloc); + + #ifndef Z7_ST + if (p->mtCoder_WasConstructed) + { + MtCoder_Destruct(&p->mtCoder); + p->mtCoder_WasConstructed = False; + } + XzEnc_FreeOutBufs(p); + #endif +} + + +CXzEncHandle XzEnc_Create(ISzAllocPtr alloc, ISzAllocPtr allocBig) +{ + CXzEnc *p = (CXzEnc *)ISzAlloc_Alloc(alloc, sizeof(CXzEnc)); + if (!p) + return NULL; + XzEnc_Construct(p); + XzProps_Init(&p->xzProps); + XzProps_Normalize(&p->xzProps); + p->expectedDataSize = (UInt64)(Int64)-1; + p->alloc = alloc; + p->allocBig = allocBig; + return (CXzEncHandle)p; +} + +// #define GET_CXzEnc_p CXzEnc *p = (CXzEnc *)(void *)pp; + +void XzEnc_Destroy(CXzEncHandle p) +{ + // GET_CXzEnc_p + XzEnc_Free(p, p->alloc); + ISzAlloc_Free(p->alloc, p); +} + + +SRes XzEnc_SetProps(CXzEncHandle p, const CXzProps *props) +{ + // GET_CXzEnc_p + p->xzProps = *props; + XzProps_Normalize(&p->xzProps); + return SZ_OK; +} + + +void XzEnc_SetDataSize(CXzEncHandle p, UInt64 expectedDataSiize) +{ + // GET_CXzEnc_p + p->expectedDataSize = expectedDataSiize; +} + + + + +#ifndef Z7_ST + +static SRes XzEnc_MtCallback_Code(void *pp, unsigned coderIndex, unsigned outBufIndex, + const Byte *src, size_t srcSize, int finished) +{ + CXzEnc *me = (CXzEnc *)pp; + SRes res; + CMtProgressThunk progressThunk; + + Byte *dest = me->outBufs[outBufIndex]; + + UNUSED_VAR(finished) + + { + CXzEncBlockInfo *bInfo = &me->EncBlocks[outBufIndex]; + bInfo->totalSize = 0; + bInfo->unpackSize = 0; + bInfo->headerSize = 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; + MtProgressThunk_INIT(&progressThunk) + + { + CXzEncBlockInfo blockSizes; + int inStreamFinished; + + res = Xz_CompressBlock( + &me->lzmaf_Items[coderIndex], + + NULL, + dest, + dest + XZ_BLOCK_HEADER_SIZE_MAX, me->outBufSize - XZ_BLOCK_HEADER_SIZE_MAX, + + NULL, + // srcSize, // expectedSize + src, srcSize, + + &me->xzProps, + &progressThunk.vt, + &inStreamFinished, + &blockSizes, + me->alloc, + me->allocBig); + + if (res == SZ_OK) + me->EncBlocks[outBufIndex] = blockSizes; + + return res; + } +} + + +static SRes XzEnc_MtCallback_Write(void *pp, unsigned outBufIndex) +{ + CXzEnc *me = (CXzEnc *)pp; + + const CXzEncBlockInfo *bInfo = &me->EncBlocks[outBufIndex]; + const Byte *data = me->outBufs[outBufIndex]; + + RINOK(WriteBytes(me->outStream, data, bInfo->headerSize)) + + { + UInt64 totalPackFull = bInfo->totalSize + XZ_GET_PAD_SIZE(bInfo->totalSize); + RINOK(WriteBytes(me->outStream, data + XZ_BLOCK_HEADER_SIZE_MAX, (size_t)totalPackFull - bInfo->headerSize)) + } + + return XzEncIndex_AddIndexRecord(&me->xzIndex, bInfo->unpackSize, bInfo->totalSize, me->alloc); +} + +#endif + + + +SRes XzEnc_Encode(CXzEncHandle p, ISeqOutStreamPtr outStream, ISeqInStreamPtr inStream, ICompressProgressPtr progress) +{ + // GET_CXzEnc_p + + const CXzProps *props = &p->xzProps; + + XzEncIndex_Init(&p->xzIndex); + { + UInt64 numBlocks = 1; + UInt64 blockSize = props->blockSize; + + if (blockSize != XZ_PROPS_BLOCK_SIZE_SOLID + && props->reduceSize != (UInt64)(Int64)-1) + { + numBlocks = props->reduceSize / blockSize; + if (numBlocks * blockSize != props->reduceSize) + numBlocks++; + } + else + blockSize = (UInt64)1 << 62; + + RINOK(XzEncIndex_PreAlloc(&p->xzIndex, numBlocks, blockSize, XZ_GET_ESTIMATED_BLOCK_TOTAL_PACK_SIZE(blockSize), p->alloc)) + } + + RINOK(Xz_WriteHeader((CXzStreamFlags)props->checkId, outStream)) + + + #ifndef Z7_ST + if (props->numBlockThreads_Reduced > 1) + { + IMtCoderCallback2 vt; + + if (!p->mtCoder_WasConstructed) + { + p->mtCoder_WasConstructed = True; + MtCoder_Construct(&p->mtCoder); + } + + vt.Code = XzEnc_MtCallback_Code; + vt.Write = XzEnc_MtCallback_Write; + + p->checkType = props->checkId; + p->xzProps = *props; + + p->outStream = outStream; + + p->mtCoder.allocBig = p->allocBig; + p->mtCoder.progress = progress; + p->mtCoder.inStream = inStream; + p->mtCoder.inData = NULL; + p->mtCoder.inDataSize = 0; + p->mtCoder.mtCallback = &vt; + p->mtCoder.mtCallbackObject = p; + + if ( props->blockSize == XZ_PROPS_BLOCK_SIZE_SOLID + || props->blockSize == XZ_PROPS_BLOCK_SIZE_AUTO) + return SZ_ERROR_FAIL; + + p->mtCoder.blockSize = (size_t)props->blockSize; + if (p->mtCoder.blockSize != props->blockSize) + return SZ_ERROR_PARAM; /* SZ_ERROR_MEM */ + + { + size_t destBlockSize = XZ_BLOCK_HEADER_SIZE_MAX + XZ_GET_MAX_BLOCK_PACK_SIZE(p->mtCoder.blockSize); + if (destBlockSize < p->mtCoder.blockSize) + return SZ_ERROR_PARAM; + if (p->outBufSize != destBlockSize) + XzEnc_FreeOutBufs(p); + p->outBufSize = destBlockSize; + } + + p->mtCoder.numThreadsMax = (unsigned)props->numBlockThreads_Max; + p->mtCoder.expectedDataSize = p->expectedDataSize; + + RINOK(MtCoder_Code(&p->mtCoder)) + } + else + #endif + { + int writeStartSizes; + CCompressProgress_XzEncOffset progress2; + Byte *bufData = NULL; + size_t bufSize = 0; + + progress2.vt.Progress = CompressProgress_XzEncOffset_Progress; + progress2.inOffset = 0; + progress2.outOffset = 0; + progress2.progress = progress; + + writeStartSizes = 0; + + if (props->blockSize != XZ_PROPS_BLOCK_SIZE_SOLID) + { + writeStartSizes = (props->forceWriteSizesInHeader > 0); + + if (writeStartSizes) + { + size_t t2; + size_t t = (size_t)props->blockSize; + if (t != props->blockSize) + return SZ_ERROR_PARAM; + t = XZ_GET_MAX_BLOCK_PACK_SIZE(t); + if (t < props->blockSize) + return SZ_ERROR_PARAM; + t2 = XZ_BLOCK_HEADER_SIZE_MAX + t; + if (!p->outBufs[0] || t2 != p->outBufSize) + { + XzEnc_FreeOutBufs(p); + p->outBufs[0] = (Byte *)ISzAlloc_Alloc(p->alloc, t2); + if (!p->outBufs[0]) + return SZ_ERROR_MEM; + p->outBufSize = t2; + } + bufData = p->outBufs[0] + XZ_BLOCK_HEADER_SIZE_MAX; + bufSize = t; + } + } + + for (;;) + { + CXzEncBlockInfo blockSizes; + int inStreamFinished; + + /* + UInt64 rem = (UInt64)(Int64)-1; + if (props->reduceSize != (UInt64)(Int64)-1 + && props->reduceSize >= progress2.inOffset) + rem = props->reduceSize - progress2.inOffset; + */ + + blockSizes.headerSize = 0; // for GCC + + RINOK(Xz_CompressBlock( + &p->lzmaf_Items[0], + + writeStartSizes ? NULL : outStream, + writeStartSizes ? p->outBufs[0] : NULL, + bufData, bufSize, + + inStream, + // rem, + NULL, 0, + + props, + progress ? &progress2.vt : NULL, + &inStreamFinished, + &blockSizes, + p->alloc, + p->allocBig)) + + { + UInt64 totalPackFull = blockSizes.totalSize + XZ_GET_PAD_SIZE(blockSizes.totalSize); + + if (writeStartSizes) + { + RINOK(WriteBytes(outStream, p->outBufs[0], blockSizes.headerSize)) + RINOK(WriteBytes(outStream, bufData, (size_t)totalPackFull - blockSizes.headerSize)) + } + + RINOK(XzEncIndex_AddIndexRecord(&p->xzIndex, blockSizes.unpackSize, blockSizes.totalSize, p->alloc)) + + progress2.inOffset += blockSizes.unpackSize; + progress2.outOffset += totalPackFull; + } + + if (inStreamFinished) + break; + } + } + + return XzEncIndex_WriteFooter(&p->xzIndex, (CXzStreamFlags)props->checkId, outStream); +} + + +#include "Alloc.h" + +SRes Xz_Encode(ISeqOutStreamPtr outStream, ISeqInStreamPtr inStream, + const CXzProps *props, ICompressProgressPtr progress) +{ + SRes res; + CXzEncHandle xz = XzEnc_Create(&g_Alloc, &g_BigAlloc); + if (!xz) + return SZ_ERROR_MEM; + res = XzEnc_SetProps(xz, props); + if (res == SZ_OK) + res = XzEnc_Encode(xz, outStream, inStream, progress); + XzEnc_Destroy(xz); + return res; +} + + +SRes Xz_EncodeEmpty(ISeqOutStreamPtr outStream) +{ + SRes res; + CXzEncIndex xzIndex; + XzEncIndex_Construct(&xzIndex); + res = Xz_WriteHeader((CXzStreamFlags)0, outStream); + if (res == SZ_OK) + res = XzEncIndex_WriteFooter(&xzIndex, (CXzStreamFlags)0, outStream); + XzEncIndex_Free(&xzIndex, NULL); // g_Alloc + return res; +} diff --git a/libraries/lzma/C/XzEnc.h b/libraries/lzma/C/XzEnc.h new file mode 100644 index 000000000..77b78c014 --- /dev/null +++ b/libraries/lzma/C/XzEnc.h @@ -0,0 +1,61 @@ +/* XzEnc.h -- Xz Encode +2023-04-13 : Igor Pavlov : Public domain */ + +#ifndef ZIP7_INC_XZ_ENC_H +#define ZIP7_INC_XZ_ENC_H + +#include "Lzma2Enc.h" + +#include "Xz.h" + +EXTERN_C_BEGIN + + +#define XZ_PROPS_BLOCK_SIZE_AUTO LZMA2_ENC_PROPS_BLOCK_SIZE_AUTO +#define XZ_PROPS_BLOCK_SIZE_SOLID LZMA2_ENC_PROPS_BLOCK_SIZE_SOLID + + +typedef struct +{ + UInt32 id; + UInt32 delta; + UInt32 ip; + int ipDefined; +} CXzFilterProps; + +void XzFilterProps_Init(CXzFilterProps *p); + + +typedef struct +{ + CLzma2EncProps lzma2Props; + CXzFilterProps filterProps; + unsigned checkId; + UInt64 blockSize; + int numBlockThreads_Reduced; + int numBlockThreads_Max; + int numTotalThreads; + int forceWriteSizesInHeader; + UInt64 reduceSize; +} CXzProps; + +void XzProps_Init(CXzProps *p); + +typedef struct CXzEnc CXzEnc; +typedef CXzEnc * CXzEncHandle; +// Z7_DECLARE_HANDLE(CXzEncHandle) + +CXzEncHandle XzEnc_Create(ISzAllocPtr alloc, ISzAllocPtr allocBig); +void XzEnc_Destroy(CXzEncHandle p); +SRes XzEnc_SetProps(CXzEncHandle p, const CXzProps *props); +void XzEnc_SetDataSize(CXzEncHandle p, UInt64 expectedDataSiize); +SRes XzEnc_Encode(CXzEncHandle p, ISeqOutStreamPtr outStream, ISeqInStreamPtr inStream, ICompressProgressPtr progress); + +SRes Xz_Encode(ISeqOutStreamPtr outStream, ISeqInStreamPtr inStream, + const CXzProps *props, ICompressProgressPtr progress); + +SRes Xz_EncodeEmpty(ISeqOutStreamPtr outStream); + +EXTERN_C_END + +#endif diff --git a/libraries/lzma/C/XzIn.c b/libraries/lzma/C/XzIn.c new file mode 100644 index 000000000..d0fc76367 --- /dev/null +++ b/libraries/lzma/C/XzIn.c @@ -0,0 +1,340 @@ +/* XzIn.c - Xz input +2023-04-02 : Igor Pavlov : Public domain */ + +#include "Precomp.h" + +#include + +#include "7zCrc.h" +#include "CpuArch.h" +#include "Xz.h" + +/* +#define XZ_FOOTER_SIG_CHECK(p) (memcmp((p), XZ_FOOTER_SIG, XZ_FOOTER_SIG_SIZE) == 0) +*/ +#define XZ_FOOTER_SIG_CHECK(p) ((p)[0] == XZ_FOOTER_SIG_0 && (p)[1] == XZ_FOOTER_SIG_1) + + +SRes Xz_ReadHeader(CXzStreamFlags *p, ISeqInStreamPtr inStream) +{ + Byte sig[XZ_STREAM_HEADER_SIZE]; + size_t processedSize = XZ_STREAM_HEADER_SIZE; + RINOK(SeqInStream_ReadMax(inStream, sig, &processedSize)) + if (processedSize != XZ_STREAM_HEADER_SIZE + || memcmp(sig, XZ_SIG, XZ_SIG_SIZE) != 0) + return SZ_ERROR_NO_ARCHIVE; + return Xz_ParseHeader(p, sig); +} + +#define READ_VARINT_AND_CHECK(buf, pos, size, res) \ + { unsigned s = Xz_ReadVarInt(buf + pos, size - pos, res); \ + if (s == 0) return SZ_ERROR_ARCHIVE; \ + pos += s; } + +SRes XzBlock_ReadHeader(CXzBlock *p, ISeqInStreamPtr inStream, BoolInt *isIndex, UInt32 *headerSizeRes) +{ + Byte header[XZ_BLOCK_HEADER_SIZE_MAX]; + unsigned headerSize; + *headerSizeRes = 0; + RINOK(SeqInStream_ReadByte(inStream, &header[0])) + headerSize = (unsigned)header[0]; + if (headerSize == 0) + { + *headerSizeRes = 1; + *isIndex = True; + return SZ_OK; + } + + *isIndex = False; + headerSize = (headerSize << 2) + 4; + *headerSizeRes = headerSize; + { + size_t processedSize = headerSize - 1; + RINOK(SeqInStream_ReadMax(inStream, header + 1, &processedSize)) + if (processedSize != headerSize - 1) + return SZ_ERROR_INPUT_EOF; + } + return XzBlock_Parse(p, header); +} + +#define ADD_SIZE_CHECK(size, val) \ + { UInt64 newSize = size + (val); if (newSize < size) return XZ_SIZE_OVERFLOW; size = newSize; } + +UInt64 Xz_GetUnpackSize(const CXzStream *p) +{ + UInt64 size = 0; + size_t i; + for (i = 0; i < p->numBlocks; i++) + { + ADD_SIZE_CHECK(size, p->blocks[i].unpackSize) + } + return size; +} + +UInt64 Xz_GetPackSize(const CXzStream *p) +{ + UInt64 size = 0; + size_t i; + for (i = 0; i < p->numBlocks; i++) + { + ADD_SIZE_CHECK(size, (p->blocks[i].totalSize + 3) & ~(UInt64)3) + } + return size; +} + +/* +SRes XzBlock_ReadFooter(CXzBlock *p, CXzStreamFlags f, ISeqInStreamPtr inStream) +{ + return SeqInStream_Read(inStream, p->check, XzFlags_GetCheckSize(f)); +} +*/ + +static SRes Xz_ReadIndex2(CXzStream *p, const Byte *buf, size_t size, ISzAllocPtr alloc) +{ + size_t numBlocks, pos = 1; + UInt32 crc; + + if (size < 5 || buf[0] != 0) + return SZ_ERROR_ARCHIVE; + + size -= 4; + crc = CrcCalc(buf, size); + if (crc != GetUi32(buf + size)) + return SZ_ERROR_ARCHIVE; + + { + UInt64 numBlocks64; + READ_VARINT_AND_CHECK(buf, pos, size, &numBlocks64) + numBlocks = (size_t)numBlocks64; + if (numBlocks != numBlocks64 || numBlocks * 2 > size) + return SZ_ERROR_ARCHIVE; + } + + Xz_Free(p, alloc); + if (numBlocks != 0) + { + size_t i; + p->numBlocks = numBlocks; + p->blocks = (CXzBlockSizes *)ISzAlloc_Alloc(alloc, sizeof(CXzBlockSizes) * numBlocks); + if (!p->blocks) + return SZ_ERROR_MEM; + for (i = 0; i < numBlocks; i++) + { + CXzBlockSizes *block = &p->blocks[i]; + READ_VARINT_AND_CHECK(buf, pos, size, &block->totalSize) + READ_VARINT_AND_CHECK(buf, pos, size, &block->unpackSize) + if (block->totalSize == 0) + return SZ_ERROR_ARCHIVE; + } + } + while ((pos & 3) != 0) + if (buf[pos++] != 0) + return SZ_ERROR_ARCHIVE; + return (pos == size) ? SZ_OK : SZ_ERROR_ARCHIVE; +} + +static SRes Xz_ReadIndex(CXzStream *p, ILookInStreamPtr stream, UInt64 indexSize, ISzAllocPtr alloc) +{ + SRes res; + size_t size; + Byte *buf; + if (indexSize > ((UInt32)1 << 31)) + return SZ_ERROR_UNSUPPORTED; + size = (size_t)indexSize; + if (size != indexSize) + return SZ_ERROR_UNSUPPORTED; + buf = (Byte *)ISzAlloc_Alloc(alloc, size); + if (!buf) + return SZ_ERROR_MEM; + res = LookInStream_Read2(stream, buf, size, SZ_ERROR_UNSUPPORTED); + if (res == SZ_OK) + res = Xz_ReadIndex2(p, buf, size, alloc); + ISzAlloc_Free(alloc, buf); + return res; +} + +static SRes LookInStream_SeekRead_ForArc(ILookInStreamPtr stream, UInt64 offset, void *buf, size_t size) +{ + RINOK(LookInStream_SeekTo(stream, offset)) + return LookInStream_Read(stream, buf, size); + /* return LookInStream_Read2(stream, buf, size, SZ_ERROR_NO_ARCHIVE); */ +} + +static SRes Xz_ReadBackward(CXzStream *p, ILookInStreamPtr stream, Int64 *startOffset, ISzAllocPtr alloc) +{ + UInt64 indexSize; + Byte buf[XZ_STREAM_FOOTER_SIZE]; + UInt64 pos = (UInt64)*startOffset; + + if ((pos & 3) != 0 || pos < XZ_STREAM_FOOTER_SIZE) + return SZ_ERROR_NO_ARCHIVE; + + pos -= XZ_STREAM_FOOTER_SIZE; + RINOK(LookInStream_SeekRead_ForArc(stream, pos, buf, XZ_STREAM_FOOTER_SIZE)) + + if (!XZ_FOOTER_SIG_CHECK(buf + 10)) + { + UInt32 total = 0; + pos += XZ_STREAM_FOOTER_SIZE; + + for (;;) + { + size_t i; + #define TEMP_BUF_SIZE (1 << 10) + Byte temp[TEMP_BUF_SIZE]; + + i = (pos > TEMP_BUF_SIZE) ? TEMP_BUF_SIZE : (size_t)pos; + pos -= i; + RINOK(LookInStream_SeekRead_ForArc(stream, pos, temp, i)) + total += (UInt32)i; + for (; i != 0; i--) + if (temp[i - 1] != 0) + break; + if (i != 0) + { + if ((i & 3) != 0) + return SZ_ERROR_NO_ARCHIVE; + pos += i; + break; + } + if (pos < XZ_STREAM_FOOTER_SIZE || total > (1 << 16)) + return SZ_ERROR_NO_ARCHIVE; + } + + if (pos < XZ_STREAM_FOOTER_SIZE) + return SZ_ERROR_NO_ARCHIVE; + pos -= XZ_STREAM_FOOTER_SIZE; + RINOK(LookInStream_SeekRead_ForArc(stream, pos, buf, XZ_STREAM_FOOTER_SIZE)) + if (!XZ_FOOTER_SIG_CHECK(buf + 10)) + return SZ_ERROR_NO_ARCHIVE; + } + + p->flags = (CXzStreamFlags)GetBe16(buf + 8); + + if (!XzFlags_IsSupported(p->flags)) + return SZ_ERROR_UNSUPPORTED; + + { + /* to eliminate GCC 6.3 warning: + dereferencing type-punned pointer will break strict-aliasing rules */ + const Byte *buf_ptr = buf; + if (GetUi32(buf_ptr) != CrcCalc(buf + 4, 6)) + return SZ_ERROR_ARCHIVE; + } + + indexSize = ((UInt64)GetUi32(buf + 4) + 1) << 2; + + if (pos < indexSize) + return SZ_ERROR_ARCHIVE; + + pos -= indexSize; + RINOK(LookInStream_SeekTo(stream, pos)) + RINOK(Xz_ReadIndex(p, stream, indexSize, alloc)) + + { + UInt64 totalSize = Xz_GetPackSize(p); + if (totalSize == XZ_SIZE_OVERFLOW + || totalSize >= ((UInt64)1 << 63) + || pos < totalSize + XZ_STREAM_HEADER_SIZE) + return SZ_ERROR_ARCHIVE; + pos -= (totalSize + XZ_STREAM_HEADER_SIZE); + RINOK(LookInStream_SeekTo(stream, pos)) + *startOffset = (Int64)pos; + } + { + CXzStreamFlags headerFlags; + CSecToRead secToRead; + SecToRead_CreateVTable(&secToRead); + secToRead.realStream = stream; + + RINOK(Xz_ReadHeader(&headerFlags, &secToRead.vt)) + return (p->flags == headerFlags) ? SZ_OK : SZ_ERROR_ARCHIVE; + } +} + + +/* ---------- Xz Streams ---------- */ + +void Xzs_Construct(CXzs *p) +{ + p->num = p->numAllocated = 0; + p->streams = 0; +} + +void Xzs_Free(CXzs *p, ISzAllocPtr alloc) +{ + size_t i; + for (i = 0; i < p->num; i++) + Xz_Free(&p->streams[i], alloc); + ISzAlloc_Free(alloc, p->streams); + p->num = p->numAllocated = 0; + p->streams = 0; +} + +UInt64 Xzs_GetNumBlocks(const CXzs *p) +{ + UInt64 num = 0; + size_t i; + for (i = 0; i < p->num; i++) + num += p->streams[i].numBlocks; + return num; +} + +UInt64 Xzs_GetUnpackSize(const CXzs *p) +{ + UInt64 size = 0; + size_t i; + for (i = 0; i < p->num; i++) + { + ADD_SIZE_CHECK(size, Xz_GetUnpackSize(&p->streams[i])) + } + return size; +} + +/* +UInt64 Xzs_GetPackSize(const CXzs *p) +{ + UInt64 size = 0; + size_t i; + for (i = 0; i < p->num; i++) + { + ADD_SIZE_CHECK(size, Xz_GetTotalSize(&p->streams[i])) + } + return size; +} +*/ + +SRes Xzs_ReadBackward(CXzs *p, ILookInStreamPtr stream, Int64 *startOffset, ICompressProgressPtr progress, ISzAllocPtr alloc) +{ + Int64 endOffset = 0; + RINOK(ILookInStream_Seek(stream, &endOffset, SZ_SEEK_END)) + *startOffset = endOffset; + for (;;) + { + CXzStream st; + SRes res; + Xz_Construct(&st); + res = Xz_ReadBackward(&st, stream, startOffset, alloc); + st.startOffset = (UInt64)*startOffset; + RINOK(res) + if (p->num == p->numAllocated) + { + const size_t newNum = p->num + p->num / 4 + 1; + void *data = ISzAlloc_Alloc(alloc, newNum * sizeof(CXzStream)); + if (!data) + return SZ_ERROR_MEM; + p->numAllocated = newNum; + if (p->num != 0) + memcpy(data, p->streams, p->num * sizeof(CXzStream)); + ISzAlloc_Free(alloc, p->streams); + p->streams = (CXzStream *)data; + } + p->streams[p->num++] = st; + if (*startOffset == 0) + break; + RINOK(LookInStream_SeekTo(stream, (UInt64)*startOffset)) + if (progress && ICompressProgress_Progress(progress, (UInt64)(endOffset - *startOffset), (UInt64)(Int64)-1) != SZ_OK) + return SZ_ERROR_PROGRESS; + } + return SZ_OK; +} diff --git a/libraries/lzma/CMakeLists.txt b/libraries/lzma/CMakeLists.txt index 462419226..bfb8e327e 100644 --- a/libraries/lzma/CMakeLists.txt +++ b/libraries/lzma/CMakeLists.txt @@ -5,26 +5,52 @@ set( CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -D_7ZIP_PPMD_SUPPPORT" ) find_package(Threads) add_library( lzma STATIC + C/7zAlloc.c C/7zArcIn.c C/7zBuf.c + C/7zBuf2.c C/7zCrc.c C/7zCrcOpt.c C/7zDec.c + C/7zFile.c C/7zStream.c + C/Aes.c + C/AesOpt.c + C/Alloc.c C/Bcj2.c + C/Bcj2Enc.c C/Bra.c C/Bra86.c C/BraIA64.c C/CpuArch.c C/Delta.c + C/DllSecur.c C/LzFind.c C/LzFindMt.c C/LzFindOpt.c C/Lzma2Dec.c + C/Lzma2DecMt.c + C/Lzma2Enc.c + C/Lzma86Dec.c + C/Lzma86Enc.c C/LzmaDec.c C/LzmaEnc.c + C/LzmaLib.c + C/MtCoder.c + C/MtDec.c C/Ppmd7.c C/Ppmd7Dec.c + C/Ppmd7Enc.c + C/Sha256.c + C/Sha256Opt.c + C/Sort.c + C/SwapBytes.c C/Threads.c + C/Xz.c + C/XzCrc64.c + C/XzCrc64Opt.c + C/XzDec.c + C/XzEnc.c + C/XzIn.c ) target_link_libraries( lzma Threads::Threads ) diff --git a/src/common/filesystem/include/fs_files.h b/src/common/filesystem/include/fs_files.h index 06160d258..756a15007 100644 --- a/src/common/filesystem/include/fs_files.h +++ b/src/common/filesystem/include/fs_files.h @@ -81,6 +81,7 @@ enum METHOD_DEFLATE = 8, METHOD_BZIP2 = 12, METHOD_LZMA = 14, + METHOD_XZ = 95, METHOD_PPMD = 98, METHOD_LZSS = 1337, // not used in Zips - this is for Console Doom compression METHOD_ZLIB = 1338, // Zlib stream with header, used by compressed nodes. diff --git a/src/common/filesystem/source/file_7z.cpp b/src/common/filesystem/source/file_7z.cpp index 9201ead99..1be0d6a1c 100644 --- a/src/common/filesystem/source/file_7z.cpp +++ b/src/common/filesystem/source/file_7z.cpp @@ -121,7 +121,7 @@ struct C7zArchive file.Seek(0, FileReader::SeekSet); LookToRead2_CreateVTable(&LookStream, false); LookStream.realStream = &ArchiveStream.s; - LookToRead2_Init(&LookStream); + LookToRead2_INIT(&LookStream); LookStream.bufSize = sizeof(StreamBuffer); LookStream.buf = StreamBuffer; SzArEx_Init(&DB); diff --git a/src/common/filesystem/source/file_zip.cpp b/src/common/filesystem/source/file_zip.cpp index 28c2dedec..9aa34095a 100644 --- a/src/common/filesystem/source/file_zip.cpp +++ b/src/common/filesystem/source/file_zip.cpp @@ -65,6 +65,7 @@ static bool UncompressZipLump(char *Cache, FileReader &Reader, int Method, int L case METHOD_DEFLATE: case METHOD_BZIP2: case METHOD_LZMA: + case METHOD_XZ: { FileReader frz; if (frz.OpenDecompressor(Reader, LumpSize, Method, false, exceptions)) @@ -349,7 +350,8 @@ bool FZipFile::Open(LumpFilterInfo* filter, FileSystemMessageFunc Printf) zip_fh->Method != METHOD_LZMA && zip_fh->Method != METHOD_BZIP2 && zip_fh->Method != METHOD_IMPLODE && - zip_fh->Method != METHOD_SHRINK) + zip_fh->Method != METHOD_SHRINK && + zip_fh->Method != METHOD_XZ) { Printf(FSMessageLevel::Error, "%s: '%s' uses an unsupported compression algorithm (#%d).\n", FileName, name.c_str(), zip_fh->Method); skipped++; diff --git a/src/common/filesystem/source/files_decompress.cpp b/src/common/filesystem/source/files_decompress.cpp index 1b6c373b3..28046ad41 100644 --- a/src/common/filesystem/source/files_decompress.cpp +++ b/src/common/filesystem/source/files_decompress.cpp @@ -36,6 +36,9 @@ // Caution: LzmaDec also pulls in windows.h! #define NOMINMAX #include "LzmaDec.h" +#include "Xz.h" +// CRC table needs to be generated prior to reading XZ compressed files. +#include "7zCrc.h" #include #include #include @@ -505,6 +508,142 @@ public: }; +//========================================================================== +// +// DecompressorXZ +// +// The XZ wrapper +// reads data from a XZ compressed stream +// +//========================================================================== + +// Wraps around a Decompressor to decompress a XZ stream +class DecompressorXZ : public DecompressorBase +{ + enum { BUFF_SIZE = 4096 }; + + bool SawEOF = false; + CXzUnpacker Stream; + size_t Size; + size_t InPos, InSize; + size_t OutProcessed; + uint8_t InBuff[BUFF_SIZE]; + +public: + + bool Open(FileReader *file, size_t uncompressed_size) + { + if (File != nullptr) + { + DecompressionError("File already open"); + return false; + } + + uint8_t header[12]; + int err; + File = file; + + Size = uncompressed_size; + OutProcessed = 0; + + // Read zip XZ properties header + if (File->Read(header, sizeof(header)) < (long)sizeof(header)) + { + DecompressionError("DecompressorXZ: File too short\n"); + return false; + } + + File->Seek(-sizeof(header), FileReader::SeekCur); + + FillBuffer(); + + XzUnpacker_Construct(&Stream, &g_Alloc); + XzUnpacker_Init(&Stream); + + return true; + } + + ~DecompressorXZ () + { + XzUnpacker_Free(&Stream); + } + + ptrdiff_t Read (void *buffer, ptrdiff_t len) override + { + if (File == nullptr) + { + DecompressionError("File not open"); + return 0; + } + + if (g_CrcTable[1] == 0) + { + CrcGenerateTable(); + } + + int err; + Byte *next_out = (Byte *)buffer; + + do + { + ECoderFinishMode finish_mode = CODER_FINISH_ANY; + ECoderStatus status; + size_t out_processed = len; + size_t in_processed = InSize; + + err = XzUnpacker_Code(&Stream, next_out, &out_processed, InBuff + InPos, &in_processed, SawEOF, finish_mode, &status); + InPos += in_processed; + InSize -= in_processed; + next_out += out_processed; + len = (long)(len - out_processed); + if (err != SZ_OK) + { + DecompressionError ("Corrupt XZ stream", err); + return 0; + } + if (in_processed == 0 && out_processed == 0) + { + if (status != CODER_STATUS_FINISHED_WITH_MARK) + { + DecompressionError ("Corrupt XZ stream"); + return 0; + } + } + if (InSize == 0 && !SawEOF) + { + FillBuffer (); + } + } while (err == SZ_OK && len != 0); + + if (err != Z_OK && err != Z_STREAM_END) + { + DecompressionError ("Corrupt XZ stream"); + return 0; + } + + if (len != 0) + { + DecompressionError ("Ran out of data in XZ stream"); + return 0; + } + + return (long)(next_out - (Byte *)buffer); + } + + void FillBuffer () + { + auto numread = File->Read(InBuff, BUFF_SIZE); + + if (numread < BUFF_SIZE) + { + SawEOF = true; + } + InPos = 0; + InSize = numread; + } + +}; + //========================================================================== // // Console Doom LZSS wrapper. @@ -748,6 +887,18 @@ bool FileReader::OpenDecompressor(FileReader &parent, Size length, int method, b } break; } + case METHOD_XZ: + { + auto idec = new DecompressorXZ; + dec = idec; + idec->EnableExceptions(exceptions); + if (!idec->Open(p, length)) + { + delete idec; + return false; + } + break; + } case METHOD_LZSS: { auto idec = new DecompressorLZSS;